VIII. THE RESULTS OF CAESAR’S INVASIONS OF BRITAIN (part 2 of 4)

VIII. THE RESULTS OF CAESAR’S INVASIONS OF BRITAIN (part 2 of 4)

The most dexterous advocate of the theory that Caesar landed on Romney Marsh was Thomas Lewin; and it says a great deal for his persuasiveness that not one of his critics appears to have detected the inconsistencies with which his work abounds. Those which vitiate his argument, in so far as it relates to the tides, I have mentioned already.[3072] The rest all spring from one and the same source. When Lewin wrote his book, he adopted a theory as to the configuration of Romney Marsh which, after obtaining what he considered ‘more accurate information’,[3073] he discarded. This information he embodied in an appendix to his second edition; but at the same time he allowed the statements based upon his former researches to stand. Thus on page 65, note 4, he implies that ‘the heart of the marsh’ was inhabited; but in his preface (pages v-vi) he affirms that ‘the eastern end of the Marsh where Caesar arrived was as much terra firma in his day as in our own’, virtually admitting, as the context and the map which faces page liii alike show, that the rest was inundated at every high tide. On page 92 he says that ‘the sea, as is proved incontestably by the fragments of ships and anchors which have been dug up, flowed to the very base of the hill, and formed there the port of Limne. Stutfall [castle], therefore, was formed for the protection of the shipping.’ On pages lxviii and lxix the incontestable proof is not only contested but flung to the winds: ‘the fragments of ships and anchors’ are silently annihilated; Stutfall Castle, it now appears, ‘was for the protection, not of the port, but of Saxonicum littus’; and the ‘Portus Limanis’ (sic) becomes a ‘narrow gut’, extending from a point near Shorncliffe to West Hythe behind a bank of shingle broken by a narrow entrance nearly opposite Hythe. In an article which Lewin contributed to the fortieth volume of Archaeologia he remarks that if the Portus Lemanis had been at the foot of Lympne Hill, ‘we should expect to find some vestiges, however faint, of the port itself’: but, he adds, ‘I have never heard or read (though I have often inquired) that any remnant of a pier or sunken vessel, or even any anchor or other part of a ship’s tackle, was ever discovered in this part.’[3074] Are we to infer, then, Mr. Lewin, that when you told us on page 92 of your book that ‘fragments of ships and anchors’ had been ‘dug up’ at the foot of the hill, you were romancing? On turning back to page 42, we find that the above-mentioned ‘narrow gut’ first came into existence in the time of the Saxons; but on page lxix we learn with bewilderment that it ‘must have continued such until the abandonment of Britain by the Romans or nearly so’. In the article which he contributed to Archaeologia Lewin changed his mind again. In the map which illustrates this article the ‘narrow gut’ extends no further westward than Hythe. On page 44 of the book we read that Caesar landed on ‘the western side’ of ‘the creek of Limne’ or ‘port of Limne’, the very existence of which the author’s later and ‘more accurate information’ led him to deny. On pages lxxii and lxxiv Caesar’s landing-place is silently transferred from ‘the western side’ of ‘the creek of Limne’ to Hythe. On page 44 two islands mentioned (according to Lewin) by Valerius Maximus are identified with ‘two islands’ depicted on ‘old maps’ of ‘the bay of Limne’: on page lxxiii we are asked to identify them with two ‘islands’ in the ‘narrow gut’ above mentioned.

The reader now understands that, according to the theory of the ancient configuration of Romney Marsh which Lewin adopted in his Appendix and illustrated in the map facing page liii of his book, it would have been impossible for Caesar to land opposite Lympne, because on that theory the marsh between Lympne Hill and the shingle beach was flooded by the sea at high tide. Nevertheless, I shall consider the arguments by which Lewin defended his original view--that Caesar landed opposite Lympne--because distinguished scholars still hold that there was a harbour there in Roman times.

When we come to examine Lewin’s final view--that Caesar landed at Hythe--we shall find some difficulty in doing justice to it; for he carefully avoids committing himself to any clear explanation of his meaning. If we look at his map[3075] of the ‘narrow gut’, which he believed to have extended from West Hythe to a point opposite Shorncliffe, we shall see that, on his theory, Caesar must have done one of two things. Either he must have landed on the shingle west of the mouth of the gut, or he must have landed on the shingle east of that mouth; for Lewin clearly gives us to understand that the Roman ships did not sail into the harbour.[3076] He maintains that on the day on which the first landing occurred a fierce struggle took place between the Romans and the Britons in ‘the field south of Hythe’.[3077] In order to reach this field, the Romans would have had to walk along the shingle either westward or eastward, and then along the northern shore of the ‘narrow gut’. But Caesar distinctly states that as soon as the Romans stood on dry land, that is to say, on Lewin’s own showing, on the shingle beach, they put the Britons to flight.[3078] What becomes, then, of the imaginary combat in ‘the field south of Hythe’? Furthermore, Lewin, while he is constrained to admit that this field is ‘below high-water mark’, assures us that it was ‘certainly dry at low water’.[3079] But he himself strenuously maintains that the Romans began to land three hours after low tide.[3080] Perhaps he was uneasily conscious that he had contradicted himself when he suggested that the Britons ‘would unquestionably have possessed the skill to embank the port and drain the land in the immediate neighbourhood’.[3081]

The theory that Caesar landed at or near Hythe involves another mystery, which Lewin does not attempt to clear up. Where was the camp which, in 54 B.C., Caesar linked by ‘a single defensive work’ (una munitione[3082]) with the ships which he found it necessary to haul up on dry land, and how was the defensive work constructed? Lewin tells us that the ships could not possibly have been drawn up opposite Shorncliffe, because the shore there ‘is rocky and precipitous’.[3083] Therefore, if the landing took place near Hythe, they must have been drawn up on the beach west of Shorncliffe, and on the seaward side of the ‘narrow gut’,--as he suggests, opposite Hythe. If so, of what was the ‘defensive work’ composed? Surely not of shingle? The entrenchments which a child constructs with his toy spade at Margate would have been just as effective. But if not of shingle, what other material was available on a shingle beach? And what was the direction of the ‘defensive work’ which connected the ships with the camp? Presumably the camp was on dry land behind the ‘narrow gut’, and constructed not of shingle but of earth. Now the entrenchment which protected the ships and connected them with this camp could hardly have been carried across the ‘narrow gut’, which was deeply submerged at every high tide! I can only suppose, then, that the connecting work was really constructed, by some occult process, of shingle; traced out along the shingle beach either eastward towards Shorncliffe, or westward to West Hythe Oaks, and then along the northern shore of the ‘narrow gut’ until it joined the camp. Or if, as Lewin suggests, the camp with which the ships were connected was distinct from the camp which Caesar marked out on an ‘advantageous position’ (loco idoneo[3084]) immediately after his second landing; and if, as he affirms,[3085] it ‘must have stood upon the seashore’, and its site ‘must long since have disappeared’; then it can only be concluded that camp and connecting work were both constructed of shingle!

Again, it is incredible that a gale which drove some of Caesar’s eighteen cavalry transports past Hythe or Lympne to a more westerly part of the island should have carried the rest back to the port from which they had started, which Lewin rightly identified with Ambleteuse. If the ships had been approaching Hythe when the storm arose, they would have been obliged, in order to return to Ambleteuse, to steer SE. by E. 9° S.[3086] Now Lewin himself maintains that the gale blew from the north-east.[3087] As a matter of fact, if it had blown from this quarter at Lympne or Hythe, it would have been practically innocuous; for off either of those places vessels would have been sheltered from it by the hills, and the transports which were driven westward would not have been, as they were, swept by the waves: if the gale had blown from the north-east off Walmer, its direction off Hythe would have been about east-north-east.[3088] Still, for the sake of argument, I will accept Lewin’s view. According to it, the ships would have been forced to sail within less than eight points of the wind. But in a heavy gale a ship will make as much as six points and a half of lee-way.[3089] Let us reduce this estimate to four, which is certainly a liberal reduction.[3090] It follows that the hapless transports would have been required, in order to reach their destination, to lie within less than four points of the wind, a feat which, I need hardly say, would have been utterly impossible.[3091] Furthermore, Lewin affirms, in an unguarded moment, that the gale, when a few hours later it wrecked Caesar’s ships as they rode at anchor, was still blowing from the north-east: but he does not explain how a north-easterly gale could have driven ashore, that is to say, driven in a northerly direction, ships anchored off Lympne or Hythe!

Such are a few of the absurdities in which Lewin’s theory plunges him. However, he shall be heard in his own defence.

The argument upon which Lewin lays the most stress has been already refuted.[3092] He maintains that at the time when Caesar, on the day of his first voyage, quitted his anchorage and sailed with the stream to the spot where he was to land, the stream must have been running down the Channel.[3093] Let us assume, in order to allow more than its due weight to his argument, that Caesar must have weighed anchor as early as 3.30 p.m.,[3094] though I shall afterwards prove that the assumption is both unnecessary and false; and let us also suppose that the tide did not actually turn eastward until after 5 p.m. Even then, unless Caesar could have calculated the exact time which it had still to run with more than the certainty of the experts who prepare the Admiralty Tide Tables, he would have had to face the risk that, before he could reach his landing-place, it might turn against him. Even if Heller[3095] is not right in interpreting the words ‘getting wind and tide simultaneously in his favour’ (et ventum et aestum uno tempore nactus secundum) as meaning that Caesar weighed anchor just after the stream had turned, there can be no doubt that this would have been the wisest course to pursue.

1. Having disposed, to his own satisfaction, of the question of the tides, Lewin remarks that ‘Hythe would be much nearer than Deal, and was the natural port for vessels from Boulogne’.[3096]

Yes, nearer by a bare two miles, and a few yards nearer than Walmer! But, as Lewin himself insists,[3097] Caesar, when he sailed from Boulogne, did not intend to land either at Hythe or at Deal; and unless he intended to sail direct to one of those places, Lewin’s argument collapses. Moreover, since the greater part of the voyage was made upon the easterly-going stream, it would have taken longer to sail to Hythe than to the coast between Walmer and Deal. And when Lewin affirms that Hythe ‘was the natural port for vessels from Boulogne’, he apparently forgets that he has already told us that ‘Folkestone ... would be the natural port for Boulogne’,[3098] that Caesar’s fleet never entered Hythe harbour, and that ‘the mooring of the Roman vessels within it would be certain destruction’.[3099]

2. Lewin attempts to show that the wind which carried Caesar from Boulogne to his anchorage on his first voyage, and which would, if it had continued, have been in his favour if he had intended to sail on towards Deal, veered round before he quitted his anchorage.[3100] His argument runs as follows:--‘Caesar says that he started from his anchorage ... having got the wind in his favour, and the Latin word nactus implies that the wind had undergone a change.... When he embarked at Boulogne he despatched the cavalry to Ambleteuse ... with orders to follow him with all haste; but ... they did not leave that haven for Britain until the fourth day after,[3101] and no plausible reason can be given for this except that, for the whole of this interval, the wind was contrary; that is ... had shifted.’ In a later passage[3102] he maintains that if the wind had not shifted during the voyage, the length of the passage, and especially the tardy arrival of the transports, would be inexplicable. Moreover, he says, the word nactus ‘implies a change either in the wind or the tide: the tide had not changed, and therefore the change alluded to must have been in the wind’.[3103] While, however, he argues that the wind must have shifted, he endeavours to secure his retreat by affirming that, if it had not shifted, Caesar could nevertheless have sailed to Hythe as easily as to Deal. ‘Supposing,’ he says,[3104] ‘the wind to have blown from the south, it would have been favourable to a movement, from a point opposite Dover, either to the east or west.’

I freely admit--indeed I have myself maintained--that the wind had shifted during the voyage. It had shifted to a point unfavourable, or comparatively unfavourable, to ships sailing from North-Eastern Gaul to Britain. I also, like Lewin, maintain that it shifted again before Caesar quitted his anchorage. The fact is obvious. But what then? How can Lewin prove that before Caesar quitted his anchorage the wind did not shift to a quarter which would have been favourable to a run from a point off the cliffs of Dover towards Deal? Moreover, it is not true that if the wind had ‘blown from the south, it would have been favourable to a movement, from a point opposite Dover, either to the east or west’. If the wind had blown exactly from the south, it would certainly not have been called favourable to a movement from a point opposite Dover to Hythe, that is to say, within less than seven points of the wind; and if the wind had blown from any point west of south, the word ‘favourable’ would have been still less appropriate.[3105]

3. Lewin maintains that the wisest course which Caesar could have pursued when he sailed on from his anchorage in 55 B.C. would have been to steer westward. ‘Let us first consider,’ he says,[3106] ‘a priori, what a prudent commander might be expected to do under similar circumstances.... To the right he would see the precipitous chalk cliffs stretching away ... till they terminated at the South Foreland.... The lowlands about Walmer and Deal would not be visible; and it is at least doubtful whether Volusenus had included them in his survey ... tracing the line of cliffs westward, he could not fail to see that they terminated at Sandgate, and that a broad level plain there succeeded.’ At Hythe, he adds, there was a landing-place ‘distinctly visible from his moorings’.[3107]

This argument rests upon the absolutely groundless assumption that Volusenus had not reconnoitred the coast on the north of the South Foreland;[3108] in other words, that Volusenus, whom Caesar specially selected as a thoroughly competent man, had grossly neglected his duty and disobeyed his instructions. Besides, even if the landing-place at Hythe had been ‘distinctly visible’ from Caesar’s moorings,[3109] nothing would have been gained; for Caesar acted, not upon what he could see, but, as he tells us himself,[3110] upon what he had learned from the report of Volusenus. Whether he could or could not see ‘from his moorings’ the place where he was to land, he knew the direction in which he intended to sail. Finally, when Lewin considers ‘a priori’ that ‘a prudent commander might be expected under similar circumstances’ to land at Hythe or Lympne, he only advertises his own ignorance of a commander’s business. As I shall show presently,[3111] no commander who was not hopelessly incompetent would have dreamed, in the circumstances of ancient warfare, of attempting to land either at Hythe or at Lympne.

4. Lewin contends in his text that at Lympne and in his appendix that at Hythe there was a landing-place which exactly corresponded with Caesar’s description; and he denies that any such landing-place exists at Deal. He affirms[3112] that the shore on the western side of ‘the creek of Lympne’ was ‘apertum or open, for the heights to the north were at least a mile distant. The sea-beach was also molle or soft ... in a sailor’s sense, i.e. it consisted of shingle, than which nothing can be more favourable to the security of vessels.... Sand, on the contrary, is, in naval phraseology, of the hardest kind, as it has no “give”, and a ship beating against it would soon be dashed to pieces.’ He says that, according to Lucan,[3113] Plutarch,[3114] and Dion Cassius,[3115] there was ‘marshy ground’ at the place where Caesar landed, and he assures us that marshes were formed by the streams which entered the port of Hythe.[3116] He remarks that, according to Valerius Maximus,[3117] there were two small islands near the landing-place, which were the scene of an exploit performed by one of Caesar’s centurions: ‘on looking,’ he says, ‘at the old maps of this part of the coast, I find ... that the bay of Limne contained ... two islands.’ Like other people who had learned to measure the trustworthiness of Valerius Maximus, Lewin had himself been sceptical about the anecdote of the centurion; but how could he resist the testimony of the ‘old maps’? He admits that he was converted:--‘the circumstance, so apocryphal before, becomes thus no inconsiderable argument for placing the descent in this locality.’[3118] Afterwards he changes his mind, and transfers the islands some distance to the east of Lympne. They actually existed, he tells us, a few years before the publication of his book, ‘either near to or in the ancient port of Hythe’; they also are depicted in various old maps; and they were ‘carted away’ by the late James Elliott, the engineer of Romney Marsh.[3119] Furthermore, he assures us[3120] that the incident described by Valerius Maximus is also noticed by Plutarch,[3121] who ‘lays the scene in sight of Caesar himself, and therefore close to the camp; and in a marsh or swamp, which, with the light afforded by the account of Valerius, must be taken to mean a lagoon subject to the alternations of the high and low tides’.

Now as to the exact meaning of the word apertum in the passage to which Lewin refers, the commentators are not agreed. While he insists that the beach on which he believes Caesar to have landed was ‘open’, because ‘the heights to the north were at least a mile distant’, Dr. Guest[3122] denies that they were ‘open’, because ‘there is a range of heights at no great distance’. According to Long,[3123] ‘“open” means that from the beach he could see into the country.’ Now any one who has read the Commentaries attentively will see that all these explanations are wrong. For in his narrative of the second expedition[3124] Caesar tells us that he ‘felt little anxiety for the ships, as he was leaving them at anchor on a nice open shore’[3125] (eo minus veritus navibus, quod in litore molli atque aperto deligatas ad ancoras relinquebat). If the explanations which I have quoted were correct, the word apertum in this passage would be irrelevant. Whether the ‘heights’ were ‘at no great distance’, or ‘from the beach Caesar could see into the country’, the security of the ships would not have been affected in the slightest degree. The word apertum does not describe the country near the shore: it describes the shore itself; and, as C. Schneider[3126] says, apertum litus means a shore free from such obstacles or dangers as rocks, boulders, and the like.[3127] But, even assuming that Lewin was right in his interpretation of the word apertum, this much is certain:--the existence of the heights to which he refers is alone sufficient to prove that Caesar did not land either at Lympne or at Hythe or at any point between those two places. This is a matter on which I confidently appeal to any military expert who has studied the records of ancient warfare. Caesar might no doubt have landed at Hythe without any extraordinary difficulty; but if he had been so foolish as to land there, he would have found that his difficulties were only beginning. Never, even when fighting against an uncivilized enemy, did he attempt to force his way up a hill if it was possible to avoid doing so.[3128] If he had landed either at Lympne or at Hythe, he could not have turned the line of heights which extends behind those two places: he could not have penetrated into the interior of the country unless he had passed them; and he could not have passed them except at a cost of life which the least experienced of his officers would have been too prudent to incur. Furthermore, however ‘open’ the country may have been on the western side of ‘the [imaginary] creek of Lympne’, he could not have landed there if, as Lewin admits,[3129] the country was under water at high tide.

The argument which Lewin bases upon the word mollis has no value; for he does not fully understand the meaning of the word. Mollis, in the passage which we are considering, simply means that the shore was one where the anchorage was good, and where the ships, if they were driven aground, would suffer comparatively little:[3130] it probably also implies that the shore was gently sloping.[3131] Moreover, even if the word mollis implied that the shore on which Caesar landed was composed of shingle, Lewin would not be justified in concluding that Caesar landed at Hythe or Lympne unless he could prove that no other shingle beach in Kent satisfied the requirements of Caesar’s narrative.

The statements of Lucan, Valerius Maximus, Plutarch, and Dion Cassius, upon which Lewin lays so much stress, are not really evidence at all; and if he had been a classical scholar, he would never have quoted them. Nor, indeed, was his scholarship sufficient to enable him to understand what they meant. The passage quoted from Lucan occurs in a rhetorical speech which he puts into the mouth of Pompey:--

Rheni gelidis quod fugit ab undis Oceanumque vocans incerti stagna profundi Territa quaesitis ostendit terga Britannis.

Stagna of course simply denotes the English Channel, about which Lucan’s ideas were vague; and the point of the line is the suggestion that Caesar, in order to magnify the difficulties of his expedition, mendaciously described ‘the pools of a shifting sea’ as an ocean.[3132] As to Valerius Maximus, the idea of going to his collection of anecdotes for the details of Caesar’s military operations is really funny;[3133] but if he is to be counted an authority, he does not support Lewin’s theory, but overthrows it.[3134] Lewin[3135] insists that ‘the vada described by Valerius Maximus as caused by the flux and reflux of the tide are evidently the vada referred to by Caesar (iv, 26)’. Now ‘the vada referred to by Caesar’ were the shallow places of the shore on which he landed; and ‘the vada described by Valerius Maximus’ were, according to Lewin’s final view, in Hythe harbour. Yet Lewin emphatically denies that Caesar landed in Hythe harbour! Furthermore, Lewin assures us that the ‘islands’ of which Valerius Maximus speaks were composed of earth, and have been ‘carted away’. But Valerius Maximus says nothing about islands: the only island which he mentions is Britain. As to the marshes, Heller[3136] points out that Lewin is wrong in concluding from the narratives of Plutarch and Dion that the coast on which Caesar landed was marshy. Dion’s word τενάγη is simply his translation of Caesar’s vada, and means not ‘marshes’ but ‘shallows’.[3137] Plutarch, as the words εἰς τόπον ἑλώδη καὶ μεστὸν ὑδάτων (‘into a marshy place full of water’), which are explained by the following words ῥεύματα τελματώδη (‘marshy [or muddy] streams’), clearly show, was not speaking of a combat on the seashore, but of one which took place inland; and when Lewin identifies Plutarch’s ῥεύματα τελματώδη with Dion’s τενάγη, he simply exposes his own ignorance of Greek. Moreover, Plutarch is so deficient in accuracy and precision that his description of an incident of this kind is useless for the purpose of topographical identification.

There is, therefore, no evidence that there was any marshy ground at or near Caesar’s landing-place; and if there was, the fact does not prove that he landed at Hythe.[3138]

5. Lewin[3139] next proceeds to examine Caesar’s narrative of the events which followed his first disembarkation,--the hauling up of the galleys on dry land, the wreck caused by the storm which occurred on the night of the full moon, and the subsequent attack by the British charioteers. The whole account, he tells us, corresponds with the topography of Lympne and Hythe, but of no other place. The shingle field between Dymchurch Wall and Shorncliffe, being ‘sound and dry, without any mixture of ooze’, deserves Caesar’s epithet, aridum:[3140] the ‘steep place’ (declivis ac praeceps locus[3141]), down which the British chariots charged, is ‘in this part of the marsh on the north’; the spring tide, driven by ‘a strong south-east wind’, would have poured over the shingle, and waterlogged Caesar’s galleys. ‘It is a singular confirmation,’ says Lewin, ‘of our hypothesis of the debarcation at Romney Marsh that the range of high water is greater here than at any other point of the southern coast. At Dungeness, for example, the mean range is twenty-one feet three quarters, while at Deal it is only sixteen feet.’ Finally, he anticipates the objection that if Caesar had landed near Hythe, he would have moored his ships within Hythe harbour, and thereby avoided the destruction which overtook so many of them. He explains that ‘the narrow and winding gut which constituted the port’ was ‘little capable of receiving a fleet’, and that, ‘as it ... could only be entered and quitted at high water, and as its banks were lined by a hostile population, the mooring of the Roman vessels within it would be certain destruction,’ but ‘it would be highly useful for keeping up his communication with the Continent’.

That Caesar’s ships could have been hauled up on the shingle between Dymchurch Wall and Shorncliffe; that the said shingle might have been described as ‘aridum’; that a ‘steep place’ existed ‘in this part of the marsh on the north’;[3142] and that the tide, driven by ‘a strong south-easterly wind’, might have waterlogged Caesar’s ships if they had been drawn up on the beach at Hythe,--all these things may be admitted: but they avail nothing to establish Lewin’s theory unless he can prove that on the east coast of Kent there was no place which answered equally well to Caesar’s narrative; and it is amusing to find that in another passage[3143] he rightly insists that the wind which he here calls ‘south-easterly’ was ‘from the north-east’. The comparison which he makes between ‘the range of high water’ at Romney Marsh and at Deal is irrelevant; for I doubt whether he would have seriously maintained that a spring tide between Walmer and Deal, heightened by a storm, would have been insufficient to cause the damage which Caesar described.

The objection which Lewin anticipates, and waves aside, is not thereby disposed of. If the reader will examine Lewin’s map, he will see that Hythe harbour would have been amply large enough to accommodate Caesar’s fleet;[3144] and if it could only be entered at high water, the Portus Itius, according to Lewin himself,[3145] was in the same case. That Caesar would have landed close to a harbour which he afterwards found ‘highly useful’, without mentioning its existence, is, to say the least, unlikely. Moreover, Lewin omits to tell us in what respect Caesar could have found a harbour ‘highly useful’ even for ‘keeping up his communication with the Continent’, if the dispatch vessels which kept up the communication never entered the harbour; nor does he explain what would have been the use of their entering the harbour when, on his own showing, they could not have remained there without ‘certain destruction’, but after they had entered it, must have forthwith gone out again, and have been hauled up on the beach outside along with the others.

6. It remained for Lewin to point out the spot on the banks of a river about twelve miles from the Roman camp, where the Britons disputed Caesar’s advance on the day after his second landing in Britain. He maintained that Wye on the Stour, between Ashford and Canterbury, answered in every detail to Caesar’s account, and denied that any such place could be found on the theory that Caesar landed near Deal.[3146]

Airy[3147] objects that the Stour at Wye is ‘little wider than a wide ditch’, and that, as it flows ‘between sound meadows, where there is not, and never has been, any marsh or broad stream’, it never could have been wider. Lewin[3148] replies that he has himself seen the Stour at Wye when it was so full of water that ‘the mill had constantly at work four pairs of stones from 5.30 a.m. till 8 p.m., except for a short time at noon’; and I may add that it is more than ten yards wide.[3149] Caesar does not describe the stream at all: he merely calls it a flumen; and he calls the Oze and the Ozerain, the two streams which encompass Alesia[3150] (the modern Mont Auxois), and which are narrower than the Stour at Wye, by the same name. Moreover, as Lewin[3151] justly remarks, Caesar, in describing the combat which took place on the banks of the stream in question, and the subsequent attack upon the British stronghold,[3152] ‘does not make the river the important part of the defence.’ The Stour at Wye and the features of the surrounding country correspond sufficiently well with Caesar’s account; but this does not constitute a positive argument in favour of Lewin’s theory unless it can be proved that, on the theory that Caesar landed in East Kent, it is impossible to discover a stream which satisfies the requirements of his narrative.

D’Anville, the famous French geographer, added little or nothing to the case for Hythe. He maintained that the ‘higher ground’ to which the Britons withdrew on descrying the Roman flotilla in 54 B.C. must have been the line of heights which extends just behind Hythe.[3153] But the passage in which Caesar describes the retirement of the Britons, so far from implying that the ‘higher ground’ was close to the landing-place, implies the very reverse; for Caesar, immediately after telling us that the Britons had retired to higher ground, expressly states that the position on which he found them posted was about twelve miles from his camp near the sea.[3154]

Lewin has now had a fair hearing; and those who are interested in the question will decide whether he has made out his case. Over and above what has been said in refutation of his arguments there remain other facts which make it absolutely certain that Caesar did not land either at Hythe or at Lympne.

1. Caesar, as we have seen, tells us that when he sailed from the Portus Itius for Britain in 54 B.C. the wind was from the south-west.[3155] Now a south-west wind would not have been favourable to a voyage from Boulogne to Hythe, much less to Lympne. The wind would have been abeam even if it had been possible to keep the ships heading in the direction of Hythe, that is to say, even if they had made no lee-way: the ships, being shallow and flat-bottomed,[3156] would have made considerable lee-way; and two or three hours after the voyage began the current turned east-north-east, and continued to run in that direction until daybreak[3157]. It is curious that so adroit a controversialist as Lewin should have inadvertently quoted an opinion which damaged his own case. He tells us[3158] that a Mr. John Dougall, in an unpublished tract, remarked that Caesar would have called the south-west wind favourable if he had sailed from Boulogne for the South Foreland. But, on Lewin’s theory, Caesar sailed, not for the South Foreland but for Hythe; and Lewin naïvely tells us that, in Dougall’s opinion, Caesar would never have called the wind favourable if it had been on the beam. Certainly he would not have done so when the tide was setting in the wrong direction.

2. Another objection to the theory that Caesar landed at Hythe or Lympne is that it involves the assumption that when he started on his first voyage, he steered for the unsuitable port of Folkestone, or else that, although he was in possession of Volusenus’s report, he did not know where he intended to land, and steered at haphazard. According to Lewin’s original idea,[3159] he first anchored off the cliffs of Dover. This is the view which naturally commends itself to every unbiassed reader of the Commentaries; but, as Lewin afterwards saw, it is irreconcilable with the theory that Caesar sailed from his anchorage either to Hythe or to Lympne; for Caesar tells us that he sailed 7 Roman miles, and Hythe is almost 11, Lympne about 14, Roman miles from the nearest point of the Dover cliffs.[3160] Lewin thus found himself obliged to relinquish his original view, and to maintain that Caesar anchored off Folkestone,[3161] or, as he suggests in another passage, and implies in his map,[3162] off East Wear Bay. But second thoughts are not always best. Lewin insists that Folkestone ‘would be the natural port for Boulogne’.[3163] Why? Surely not because the steamboats of the South Eastern and Chatham Railway Company discharge their passengers there! ‘The natural port for Boulogne’ would be the port to which, having regard to the prevailing winds and the currents, it would be easiest to sail, and which would best accommodate Caesar’s fleet. That port was Dover. Caesar’s narrative shows that he required a capacious harbour;[3164] and the port of Folkestone was too small.

3. Another consideration--one of several which, strange to say, have hitherto been overlooked--is alone fatal to the theory that Caesar landed either at Hythe or at Lympne. He tells us, in regard to his first landing, that his troops were not able to pursue the enemy far ‘because the cavalry had not been able to keep their course and make the island’.[3165] These words unmistakably imply that if he had had his cavalry with him, he would have been able to make good use of them; and in fact he says expressly that, in repelling the attack which the Britons made upon his camp just before his departure from Britain in 55 B.C., he actually did employ Commius’s small troop of cavalry.[3166] The camp, so Lewin assures us when he is advocating the cause of Lympne, was ‘in the marsh’.[3167] How, then, would Caesar have been able to utilize his cavalry? Would he have sent them up the steep slopes on which stand the ruins of Stutfall Castle? Even on the incredible hypothesis that Romney Marsh had been embanked by the Britons[3168] the cavalry would have been useless; for how could they have acted in a country intersected by sluices?[3169] It is possible, indeed, that if Caesar had landed at Hythe, his cavalry might have acted in ‘the field to the south and east of Hythe’, in which Lewin finally assumes that the combat took place on the day of the landing,--if the Britons had patiently waited to receive their charge, and if the field was not under water.[3170] But before the cavalry could have come into action the Britons would have been on the hills behind.

I do not deny that the cavalry could have managed to trot up Lympne Hill or the hills behind Hythe, if Caesar had been so foolish as to give the order.[3171] But the point is that the absence of the cavalry prevented Caesar from pursuing the fugitives far; and that the hills would have effectually concealed from him the nature of the country that lay beyond them, its woods, defiles, and other obstacles. Does he not tell us himself that when he had gained a victory on the day after his second landing in Britain, ‘he forbade [his troops] to pursue the fugitives far, partly because he had no knowledge of the ground’ (eos fugientes longius Caesar prosequi vetuit, et quod loci naturam ignorabat,[3172] &c.)? Worse ground for the manœuvring of cavalry than the wooded heights which extended behind Hythe and West Hythe it would have been hard for Volusenus to find.

4. Heller[3173] has acutely seen that Caesar’s account of the movements of one division of the ships which carried his cavalry is irreconcilable with the theory that he landed at any point on the south coast of Britain. Caesar[3174] tells us that the storm which arose when these vessels were sighted from his camp drove some of them ‘to the lower and more westerly part of the island’ (aliae ad inferiorem partem insulae, quae est propius solis occasum, ... deicerentur). ‘This,’ says Heller, ‘can only mean a different side of the island from that on which Caesar was: if he had meant to designate a point on the same side, he would have said, paulo infra ac propius solis occasum.[3175] If one compares the expression which he actually uses in the thirteenth chapter of his Fifth Book--unum latus est contra Galliam. Huius lateris alter angulus, qui est ad Cantium, quo fere omnes ex Gallia naves appelluntur, ad orientem solem, inferior ad meridiem spectat[3176]--it becomes clear that in the passage in which the fate of the eighteen ships is described inferior pars insulae means the southern side of the island. Consequently Caesar implies that his camp was on the eastern angle.’

5. One of the episodes which Caesar describes in his narrative of the first expedition is, in spite of the ingenuity with which Lewin has tried to make it fit in with his theory, irreconcilable with the view that he landed either at Hythe or Lympne. One day, when the 7th legion had gone out to cut corn, Caesar learned from the troops on guard in front of the camp that an extraordinary quantity of dust was visible in the direction in which the legion had gone. He marched to the rescue, and, after he had advanced some little distance (paulo longius), he found that the legion was hard pressed by the enemy. The place where the legionaries had been reaping was the only one in which the corn had not yet been cut; and the enemy, anticipating that they would come there, posted themselves in ambush in a wood close by.[3177] Lewin[3178] argues that the camp must have been in Romney Marsh, ‘probably on the seaside’; but he stultifies his own argument by admitting, in his appendix, that the marsh was inundated at every high tide.[3179] In his text he admits that neither the cornfield nor the wood could have been in the marsh; and he could not at first conceive how, if they had been anywhere else, the dust raised by the combatants could have been seen from the ground in front of the camp. But, he continues, ‘when I visited Hythe by land, and walked from it to the old port of Limne, and mounted the hill, I discovered the explanation. On reaching the top I stepped at once into a cornfield ... and on my right was Park Wood.... What I had taken from the sea for a hill ... had no descent on the north side ... and corn growing so near to the edge that even the reapers, if labouring in that part of the field, might have been seen from the camp. The whole narrative was now realized to the mind’s eye.... The legion had marched up to the standing fields of corn on the high ground, and the Britons, starting from their lurking-place at the side, had intercepted their retreat, and surrounded them at just such an interval from the edge that the combatants were out of sight and hearing, but the dust flying in the air had attracted the attention of the guard ... at a mile’s distance below.’

Now observe what becomes of Lewin’s explanation. First, ‘the old port of Limne,’ to which he walked, became, before the publication of his second edition, a figment of the imagination! Having obtained, as he tells us, ‘more accurate information,’ he strenuously denied its existence, and accordingly transferred Caesar’s landing-place to Hythe.[3180] Yet in this same second edition the ‘explanation’ is offered as confidently as ever! Secondly, he asks us to believe that the only cornfield which the Romans had left unreaped was the one nearest them! Thus the ‘explanation’ which Lewin discovered with such pride collapses; and his theory, which cannot stand without it, falls like a house of cards. What explanation he would have discovered on the theory that Caesar landed at Hythe, he wisely omits to say.

IX. THE THEORY THAT CAESAR LANDED AT HURST

The latest supporters of the theory that Caesar landed on Romney Marsh are Mr. H. E. Malden and, tentatively, Mr. Warde Fowler[3181] and Professor Pelham.[3182] Mr. Malden relies upon the argument, drawn from the study of the tides, which I have already refuted.[3183] But on certain points of detail he differs from Lewin. He maintains that Lewin’s description of Romney Marsh, as it existed in the time of Caesar, is incorrect, and that ‘the coast-line then ran nearly east and west from Sandgate towards Appledore’. In other words, he maintains that the hills which bound Romney Marsh on the north were accessible ‘by ships sailing over what is now embanked land’. Accordingly he believes that Caesar landed neither at Hythe nor at Lympne, but ‘on a broad flat muddy shore’ near Hurst, that is to say, about three miles west of Lympne. This, he assures us, ‘was a landing-place second to none’; so presumably the mud was a recommendation. Mr. Malden notes that there was ‘good camping-ground, wood and water on the slope above’; but a few lines lower down he observes that ‘there would be good camping-ground on the slope where Stutfall castle now stands’; that is to say, three miles off on the east! He assures us, further, that ‘the place agrees singularly with the account of the battle on the shore’; that ‘the passage into the inner country would be easy by the break in the hill above West Hythe’; and that ‘the hill is anywhere accessible’. Finally, he remarks that ‘in A.D. 893 Hastings the pirate came here with his fleet’.[3184]

Now Lewin, as I have already shown,[3185] gives three successive and different descriptions of Romney Marsh as it existed in the time of Caesar; and Mr. Malden does not say to which of the three he refers. It is evident, however, that he is thinking of the first; and I am afraid that he did not take the trouble to read the second edition of Lewin’s book. In a previous article[3186] I have examined all Lewin’s theories. I will here assume that Mr. Malden is right; that the coast-line ‘ran nearly east and west from Sandgate towards Appledore’; and that the northern fringe of Romney Marsh was accessible ‘by ships sailing over what is now embanked land’. On this hypothesis and on any other it is absolutely certain that Caesar could not have landed at the point which Mr. Malden indicates. He does not know what ‘good camping-ground’, in the circumstances of ancient warfare, was. If he knew the Commentaries as intimately as a man who professes to explain them should do, he would see that neither at Hurst, nor ‘on the slope where Stutfall castle now stands’, nor at any intermediate point did good camping-ground exist; for everywhere along this line a camp would have been dominated by higher ground above. Does Mr. Malden not remember the words in which Caesar describes the camping-ground which Reginus and Rebilus were compelled by adverse circumstances to occupy at Alesia? They ‘were obliged’, he says, ‘to make the camp on a gentle slope, which gave an assailant a slight advantage’ (necessario paene iniquo loco et leniter declivi castra fecerunt[3187]). The slope between Hurst and Stutfall Castle would have given an assailant not a slight but a considerable advantage. Where, then, could Caesar have found the ‘advantageous position’ (loco idoneo[3188]) which he selected for his camp in 54 B.C.? ‘Hastings the pirate,’ as I have already proved,[3189] did not ‘come here with his fleet’; and what good Caesar’s cavalry, the absence of which he deplored, could have done by pounding up the hill, which was overgrown by woods,[3190] Mr. Malden does not explain. The hill, as he says, is ‘everywhere accessible’,--to a pedestrian: but, as I have already shown,[3191] ‘the passage into the inner country,’ in the face of armed resistance, could only have been effected with heavy loss; and it would be interesting to hear Mr. Malden account for the fact that the Britons retired in 54 B.C. to ‘higher ground’ twelve miles from the landing-place, when, just above the landing-place which he calls ‘second to none’, there was ground 312 feet above the level of the sea, and 300 feet above the present elevated level of the marsh.[3192] I should also like to hear him explain why Volusenus, a trained soldier who thoroughly understood his business, should have advised Caesar to land at Hurst rather than between Walmer and Deal; what motive could have induced Caesar to attempt to sail from Wissant (which he identifies with the Portus Itius) to Hurst with a south-west wind, that is to say, to attempt to sail west-north-west, on an easterly-going stream, within six points of the wind; how his cavalry transports contrived to sail back from a point near Hurst to Sangatte, or even to Ambleteuse, against a gale which drove their sister ships towards the west; or by what miracle this gale dashed his ships ashore in a northerly direction!

X. THE THEORY THAT CAESAR LANDED BETWEEN HURST AND KENNARDINGTON

The late Francis Hobson Appach published in 1868 a book called Caius Julius Caesar’s British Expeditions, the principal object of which was to prove that Caesar’s troops disembarked in 55 B.C. between ‘the foot of the spur from Aldington Knoll’ (which is about four miles west of Lympne) on the right, and a point ‘about half way between Bonnington and Bilsington’ on the left;[3193] and that in the following year the left of the line extended as far westward as Kennardington. This theory differs but slightly in locality from Mr. Malden’s, but is defended by different arguments; and, moreover, from Bonnington westward the slope of the hills which bound Romney Marsh on the north becomes much gentler. Appach maintains that at least the northern portion, if not the whole, of the marsh was, in Caesar’s time, submerged, and that its northern fringe was accessible to ships sailing from the east. His arguments have already been refuted in an earlier section of this book;[3194] but for the purpose of the present inquiry I will assume that he made out his case, and that, in Caesar’s time, ‘the Bay of Apuldore,’ to which he constantly alludes, was, as he assures us, deep enough to float ‘the heaviest of Caesar’s ships at the lowest spring tides’.[3195]

Appach, of course, like Lewin and Airy, bases his theory upon the hypothesis that the tidal stream which Caesar had in his favour when he sailed from his anchorage to his landing-place must have set towards the west. Although I have proved that this hypothesis is untenable, I will assume, in order to do justice to Appach’s arguments, that Caesar may have sailed with the westward stream.

1. Appach begins by arguing that Volusenus, when he was reconnoitring the British coast, found on the west of Aldington Knoll ‘the very ground of which he was in search’, namely, ‘a low level coast’, which ‘extended as far as Apuldore ... forming the head of a bay protected from the waves in the Channel by the cliff at Fairlight ... and the cliff at Folkestone’.[3196]

It may be freely admitted that this coast, if it was accessible by sea, would have been a convenient landing-place, if Caesar had only desired to land, and then to sail away again. But his object was to invade Britain; and the landing-place which Appach believes Volusenus to have selected would have led Caesar into the inhospitable forest of Anderida,[3197] where no corn was to be procured. This objection, indeed, is equally fatal to Mr. Malden’s theory; and neither he nor Appach attempts to explain how there could have been ‘buildings far and wide’ (omnibus longe lateque aedificiis incensis) on those lonely slopes, on which, even now, woods are abundant, and buildings very few and very far between.[3198]

2. Appach tells us that Caesar’s original intention was to disembark between Sandgate and Hythe; and it must be borne in mind that, according to him, Hythe harbour, as described by Lewin, did not then exist. ‘This landing place,’ he says, ‘though open to the south and east, was sheltered on the north and west. It was also sheltered from the force of the sea and stream in the Channel, and must therefore have been the principal British port on this part of the coast in the time of Caesar.’[3199]

Perhaps (if it had existed),--‘on this part of the coast’. But then Caesar[3200] says that ‘almost all ships from Gaul’ used to make for what he calls the eastern ‘corner’ of Britain, ‘by Kent.’ He may have been thinking of Dover, or of Richborough, perhaps of both; but I doubt whether it is possible to force into his words the meaning that the harbour where ‘almost all ships from Gaul’ discharged their freight was between Sandgate and Hythe.

3. But let us assume that Caesar intended, for some inscrutable reason, to land between Sandgate and Hythe. What, then, were the ‘precipitous heights’ off which he anchored,--the heights from which it was possible to throw a missile right on to the shore? Can Appach find them? Yes! He tells us that they are there, and that if we look for them anywhere else, we shall look in vain.[3201] He bids us note carefully the passage in which Caesar describes them:--‘there, standing in full view on all the heights, he saw an armed force of the enemy. The formation of the ground was peculiar, the sea being so closely walled in by abrupt heights that it was possible to throw a missile from the ground above on to the shore’ (in omnibus collibus expositas hostium copias armatas conspexit. Cuius loci haec erat natura, atque ita montibus angustis mare continebatur, uti ex locis superioribus in litus telum adigi posset[3202]). You are to observe, says Appach, that ‘Caesar uses the word “all” ... thus pointedly implying, if not ... asserting that there were more than two hills’. We are then told that, ‘it being about half an hour after high water, the sea reached quite up to the mouths of the three valleys between Hythe and Sandgate, so that the bay of Hythe appeared to Caesar to be shut in by the hills exactly as he describes it.’

Now these words alone are evidence that Appach did not know enough Latin to qualify him for the task which he had undertaken. That he should have misconstrued the words montibus angustis mare continebatur is not so surprising; but to a man who is familiar with Caesar’s usus loquendi, the conclusion which Appach draws from the words omnibus collibus--that Caesar must have seen at least three separate hills--will appear somewhat forced.[3203] If he had engaged a boat at Dover, and looked at the cliffs stretching away eastward towards the South Foreland, he would have seen that they answered exactly to Caesar’s description. The upper edge of the chalk is not a straight line: it rises and falls in a succession of deep curves, corresponding with the rolling downs above, which are, so to speak, divided into a series of heights by well-defined depressions. Here are the omnes colles, as plain as can be, not three of them but six,[3204] between the Priory Valley and the South Foreland. And Appach wants us to believe that the hills between Sandgate and Hythe are ‘precipitous heights’ (angusti montes), from the summits of which it would have been possible to throw a missile right on to the shore. Precipitous heights! Go, reader, and look at them.[3205]

4. Appach candidly admits that Caesar, on his imaginary voyage from Hythe to Bonnington, lost off Lympne the benefit of the westward current, and ‘met the eddy in the bay of Apuldore running from west to east along its northern shore’:[3206] but by this candour he gives away his whole case; for who will believe that Caesar would have told us that he ‘got the stream in his favour’ if the stream had turned against him before he had completed half his journey?

5. Again, Appach observes that Caesar’s account of his first landing implies that the Britons who opposed him used their chariots as well as their cavalry; and he scoffs at the notion that the chariot-horses or even the cavalry could have acted on shingle. Therefore, he concludes, ‘there could have been no shingle at the place where Caesar landed. Bonnington fulfils this condition.’[3207]

Quite so; and the coast north of Sandown Castle also practically fulfils, or once fulfilled, this condition.[3208] If Appach’s argument shows that Caesar did not land on shingle, it does not show that he landed at Bonnington. Moreover, granting that Caesar’s narrative implies that chariots were used by the Britons in opposing the landing, Appach fails to realize the situation. The horses were not required to gallop into the sea; nor could they have galloped through the waves at Bonnington any more than at Lympne, or Hythe, or Deal: but at any of those places they could have seriously obstructed men who could hardly keep their footing, and who were encumbered by their armour in the way which Caesar[3209] so graphically describes. All that the horses had to do was to convey their masters ‘a little way into the water’ (paulum in aquam [progressi]), so as to enable them to throw their missiles with effect; and this they could easily have done if they had drawn their light cars no faster than the ancient steeds which drag, or used to drag, bathing-machines over the shingle at Eastbourne.

6. Finally, Appach cites the well-known passage in which Caesar tells us that at daybreak on his second voyage he ‘descried Britain lying behind on the port quarter’ (sub sinistra Britanniam relictam conspexit). He argues that Caesar’s words would be meaningless ‘if the configuration of the coast had been the same in his day as it is at the present time, for Britain could not ... have been in any other position. The expression, however, is peculiarly appropriate if the sea then filled the Bay of Apuldore; for Caesar, sailing, as he thought, from Boulogne to Kennardington, of course expected to see Britain on his right.’[3210]

Yes, we of course see that, as Caesar was drifting up channel, ‘Britain could not have been in any other position’ than on his left. But Caesar was not writing for us: he was writing for his countrymen, who did not know anything about the configuration of Britain until they had read as far as the thirteenth chapter of his Fifth Book.[3211] Moreover, if Appach had known his Polybius, he would have remembered the passage in which readers who had passed their lives in the Mediterranean basin were informed of the self-evident fact that Hannibal, marching eastward through Southern Gaul, had the Mediterranean on his right.[3212] Furthermore, if Caesar had steered for Kennardington, he would have had to drift more than fifteen nautical miles in order to reach a point opposite the South Foreland: it would have been impossible to drift nearly so far between ‘about midnight’ and daybreak;[3213] and unless he had drifted further, he could not have seen Britain ‘lying behind’ on his left.[3214]

I need hardly add that Appach does not explain how those cavalry transports got back, in spite of the north-easterly or east-north-easterly gale, from off Bonnington to Ambleteuse.

Enough has been said to show that, even if ‘the Bay of Apuldore’, as Appach describes it, existed in Caesar’s time, he never sailed its waters. We now know where to look for his landing-place: he must have first set foot in Britain on the eastern coast of Kent.

XI. THE THEORY THAT CAESAR LANDED BETWEEN WALMER AND DEAL

The oldest English writers whose works have come down to us believed that Caesar had landed on the north of the South Foreland. This was certainly the view of Nennius, or of an author whose work Nennius edited;[3215] and Dr. Guest attached great weight to his testimony.[3216] I think that, in doing so, he showed less than his usual judgement; and perhaps he was not aware that Maistre Wace, who lived in the twelfth century, had anticipated the modern theory that Caesar landed on Romney Marsh.[3217] For some centuries, however, the prevailing view, first definitely stated, if I am not mistaken, by Leland,[3218] who lived in the reign of Henry the Eighth, was that the disembarkation took place near Deal.[3219]

1. We know that Caesar, before he set sail from Gaul, intended to land, if possible, in one of the harbours of Britain; for he instructed Volusenus to ascertain what harbours were capable of accommodating a numerous fleet, and, before he set sail, a deputation came to him from numerous British tribes to promise submission.[3220] The choice, as we have seen, lies between the harbours of Dover and Folkestone; and it is not credible that Caesar should have deliberately preferred the small harbour of Folkestone to the more spacious and more easily accessible harbour of Dover.[3221] And if he steered for Dover harbour, the ‘precipitous heights’ off which he anchored in 55 B.C. must have been, not the cliffs of East Wear Bay, but the cliffs between Dover and the South Foreland.[3222]

Airy,[3223] indeed, denies that Caesar ever intended to land in a harbour. ‘This,’ he says, ‘is not the manner of attempting debarkation in a country possessed by an enemy.... Sir Arthur Wellesley made no attempt at Lisbon, but put his troops on shore at the Mondego Beach.’ I wonder whether Airy knew the circumstances and the motives which determined Wellesley’s action: at all events his argument will not mislead any reader who has studied the history of the Peninsular War. Why did Wellesley make ‘no attempt at Lisbon’? Simply because, as Napier[3224] says, ‘the strength of the French, the bar of the river, the disposition of the forts, the difficulty of landing in the immediate neighbourhood, where a heavy surf broke in all the undefended creeks and bays, convinced him such an enterprise was unadvisable if not impracticable.... It was difficult to find a place to land. The coast, from the Minho to the Tagus, save at a few points, is rugged and dangerous; all the river harbours have bars and are difficult of access even for boats.... Seventy miles northward of the Lisbon Rock, the small peninsula of Peniché offered the only safe and accessible bay adapted for a disembarkation; but the anchorage was within range of the fort, which contained a hundred guns and a garrison of a thousand men. The next best place was the Mondego river; there the little fort of Figueras, now occupied by English marines, secured a free entrance, and Sir Arthur adopted it.’ The reader will have noticed that Airy not only ignores the circumstances which forbade Wellesley to attempt a landing at Lisbon, but also ignores the existence of the Mondego river.[3225] Caesar put his troops ashore on a beach; but he had intended to put them on shore in the best available harbour, because, unlike Wellesley, he had had reason to hope that his landing would not be opposed. If he did not intend to land in a harbour, why did he instruct Volusenus to report upon the Kentish harbours? And why did he steer towards ‘precipitous heights’, at the foot of which the dullest soldier in his army knew that it would be madness to land, if he did not intend to enter the harbour formed by the gap in those cliffs? Why did the enemy occupy those heights unless they believed that this was his intention? It might, indeed, be argued that after he received Volusenus’s report, he decided that Dover harbour was unsuitable, and therefore abandoned all idea of landing in a harbour. If so, he would not have steered towards the Dover cliffs; but neither would he have steered towards those of Folkestone or East Wear Bay. The only alternative would be to assume that he anchored off the lower cliffs north of St. Margaret’s Bay in order to wait for the overdue ships, having intended from the outset of his voyage to sail on and land north of Walmer.[3226] But when he says that ‘he thought the place [off which he anchored] most unsuitable for landing’ (hunc ad egrediendum nequaquam idoneum locum arbitratus),[3227] he unmistakably implies that he had contemplated the possibility of landing there; and, as he could have decided with his eyes shut that it would be absurd to land at the foot of ‘precipitous heights’, he must have concluded that it would be unwise to attempt to land, in the face of an enemy, in the harbour which was formed by the gap in the cliffs. Probably, before he knew that his landing would be opposed, he intended to observe the harbour with his own eyes and decide upon its merits himself.

Mr. H. E. Malden, however, remarks that, according to Dion Cassius, ‘Caesar made the land on the first occasion where he ought not, οὐ μέντοι καὶ ᾗ ἔδει προσέσχεν.’ ‘Surely,’ Mr. Malden continues, ‘this implies not that Caesar aimed at a certain point for a landing place and then abandoned it upon a nearer view, but that something like what befell him on the second voyage happened on the first also, and that he drifted out of his course to a point which he did not intend to reach. If so, this disposes of all idea of his aiming at Dover as the usual port of landing.’[3228]

Yes,--‘if so’. According to Mr. Malden,[3229] Caesar landed near Hurst in Romney Marsh. On this theory, he would have anchored off the coast near Sandgate, which was very little out of his course if he intended to land near Hurst, and not at all out of his course if, as Mr. Malden wrongly supposes, he sailed from Wissant. It is universally admitted, and it is certain that when Caesar was approaching Britain, and for some hours previously, the tidal stream was running up the Channel. Will Mr. Malden explain what that current could have been which would have caused Caesar’s ships, while steering either from Wissant or from Boulogne for Hurst, to ‘drift’ towards Sandgate on the eastward stream?

2. The reader is already familiar with the subject of the tidal streams, in so far as it relates to the present discussion.[3230] As we have seen,[3231] Caesar may have landed in Britain on the 26th of August, 55 B.C.; and on that day high water should have occurred at Dover at 6.21 a.m. I say ‘should have occurred’, because observations have shown that high tide sometimes occurs a few minutes earlier or later than the time predicted in the Admiralty Tide Tables.[3232] Therefore, accepting the general rule laid down by Admiral Beechey and Surveyor Calver,[3233] the stream off the Dover cliffs would have turned to the west some time between 10.21 and 11.9 a.m., and again to the east at 5.24 p.m.

But Caesar, we are assured by Airy and Lewin, weighed anchor at 3 p.m.; and at 3 p.m. the stream would still have been running towards the west. The argument has imposed upon weak minds: the reader will see that it has not the slightest force. There is no evidence that Caesar weighed anchor at 3 p.m. What he says is simply that he awaited at anchor the arrival of the rest of the ships--the ships which had failed to keep pace with the leading division of the fleet--until the ninth hour. The ninth hour in the latitude of Dover lasted on the 26th of August from 2.20 till 3.30 p.m.[3234] Therefore Caesar, assuming that his statement was literally correct, may have waited at anchor for the arrival of the laggard ships until 3.30 p.m. Now, as the reader will remember, I have demonstrated, from the evidence to which Airy appeals--the evidence supplied by Admiral Beechey and Surveyor Calver--that on the day of Caesar’s landing the tidal stream may have turned eastward earlier than 3.54 p.m.;[3235] and if it turned twenty-five minutes earlier, it turned in the ninth hour.

Hitherto we have assumed that Caesar’s statements of the hour up to which he awaited the arrival of his overdue ships was literally correct. But now let me beg any one who still feels a doubt whether his narrative agrees with the hypothesis that he sailed eastward from his anchorage, to use his common sense and to remember that he has a sense of humour. Airy, Lewin, Mr. Malden, and the rest argue in this strain:--Caesar waited at anchor till 3 [or 3.30] p.m., and not a minute later: the stream was then running westward; therefore Caesar sailed towards the west.--Q.E.D. This is the sort of argument that might have been expected, not from an Astronomer Royal, or from a barrister like Lewin who knew the world, but from a clever schoolboy. Yet not a single commentator has ever pointed out its absurdity. Had Airy forgotten the discrepant statements that were made by officers who had watches in their pockets as to the hour at which this or that episode occurred in the campaign of Waterloo?[3236] Did Airy or Lewin imagine that Caesar had a Dent’s chronometer on board his galley, and, the moment after he weighed anchor, noted down in his diary the words, Hora nona ancorae sublatae sunt? It is possible that he may have had a water-clock (clepsydra)[3237] on board, which would have enabled him, if it had been duly corrected for the latitude of Dover, and if the sea had been so smooth that the ship was motionless, to tell the time approximately; but surely it is probable that he roughly estimated the time from his observation of the altitude of the sun?[3238] And is it not equally probable that he trusted not to a diary but to his memory? His estimate may have been right: but also it may have been wrong; and anyhow it is folly to stake the whole argument upon its accuracy.

Nor, again, is it even certain that Caesar did weigh anchor at the time which he called horam nonam. As he says that he waited for the arrival of his overdue ships till the ninth hour, it may be presumed that they did not arrive earlier. When they arrived, their captains had, I suppose, to receive instructions, as the generals and military tribunes had done already. As Heller[3239] and de Saulcy have argued, we have no right to infer from the words in ancoris exspectavit that when the period of waiting was over, Caesar ceased to remain at anchor.[3240] And, even if he gave no instructions to the captains of the laggard ships, but left them to their own devices, it remains certain that to get the ships into order, to give the signal for starting, and to weigh anchor, consumed an appreciable time.

But we need not insist upon this argument. We may rest satisfied with the knowledge that the stream may have turned earlier than 3.54 p.m.[3241] We may be assured that Caesar had no means of knowing exactly at what time he weighed anchor; and therefore, even if he intended to convey that he weighed anchor in the ninth hour, we are not compelled to assume that he did so before 3.30 p.m. We know that Caesar was too wise to start for a seven miles’ sail on the last of the ebb tide, when the current was slowest, and when he would have had no certainty that it might not turn against him before he had completed his voyage; and we are confirmed in this conviction by his own words, which tell us that he started when he had got wind and tide in his favour. Finally, we have proved that it is not improbable but impossible that he landed either at Pevensey or on any part of Romney Marsh.[3242] Thus the question of the direction in which the tide was running when Caesar sailed from his anchorage, which in itself would be doubtful, is settled by other considerations. There is but one conclusion to which we can come, and that conclusion is absolutely certain: when Caesar weighed anchor off the Kentish cliffs, he sailed towards the north-east.[3243]

3. It will be remembered that Caesar, in describing the final stage of his first voyage, says that he ‘moved on’ (progressus[3244]), seven miles from his anchorage. Both Long[3245] and Heller[3246] regard the word progressus as a proof that Caesar must have landed on the coast of East Kent. Had he anchored off East Wear Bay, and sailed thence to Hythe or to Lympne, he would not, says Heller, have continued in a straight line the direction of his voyage from Gaul to Britain; but he would have done so if he had anchored off the South Foreland and sailed thence to Deal. I do not commit myself to absolute agreement with this argument, as it stands, although every one who is familiar with Caesar’s use of the word progredi[3247] will admit that it would apply much better to a run from the Dover cliffs to Walmer, made in continuation of a voyage from Boulogne towards the South Foreland, than to a run from East Wear Bay or from a point off Folkestone to Hythe or Lympne.

4. Dion Cassius says that Caesar, in sailing from his anchorage to the place where he landed, rounded a promontory.[3248] I attach some importance to these words for the following reasons. Although Dion is no authority in regard to such matters as the details of a battle, and although, from carelessness or love of meretricious ornament, he constantly misunderstood or misused his authorities, it is not conceivable that he should have made so definite and simple a statement as this without authority, unless he had invented it. I am sure that every one who is capable of weighing the credibility of an historical writer will agree with me that this is not such a statement as Dion would have invented, and that it is such a statement as the most careless and rhetorical of writers, if he had found it in an authority, would have followed correctly. I therefore believe that Dion took it from some authority which is now lost, and that it is true. If so, the promontory which Caesar rounded can only have been the South Foreland. Lewin[3249] admits that the statement, ‘if taken literally, looks as if he went round the South Foreland; but,’ he adds, ‘I am satisfied that if he had done so, Caesar would have mentioned so remarkable a promontory’. This observation only shows that Lewin was not familiar with Caesar’s style. Caesar did not trouble himself about picturesque details, however remarkable they might be, the mention of which was not essential to the clearness of his narrative. ‘If,’ continues Lewin, ‘the descriptive words of so late a writer as Dion are to have any weight, I should interpret them as meaning only that Caesar sailed round the bend of the precipitous shore between Folkestone and Sandgate ... or else that Caesar arrived at first off Eastweir Bay ... and then sailed round the cliff which shuts in the bay on the west.’ I doubt, however, whether either ‘the bend’ or ‘the cliff’ could fairly have been described as ‘a projecting headland’ (ἄκραν προέχουσαν); and Lewin virtually admits that neither would have been worth mentioning. But if any one gainsays this argument, we can well afford to dispense with it.

5. If Volusenus deserved the confidence which Caesar reposed in him, there cannot be the faintest doubt in the mind of any unprejudiced man who has studied military history or even Caesar’s Commentaries, and has taken the trouble to observe the features of the Kentish coast, as to the landing-place which he would have selected. He would not have wasted his time by cruising down the coast of Sussex: the first sight of the lofty hills which hem in East Wear Bay, of the cliffs which extend from Folkestone to Sandgate, and of the heights which back the coast from Shorncliffe to Lympne and beyond, would have warned him not to advise the great captain whom he served to land beneath them; but, as his galley glided past Kingsdown and neared Walmer, he saw, stretching away towards the neighbourhood of Sandwich, the open coast, offering easy access into the interior, of which he was in search. Here and here only, if Dover Harbour were held by an enemy, it would be safe to land. Here and here only it would be possible to follow up a victory by an effective cavalry charge.

6. Assuming that Caesar landed in East Kent, his narrative of the adventures of his cavalry transports and of the shipwrecks in 55 B.C., which, as we have seen,[3250] is inexplicable on any other theory, presents no difficulty. The transports which returned to the port whence they started would have been laid to on the port tack, and would have been carried back to Ambleteuse by the same wind which drove their sister ships westward down the Channel.[3251] This wind, which could not have driven ashore, in the immediate neighbourhood of their anchorage, vessels anchored off Lympne or Hythe, would inevitably have driven ashore, once they had parted their anchors, vessels lying off the coast between Walmer and Deal.[3252]

7. Finally, Caesar sailed on his second voyage to Britain with a south-west wind; and a south-west wind would obviously have been much more favourable to a voyage either from Boulogne or from Wissant to Deal or to any point on the coast of East Kent than to a voyage from Boulogne to Lympne or even Hythe.

It remains to consider objections.

1. I have already refuted Airy’s criticism of the view that Caesar anchored off the cliffs of Dover; and we have seen that Airy is compelled by his own argument to identify the cliffs which Caesar describes as ‘precipitous heights’, and Cicero as ‘astonishing masses of cliff’[3253] with cliffs ‘ten to thirty feet high’.[3254] He maintains that our interpretation of the words angustis montibus ‘must be guided by consideration of the character of place under which an officer would think of attempting to land’, and ‘must also depend upon the possibility of aiming a javelin from the heights’. ‘Both considerations,’ he insists, ‘exclude such lofty cliffs as those of Dover and Folkestone.’[3255] One might have expected that a man so clever as Airy would have perceived that Caesar never intended to land under angusti montes at all, whether they were 300 or, as Airy will have it, 10 feet high. He intended to land in the gap between the angusti montes, that is to say, in Dover Harbour.[3256] Airy, indeed, contends that ‘neither Caesar nor Volusenus would think for a moment of pushing his boats into a creek whose defenders could attack them on both sides’;[3257] but he forgets that Caesar had reason to expect that his landing would not be opposed.[3258] As a matter of fact, Caesar says nothing about javelins; but if Airy had stood on the beach at the foot of Dover cliffs and allowed an army standing above to pelt him with missiles, he would have speedily realized that it was possible to take aim from those ‘lofty cliffs’; and, if he had survived the experiment, he would have been a wiser and a sadder man.

2. Lewin[3259] says that ‘if we assume that Caesar was wholly ignorant of the British coast ... he could not have discovered the level at Deal’.

But what is the use of making absurd assumptions? As we have already seen, it is certain that before Caesar set foot in Britain he knew everything about the coast of East Kent that Volusenus could tell him.[3260]

3. It has been objected that the distance from Dover to Walmer exceeds the seven miles which separated Caesar’s anchorage from his landing-place. But no sensible man maintains that Caesar anchored exactly opposite Dover Harbour. He anchored off the cliffs east of Dover. He says that he sailed ‘about seven miles’ (circiter milia passuum VII[3261]) to his landing-place. The distance by sea from off Dover to a point just north of Walmer Castle is 7½ English miles, or a little less than 8⅛ Roman miles; from off the South Foreland to the same point 4⅞ English miles, or a little more than 5¼ Roman miles. Accordingly, if we assume that Caesar’s landing-place extended for about one mile north of Walmer Castle, and that his anchorage extended eastward for about one mile towards the South Foreland from a point about one mile east of Dover, the requirements of the Commentaries are satisfied absolutely; and even if the ship on the left of the line anchored just east of the harbour, off the Castle cliff, Caesar’s rough estimate, which he qualified by the word ‘about’ (circiter), is hardly violated. Surely the difference between ‘about seven’ and 8⅙ is not worth cavilling over.[3262]

4. Lewin[3263] maintains that the beach at Deal does not correspond with Caesar’s account of his disembarkation. The Romans, he argues, evidently landed on a gradually shelving shore; whereas ‘at Deal the beach ... descends so steeply that with a three hours’ flood transports can come up to the water’s edge’.[3264] Again, remarking that Caesar describes the shore on which he landed as ‘open’ (apertum) and ‘level’ (planum), he says that ‘between Walmer and Deal ... the ground is uneven, and cannot be called flat’. Lastly, he recalls the witnesses whose evidence he had cited in support of his own theory. ‘Caesar, Dion Cassius, Plutarch, and Valerius Maximus, all ... refer to the marshes at the place of landing ... who has ever heard of a marsh at Deal.’[3265] And again, ‘Valerius Maximus speaks of two small islands at the landing-place.... It was never suggested that islands did exist or could have existed at Deal.’[3266]

Now, to begin with, Lewin assumes that the beach between Walmer and Deal has undergone no change, and was as steep 2,000 years ago as it is to-day. In a previous article[3267] I have shown that this assumption is untenable. Secondly, Lewin misunderstands Caesar’s narrative. As some of the Britons threw their missiles from the shore,[3268] it is evident that the deep water[3269] in which the Roman ships grounded was within the range of a sling or of an arrow, that is to say, quite close to the shore. Thirdly, no shore is or could conceivably be ‘level’: the shore between Walmer and Deal is, as anybody may satisfy himself by the evidence of his own eyes, free from obstructions[3270] (apertum) and, speaking generally, evenly shelving (planum)[3271]. Caesar applied these epithets to the shore on which he ran his ships aground, not to ‘the ground’ behind it. As to the arguments which Lewin bases upon the statements of Plutarch, Dion Cassius, and Valerius Maximus, I have already shown[3272] that they are worthless; but if the story told by Valerius Maximus were worth anything, it would not support, but destroy Lewin’s theory. For, as we have seen, Valerius Maximus[3273] speaks not of ‘two small islands’ but of one large island (Britannica insula), and a rock (scopulus); and there are no rocks off Hythe or Lympne.[3274] Moreover, although the fact is of no consequence, there are rocks, called the ‘Malms’, which are visible at low water, during spring tides, opposite Deal;[3275] and in the time of Caesar there were marshes behind the sand-hills north of Deal.

5. General Creuly,[3276] referring to Caesar’s description of the storm by which some of his cavalry transports, after they were descried from his camp in 55 B.C., were driven further down the coast,[3277] maintains that the only point to which they could have been driven was Dungeness. It follows, he says, that the camp could not have been at Deal; for its distance from Dungeness is far too great.

The answer to this curious argument is, first, that it is plainly impossible to indicate the exact point to which the ships were driven; secondly, that it was certainly not Dungeness, for Dungeness did not exist in Caesar’s time;[3278] thirdly, that there is nothing in Caesar’s narrative to show how far the point to which the ships were driven was from his camp; and, fourthly, that if the camp was at or near Deal, there is nothing in his narrative to show that the ships could not have run before the gale as far as the longitude of Dungeness, or even a good deal further. From Deal to Dungeness is only about 28 nautical miles; and if the ships approached the British coast in the morning,[3279] they would have had the greater part of the day in which to make the run.

6. General Creuly[3280] calls attention to the passage, which I have already more than once quoted, in which Caesar tells us that at daybreak, on his second voyage, he ‘saw Britain lying behind on the port-quarter’. If, says the general, Caesar had steered for Deal, he would have had the coast of Britain on the port side throughout the voyage, and there would have been no point in the words sub sinistra. He insists that these words are simply the correlative of longius: Caesar saw Britain on the left because the flood had carried him too far.

It will be remembered that Appach used this passage to prove that Caesar landed in the (assumed) Bay of Appledore.[3281] Creuly uses it to prove that Caesar landed near Hythe. I have shown the futility of Appach’s argument; and to answer Appach is to answer Creuly. I may remark, however, that Caesar does not merely say that he saw Britain ‘on the left’ (sub sinistra): he says that he saw it ‘lying behind on the left’ (sub sinistra relictam). The coast could not have been described as ‘lying behind on the left’ until after Caesar had passed the South Foreland;[3282] and, although he was steering not for Hythe, but for some point on the coast of East Kent, the flood tide did carry him ‘too far’.

7. Many commentators have argued that Caesar’s account of the last stage of his second voyage is inconsistent with the theory that he landed near Deal. It will be remembered that, after telling us that ‘he saw Britain lying behind on the port quarter’, he goes on to say that he then followed the turn of the tide, and that all his ships reached the landing-place by rowing towards midday. The tide which he followed was of course the westward stream: strictly speaking, its direction, if he had drifted to some point east of the Goodwin Sands, was south-west[3283] (magnetic), or between south-west by south and south-south-west (true). It is very doubtful, Lewin thinks, whether, with a current running at the rate of 3¾ miles an hour, the fleet could, by the mere use of oars, have reached Deal at all; and it is certain that, in order to do so, it would have been necessary to steer ‘across, if not actually against the current’. Caesar ‘could not’, Lewin concludes, ‘be said to follow the tide when he was steering athwart it. Besides, as it must necessarily have been almost low water when the tide turned, had he held on for Deal he would infallibly have struck on the Goodwin Sands.’[3284] Airy, on the other hand, asserts that Caesar ‘must have been cast upon the Goodwin Sands’ during the drift.[3285]

Now the reader will have already perceived that if Lewin’s argument tells against the view that Caesar landed near Deal, it is fatal to the view which Lewin himself defends. For Lewin’s theory compels him to assume that Caesar had drifted no further than a point off the South Foreland:[3286] in order to reach Hythe he would have had to row less than thirteen nautical miles, with the stream throughout; and it is therefore not easy to understand why his rowers should have been called upon to make any extraordinary efforts. This point he ignores. Again, he admits, or rather insists, that Caesar must have drifted as far as the South Foreland; but it is easy to demonstrate that if Caesar had been steering for Hythe, he could not have drifted so far. For, as Lewin himself says, the length of the drift must, in that case, have been not less than twelve nautical miles.[3287] Now the drift only lasted from about midnight till daybreak; and, assuming that it lasted four hours--a liberal estimate--the stream actually travelled, at the most, nine miles:[3288] of course the ships would not have travelled so far if they had merely drifted,[3289] but the faint breeze, without which they would not have had steerage-way,[3290] may have made up the deficit. But, in order to give Lewin every chance, let us accept the most favourable of three estimates with which he himself supplies us. He says[3291] that ‘the greatest velocity of the tide is, according to the Tidal Tables, 3·3 knots an hour.... The drift would, of course, be less than the velocity.... From midnight till daybreak at 4 a.m., would, therefore, give a drift of twelve miles’. But Caesar’s voyage took place on or about the 7th of July;[3292] and daybreak was about 3.15 a.m. This consideration alone compels us to reduce Lewin’s estimate to ten miles and a half; and, moreover, he forgets that the tide never runs for four consecutive hours, much less for the last four hours, at its greatest velocity.[3293] Thus the argument upon which he relies to prove that Caesar must have landed at Hythe turns and pulverizes his already shattered theory.

On the hypothesis that Caesar landed anywhere between Walmer and Sandwich, the statement that he ‘followed the turn of the stream’ presents no difficulty unless he meant that from the time when the tide turned until the time when he reached Britain his men were rowing hard exactly in the direction of the stream. This is what Lewin assumes. But Caesar says no such thing. What he says is that, having followed the turn of the tide, he rowed hard in order to gain the desired landing-place. So long as the tide served, hard rowing was obviously unnecessary,--if the Goodwin Sands existed. In that case the true explanation is a modification of that offered by C. Schneider,[3294] who says, ‘As long as it was possible to follow the turn of the tide, rowing was unnecessary. But after they had reached a point where they could do so no longer without being carried past their destination, they took to rowing.’ Of course they would not have trusted to the current alone at any time; but, supposing that the Goodwin Sands then existed or that an island occupied their site, they travelled south-westward with the current until they had turned the obstacle, and then rowed hard in a north-westerly direction, across the current,[3295] till they reached the landing place.[3296]

The question of the Goodwin Sands has been discussed in an earlier part of this book.[3297] Either they did not exist in the time of Caesar, the substratum on which they rest being covered by the sea; or they did exist, but had not accumulated to their present height, or perhaps to their present extent; or they were virtually identical with the present sands, though their limits, which are not constant, may not have been the same as they are now; or, finally, as Sir Charles Lyell suggested, their place may have been occupied by an island. The question cannot be positively settled; but, for reasons which I have already given, I am rather inclined to believe that either Sir Charles Lyell’s suggestion was right, or the sands had accumulated sufficiently to be visible, or at all events dangerous, at certain points at low water. I shall, however, presently take account of the possibility that neither of these alternatives is true.

Much depends upon the answer which is to be given to the question, To what point had Caesar drifted when he saw the coast of Britain? Heller argues that this point must have been about nine miles due east of Ramsgate; but, as I have shown in the previous article, his reasoning is unsound.[3298] If Caesar had drifted to the point which Heller indicates, his course to Deal, where, according to Heller, he disembarked, would have been nearly south-west, that is to say, nearly identical with the direction of the Gull Stream, which is described in the Admiralty Tide Tables as SW. ½ W. (magnetic), or, approximately, SW. by S. true.[3299] But, on this hypothesis, it would be inexplicable that his soldiers were obliged to row hard. Heller, indeed, conjectures that Caesar steered for the nearest point of the coast, that is to say, nearly due west, intending to keep close inshore until he found the landing-place; and he remarks that this would explain why his men were obliged to use their oars instead of committing themselves to the stream alone.[3300] But Caesar must have known the whereabouts of his landing-place; and Heller’s explanation seems to be far-fetched. Besides, as I shall presently show, Caesar, on his second expedition, did not land between Walmer and Deal, but in the neighbourhood of Sandwich.

For reasons which I have given in the article on the Portus Itius[3301] I think we must conclude with Napoleon the Third,[3302] that Caesar could hardly have drifted much further than a point on the latitude of Deal and east of the Goodwin Sands. When the tide turned soon after daybreak--about 4.30 a.m., if, as is probable, the day was the 7th of July[3303]--he would have dropped down with the ebb as I have already explained. If, after he had passed the sands or the island, he had waited till about 9.30 a.m., the stream would have turned, and have begun to flow NE. ½ N.[3304] magnetic, or, approximately, NNE. true: if he had not waited, he would have had to row hard, as I have shown above,[3305] athwart a stream which was flowing at a rate varying from three to two knots, until it turned about 9.30.

The objection that, during the drift, Caesar ‘must have been cast upon the Goodwin Sands’ is as groundless as the objection, which has just been met, that if he had attempted to row to Deal [or to the coast between Sandown Castle and Sandwich] ‘he would infallibly have struck upon the Goodwin Sands’.[3306] Captain Iron, the harbour-master of Dover, traced out upon the chart in my presence the course which the Roman flotilla would naturally have steered from Boulogne, and showed that, after the south-west wind dropped, it would have drifted east of the Sands. If, as is probable, the flat-bottomed vessels had made so much lee-way that, even before the wind dropped, they had got a little out of their course, they would have drifted still further eastward. It may also be objected that if Caesar, after he had followed the tide south-westward, had turned the sands or the island which may have occupied their place, he would have mentioned the fact. The objection would be quite natural if it came from a writer who had merely ‘got up’ so much of Caesar’s narrative as he thought would be necessary to enable him to study the question; but every one who is really familiar with the Commentaries knows that Caesar often omitted to mention matters, especially geographical, which a modern historian would feel bound to record. There remains the possibility--perhaps the probability--that neither the Goodwin Sands nor the hypothetical island then existed. In this case Caesar would have had to row across the current: still he might fairly have said that he followed the turn of the tide. He was bound for Britain, and could not begin to row until the tide began to set towards Britain, though in a different direction from his: what other expression could he have used than aestus commutationem secutus? Lewin doubted whether he could have reached Britain by rowing at all; but Lewin did not understand what he was writing about, and ought to have consulted a treatise on practical navigation.[3307] On the other hand, it might possibly be objected that he would not have taken till noon to reach his destination: that depends upon the exact direction of the current, which often varies from the direction indicated in the Admiralty Tide Tables,[3308] and upon the rate at which the vessels could have been rowed in still water. What he says is that all his ships reached Britain by about noon; and doubtless there were stragglers. Of course it may be argued that there were not, and that all the ships were actually rowed for seven hours. But if any one thinks that the possible objection which I have anticipated is valid he will find himself confronted by another which is absolutely insuperable. For he must needs accept Lewin’s alternative theory, to which the objection would apply with redoubled force,--that Caesar took seven hours to row from a point off the South Foreland (though he could not have drifted so far) to Hythe; in other words, that by rowing hard Caesar could only manage to travel two knots an hour with the stream; and that he took twice as long to row less than thirteen knots with the stream as he had taken to drift twelve knots without rowing!

8. Airy,[3309] remarking that it is evident from the Commentaries that ‘there were forests and cornfields near’ the Roman camp, maintains, first, that if Caesar had landed near Deal ‘he would have had for seven miles all round his camp bare chalk-downs, on which in those days there probably was neither a tree nor a ploughed field’; secondly, that a night march, such as that which Caesar made after he landed in 54 B.C., can only be made upon good roads; that ‘the roads in a woodland and clay-ground country are almost invariable’; that accordingly the roads of East Kent ‘are in the very same tracks as in the days of Julius Caesar’; and that Caesar’s night march, the length of which was 12 Roman miles, would have brought him to the marshes of the Stour, whereas, if the Britons had been posted on that river, he would have crossed it ‘at the sound ground of Canterbury or above it, and would have attacked their flank’; and thirdly, that if Caesar’s march to the point where he crossed the Thames had begun near Deal, ‘his course would have been all the way parallel to the Thames, and the expression “ad Tamesin”[3310] could scarcely have been used.’

The first of these objections is summarily disposed of by Long,[3311] who points out that at Worth, between Deal and Sandwich, there is ‘some of the best wheat land in England’; and by Dr. Guest,[3312] who remarks that ‘the uplands round Deal are every autumn white with corn’;[3313] and that, as many of the great forests which once existed in England have disappeared, the absence of woods in the neighbourhood of Deal is no proof that there were none there in Caesar’s time. But, as a matter of fact, there are no less than five woods at distances varying from about a mile to three miles and a quarter from Upper Deal;[3314] and Caesar only speaks of one wood as having existed in the neighbourhood of his camp, and implies that it was a considerable distance off.[3315] The second objection is valid against the theory that Caesar landed between Walmer and Deal on his second expedition, unless he encountered the Britons on the Little Stour; but I shall presently show that on that occasion he landed near Sandwich, and that his march of ‘about twelve miles’ did bring him either to ‘the sound ground of Canterbury or above it’, or to Fordwich or Sturry below Canterbury, but of course not to ‘the marshes of the Stour’. The third objection, from the point of view of Airy, according to whom Caesar landed at Pevensey, may to some minds appear plausible; but the view that Caesar landed at Pevensey is out of the question. But the same objection has been urged by the advocates of Hythe; and they hardly deserve an answer. Supposing Caesar’s march had been nearly parallel with the Thames, what then? If he had landed at or near Hythe, he must have first encountered the Britons at Wye on the Stour; and from Wye to Brentford[3316] his march would have been hardly less parallel (if the expression may be pardoned) to the Thames than from the neighbourhood of Sandwich. In writing the words ad Tamesim he simply intended to indicate approximately the distance from his naval camp, either to the point where he crossed the Thames or to the nearest frontier of the territory of Cassivellaunus; and the direction of his march is nothing to the purpose. Whether that direction was nearly parallel to the Thames or at right angles to it, the distance was about 80 Roman miles.

9. It will be remembered that at the close of Caesar’s second campaign his camp was attacked by the four chieftains of Kent.[3317] General Creuly argues that if the camp had been at Deal, Caesar, when he was marching to the place where he crossed the Thames, must have traversed the country of the four chieftains without having first subdued them. If, on the other hand, says the general, he had started from Hythe, he would have marched not through the heart of their territory, but close to their frontier; and for this very reason they would not have thought it necessary to submit.[3318]

Heller[3319] has taken the trouble to answer this nebulous argument. He points out that Caesar, having defeated the chieftains in the engagements which immediately followed his arrival, evidently did not think them sufficiently dangerous to wait until he had secured their complete submission. Indeed, if their territory comprised the whole of Kent, or even that part of it which lies east of Maidstone, it is evident that Caesar, marching northward from Hythe to the Stour, and then turning westward or north-westward, would have traversed the heart of their country. Moreover, as Heller might have added, it would have been just as hazardous for Caesar to leave the chiefs unsubdued if he had marched from Hythe as if he had marched from the neighbourhood of Deal or of Sandwich. Furthermore, he had no time to spare; and unless he had completely laid waste their country, and treated their people with the ruthless severity with which he afterwards treated the Eburones[3320]--and to do this would have required the greater part of the time which he had to spend in Britain--it would have been utterly impossible for him to subdue them so thoroughly as to prevent them from attacking his camp in his absence. Think of the Boers!

10. Lastly, it has been objected[3321] that Caesar could not have landed near Deal, or indeed at any point on the coast of East Kent, because, if he had marched against Cassivellaunus from that neighbourhood, he must have passed through tracts abounding in beech-woods, whereas he says expressly that there were no beech-trees in Britain.[3322] But Dr. Guest[3323] disposed of this objection by pointing out that ‘at whatever point on the south coast Caesar landed ... he must have crossed the North Downs on his way to the Thames, and so have passed through “tracts abounding in beech-woods”’. Mr. Mackinder,[3324] indeed, asserts, without giving any authority, that the beech was introduced into this country by the Romans; but it has been found in submerged forests and in deposits of the Bronze Age.[3325] If by the word fagum Caesar meant the beech, his statement was incorrect.[3326]

XII. THE THEORY THAT CAESAR LANDED AT RICHBOROUGH OR SANDWICH

The commentators who believe that Caesar landed in the neighbourhood of Sandwich are not agreed among themselves. Napoleon III, who holds that he landed between Walmer and Deal in 55, and at or near Sandwich in 54 B.C., argues that the disaster which befell his ships on the night of the full moon in August, 55 B.C., must have taught him the danger to which they would be exposed on the beach near Deal, and that accordingly he must have selected a better landing-place ‘some kilometres further north’.[3327] Others, like the late George Dowker, maintain that he landed near Sandwich on both occasions.

Dowker assumes that Caesar’s own ship, if not the rest of his fleet, anchored in 55 B.C. off the South Foreland; and he goes on to say that ‘from the South Foreland ... seven miles would bring him near the mouth of Sandwich Haven’. He decides for Sandwich instead of Deal because, in his opinion, the Commentaries show that Caesar landed ‘at or near a point whence he could get his long vessels on the flank of an enemy’, and at Deal ‘no such bay existed’, whereas at Sandwich the very bay which he wanted was formed by the mouth of the Stour.[3328]

That Caesar did ‘get his long vessels on the flank of an enemy’[3329] is unquestionable: but he does not say that he placed them in a bay or in the mouth of the Stour or any other river; nor is it easy to understand why a bay or the mouth of a river should have been necessary for his purpose. The object of placing the ‘long vessels’ on the enemy’s right flank was that the artillerymen, slingers, and archers who manned them might drive away the enemy who were trying to stop the disembarkation; and some of the enemy were standing on the shore, while the rest were wading, or mounted on horseback, or perhaps standing in their cars in the sea close to the water’s edge.[3330] Why should not the ‘long vessels’ have been in the sea too? What was to be gained by sending them into the mouth of a river? When Dowker said that a run of 7 miles would have brought Caesar from the South Foreland nearly to ‘the mouth of Sandwich Haven’, his eagerness to prove his point prevented him from making an accurate measurement. From the South Foreland to the place which was once the mouth of Sandwich Haven, as measured within half a mile from the shore on Sheet 290 of the One Inch Ordnance Map, is just over 11 statute, or 12 Roman miles. Nor does Battely contribute much to the argument when he pleads that VIII milia passuum ‘does not occur invariably in all the editions’, and that Caesar may have made a mistake. We are not concerned with ‘the editions’, but with the MSS.; and VIII milia passuum does not occur in any of them, but VII (or septem) milia passuum in all.[3331] It would be strange if Caesar had not made a slight mistake; but it would be stranger still if he had mistaken twelve miles for seven. Battely[3332] argues that he must have landed at Richborough, (a) because he says that Cantium, where the Gauls generally landed, has an easterly aspect, whereas Dover looks south; (b) because Dion’s description of the landing-place, ‘which, as to the nature of the shore, directly contradicts Caesar’s narrative,’ is applicable to Richborough, where there ‘was a marshy and muddy shore, on which Caesar’s soldiers ... could not keep their footing’;[3333] and (c) because ‘all the time the Romans were masters of our island, Rutupiae ... was the only port where they disembarked’. Now Caesar, who never talks nonsense, does not say that the whole of Cantium, or even that part of it in which the Gauls used to land, faces the east. He merely says that of the side of Britain which faces Gaul ‘one corner, by Kent--the part which almost all ships from Gaul make for--has an easterly ... aspect’ (huius lateris alter angulus, qui est ad Cantium, quo fere omnes ex Gallia naves adpelluntur, ad orientem solem ... spectat[3334]). Besides, nobody would argue that Caesar landed at Dover; and the coast between Walmer and Deal has an easterly aspect no less than Richborough. Secondly, when Battely says that Dion’s description of the landing-place ‘directly contradicts’ Caesar’s narrative, and then elects to believe the inaccurate and rhetorical Greek, who wrote two hundred years after the event, rather than the eye-witness, he shows that he is incapable of serious criticism. Besides, it is not true that Dion’s description of the landing-place[3335] contradicts Caesar’s narrative either directly or indirectly; and neither of them says that the shore was either ‘marshy’ or ‘muddy’.[3336] Thirdly, Rutupiae was not ‘the only port’ where the Romans disembarked after they had become masters of the island. It is certain that Dover Harbour was in use during the Roman occupation of Britain, for it is mentioned in the Itinerary of Antonine;[3337] inscribed tiles found at Dover prove that it was one of the stations of the Classis Britannica,[3338]--the Roman ‘Channel Fleet’; and the mere fact that two disembarkations after the time of Caesar are recorded to have taken place at Rutupiae[3339] does not prove the contrary.

The conclusion is that Caesar, on his first voyage, did not land either in the neighbourhood of Sandwich or at Richborough, and that he did land between Walmer and Deal. The disembarkation must of course have taken place along a front of considerable extent; and the most southerly point at which it would have been easily practicable is quite close to Walmer Castle.

In the following year, however, the fleet must have made the land somewhere between the site of Deal Castle and the latitude of Sandwich. For the Romans, as we have seen, on the morning after their disembarkation, fought a cavalry action on the banks of a river, at a spot about twelve miles from the camp which Caesar had constructed near the place of landing: the river, as we shall afterwards see,[3340] was the Stour; and if the length of the march was estimated with tolerable accuracy, the camp must have been in the neighbourhood of Sandwich.[3341] Besides, Caesar tells us that ‘he felt little anxiety for the ships, as he was leaving them at anchor on a nice open shore’ (eo minus veritus navibus, quod in litore molli atque aperto deligatas ad ancoras relinquebat[3342]). The meaning of the epithet, mollis, has been already explained;[3343] and, moreover, although no commentator has noticed the fact, it does not need much acumen to see that Caesar was here excusing himself for having left his ships at anchor, in spite of the severe lesson which the storm of the previous year had given him, by the plea that he had selected a more favourable anchorage.[3344]

* * * * *

I began this inquiry early in 1900 with a mind absolutely unbiassed, resolved to do one of two things,--either to solve the problem, or, if that could not be done, to show, once for all, that it was insoluble. The reader knows that I have not neglected any means of ascertaining the truth; and I have provided him with the means of controlling every statement which I have made. I have set down fully and fairly the arguments of those from whom I differ; and I have kept back nothing, I have called attention to everything, that might appear to tell against the conclusion to which the evidence inevitably led. I need not say anything by way of recapitulation, for no man who has read this article attentively can be lacking either in patience or in intelligence; and I am sure that the reader is by this time convinced of these things:--that it has been demonstrated that Caesar did not land at Pevensey, or anywhere in Sussex; that it has been demonstrated that he did not land at Hythe, or anywhere in Romney Marsh; and that it has been demonstrated that he did land both in 55 and in 54 B.C. in East Kent,--in the former year between Walmer Castle and Deal Castle, in the latter north of Deal Castle. That some will still for a time dispute these conclusions is likely enough; but not those whose judgements count. For them the problem is solved.

NOTE.--The following is a transcript of the report of Messrs. Doak, Hudson, and Sprigge, sent to me from the Nautical Almanac Office, and alluded to on page 610, supra:--

‘The calculations have proved somewhat more lengthy and complicated than was at first anticipated, since it was found on examination that no approximate process would give results of a satisfactory degree of accuracy. In consequence the following work has been done:--

1. Twelve complete places of the Moon were determined from Hansen’s “Tables de la Lune”, embodying all the inequalities. Longitude, Latitude, and Parallax were thus obtained.

2. The Sun’s Longitude, the Obliquity of the Ecliptic and the Sidereal Time were then computed from Newcomb’s Tables for seven successive Greenwich Mean Noons. A slight shortening was possible in the case of the Longitude, but the resulting error cannot, in the opinion of Professor Newcomb, exceed ±30 seconds of arc, and is probably about ±12 seconds.

3. The Moon’s position was then converted from Longitude and Latitude to Right Ascension and Declination.

4. The Greenwich Mean Time of Moon’s Transit was computed for twelve transits.

5. The times of High Water at Dover were obtained by applying the proper quantities from the Admiralty Tidal Curve for Dover to the times of Moon’s Transit.

6. From the Longitudes of Moon and Sun the time of Full Moon was determined.

7. The whole of the calculations were then examined and duplicated where desirable.

With the one exception of the Sun’s Longitude the calculations have been rigorous, and, so far as the computing is concerned, are of the same degree of accuracy as those published annually in the Nautical Almanac. A very slight divergence from the truth is, however, possible owing to the fact that the tables of the Moon and Sun are used for an epoch nineteen-and-a-half centuries ago; but it is very unlikely that this is large enough to affect the times of high water or of full moon.

4th October, 1902.’

THE CREDIBILITY OF CAESAR’S NARRATIVE OF HIS INVASIONS OF BRITAIN

I have already published in Caesar’s Conquest of Gaul an essay upon the credibility of Caesar’s Commentaries; but, as it only deals incidentally with the invasions of Britain, I intend in this place to examine certain charges which have been made against that section of Caesar’s narrative. It has been objected, perhaps justly, to the essay by competent critics in this country (though not on the Continent[3345]) that it devotes an undue amount of space to the refutation of inferior writers; but at all events those who have read it will believe that the charges which I leave unnoticed in the present article may safely be disregarded.

I agree with Dr. F. Vogel that, apart from internal evidence, virtually the only authority by which we can test the Commentaries is the correspondence of Cicero; and the opinion which I have already expressed of Dion Cassius[3346] is supported by the German scholar.[3347]

1. Dr. Vogel points to a passage in one of Cicero’s letters to Atticus,[3348] which, he says, has not yet received due attention or been correctly explained:--Britannici belli exitus exspectatur; constat enim aditus insulae esse #muratos#[3349] mirificis molibus; etiam illud iam cognitum est, neque argenti scripulum esse ullum in illa insula neque ullam spem praedae nisi ex mancipiis, &c. The following translation will be generally accepted as accurate:--‘The result of the British war is a source of anxiety. For it is notorious that the approaches to the island are ramparted by astonishing masses of cliff; and, besides, it is now known that there isn’t a pennyworth of silver in the island, nor any hope of loot except from slaves.’ But Dr. Vogel insists that the words Britannici belli exitus exspectatur ‘can only mean “the end of the British war is expected”, that is to say, it is expected that the war, being too hazardous and too unremunerative, will be entirely abandoned’.[3350] The letter in question was written, as Dr. Vogel remarks, before the 3rd of July, A.U.C. 700, that is to say, before the 8th of June of the Julian calendar.[3351] He goes on to say that for some time afterwards it must have continued an open question whether the expedition was not to be given up; for in a letter written after the 15th of July,[3352] Cicero says to his brother, ‘I wish that you could come at the time you arranged’ (cuperem te ad id tempus venire quod dixeras). But, says Dr. Vogel, Caesar says not a word to show that the proposed expedition was almost abandoned. Not a word directly: but Dr. Vogel believes that he can detect a significant hint in the Commentaries. Caesar relates that, before he sailed for Britain, he was obliged to march into the country of the Treveri, who were said to be preparing for rebellion. Soon after the legions entered the country, Indutiomarus, the leader of the malcontents, finding that resistance would, for the time being, be hopeless, sent envoys to Caesar to assure him of his loyalty and to make excuses for not having come in person to pay his respects. On which Caesar remarks, ‘Caesar was aware of his motive for saying this, and of the circumstances that deterred him from prosecuting the design which he had formed: still, in order to avoid having to waste the summer in the country of the Treveri after having made all his preparations for a campaign in Britain, he told Indutiomarus to present himself with two hundred hostages’ (Caesar, etsi intellegebat qua de causa ea dicerentur quaeque eum res ab instituto consilio deterreret, tamen, ne aestatem in Treveris consumere cogeretur omnibus ad Britannicum bellum rebus comparatis, Indutiomarum ad se cum ducentis obsidibus venire iussit[3353]). Now, says Dr. Vogel, although Caesar is silent as to his real motive, we may divine from the words ‘in order to avoid having to waste the summer in the country of the Treveri after having made all his preparations for a campaign in Britain’ that he hesitated for some time to carry out his resolve of making a second British expedition.[3354]

This charge rests upon a mistranslation.[3355] The words Britannici belli exitus exspectatur will certainly bear the interpretation which I have put upon them: why, then, suggest another interpretation, which, even if it could be got out of the Latin, would necessitate the assumption that Caesar did not know his own mind, and that, after he had kept his army busy for six months building six hundred ships, he told his staff that, all things considered, it would be better not to make use of them, and presently changed his mind again, and did make use of them? Again, in his quotation from Cicero’s letter to Quintus, Dr. Vogel leaves out the important part. The passage runs as follows:--Qua re suavitatis equidem nostrae fruendae causa cuperem te ad id tempus venire quod dixeras, sed illud malo tamen, quod putas magis #e re tua; magis# ... illa etiam magni aestimo, ἀμφιλαφίαν illam tuam et explicationem debitorum tuorum. Read Professor Tyrrell’s translation:--‘Wherefore I should indeed wish that you could come at the time you arranged, for the sake of our pleasure in each other’s society; but yet I desire more that you should do what you think your interests demand [and stay in the camp of Caesar]; still more do I value other considerations, your being in easy circumstances, and free from embarrassments.’ Professor Tyrrell goes on to point out that ‘the words printed in italics [#e re tua; magis#], or some such words, must, as Wesenburg suggested, have fallen out’,[3356] &c. As for Caesar’s words, ne aestatem in Treveris consumere cogeretur omnibus ad Britannicum bellum rebus comparatis, I confess that my powers of divination are not equal to those of Dr. Vogel. I am unable to infer from these words that Caesar hesitated for a moment to carry out his matured resolve.

Dr. Vogel then calls our attention to the well-known letter in which Cicero complained that he had had no news either from Quintus or from Caesar for more than fifty days, by which he meant that the last letters which he had received were dated more than fifty days before the time at which he was himself writing.[3357] Caesar’s last letter had been written on the 1st of September, and Quintus’s apparently a few days earlier.[3358] On the 25th of September they both wrote again.[3359] Dr. Vogel admits that from the 1st to the 25th Caesar and Quintus were engaged in operations against Cassivellaunus. But, he says, this does not satisfactorily explain the long break in the arrival of letters. In order to understand it, one must take into account a circumstance which Caesar himself relates,[3360] but the importance of which he minimizes, namely, that while he was campaigning on the north of the Thames, the four kings of Kent made an attack on his naval camp. It is true, says Dr. Vogel, that the attack was repulsed; but what did the Roman success amount to? All that Caesar can say for his troops is that they returned from their sortie unhurt. How long, owing to this outbreak in Kent, the communication between his army and his fleet was interrupted he tries to conceal by the meaningless expression, ‘while the operations above mentioned were going on in this district’ (Dum haec in his locis geruntur[3361]). As he goes on to say that Cassivellaunus, owing to the failure of the attack on the naval camp, sent envoys, one may read between the lines of his narrative that it was the action of the four kings which induced him to accept the embassy, even if he did not actually invite it.[3362]

To ‘read between the lines’ is always easy: the only difficulty is to avoid reading what is not there. The reason for ‘the complete break in the arrival of letters’ to which Dr. Vogel refers is intelligible to any reader who is not determined to convict Caesar of suppressio veri: the reason is either that, as I have elsewhere shown, Caesar was engaged, first in marching back from the coast into the interior, after a temporary visit to his naval camp, then in negotiations with Cassivellaunus, and, finally, in leading his army back to the sea; or, if he remained in his naval camp from the 1st to the 25th of September,[3363] that he saw no reason for writing to Cicero, or did not think it worth while to send a ship to Gaul for the sole purpose of conveying a letter. What Dr. Vogel calls ‘the meaningless expression’, Dum haec in his locis geruntur, is one of a class of expressions, all containing the words dum haec geruntur, which Caesar uses thirteen times:[3364] like our ‘meanwhile’, it is doubtless wanting in chronological precision, but it is not meaningless. Dr. Vogel asserts that ‘the complete break in the arrival of letters’ was due to ‘the outbreak in Kent’ on the part of the four kings. Yet he admits that the attack which the four kings made on the naval camp was repulsed; and Caesar adds that the garrison killed many of the enemy, and captured their leader, Lugotorix. In other words, their success was complete. All that they had to do was to beat off an attack, and this they effectually did. Even assuming that ‘the complete break in the arrival of letters’ was due to the action of the four kings, what then? It was not Caesar’s business to chronicle postal irregularities, but simply to describe his campaign.

To the very end of the narrative Dr. Vogel continues to read between the lines. ‘Thoroughly characteristic,’ he tells us, ‘is the way in which Caesar describes the results of the expedition. It is true that his account[3365] substantially tallies with what Cicero writes to Atticus,--“On the 24th of October I received letters from my brother Quintus and from Caesar, dated from the nearest coasts of Britain on the 25th of September. They had settled affairs in Britain, received hostages, and imposed tribute, though they had got no booty.” (A Quinto fratre et a Caesare accepti a. d. IX Kal. Nov. litteras, datas a littoribus Britanniae proximis a. d. VI Kal. Octobr. Confecta Britannia, obsidibus acceptis, nulla praeda, imperata tamen pecunia,[3366] &c.). But what about Cassivellaunus? How did Caesar balance accounts with him?’ Dr. Vogel reminds us that Caesar’s words have led Mommsen to believe that Cassivellaunus promised to pay tribute and to give hostages. But, says Dr. Vogel, Caesar nowhere says this: he only leads the reader to imagine it. What he says is this:--‘On receiving news of the action [namely, the repulse of the four kings], Cassivellaunus, who was greatly alarmed by the defection of the tribes, following the numerous disasters which he had sustained and the ravaging of his country, availed himself of the mediation of the Atrebatian, Commius, and sent envoys to Caesar, to propose surrender. Caesar had resolved to winter on the Continent because disturbances had broken out suddenly in Gaul: not much of the summer remained; and the enemy, as he knew, could easily spin out the time. Accordingly he ordered hostages to be given, and fixed the tribute which Britain was to pay annually to the Roman People, at the same time expressly forbidding Cassivellaunus to molest Mandubracius or the Trinovantes.’ (Cassivellaunus, hoc proelio nuntiato, tot detrimentis acceptis, vastatis finibus, maxime etiam permotus defectione civitatum, legatos per Atrebatem Commium de deditione ad Caesarem mittit. Caesar, cum constituisset hiemare in continenti propter repentinos Galliae motus neque multum aestatis superesset atque id facile extrahi posse intellegeret, obsides imperat et quid in annos singulos vectigalis populo Romano Britannia penderet constituit; interdicit atque imperat Cassivellauno ne Mandubracio neu Trinovantibus noceat.[3367]) Now, observes Dr. Vogel, in the first sentence Cassivellaunus is grammatically the subject, and in the last the object; but the reader involuntarily supposes him to be the object in the intermediate sentence as well. In other words, the reader takes for granted that Cassivellaunus was ordered to give hostages, though Caesar does not say so. Moreover, the first sentence, taken by itself, leads one to suspect Caesar’s good faith. For how came Caesar’s understrapper, Commius, to be with Cassivellaunus? Is it not clear that Caesar had sent Commius to Cassivellaunus as his envoy? In other words, that, whereas Caesar represents Cassivellaunus as having been driven by a series of reverses to offer submission, Caesar had in reality himself made overtures to Cassivellaunus?[3368]

If I were Caesar’s advocate I should merely reply that as Cicero states, on the testimony of his brother, hostages were given. But, as Caesar was undoubtedly not incapable of misrepresentation, Dr. Vogel’s suspicion is possibly not groundless. Caesar may have sent Commius to Cassivellaunus with an offer of terms; and if so, his narrative is so far misleading. On the other hand, Cassivellaunus may first have signified his willingness to submit; and Caesar may then have employed Commius as his agent.

Dr. Vogel’s general conclusion is that, although Caesar’s narrative is not expressly contradicted by Cicero’s letters, yet it was, from first to last, written for effect. Always literally true, it is often substantially false. His most effective weapon is the apparent clearness and candour of his style, which puts the reader off his guard, and prevents him from noticing how very ambiguous many of the statements are. He conceals essential facts and exaggerates the importance of trivial successes; and he prevents the unwary reader from noticing the slowness of his progress in Britain by inserting in the twelfth and the two following chapters of his Fifth Book a general description of the country and its inhabitants, from which point of view this otherwise very inartistic interpolation must be regarded as a masterpiece of ingenuity.

My general conclusion is that the charges which Dr. Vogel has brought against Caesar’s narrative for the most part break down, but that in the one instance which I have noted he may have detected a flaw.

2. Thomas Lewin, who was a writer of considerable ability, remarked, in regard to Caesar’s narrative of the events that immediately preceded his departure from Britain in 55 B.C., that ‘it is easy to see, notwithstanding the veil attempted to be thrown over the transaction, that he wanted only a plausible pretext for transporting himself and his army back to Gaul’.[3369]

Caesar’s account runs as follows:--‘On the same day the enemy sent envoys, who came to Caesar to sue for peace. Caesar ordered them to furnish twice as many hostages as before and take them across to the Continent; for the equinox was near, and, as his ships were unsound, he did not think it wise to risk a voyage in stormy weather. Taking advantage of favourable weather, he set sail,’ &c. (Eodem die legati ab hostibus missi ad Caesarem de pace venerunt. His Caesar numerum obsidum quem ante imperaverat duplicavit eosque in continentem adduci iussit, quod propinqua die aequinoctii infirmis navibus hiemi navigationem subiciendam non existimabat. Ipse idoneam tempestatem nactus ... naves solvit,[3370] &c.). Where is ‘the veil attempted to be thrown over the transaction’? Dion’s account,[3371] so far as it goes, confirms that of Caesar. ‘From Dion,’ remarks Lewin,[3372] ‘we learn that negotiations were opened by the intervention of some Morini who were friends of the Britons.’ Evidently he had not read Dion with due attention: the negotiations of which Dion speaks[3373] were opened immediately after Caesar had, in spite of British resistance, effected his landing; and Lewin forgets to add that the Britons induced the Morini to intervene. The truth of Caesar’s narrative is confirmed by the humorous frankness with which he avows that the Britons did not all obey his orders. After describing his return to Gaul, he says, ‘Two British tribes and no more sent hostages: the rest neglected to do so’ (Eo duae omnino civitates ex Britannia obsides miserunt, reliquae neglexerunt[3374]).

But I am not concerned to maintain that Caesar’s object was to tell the truth, the whole truth, and nothing but the truth.[3375] The most serious omission in his account of his British campaigns has passed almost unnoticed. In the section of this book which deals with the chronology of his operations I have demonstrated that either he passed over in silence a hurried temporary visit to his naval camp which he made just before the 1st of September (the 5th of August of the Julian calendar), 54 B.C., or, if he remained in the camp from that day until his final departure from Britain, his account of his negotiations with Cassivellaunus and of the return march of his army to the coast is misleading.[3376] Moreover, as I have pointed out in the seventh chapter,[3377] there is reason to suspect that he and his officers may have known more than he would admit about the connexion between the tides and the moon’s age. But he told the truth, so far as he could ascertain it, when he had no solid motive for falsification; and when he wrote the Commentaries on the Gallic War, he could generally afford to be true.

THE DISEMBARKATION OF THE ROMANS IN 55 B. C.

‘The men ... weighed down with their heavy cumbrous armour, had to leap down from the ships and keep their foothold in the waves,’ &c. (militibus autem ... magno et gravi onere armorum oppressis simul et de navibus desiliendum et in fluctibus consistendum, &c.[3378]) This passage has occasioned needless perplexity to commentators who forgot that the ships’ bows may have projected considerably, and also that when they were run aground they would necessarily have been buried for a considerable depth.[3379] Thus it would have been possible to jump into four feet six inches or five feet of water from the bow of a ship whose draught was a good deal more.

THE SITE OF CAESAR’S CAMP IN 55, AND OF HIS NAVAL CAMP IN 54 B. C.

I have proved in an earlier article that Caesar landed in 55 B.C. between Walmer Castle and Deal Castle, and in the following year between Deal Castle and Sandwich.[3380] Camden,[3381] who assumed that Caesar landed at the same place on both his expeditions, remarked that ‘for a considerable length under this shore [in the neighbourhood of Deal] are a number of heaps like banks which some suppose to have been blown up by the wind; but’, he added, ‘I rather take them for the fortifications or defences for ships which Caesar was ten days and nights throwing up.’ If Camden was referring to the sand-dunes between Sandown and Sandwich, it is of course unnecessary to discuss his conjecture. Hasted[3382] thought that the camp of 55 B.C. might be represented by ‘remains of entrenchments still visible’, (1) ‘close to the shore between Deal and Walmer Castle’, or (2) ‘within the country round Walmer Church’, or (3) ‘upon a rise ... between Deal and Upper Deal’. But these are mere guesses; and Professor Flinders Petrie, in his ‘Notes on Kentish Earthworks’,[3383] says nothing which can support either Hasted or Camden, except that ‘there is said to be a fosse around Walmer Church’. I can only add that nobody will learn anything by going to look for it. Dowker,[3384] who deluded himself into the belief that Caesar had landed near Sandwich in 55 as well as in 54 B.C., held that he had encamped in the first year between Sandwich and Worth, in the second on Richborough Hill. But Dowker himself maintained that Richborough Hill was isolated at spring tides; and would Caesar have omitted to mention that he had sailed into the Rutupian estuary? I am inclined to believe that he encamped in 54 B.C. on the gently rising ground near Worth.

Last year (1902) I spent two days in examining the coast between Kingsdown and Sandown.[3385] The conclusion to which I came was that in 55 B.C. Caesar must have encamped either on the plateau between Walmer Castle and Kingsdown, or, much more probably, on the rising ground north-west of the plateau which is now covered by part of the town of Walmer. On returning to London, I opened the second volume of Napoleon’s Histoire de Jules César, and found on pages 160 and 161 that Colonel Stoffel had adopted the latter alternative.

THE WAR-CHARIOTS OF THE BRITONS

‘Chariots,’ says Caesar,[3386] ‘are used in action in the following way. First of all the charioteers drive all over the field, the warriors hurling missiles; and generally they throw the enemy’s ranks into confusion by the mere terror inspired by their horses and the clatter of the wheels. As soon as they have penetrated between the troops of cavalry the warriors jump off the chariots and fight on foot. The drivers meanwhile gradually withdraw from the action, and range the cars in such a position that if the warriors are hard pressed by the enemy’s numbers, they may easily get back to them. Thus they exhibit in action the mobility of cavalry combined with the steadiness of infantry; and they become so expert from constant habit and practice that they will drive their horses at full gallop, keeping them well in hand, down a steep incline, check and turn them in an instant, run along the pole, stand on the yoke, and step backwards again to the car with the greatest nimbleness’ (Genus hoc est ex essedis pugnae. Primo per omnes partes perequitant et tela coiciunt atque ipso terrore equorum et strepitu rotarum ordines plerumque perturbant, et cum se inter equitum turmas insinuaverunt, ex essedis desiliunt et pedibus proeliantur. Aurigae interim paulatim ex proelio excedunt atque ita currus conlocant ut, si illi a multitudine hostium premantur, expeditum ad suos receptum habeant. Ita mobilitatem equitum, stabilitatem peditum in proeliis praestant, ac tantum usu cotidiano et exercitatione efficiunt uti in declivi ac praecipiti loco incitatos equos sustinere et brevi moderari ac flectere et per temonem percurrere et in iugo insistere et se inde in currus citissime recipere consuerint).

M. G. Lafaye[3387] gathers from this description that the object of the warrior in running along the pole was to jump down in order to throw his javelin and to avoid being impeded in his movements by the proximity of the driver. M. Lafaye assures us that certain coins represent warriors moving on to the poles of their chariots in order to hurl their javelins: but they do not represent them as about to jump down;[3388] and Caesar says that the warrior, after he had run along the pole, stepped back again on to the car.

Caesar[3389] tells us that, after he had crossed the Thames in 54 B.C., Cassivellaunus dismissed the whole of his forces except 4,000 essedarii. Most commentators have inferred, I think rightly, from this statement that Cassivellaunus had 4,000 chariots; but it seems possible that by 4,000 essedarii Caesar may have meant 2,000 warriors and 2,000 drivers.[3390] Napoleon the Third,[3391] on the other hand, assumes that there were ‘six essedarii par char’. It is unnecessary to make any assumptions; for, according to Diodorus Siculus,[3392] who derived his information from Posidonius, every Gallic chariot carried a driver and one warrior. Furthermore, a coin of the Hostilian family, which was struck between 49 and 46 B.C.,[3393] depicts a chariot drawn by two horses, and driven by a charioteer, who is accompanied by one warrior, armed with an oblong buckler.

Professor E. B. Tylor,[3394] referring to Pomponius Mela,[3395] Lucan,[3396] and Silius Italicus,[3397] argues that the Britons used chariots armed with scythes: Tacitus,[3398] who derived his information from Agricola, says that the British army which encountered the latter in the Grampians included covinnarii, who, according to Pomponius Mela, were warriors who fought in scythed chariots;[3399] and Jornandes[3400] says that the chariots of the Britons were armed with scythes: but if the Britons whom Caesar encountered had used such chariots, he would certainly have mentioned the fact;[3401] and no scythes are to be seen on the Roman coins which depict war-chariots. Moreover, of the numerous interments of warriors with chariots that have been discovered in the department of the Marne not one showed any traces of scythes;[3402] nor have any such traces ever been found in Great Britain.[3403]

I have remarked in my narrative of Caesar’s second invasion[3404] that very few chariot-burials have been found in this country. It is noteworthy that of the whole number--not more than a dozen--all but two were in Yorkshire, and not one in Scotland; and also that whereas in many of the Gallic interments the chariot was placed in the grave entire, only the wheels and other detached parts were buried in Britain. The most famous of these discoveries was made nearly a century ago in a barrow on Arras Farm, close to the road between Beverley and York. Here in a large round grave in the chalk was found the skeleton of a man, inclining from which, one on each side, were two wheels, each two feet eleven inches in diameter.[3405] ‘Under and adjoining to each wheel,’ writes Thurnam,[3406] ‘were the remains of the skeletons of two small horses, neither of them exceeding thirteen hands.’[3407]

In the paragraph in which Caesar describes the tactics of the charioteers he says that ‘as soon as they have penetrated between the troops of cavalry, the warriors jump off the chariots and fight on foot’ (cum se inter equitum turmas insinuaverunt, ex essedis desiliunt et pedibus proeliantur). The editors generally assume that equitum turmas means ‘the hostile troops of cavalry’; but von Göler,[3408] with whom Napoleon III[3409] agrees, rejects this view. He argues that in Caesar’s first campaign in Britain, in his account of which the passage in question occurs, the Britons were not opposed by any cavalry, for Caesar had none with him; and that the paragraph is not to be regarded simply as a general description of the tactics of the charioteers, but also as an explanation of the tactics which they had pursued in the combat described in the preceding paragraph. Moreover, he insists that if the warriors had jumped off their chariots when they had penetrated between troops of hostile cavalry, and had then allowed the drivers to turn round and move back, it would have been impossible for them to get on to the chariots again in case of need: hostile cavalry which allowed them to do this would have been worthless. ‘According to my interpretation,’ von Göler concludes, ‘we are to understand by et cum se inter equitum turmas insinuaverunt that the warriors had penetrated within the intervals of their own cavalry ... the moment of jumping down, always hazardous, was protected by their own cavalry, just as nowadays cavalry protect the limbering up and unlimbering of the horse-artillery associated with them.’ See also pp. 136-7, and Taf. vii, fig 7, of von Göler’s book, and pp. 688-91, infra.

THE OPERATIONS OF THE BRITONS DURING THE LAST FEW DAYS OF CAESAR’S FIRST EXPEDITION

After describing how he rescued the 7th legion, which had been sent out on a foraging expedition and surprised by a British force, Caesar tells us that he led this legion and the force with which he had marched to its assistance back to camp. ‘Meanwhile,’ he continues, ‘our people were all busy, and the Britons who were still in their districts moved off’ (dum haec geruntur, nostris omnibus occupatis, qui erant in agris reliqui discesserunt[3410]). The words ‘the Britons who were still in their districts’ (qui erant in agris reliqui) evidently refer back to two passages in the thirtieth and thirty-second chapters of the Fourth Book of the Commentaries. In the former we read that after the storm which wrecked several of Caesar’s ships the British chiefs who had disbanded their levies and come into the Roman camp ‘renewed their oaths of mutual fidelity, and began to move away one by one from the camp and to fetch their tribesmen secretly from the districts’ (itaque rursus coniuratione facta, paulatim ex castris discedere et suos clam ex agris deducere coeperunt). In the thirty-second chapter Caesar says that, just before the 7th legion was attacked, ‘some of the natives still remained in the districts’ (pars hominum in agris remaneret). Evidently, then, the meaning of the passage which I quoted at the beginning of this note is that during and after the attack on the 7th legion, and while the Roman soldiers were employed in various duties, those Britons who had not yet left their respective districts in order to rally round their leaders did so. However, the meaning which is obvious to the ordinary mind does not satisfy von Göler,[3411] who insists that the MS. reading, nostris omnibus occupatis, qui erant in agris reliqui discesserunt, yields no satisfactory sense, and offers in place of it one of his conjectural emendations:--(nostris omnibus occupatis,) #quae erant in agris relicta# (discesserunt). After a few moments of bewilderment the reader suddenly apprehends von Göler’s meaning. He fancied that occupatis meant ‘having been taken possession of’, and was ignorant that nostris omnibus in Caesarian Latin could not mean ‘all our belongings’; so he persuaded himself that Caesar intended to convey that the Britons, ‘having appropriated all our property, which had been left in the fields, made off’! ‘The Romans,’ he explains, ‘had not only not been able to convey into camp the corn which they had cut, but, on account of the surprise, they must even have abandoned their tools for cutting and gathering the corn.’

Comment is needless.

WHERE DID CAESAR ENCOUNTER THE BRITONS ON THE MORNING AFTER HIS SECOND LANDING IN BRITAIN?

‘Caesar disembarked the army, and chose a suitable spot for a camp. Having ascertained from prisoners where the enemy’s forces were posted, he marched against them about the third watch.... After a night march of about twelve miles he descried the enemy’s force. Advancing with their cavalry and chariots from higher ground towards a river,[3412] they attempted to check our men, and forced on an action. Beaten off by the cavalry, they fell back into the woods and occupied a well-fortified post of great natural strength, which they had apparently prepared for defence some time before with a view to war with their neighbours; for all the entrances were blocked by felled trees laid close together’ (Caesar exposito exercitu et loco castris idoneo capto, ubi ex captivis cognovit quo in loco hostium copiae consedissent ... de tertia vigilia ad hostes contendit ... ipse noctu progressus milia passuum circiter XII hostium copias conspicatus est. Illi equitatu atque essedis ad flumen progressi ex loco superiore nostros prohibere et proelium committere coeperunt. Repulsi ab equitatu se in silvas abdiderunt locum nacti egregie et natura et opere munitum, quem domestici belli, ut videbatur, causa iam ante praeparaverant: nam crebris arboribus succisis omnes introitus erant praeclusi[3413]).

Such is the description which Caesar gives of his first encounter with the Britons in 54 B.C. The question of the site is closely connected with the question of the place where he landed. I have proved that this place was between Walmer and Sandwich.[3414] It is clear, therefore, that Caesar could not have encountered the Britons either at Robertsbridge on the Rother, where Airy believed that he had discovered the battle-field,[3415] or at Wye on the Great Stour, the site adopted by Lewin.[3416] It is certain, however, that the battle was fought either on the Great Stour or on the Little Stour.

1. Napoleon the Third[3417] asserts that the river mentioned by Caesar was ‘unquestionably the Little Stour’; and he maintains that the left bank in the neighbourhood of Barham and Kingston corresponds with the description of the combat. The rising ground on this bank is not, he remarks, too uneven to have prevented war-chariots and cavalry from acting, and, as the text of the Commentaries requires, the Britons would have occupied a commanding position (locus superior) on the gentle slopes which terminate at the water’s edge. We may safely conclude, he adds, from Caesar’s narrative that the combat was unimportant, and that his cavalry crossed the river without difficulty.

Lewin[3418] dismisses this view with the remark that the Little Stour is ‘too insignificant to have been designated by Caesar as a river’; but the Emperor, who had anticipated this objection, replied that Caesar used the same word (flumen) to denote the Oze and the Ozerain,--the two rivulets which encompass Mont Auxois in the Côte-d’Or, where Vercingetorix made his final stand.[3419] No one, however, who has seen the Little Stour at Kingston, the Oze, and the Ozerain, will admit that the Kentish ‘nailbourne’ deserves to be treated as respectfully as the two Burgundian streams.

On the 1st of May, 1902, I walked along the Little Stour from Barham to Bekesbourne. There was not so much as a teaspoonful of water in the channel; and a policeman whom I met near Kingston told me that it had been dry for the last five years. On the other hand, an old labourer, who had lived in the valley for sixty-four years, remarked that he could remember a time when the rivulet often overflowed its banks: on the 18th of April of this year (1904) I myself saw it running past Barham with a strong stream; and a porter at Barham station said, ‘Yes, and it isn’t half as strong as it was a month ago.’ Moreover, Bryan Faussett,[3420] writing between 1767 and 1773, said that the Little Stour about a mile below Kingston was ‘seldom or never dry’; and Philippott,[3421] speaking of Bekesbourne, affirms that in the reign of Edward the Third ‘there was a small navigation out of the river of Stoure up to this place’.

There need be no difficulty, then, in believing that the Little Stour, in Caesar’s day, was a running stream, which he might perhaps have called a flumen, though it must not be forgotten that both in 55 B.C. and in the following year, the summer, at all events in Gaul, was exceptionally dry.[3422] Nevertheless, it is certain that the Britons did not encounter Caesar on the Little Stour. Consider the meaning of his words:--illi equitatu atque essedis ad flumen progressi ex loco superiore nostros prohibere et proelium committere coeperunt.[3423] At one time I thought that this passage meant, ‘Advancing towards a river with their cavalry and chariots, they attempted from their commanding position (ex loco superiore) to check our men,’ &c.: but a passage in the twenty-third chapter of the Second Book of the Gallic War--alia in parte diversae duae legiones, XI et VIII, profligatis Viromanduis, quibuscum erant congressae, ex loco superiore in ipsis fluminis ripis proeliabantur--in which the words ex loco superiore unquestionably belong to profligatis, leads me to believe that the former of the alternative translations which I have given--‘Advancing with their cavalry and chariots from higher ground towards a river, they attempted to check our men,’ &c.--is to be preferred. At all events the locus superior was not without tactical significance, and was either the left bank of the stream or high ground in close proximity to the left bank. Now between Barham and the northern end of Charlton Park, which is below Kingston, the depth of the channel of the Little Stour does not exceed 18 inches; and even at Bekesbourne it is only about two feet. Therefore, unless the depth of the channel was considerably greater in 54 B.C. than it is now, and unless the water flowed considerably below the level of the banks, the words locus superior could not have been applied to the bank itself anywhere between Barham and Bekesbourne.[3424] Moreover, although there are well-defined heights on the left bank between Barham and Bridge, the lowest slopes, except opposite Kingston and for a very short space on either side of it, are at a considerable distance from the channel. Assuming that Caesar crossed the Little Stour at or near Kingston, the Britons could have opposed him more effectually when he was ascending Barham Downs than by attempting to defend the passage of the rivulet. And, since he would in any case be obliged to cross the Stour itself, is it not obvious that they would have waited for him behind the river which might fairly be called an obstacle rather than on the banks of the streamlet which an active lad could have jumped over?

2. If Napoleon’s view is inadmissible, it is difficult to characterize that adopted by the Reverend Francis Vine.[3425] He assures us that Caesar ‘descried the British forces ... lining the crest of the hill (described in the ‘Commentaries’ as “superior locus”) from Garrington (near Littlebourne) on his right hand, to probably the part of Barham Downs opposite Bridge and Bishopsbourne on his left. This was the best position which the Britons could possibly have chosen for the purpose of arresting the progress of an army marching upon Caer Caint (Canterbury)’. In other words, Mr. Vine holds that the Britons awaited Caesar’s approach not on the further but on the nearer side of the river! ‘That the Britons were traditionally reported to have opposed Caesar’s progress before he reached the river, rather than after passing it, may,’ he says, ‘be inferred from the following passage from Pomponius Sabinus, out of Seneca: “And in the night marching twelve miles up into the country, Caesar finds out the Britons, who retreated as far as the river, but gave him battle there.”’ So we are to prefer the authority of Seneca to that of the general who fought the battle! Besides, Mr. Vine fails to see that the passage which he quotes (and which is not to be found in Seneca) need only mean that the Britons, when Caesar descried them, had retreated from the seashore to the banks of the river. Why Pomponius Sabinus, an Italian scholar who was born 1,470 years after Caesar died, should be summoned as a witness it is not easy to understand. The only inference which can be deduced from Caesar’s narrative--the only inference which has ever been deduced from it by any scholar--is that the Britons, when Caesar descried them, were on the further side of the river, and that they advanced to its left bank in order to dispute his passage.

Mr. Vine’s view[3426] of the route which Caesar followed appears to be partly based upon ‘traces of encampments which still remain’. But who made them? Certainly not these Britons, who, only a few hours before Caesar began his night march, had retreated from the coast into the interior in order to oppose his progress, and whose stronghold was not on the right but on the left bank of the river. Certainly not Caesar, who, in marching from the coast to encounter the Britons, made no camp at all. The tumuli which have been opened on Kingston Down were, as Roach Smith says, ‘neither more nor less than those of Saxons.’[3427] Mr. Vine[3428] asserts that ‘there were probably two large oblong castra [constructed by Caesar], the one extending along Barham Downs opposite Charlton, the other at the western extremity of the Downs extending over part of Bridge Hill, Bourne Park, and perhaps the grounds of Higham’. But without excavation it would be impossible to prove that any ‘castra’ had been erected on Barham Downs by Caesar; and without going to this trouble any man who can understand the Commentaries may conclude that he certainly did not construct two.

3. Lyon, the author of the History of the Town and Port of Dover,[3429] maintains that the combat took place near Littlebourne, about two miles lower down than Bekesbourne; but this place is on the road from Sandwich to Canterbury, and barely nine miles from the former. If Caesar marched along the line of this road, he must have encountered the Britons on the Great Stour near Canterbury. Assuming that he marched from the neighbourhood of Deal, Lyon’s view might perhaps be defended if there were any reason to believe that on the left bank of the Little Stour near Littlebourne there was a stronghold;[3430] but in a former article[3431] I have given cogent reasons for believing that in 54 B.C. Caesar encamped some miles north of Deal.

It may, then, be regarded as morally certain that the river on which the Britons encountered Caesar was the Great Stour. It has, indeed, been objected that the least distance of the Great Stour from Deal is not twelve, but fifteen miles; but while this argument may be valid against the theory that Caesar landed in 54 B.C. at Deal, the position that he defeated the Britons on the Great Stour remains unshaken.

4. The Reverend R. C. Jenkins[3432] holds that the scene of the encounter was Chilham, about six miles above Canterbury, on the Ashford road. ‘The only obstacle,’ he pleads, ‘is the increased distance, which is sixteen, instead of twelve miles ... a difference which the loss of a single stroke might account for [the scribe being supposed to have written XII instead of XVI], if, indeed, it is not sufficiently explained by the possible miscalculation of the time of the march ... everything else falls into perfect harmony with the narrative--the high wooded ground at the back, the steep banks, the wide and rapid stream,’ &c. And again, ‘Let us remember that the journey was during the night, when the ground would be rapidly passed over, and the actual distance would be less apparent ... the space traversed is only described as “milia passuum circiter duodecim”, and even then the position of the enemy was merely discerned afar off ... here we have ancient mounds and earthworks, which give silent testimony to the fact that Chilham was a military position of the highest importance even during the British period.’[3433]

These arguments have no weight. When numbers attested by the consensus of the MSS. are not manifestly wrong, we have no right to distort them into agreement with our own preferences. The distance, in a straight line, from the nearest point on which Caesar could have encamped, if he had landed near Sandwich,[3434] to Chilham, is about 20 Roman miles; from the place where he would have encamped if he had landed near Deal,[3435] approximately 16; and the actual distance which he would have had to march is of course considerably longer. Why he should have marched more rapidly by night than by day it would be difficult to explain;[3436] and the remark that ‘the position of the enemy was merely discerned afar off’ is a pure invention. If ‘we have ancient mounds and earthworks’ at Chilham, they prove nothing about Caesar;[3437] and we have them also at other places near the course of the river. Besides, why should Caesar have made a forced march in order to cross the Stour at Chilham when, by making an ordinary march of 12 miles, he could have crossed it near Canterbury?[3438]

Our search for the site is now confined within narrow limits. Below Fordwich, the Stour, in Caesar’s time, would certainly have been impassable in the face of an enemy; for it flowed through a broad morass.[3439] Between Canterbury and the bridge above Sturry the river is virtually flush with its banks. It appears to me, then, that Caesar must have crossed it either between Fordwich and Sturry, or in the neighbourhood of Thanington, just above Canterbury, or possibly at Canterbury itself. Above Canterbury it flows through nearly level meadows: its width is about 15 or 16 yards: the banks are about 2 feet high; and the depth of the water at present is apparently about 2 feet. The bottom at the bridge above Thanington is sandy and gravelly. Opposite this point and at a distance of, say, 600 yards, the ground begins to rise into wooded heights. Opposite Thanington, and east of it, the heights are considerably nearer the river; but they gradually sink as they approach Canterbury.

At Sturry the lower slopes of the low hills which extend along the northern side of the valley approach very close to the river, say to within 100 yards; but opposite Fordwich they are much further away. Just below the mill at Sturry the Stour is from 15 to 20 yards wide: the banks are 3 or 4 feet above the water: the average depth is apparently from about 18 inches to 2 feet; and the bottom is sand mixed with stones. At Fordwich the depth of the water, as seen from the bridge, is about 5 feet: the banks are 4 or 5 feet above the water; and the bottom from this point downwards is mud.

5. The eminent geographer, Major Rennell,[3440] believed that Caesar crossed at the place where ‘the western road intersects the course of the Stour’; but he gave no reasons for preferring this site to Fordwich, Sturry, or Thanington.

6. Von Göler[3441] and Guest[3442] maintain that the battle was fought at or near Sturry; and Heller[3443] appears to agree with them. So does Roach Smith,[3444] who, at Guest’s request, made ‘a survey of the vicinity of Grove Ferry’, which survey, I presume, extended as far up the river as Sturry. ‘There,’ he says, ‘I found the river with the high bank ... the woods and oppidum,’[3445] &c. Napoleon[3446] argues that the banks at Sturry are so steep that the Roman cavalry could not have forced a passage without great difficulty, whereas it would appear from Caesar’s account that they crossed easily; and also that Sturry is 15, not 12 miles from Deal. The latter objection may be disposed of at once. Napoleon himself maintains that Caesar’s anchorage in 54 B.C. was ‘some kilometres’ north of the spot where he had landed in the preceding year;[3447] and this spot he rightly fixes between Walmer and Deal. Therefore the place from which Caesar descried the Britons just before they advanced to the bank of the stream was not 12 miles from Deal, but, as I have shown in a previous article,[3448] from a point in the neighbourhood of Sandwich. Certainly it would have been more difficult to force the passage of the Stour at Sturry than of the Little Stour near Kingston; but, as we have seen, the latter would practically have presented no obstacle at all. Airy,[3449] who holds that if the Britons had been posted on the Stour at all, ‘Caesar would have crossed at the sound ground of Canterbury or above it,’ observes that ‘the place had been selected by the Britons as a defensive post at least two days previously, and may therefore be presumed to have had the qualifications necessary for a defensive post, namely that it could not be turned, and that enemies could attack it in front only at a disadvantage’; and, remarking that ‘there can scarcely be a doubt that Canterbury existed then as an important town’, he adds that ‘of this there is no mention in Caesar’. But why should Caesar have mentioned Canterbury? It was not a strategical point: there was nothing to be gained by attacking it except perhaps a little plunder; and anything worth plundering would certainly have been removed into the stronghold which he did attack. ‘The place [on the banks of the river] had been selected [or, at all events, occupied] as a defensive post’ not two days but one day previously; and, generally speaking, to select a defensive post on a river which cannot be turned is impossible.[3450] The stronghold to which the Britons retreated was probably, as Mr. George Payne holds,[3451] and as I have shown in my narrative,[3452] the British oppidum in Bigberry (or Bigbury) woods, about a mile and a half west of Canterbury, of which traces still exist; and it seems most likely that the passage of the river took place at some point between Canterbury and Thanington.

CAESAR’S EARLIER OPERATIONS IN 54 B. C. (B. G., V, 9-11)

Caesar’s account of the events which occurred on the day after his first encounter with the Britons in 54 B.C. has been interpreted in several different ways; and yet his narrative is so clear that one would have thought it impossible to misunderstand it. After describing the first encounter, he proceeds, ‘Caesar, however, forbade them [the legionaries] to pursue the fugitives far, partly because he had no knowledge of the ground, partly because the day was far spent and he wished to have time for entrenching his camp. On the following morning he sent a light division of infantry and cavalry, in three columns, to pursue the fugitives. They had advanced a considerable distance, the rear-guard being still in sight, when some troopers from Quintus Atrius came to Caesar with the news that there had been a great storm on the preceding night, and that almost all the ships had been damaged and gone ashore.... On receiving this information, Caesar recalled the legions and cavalry, ordering them to defend themselves as they marched, and went back himself to the ships.... Although it involved great trouble and labour, he decided that the best plan would be to have all the ships hauled up and connected with the camp by one entrenchment. About ten days were spent in these operations, the troops not suspending work even in the night.’ (Sed eos fugientes longius Caesar prosequi vetuit, et quod loci naturam ignorabat, et quod magna parte diei consumpta munitioni castrorum tempus relinqui volebat. Postridie eius diei mane tripertito milites equitesque in expeditionem misit, ut eos qui fugerant persequerentur. His aliquantum itineris progressis, cum iam extremi essent in prospectu, equites a Q. Atrio ad Caesarem venerunt, qui nuntiarent superiore nocte maxima coorta tempestate prope omnes naves adflictas atque in litus eiectas esse.... His rebus cognitis Caesar legiones equitatumque revocari atque in itinere resistere iubet, ipse ad naves revertitur ... Ipse, etsi res erat multae operae ac laboris, tamen commodissimum esse statuit omnes naves subduci et cum castris una munitione coniungi. In his rebus circiter dies X consumit ne nocturnis quidem temporibus ad laborem militum intermissis.[3453])

The camp for the construction of which Caesar wished to allow time was of course quite distinct from the one for which, on the previous day, he had selected a site near his landing-place,[3454] and was to be in the neighbourhood of the place where he had beaten the enemy, that is to say, twelve Roman miles or more from the sea.[3455] Next morning he dispatched three flying columns in pursuit of the fugitives; and his rear-guard was just visible when the messengers arrived with the news of the shipwreck. He sent a galloper to recall the pursuing columns; and, as their retreat would naturally encourage the fugitives to rally and return to the attack, they were to offer the best resistance they could as they marched back to the coast.

C. Schneider[3456] misunderstands the passage as far as misunderstanding is possible. He holds that cum iam extremi essent in prospectu means ‘when the rear of the enemy was just in sight’. But Caesar was not with any of the three pursuing columns, for he tells us that he sent them in pursuit: therefore, if Schneider were right, we should be forced to believe that he learned afterwards that the enemy’s rear had been just in sight. But why resort to this fanciful explanation, seeing that Caesar could by no possibility have ascertained that the rear of the fugitives was just visible to his troops at the very moment when he himself, separated from the troops, received the news of the shipwreck; and, further, that extremi must grammatically refer to his, that is to say, the pursuing columns? It has been objected that if extremi meant the Roman rear-guard, iam would not have its proper meaning, ‘already’ (ἤδη). But iam of course often means ‘by this time’, or ‘at length’ (nearly tandem). In the passage in question its sense may, I think, be illustrated by a clause in B. G., vii, 83, § 7,--cum iam meridies adpropinquare videretur (‘at length when it was evidently near noon’). Caesar meant to say that his pursuing columns had advanced so far that by the time when the news of the shipwreck reached him the rear-guard only was in sight. Again, Schneider takes the words in itinere resistere iubet to mean that Caesar ordered his troops ‘to stop in their march and halt where they were’. But, as Long[3457] points out, ‘if this is so, he returned to his ships himself (“ipse”), leaving his men in the country doing nothing for ten days. But he tells us that his soldiers were employed in making his “munitio”, and therefore they must have come back to the camp.’

So far Long is quite right; but he too makes a great mistake. He actually believes[3458] that the words munitioni castrorum tempus relinqui volebat (‘he wished to leave time for entrenching his camp’) refer to the camp near the seashore, for which he had selected a site immediately after the disembarkation. In other words, he believes that Caesar compelled his troops, who had already made a night-march of twelve miles, beaten the enemy on the banks of the Stour, stormed the stronghold to which the beaten enemy had retreated, and then pursued the fugitives for some distance, to march back all the way to the camp by the seashore, entrench it, and next morning march back again twelve miles or more into the interior, and then start in pursuit of the fugitives! The legionaries were hardy fellows; but if they did all this, they must have been as breathless as the reader will be when he has got to the end of the preceding sentence. And what had the fugitives been about? If Long is right, they must have got such a start that to pursue them would have been a wild-goose chase indeed. Needless to say, what Caesar meant was simply this:--when the Britons had been dislodged from their stronghold, he would not allow his troops to pursue them far because he wished to leave time for the construction of the temporary camp in which he intended to pass the night.

Von Göler[3459] needlessly quarrels with the text of the Commentaries, and makes matters worse by an absurd emendation. ‘The reading legiones equitatumque revocari atque in itinere resistere iubet (B. G., v, 11, § 1),’ he says, ‘is obviously corrupt. Atque requires that the following clause should involve a climax’ (Steigerung). But so it does:--Caesar ‘orders the legions and cavalry to be recalled, and [not only to come back but also] to defend themselves upon the march’. In this literal translation a climax is as evident as in the emendation, to be quoted presently, which von Göler offers; and any one who is not familiar with the Commentaries will find in Meusel’s Lexicon Caesarianum abundant evidence that atque in the passage which von Göler brands as corrupt is in accord with Caesar’s usus loquendi.

‘As finally,’ continues von Göler, ‘one may conclude with certainty from the later substance of the narrative that all the legions returned to the shore, but the necessary express mention of the march thither is nowhere to be found in the Commentaries, I believe that the passage originally ran as follows:--legiones equitatumque revocari atque in itinere persistere iubet ad naves, ipse revertitur.’ If we may ‘conclude with certainty’ that ‘the legions returned’, why should the ‘express mention’ of their return be necessary? Heller[3460] has taken the trouble to show that the text is above suspicion.

CAESAR’S SECOND COMBAT WITH THE BRITONS IN 54 B. C.

There is a passage in Caesar’s account of his second combat with the Britons in 54 B.C. which has greatly exercised the minds of the commentators; and editors have put upon it an interpretation which soldiers will not accept. The passage runs as follows:--‘Throughout this peculiar combat, which was fought in full view of every one and actually in front of the camp, it was clear that our infantry, owing to the weight of their armour, were ill fitted to engage an enemy of this kind; for they could not pursue him when he retreated, and they dared not abandon their regular formation: it was clear too that the cavalry fought at great risk, because the enemy generally fell back on purpose, and, after drawing our men a little distance away from the legions, leaped down from their chariots and fought on foot with the odds in their favour. [On the other hand, the mode in which their cavalry fought exposed the Romans, alike in retreat and in pursuit, to an exactly similar danger.[3461]]’ (Toto hoc in qenere pugnae cum sub oculis omnium ac pro castris dimicaretur, intellectum est nostros propter gravitatem armorum, quod neque insequi cedentes possent neque ab signis discedere auderent, minus aptos esse ad huius generis hostem, equites autem magno cum periculo proelio dimicare, propterea quod illi etiam consulto plerumque cederent et, cum paulum ab legionibus nostros removissent, ex essedis desilirent et pedibus dispari proelio contenderent. [Equestris autem proelii ratio et cedentibus et insequentibus par atque idem periculum inferebat[3462]]). It is the last sentence which has caused all the trouble.

1. ‘After describing the difficulties of the infantry,’ says Long,[3463] ‘Caesar explains the danger to which the cavalry was exposed, when they pursued the “essedarii”, for the Britanni quitted their “esseda” and fought on foot among the Roman “equites”.... This was an unequal kind of fighting (“dispar proelium”) for the Roman “equites”. Caesar adds, that on the other hand (“autem”) the British mode of fighting from the “esseda” (“equestris proelii ratio”) was equally dangerous to his cavalry and legions in the pursuit and the retreat. This is Schneider’s explanation, and I believe that it is right. The Britanni had no cavalry: they had only “essedarii”, to whom Caesar (iv, 33) applies the term “perequitant”. It follows that “equites hostium essedariique” (c. 15) are no more than the “essedarii” (iv, 24) [in other words that Caesar, who was not a prolix writer, used four words when one would have sufficed.] In iv, 34, however, Caesar says, “peditatus equitatusque.”’ Yes, he does; and the reader will please note the significance of this admission.

The assumption upon which Long’s explanation rests, namely, that ‘the Britanni had no cavalry’, is unsupported by any evidence, and is wholly inadmissible. Caesar says that in his first campaign in Britain a party of Roman foragers were surrounded by equitatu atque essedis.[3464] If these words do not mean ‘cavalry and chariots’, what do they mean? Warriors on horseback are depicted on ancient British coins;[3465] and the Britons certainly had cavalry as well as chariots in A.D. 61[3466] and in the time of Agricola.[3467] Besides, as von Göler points out,[3468] the charioteers could not have fought with effect unless they had been supported by cavalry. And when Long says that, according to Caesar, ‘the British mode of fighting from the “esseda” ... was equally dangerous to his cavalry and legions in the pursuit and the retreat,’ his explanation refutes itself. Caesar himself says that the legions could not pursue the charioteers; and, on the other hand, it is evident that they were not pursued by the charioteers.

2. According to von Göler,[3469] ‘by cedentibus et insequentibus the chariot-fighters only can be meant’ (Unter den ‘cedentibus et insequentibus’ können nur die Wagenstreiter verstanden werden); and, placing a comma after par, he translates the passage thus:--‘The mode of fighting of the [British] cavalry corresponded with that of their charioteers, whether they retreated or pursued, and brought the Roman cavalry into precisely the same danger’ (Der Gefechtsmechanismus der [britischen] Reiterei war aber der Fechtweise ihrer Wagenstreiter, ob sie wichen oder verfolgten, entsprechend [par] und brachte die römische Reiterei gerade in jenes nachtheilige Verhältniss[3470]). But no Latin scholar would admit that a comma could be placed after par, which would of course make it necessary to supply the word erat.

3. According to Dittenberger-Kraner,[3471] the words equestris autem proelii ratio, if they are genuine, can only mean, in opposition to ex essedis ... contenderent, the mode of fighting of the charioteers, which, while it was dangerous to the Romans when they pursued, was no less dangerous to them when they retreated. I confess that I cannot understand this comment.

4. Doberenz-Dinter[3472] also insist upon the opposition between equestris autem proelii ratio and ex essedis ... contenderent, and maintain that the former refers to ‘a regular cavalry combat (on the part of the Romans)’,--ein regelrechtes Reitertreffen (von Seiten der Römer). Now in the passage which ends with the words ex essedis ... contenderent Caesar describes a combat between the Roman cavalry and the British charioteers, who were supported by cavalry. Therefore the editors (if I have succeeded in grasping their meaning) suppose that the ‘regular cavalry combat’ to which they allude was fought between the Roman cavalry and the British cavalry alone. This must, I think, be what they intend to convey; for, according to Mr. Peskett,[3473] who may have followed them, ‘Caesar means that when the British and Roman cavalry were engaged, the danger was equalized [equestris autem proelii ratio et cedentibus et insequentibus par atque idem periculum inferebat], whereas when they used chariots the Britons were at an advantage.’ But if Caesar simply meant that the British cavalry and the Roman in the (hypothetical) ‘regular cavalry combat’ were each exposed to the same danger, what is the point of the words et cedentibus et insequentibus? Surely retreating cavalry are in greater danger than the cavalry which has forced them to retreat and is pursuing them! I do not see how Mr. Peskett’s explanation can be got out of the Latin. And who will believe that the Britons would have used their cavalry alone when, by associating them with chariots, they ‘were at an advantage’?

5. It might possibly be suggested that the words equestris ... inferebat refer to a cavalry combat between the Romans and the Britons, distinct from the combat between Roman cavalry and the British charioteers, in which the British cavalry, like that of the Germans,[3474] were associated with light infantry. But there is no evidence that the Britons had cavalry of this kind; and, as we have seen, it is certain that their cavalry acted in support of their charioteers.

6. Köchly and Rüstow[3475] offer the following explanation:--‘As the enemy were also supported by their cavalry, our men [the Romans] were exposed to the same danger, whether they advanced or retreated’ (Da aber der Feind auch die Unterstützung seiner Reiterei hatte, so war für die unsrige die Gefahr immer dieselbe, mochte sie vorgehen oder zurückgehen).

7. Napoleon the Third[3476] explains the matter thus:--‘Un désavantage plus grand encore existait pour les cavaliers. Les Bretons, par une fuite simulée, les attiraient loin des légions, et alors, sautant à bas de leurs chars, engageaient à pied une lutte inégale; car, toujours soutenus par leur cavalerie, ils étaient aussi dangereux dans l’attaque que dans la défense.’ These words, which appear to be virtually identical in sense with those of Köchly and Rüstow, undoubtedly give an accurate account of what took place. Apparently Köchly, Rüstow, and Napoleon do not take par atque idem periculum as meaning ‘an exactly similar danger’ (to that which Caesar described in the preceding sentence, Toto hoc ... contenderent), but as meaning that the danger which beset the Romans was the same whether they pursued or retreated. Now the Roman infantry, as Caesar says, did not pursue; obviously therefore et cedentibus et insequentibus can only refer to the Roman cavalry. I suppose then that what Köchly, Rüstow, and Napoleon meant was this:--if the Roman cavalry pursued the Britons, they were attacked by the charioteers, who jumped off their cars and fought as infantry: as soon as they retreated they were pursued by the British cavalry, and if they turned to bay the charioteers had time to mount their cars again, come up, and engage them anew. If this was what Caesar meant by equestris ... inferebat, his language was not lucid.

The words equestris proelii ratio, if they were really written by Caesar, must refer either to a combat between the Roman and the British cavalry or to a combat between the Roman cavalry and the combined British charioteers and cavalry. There is, as we have seen, no reason to suppose that a purely cavalry combat took place; and if it did, the idea that the Roman cavalry was as much in danger when it pursued as when it retreated is absurd. If we accept the other alternative, the meaning of the passage must be either (as Köchly, Rüstow, and Napoleon explain) ‘On the other hand, the mode in which the British cavalry fought (in co-operation with the charioteers) exposed the Romans, alike in retreat and in pursuit, to exactly the same danger’; or ‘In fact the nature of the combat of horse [that is to say, the combat between the Roman cavalry and the combined British charioteers and cavalry] exposed the Romans’, &c. In the former case autem would be an adversative, in the latter merely a connecting particle. The passage is not in the editio princeps of the Commentaries, and is bracketed in Meusel’s edition; and perhaps it is an interpolation.

THE COMBAT BETWEEN TREBONIUS AND THE BRITONS

‘At midday Caesar having sent three legions and all his cavalry on a foraging expedition under one of his generals, Gaius Trebonius, they [the enemy] suddenly swooped down from all points on the foragers, not hesitating to attack the ordered ranks of the legions’ (Sed meridie cum Caesar pabulandi causa tres legiones atque omnem equitatum cum C. Trebonio legato misisset, repente ex omnibus partibus [hostes] ad pabulatores advolaverunt, sic uti ab signis legionibusque non absisterent[3477]). To a plain man these words are perfectly intelligible; and no military commentator, so far as I know, has ever found any difficulty in them: but Kraner[3478] must needs rewrite the last clause. This is what he makes of it:--sicubi ab signis legionibusque absisterent. So, according to Kraner, the enemy attacked the foragers at every point where they were separated from the legions. The unpractical fellow fails to perceive that, as the foragers could not forage while they were in their ranks, there was no point where they were not separated from the legions. The legions, or rather a due proportion of the cohorts which composed them, were there to protect the foragers; and of course what Caesar means is that the enemy, flushed with their easy success in driving off the foragers and compelling them to rejoin their respective cohorts, had the temerity to attack the cohorts themselves.

WHERE DID CAESAR CROSS THE THAMES?

The only indications which Caesar gives as to the place where he crossed the Thames are these. At an early stage of his narrative he tells us that ‘the chief command and the general direction of the campaign had been entrusted by common consent to Cassivellaunus, whose territories are separated from those of the maritime tribes by a river called the Thames, about eighty miles [or seventy-three English miles] from the sea’ (summa imperii bellique administrandi communi consilio permissa Cassivellauno, cuius fines a maritimis civitatibus flumen dividit, quod appellatur Tamesis, a mari circiter milia passuum LXXX[3479]). In a later chapter he describes his passage of the river. ‘Having ascertained the enemy’s plans, Caesar led his army to the Thames, into the territories of Cassivellaunus. The river can only be forded at one spot, and there with difficulty. On reaching this place, he observed that the enemy were drawn up in great force near the opposite bank of the river. The bank was fenced by sharp stakes planted along its edge; and similar stakes were fixed under water, and concealed by the river. Having learned these facts from prisoners and deserters, Caesar sent his cavalry on in front, and ordered the legions to follow them speedily; but the men advanced with such swiftness and dash, though they only had their heads above water, that the enemy, unable to withstand the combined onset of infantry and cavalry, quitted the bank and fled.’ (Caesar, cognito consilio eorum, ad flumen Tamesim in fines Cassivellauni exercitum duxit; quod flumen uno omnino loco pedibus, atque hoc aegre, transiri potest. Eo cum venisset, animadvertit ad alteram fluminis ripam magnas esse copias hostium instructas. Ripa autem erat acutis sudibus praefixisque munita, eiusdemque generis sub aqua defixae sudes flumine tegebantur. Eis rebus cognitis a captivis perfugisque, Caesar praemisso equitatu confestim legiones subsequi iussit. Sed ea celeritate atque eo impetu milites ierunt, cum capite solo ex aqua extarent, ut hostes impetum legionum atque equitum sustinere non possent ripasque dimitterent ac se fugae mandarent).[3480]

When Caesar says that the Thames was only fordable at one spot, he evidently means that there was only one ford available for his purpose, that is to say, only one by which he could cross the river into the territories of Cassivellaunus. When he says that these territories were separated from those of the maritime tribes by the Thames, about 80 miles from the sea, he means, I suppose, that it was 80 miles from the place where he landed to the eastern frontier of Cassivellaunus; but, according to some commentators, he reckoned the distance from the place where he landed to the point at which he crossed the river. Heller[3481] insists that his statement of the distance was based upon hearsay: I am inclined to believe that it was a rough estimate based upon the number of the marches which he made.

1. The view which has gained most adherents is that Caesar crossed the Thames at ‘Coway Stakes’, about a furlong west of Walton Bridge. It has been said that this view is based upon a ‘tradition which has certainly prevailed for many ages’;[3482] but I can find no evidence that the tradition existed before the publication of Camden’s Britannia; and I believe that it was created by him. Camden[3483] referred, in support of his conjecture, to a well-known passage in Bede;[3484] but Bede did not mention Coway Stakes at all. He merely said that, at the point of the river where the enemy had planted their stakes, the stakes were still to be seen, and that they were as thick as a man’s thigh and cased with lead. S. Gale[3485] affirms, in support of Camden’s view, that there is ‘a large Roman encampment ... about a mile and a half distant from the ford’. This camp, which is on St. George’s Hill, is really about two miles and three-quarters[3486] from the place where the ford is assumed to have been. It is not Roman, but British; but even if it were Roman, the fact would throw no light upon the question which we are discussing unless the camp could be proved to have been constructed by Caesar. Daines Barrington, an antiquary of the eighteenth century, made an attempt, which, at first sight, would appear completely successful, to demolish Camden’s theory.[3487] A fisherman of Shepperton, he tells us, rowed him across the river in the line along which the stakes had been planted. He illustrated his story by a sketch-plan, which shows that the stakes were at right angles with the river-banks; and, he argues, ‘such stakes could not possibly have obstructed the passage of an army.’ In reply it has been asserted that ‘the line of the ford was not transversely straight across the stream, but formed a curve nearly in a semi-circle, so that ... the stakes must have twice intercepted the passage’;[3488] but who will believe that Caesar would have attempted the passage of the river, which, as he tells us, was barely practicable, by a ford so intricate as this? A more effective answer might be based upon the fact, attested by Lord Wolseley,[3489] that fords ‘almost always run diagonally across the river’: if the ford by which Caesar crossed was no exception, stakes planted in the direction indicated by Barrington would obviously have obstructed the passage. Dr. Guest has made another and most ingenious attempt to remove the difficulty.[3490] He believes that the stakes which impeded Caesar’s advance had not been planted for the purpose of stopping him, but had existed for many years. ‘I think,’ he says, ‘the stakes formed part of what may be called a fortified ford, and were distributed so as to stop all transit ... save along a narrow passage, which would bring the passenger directly under the command of the watch, stationed on the northern bank to ... receive the toll. The shallow at Coway was probably of considerable extent, and through its whole length must have extended the line of stakes which Caesar observed on the northern bank. But there must also have been two other lines of stakes across the river to ... define the passage.’[3491] The stakes to which Barrington referred were, in Guest’s opinion, the remains of these. ‘The remaining portion,’ he continues, ‘of the shallow was, no doubt, covered with the short stakes that were “concealed by the river” ... that such was really the disposition of the stakes may, I think, be gathered, not only from the reports of the fishermen, but also from Caesar’s narrative. When he saw the Britons ranged along the northern bank with the stakes in front of them, he ordered the cavalry to pass the river, and the legions to follow them. How could either cavalry or infantry cross the river if the stakes were ranged as our antiquaries assume them to have been? ... Besides, what were the Britons doing while the Roman soldiers were removing the stakes? ... Caesar says not a word about taking the Britons in flank, nor about removing the stakes.’[3492]

It may, perhaps, be thought that Guest has succeeded in showing that the particular objection which Barrington raised against Camden’s theory is not necessarily insuperable. If, however, the stakes to which Barrington referred were the remains of those which had served, on Guest’s theory, to ‘define the passage’, the ford in question was an exception to the rule that fords ‘almost always run diagonally across the river’. Again, Guest argues that because ‘Caesar says not a word about ... removing the stakes’, therefore the Romans did not remove them. But it is needless to insist that they must have removed them, unless at the particular part of the ‘shallow’ by which they crossed there were none to remove. Guest, therefore, is obliged to make the incredible assumption that the Britons, having left intact the stakes which ‘defined the passage’, and having thus pointed out to the Romans the best way of crossing the river, were so obliging as to plant no stakes to bar this passage either in the bed of the river or on the bank, while they planted them everywhere else.

Guest undertakes to prove not only that Coway Stakes may, but that it must have been the spot at which Caesar crossed the Thames. ‘From the Coway Stakes,’ he says, ‘the ground rises gradually for about three miles, and then dips almost precipitously into the valley of the Wey. On the top of the hill [St. George’s Hill] is an ancient British stronghold which commands the whole valley, and as the valley certainly belonged to the Atrebates, I infer that ... this people constructed the fortress. Aubrey tells us that “a trench” went from this fortress to Walton, and gave that village its name. A dyke still runs from the ramparts towards Walton. I have traced it for more than one-third of the distance, and I have no doubt that it once reached the village ... The ditch is towards the river. For what purpose could this dyke have been made? The only object for which I can conceive it was made, was to bar progress along the trackway which led from the Coway Stakes eastward to the maritime states. If such were its object, we have another strong proof that the great means of access to the country of Cassivellaunus was at the spot where Camden placed it.’[3493] Nor is this all. ‘From Hurleyford,’ says Guest, ‘to the sea, a distance of nearly 100 miles, taking into account the windings of the river, there is but one place on the banks of the Thames bearing a name which indicates a ford over it.[3494] This solitary place is Halliford, at the Coway Stakes. Caesar says there was but one ford on the Thames--meaning, of course, the lower Thames, with which alone he was acquainted, and we now have but one place on its banks, the name of which points to the existence of a ford. Our topography is in perfect agreement with his statement; and, to my mind, this coincidence is almost decisive of the question.’[3495] We shall see.

2. Colonel Stoffel, by whose advice Napoleon was mainly guided in his attempts to solve the topographical problems presented by the Commentaries, was informed by the Thames boatmen whom he consulted that between Shepperton and London there were eight or nine fords, the most favourable of which was at Sunbury; and here accordingly the Emperor concluded that Caesar had crossed the river. Still he (or the colonel) was sagacious enough to doubt. ‘La seule chose,’ he wrote, ‘qui nous paraisse evidente, c’est que l’armee romaine n’a point passe en aval de Teddington.’[3496]

But even this conclusion rested upon a rotten foundation. The Emperor, as Guest[3497] observes, ‘reasons from the present to the past without taking any note of the changes that have occurred during 2,000 years. In the time of Caesar the river ran ... to the sea uninterruptedly. Now, from Teddington westward it is a canal, crossed every two or three miles by weirs and locks ... as the lock [at Teddington] did not exist in the time of Caesar, any inference drawn from the fact that the tide now ends there, is beside the question.’ And, to quote Guest again, ‘the shallow at Sunbury is a mere consequence of Sunbury weir. Remove the weir, and Caesar’s ford at Sunbury would be swept away in a twelvemonth by the natural scour of the river.’ Guest[3498] thought it probable that ‘when the river was in its natural state, spring-tides ran up the river eight or nine miles further [than they do now]--in other words, to Coway: and that the deposit which they now leave at Teddington then contributed to form the shallow over which Caesar passed.’ Mr. F. C. J. Spurrell, on the other hand, has given reasons for believing that ‘the estuary did not reach so far west as at the present day’.[3499]

3. Von Goler[3500] peremptorily decides that the ford was ‘undoubtedly at Kingston’. An English engineer, he tells us, informed him that the depth of the water there was only from 3½ to 4½ feet; and he remarks further that Kingston is just 80 Roman miles from the sea. If the engineer whom he consulted could have informed him that the depth of the water at Kingston was the same in 54 B.C. as at the time when he sounded it, his information would have been more valuable. Assuming that the tide in Caesar’s time, as now, did not flow beyond Teddington, W. H. Black may have been justified in saying that Kingston ‘presents geographically the most favourable place for crossing the Thames’ to an invader coming from the neighbourhood of Deal.[3501] But we must not assume that Caesar crossed the Thames at the place which was ‘geographically the most favourable’: he could not pick and choose; he had to cross where there was a ford, and there was only one.

It has been shown that the Thames has, at various times and at certain states of the tide, been fordable at Westminster,[3502] at Chelsea,[3503] at Old Brentford,[3504] and at Petersham;[3505] and the claims of these places have been advocated by zealous antiquaries: but, with one exception, which I shall notice presently, I do not recommend students of the Commentaries to read what they have written. As Guest points out, ‘the name of Brentford had no reference to a ford over the Thames; it certainly designated the ford over the Brent by which the Roman road from London to Staines crossed the latter river.’[3506] Moreover, the fact that the Thames has occasionally been crossed on foot at various points near London proves nothing: these cases, as Guest truly says, were exceptional, and were recorded because they were exceptional; ‘Caesar,’ he concludes, ‘knew the river in its natural state, and had ... adequate means of acquiring knowledge ... he tells us distinctly that the Thames was passable on foot only in one place.’[3507]

The claims of Brentford have, however, recently been advocated by Mr. Montagu Sharpe in a pamphlet which contains some real evidence. From information supplied by Messrs. W. S. Bunting and W. Benell of the Thames Conservancy, and by Conservancy Inspector G. J. Rough, he shows that a line of stakes, of which some still remain ‘for about 400 yards below Isleworth Ferry,’ extended thirty years ago for about a mile up the river from ‘Old England’, opposite the mouth of the Brent; and that ‘no other ancient stakes have been discovered in the lower river during dredging operations’.[3508]

Except Coway Stakes and Brentford, there is no spot in the Thames valley for the identification of which with the scene of Caesar’s exploit a shred of real argument has been advanced. Guest’s arguments are of very unequal value; but the one which he founds upon the name ‘Halliford’ is worth considering, though it would be going too far to say that it is ‘almost decisive of the question’. The claim which Mr. Sharpe makes for his own discovery rests upon somewhat better grounds. [See Addenda, p. 742.]

CAESAR’S PASSAGE OF THE THAMES

The excessively laconic chapter in which Caesar describes how he crossed the Thames in the face of a British force seems at first sight hard to explain. He tells us that ‘the river can only be forded at one spot, and there with difficulty’. ‘On reaching this place,’ he continues, ‘he observed that the enemy were drawn up in great force near the opposite bank of the river. The bank was fenced by sharp stakes planted along its edge; and similar stakes were fixed under water and concealed by the river. Having learned these facts from prisoners and deserters, Caesar sent his cavalry on in front, and ordered the legions to follow them speedily; but the men advanced with such swiftness and dash, though they only had their heads above water, that the enemy, unable to withstand the combined onset of infantry and cavalry, quitted the bank and fled’ (quod flumen uno omnino loco pedibus, atque hoc aegre, transiri potest. Eo cum venisset, animadvertit ad alteram fluminis ripam magnas esse copias hostium instructas. Ripa autem erat acutis sudibus praefixisque munita, eiusdemque generis sub aqua defixae sudes flumine tegebantur. His rebus cognitis a captivis perfugisque Caesar praemisso equitatu confestim legiones subsequi iussit. Sed ea celeritate atque eo impetu milites ierunt, cum capite solo ex aqua extarent, ut hostes impetum legionum atque equitum sustinere non possent ripasque dimitterent ac se fugae mandarent).[3509]

Von Göler[3510] explains the passage as follows:--‘he first sent his cavalry against the enemy, making them swim across the adjacent part of the river, which, though deeper, was not barricaded, and by their attack so effectually protected the passage of the infantry, which was begun immediately afterwards, that they gained time to remove the stakes’; and in a note[3511] he says, ‘Only in this way can the expression praemisso equitatu confestim legiones subsequi iussit be understood, for the cavalry could by no possibility swim across or disregard the stakes or the palisade.’ This explanation, as far as it goes, is substantially identical with that of Turpin de Crissé,[3512] who supposes, further, that while part of the Roman infantry cut down the stakes in the river, they were supported by their comrades in the rear, who discharged missiles against the enemy, and, he might have added, by the auxiliary troops,--slingers and archers. Napoleon the Third[3513] hardly differs from his predecessors in suggesting that the cavalry were sent ‘à une certaine distance en amont ou en aval’. Mr. A. G. Peskett[3514] objects to Napoleon’s explanation on the ground that it involves a mistranslation of the word praemittit, which, he insists, ‘must mean that Caesar sent the cavalry across the river, ordering the infantry to follow them.’ Mr. Peskett evidently means that the infantry crossed directly in the rear of the cavalry. But, as any soldier would tell Mr. Peskett, the operation which he supposes would have been absolutely impossible; and, moreover, his rigidly literal interpretation of the word praemisso, which is not shared by that competent Caesarian scholar, C. Schneider, is irreconcilable with the word sed, which opens the next sentence. This word, as Schneider[3515] remarks, is intended to show that the infantry, in their ardour, outvied the cavalry, and crossed the river before them. The use of the word praemisso may surely be defended if the cavalry were sent into the water before the infantry. Similarly it will be evident to any one who carefully reads the last two sentences in the twenty-fifth chapter of Caesar’s Fourth Book[3516] that the word subsecuti cannot there mean ‘following directly in rear’. For in that case ‘the nearest ships’ (proximis navibus), the troops in which jumped into the sea and followed (subsecuti) their comrades, would have been drawn up in a second line behind the other ships. If so, being in deeper water, they evidently could not have been run aground; and the soldiers who descended from them would have been drowned.

THE SITE OF CASSIVELLAUNUS’S STRONGHOLD

The indications which Caesar gives as to the geographical position of the stronghold of Cassivellaunus are of the vaguest kind. After describing his passage from the southern to the northern bank of the Thames, which brought him into the territory of Cassivellaunus, he gives the following account of his operations:--

‘Cassivellaunus, abandoning, as we have remarked above,[3517] all thoughts of regular combat, disbanded all his forces, except some four thousand charioteers, watched our line of march, and, moving a little away from the track, concealed himself in impenetrable wooded spots, and removed the cattle and inhabitants from the open country into the woods in those districts through which he had learned that we intended to march. Whenever our cavalry made a bold dash into the country to plunder and devastate, he sent his charioteers out of the woods (for he was familiar with every track and path), engaged our cavalry to their great peril, and by the terror which he thus inspired prevented them from moving far afield. Caesar had now no choice but to forbid them to move out of touch with the column of infantry, and, by ravaging the country and burning villages, to injure the enemy as far as the legionaries’ powers of endurance would allow.

‘Meanwhile the Trinovantes--about the strongest tribe in that part of the country--sent envoys to Caesar, promising to surrender and obey his commands. Mandubracius, a young chief of this tribe, whose father had been their king and had been put to death by Cassivellaunus, but who had saved his own life by flight, had gone to the Continent to join Caesar, and thrown himself upon his protection. The Trinovantes begged Caesar to protect Mandubracius from harm at the hands of Cassivellaunus and to send him to rule over his own people with full powers. Caesar sent Mandubracius, but ordered them to furnish forty hostages and grain for his army. They promptly obeyed his commands, sending hostages to the number required and also the grain.

‘As the Trinovantes had been granted protection and immunity from all injury on the part of the soldiers, the Cenimagni, Segontiaci, Ancalites, Bibroci, and Cassi sent embassies to Caesar and surrendered. He learned from the envoys that the stronghold of Cassivellaunus, which was protected by woods and marshes, was not far off, and that a considerable number of men and of cattle had assembled in it. The Britons apply the name of stronghold to any woodland spot, difficult of access and fortified with a rampart and trench, to which they are in the habit of resorting in order to escape a hostile raid. Caesar marched to the spot indicated with his legions, and found that the place was of great natural strength and well fortified: nevertheless he proceeded to assault it on two sides. The enemy stood their ground a short time, but could not sustain the onset of our infantry, and fled precipitately from another part of the stronghold.’ (Cassivellaunus, ut supra demonstravimus, omni deposita spe contentionis, dimissis amplioribus copiis, milibus circiter IIII essedariorum relictis, itinera nostra servabat paulumque ex via excedebat locisque impeditis ac silvestribus sese occultabat, atque iis regionibus quibus nos iter facturos cognoverat pecora atque homines ex agris in silvas compellebat et, cum equitatus noster liberius praedandi vastandique causa se in agros eiecerat, omnibus viis notis semitisque essedarios ex silvis emittebat et magno cum periculo nostrorum equitum cum iis confligebat atque hoc metu latius vagari prohibebat. Relinquebatur ut neque longius ab agmine legionum discedi Caesar pateretur, et tantum agris vastandis incendiisque faciendis hostibus noceretur quantum labore atque itinere legionarii milites efficere poterant.

Interim Trinovantes, prope firmissima earum regionum civitas, ex qua Mandubracius adulescens Caesaris fidem secutus ad eum in continentem [Galliam] venerat, cuius pater in ea civitate regnum obtinuerat interfectusque erat a Cassivellauno, ipse fuga mortem vitaverat, legatos ad Caesarem mittunt pollicenturque sese ei dedituros atque imperata facturos: petunt ut Mandubracium ab iniuria Cassivellauni defendat atque in civitatem mittat qui praesit imperiumque obtineat. Iis Caesar imperat obsides XL frumentumque exercitui, Mandubraciumque ad eos mittit. Illi imperata celeriter fecerunt, obsides ad numerum frumentumque miserunt.

Trinovantibus defensis atque ab omni militum iniuria prohibitis, Centimagni, Segtontiaci, Ancalites, Bibroci, Cassi legationibus missis sese Caesari dedunt. Ab iis cognoscit non longe ex eo loco oppidum Cassivellaumi abesse silvis paludibusque munitum, quo satis magnus hominum pecorisque numerus convenerit. Oppidum autem Britanni vocant, cum silvas impeditas vallo atque fossa munierunt, quo incursionis hostium vitandae causa convenire consuerunt. Eo proficiscitur cum legionibus; locum reperit egregi natura atque opere munitum; tamen hunc duabus ex partibus oppugnare contedit. Hostes paulisper morati militum nostrorum impetum non tulerunt seseque alia ex parte oppidi eiecerunt.[3518])

1. Many commentators have identified the stronghold with Verulam, or Verulamium, which was situated immediately west of St. Albans.[3519] The arguments which can be adduced in support of this view are that marshes might have been formed by the river Ver; that Verulamium, under Tasciovanus, who began to reign not later than 30 B.C., was the chief town of Catuvellauni; and that the territory of the Catuvellauni belonged to Cassivellaunus.

2. Others point to Cassiobury in Hertfordshire.[3520] Cassiobury, they argue, evidently preserves the name of the Cassi,[3521] who were as evidently subject to Cassivellaunus.

3. Von Göler[3522] remarks that, ‘judging from the configuration and nature of the terrain,’ the oppidium ‘may be the hill lying on the south-western side of Wendover’. May be, or may be not; for Caesar’s vague description of ‘the configuration and nature of the terrain’ would apply to other sites as well.

4. The most interesting theory is that of T. Lewin,[3523] who maintains that the oppidium was no other that London, that is to say, the settlement which many writers believe to have existed long before the Roman conquest, in the neighbourhood of Ludgate Hill. As Cassivellaunus, he argues, had conquered the Trinovantes,[3524] ‘whose western border was the Lea,’ we may assume that his dominions extended westward from that river, and comprised Middlesex and Hertford. As Caesar says that he prohibited his soldiers from plundering the Trinovantes, it is clear that, after crossing the Thames, he marched into Essex. There he learned that the oppidum of Cassivellaunus was not far off; and ‘this situation answers to London’. Moreover British London, which was situated on the rising ground between Ludgate and Dowgate, and protected on the south by the marshes of the Thames, on the west by the marshes of the river Flete, and on the east by the marshes of the river Wallbrook, was just such a stronghold as Caesar described.

Who would not accept such an attractive theory if he could only give rein to his imagination? But unhappily the very existence, in 54 B.C., of British London is matter of inference and conjecture, however reasonable.[3525] And is it reasonable to assume that if the stronghold which Caesar captured had been situated on the banks of the Thames he would have neglected to mention the fact?

5. W. H. Black[3526] argues that the oppidum was most probably somewhere in ‘the woody lands about Pinner, Harrow, and Cashiobury Park.’

6. The Reverend H. Jenkins[3527] maintains that it was in Essex; for, he argues, in order to fulfil the compact which he had made with the Trinovantes, Caesar’s ‘chief object, after he had crossed the Thames, must have been to lead his army into Essex’. Certainly Caesar’s army, or a part of it, must have entered the country of the Trinovantes; for, as we have seen, he would not allow his soldiers to plunder that people. Jenkins’s theory is demolished by Caesar’s statement that the stronghold belonged to Cassivellaunus, which shows that it was not in the territory of the Trinovantes. And since, immediately after the sentence in which he tells us that he prohibited his soldiers from plundering the Trinovantes, he goes on to say that the stronghold of Cassivellaunus was not far off, it is fair to conclude that it was near the common frontier of the Trinovantes and of Cassivellaunus.[3528] Of the places which fulfil this condition more can be said for Verulam than for any other; but its identity with the oppidum in question has not been proved.

DID LONDINIUM EXIST IN CAESAR’S TIME?

The earliest mention of London occurs in the Annals[3529] of Tacitus, who, describing the events of the year 61, speaks of it as a busy centre of commerce. It has been argued[3530] that a settlement existed there before the Roman conquest of Britain, because the name Londinium is Celtic. Lewin[3531] maintains further that if London had been founded by the Romans, it would have been a strong military post, whereas in 61, eighteen years after the invasion of Britain by Aulus Plautius, it was attacked by the Iceni and Trinovantes because it was defenceless and wealthy.[3532] ‘It must,’ he insists, ‘have attained to this height of prosperity, not under the Romans, who did not patronise it, but by the silent progress of trade, a work that could not ... have been accomplished in ... 19 years.’ ‘We know,’ he continues, referring to Dion Cassius,[3533] ‘that Camulodunum was a flourishing British city before the ... time of Plautius, and, if so, London, which enjoyed far greater advantages, must also have been a British city.’ Lewin holds that this city stood upon the hill between the river Flete and the Wallbrook. On the west was _Lud_gate, the name of which is Celtic: on the east Dowgate,--‘a corruption of the Celtic Dwrgate or water-gate.’ The river Flete, or Fleet, entered the Thames just below the site of Blackfriars Bridge; while at Dowgate, about 1,000 yards to the east, was the mouth of the Wallbrook.[3534]

Mr. W. J. Loftie[3535] finds the site of British London on the western side of the Wallbrook. In his History of London,[3536] however, he affirmed that the British settlement stood ‘on the eastern hill [that is to say, on the gently rising ground east of the Wallbrook], if anywhere’. Canon Isaac Taylor[3537] asserts that the British hill-fort was ‘formed by Tower Hill, Cornhill, and Ludgate Hill, and effectually protected by the Thames ... the Fleet ... the great fen of Moorfields and Finsbury’, &c. Seeing that it has been proved that ‘the great fen’ did not exist in Roman times,[3538] and that the very existence of the British hill-fort can as yet only be inferred, it is plainly useless to attempt to determine its site.

Dr. Guest[3539] holds that ‘the notion ... that a British town preceded the Roman camp [of Aulus Plautius] has no foundation ... and is inconsistent with all we know of the early geography of this part of Britain.’ ‘Such town,’ he adds, ‘could not have belonged to the Trinobantes, for it lay beyond their natural limits, nor to the settled district of the Catuvellauni, for then Caesar’s statement that the Thames divided their country from the maritime states “about eighty miles from the sea” would be grossly inaccurate.’ I cannot see the force of these arguments. ‘The notion that a British town preceded the Roman camp’ has a foundation,--the solid foundation of etymology. London is indisputably a Celtic name;[3540] and if London had been founded by the Romans, why should it have had a purely Celtic name[3541] at all, and why should its Celtic name have outlasted Augusta,--the name which the Romans gave to their London? Assuming that Caesar’s statement is to be interpreted in the sense which Dr. Guest attaches to it, the distance by road of British London from Sandwich, in the neighbourhood of which Caesar landed,[3542] is 67 English miles, or nearly 73 Roman miles. Is the difference between 73 and ‘about 80’ so great as to justify the use of the words ‘grossly inaccurate’?

More recently John Richard Green has endeavoured to disprove the existence of British London.[3543] ‘Much,’ he says, ‘has been made of its name, but “Llyn-dyn” ... is as likely to be the designation of a spot as of a town on it. An almost conclusive proof, however, that no such town existed west of the [river] Fleet may be drawn from the line of the old British road from Kent (the predecessor of the Watling Street), which, instead of crossing the river as in Roman and later times at the point marked by London Bridge, passed, according to Higden, to a point opposite Westminster ... (Loftie, “Roman London”, Archaeological Journal, volume xxxiv, page 165) ... the rise of such a town [Roman London] is the best explanation of the later change in the line of this road.’

‘According to Higden!’ According to the monk of the fourteenth century who wrote the Polychronicon! And Higden does not so much as mention ‘the old British road from Kent’,--a road the very existence of which can only be conjectured. What he says is that ‘Watlingstrete’ crossed the Thames west of Westminster;[3544] and Mr. Loftie, to whom Green appeals, affirms that he is ‘driven to the conclusion that there was a British town’.[3545] Accepting the statement of Higden, he argues that the ‘Watlingstrete’ which is said to have crossed the Thames west of Westminster was a pre-Roman road and followed the line of Park Lane and Edgware Road. I will only add that, having failed to discover any paper worth reading about the direction of Watling Street in that part of its course which passed through or by Roman London, I consulted Professor Haverfield. ‘I know nothing satisfactory,’ he replied, ‘about the line of Watling Street, and nothing to suggest that it existed before A.D. 43.’

The very large number of palaeolithic implements which have been found in London and its environs prove that in the earliest times it was a centre of population;[3546] but it would hardly be safe to infer from the discoveries of bronze and iron tools and weapons and of British coins[3547] that the Romans found a town on the site. If there was such a town, it certainly had little political importance; for while numerous British coins issued from the mints of Verulamium and Camulodunum, not one has been discovered which bears the name of Londinium.[3548] Nevertheless it may reasonably be affirmed that London existed before the Roman conquest: first, because the same advantages that attracted the traders of Rome would also have commended themselves to those of Britain; and secondly, I repeat, because it is improbable that a Celtic name would have been given to a town which the Romans had built upon a virgin site.[3549]

THE JULIAN CALENDAR AND THE CHRONOLOGY OF CAESAR’S INVASIONS OF BRITAIN

I. The chronology of Caesar’s first invasion of Britain is simple enough, so far as it can be ascertained, and requires no knowledge of the intricacies of the Roman calendar. I have shown in an earlier article that the disembarkation took place on the 26th or 27th of August, 55 B.C.[3550] After describing the storm which occurred on the night of August 30-31, and the consequent loss of many of his ships, Caesar goes on to say that the Britons endeavoured to protract the war by cutting off his supplies, and that he had corn brought in daily from the open country into his camp, and ordered the materials necessary for the repair of those ships which were only partially damaged to be fetched from the Continent. While the ships were being repaired an attack was made upon the 7th legion, which was engaged in cutting corn. This attack evidently took place several days after the 31st of August; for the field in which the legion was reaping was the only one accessible from the camp in which the corn had not been cut. The day on which the legion was attacked was followed by several days of stormy weather, during which all military operations were suspended. At the end of this time the Britons attacked the camp unsuccessfully, and on the same day sued for peace, which Caesar granted on condition of their giving hostages. Instead of waiting for the arrival of the hostages he ordered the British chiefs to send them over to the Continent, ‘because the equinox was at hand, and he did not think it wise to expose his unseaworthy ships to a voyage in stormy weather.’[3551] The equinox occurred on the 26th of September.[3552] Our data, then, are as follows:--the attack on the 7th legion occurred several days after the 31st of August; and Caesar returned to Gaul several days after the attack on the 7th legion, but before--probably several days before--the 26th of September.[3553] Let us say that he returned about the middle of the month. Napoleon, to whom indefiniteness was an abomination, fixed the 12th of September as the date.[3554]

II. If we were to believe certain writers of high reputation, we should be deterred from attempting to fix any dates for the second of Caesar’s expeditions; for our principal sources of information are dates mentioned by Cicero in his letters; and, as everybody knows, the Roman calendar, before Caesar’s reform, was often in disagreement with the Julian calendar. The writers of the article CALENDARIUM in Smith’s Dictionary of Greek and Roman Antiquities[3555] affirm that ‘it is very difficult or rather quite impossible to determine the actual dates which correspond to the nominal dates of any events before the Julian reform of the calendar’. But for the period comprised between the beginning of the year 696 (58 B.C.), just after Caesar became Governor of Gaul, and his reform of the calendar, which took effect in 709 (45 B.C.), this view is quite incorrect. Every date in Cicero’s correspondence which relates to the subject of this essay can be reduced, if not with absolute precision, at all events with a possible error of not more than one day, to its corresponding date in the Julian calendar. First of all, however, we must find out the nature of the reform which Caesar initiated, and understand the chronological disturbances which made reform necessary.

Every scholar knows that, after the period of the Decemvirate, the Roman year consisted of 355 days only, and that every other year an additional month, consisting alternately of 22 and 23 days, was, or rather ought to have been, intercalated after the 23rd of February. January, April, June, August, September, November, and December each contained 29 days; February 28; and March, May, July, and October 31. If I may remind the general reader of what he learned at school, the first day of every month was called the Kalends; the fifth day of January, February, April, June, August, September, November, and December was called the Nones; and the thirteenth day of each of those months was called the Ides. But in March, May, July, and October the Nones were on the seventh, and the Ides on the fifteenth day. In mentioning dates the Romans described any given day as occurring so many days before the Kalends, Nones, or Ides, as the case might be; and in so doing they adopted the inclusive method of reckoning. Thus the last day of December was called ‘the day before the Kalends of January’, pridie Kalendas Ianuarias, or shortly prid. Kal. Ian. But the 27th of December was called, not the third, but the fourth day before the Kalends of January, ante diem quartum Kalendas Ianuarias, or shortly a. d. IV Kal. Ian.

To return to the month which was or ought to have been intercalated every other year. As the ordinary year contained 355 days, and the solar year was believed to contain 365¼ days, it is obvious that to intercalate a month of 22 and 23 days alternately every other year was to make an excessive correction, the excess amounting to 4 days in every period of 4 years. Macrobius[3556] tells us that in order to remedy this error, 24 days were omitted from every twenty-fourth year. For various reasons, however, this regulation, if indeed it ever really took effect,[3557] was not always properly carried out; and accordingly in the year of the city 563, or 191 B.C., a reform was introduced, the college of pontiffs being authorized to intercalate or to omit intercalations at their discretion.[3558] But this innovation, as we learn from Cicero, Censorinus, and other writers, only made matters worse. Speaking of the pontiffs, Censorinus complains that ‘most of them, either from hatred or from favour, to cut short or to extend the tenure of office, or that a farmer of the public revenue might gain or lose more by the length of the year, by intercalating more or less at their pleasure, deliberately made worse what had been entrusted to them to set right’.[3559]

III. In 708 (46 B.C.), which is generally called ‘the year of confusion’, Caesar intercalated a certain number of days, in order to bring the calendar into harmony with the solar year before inaugurating the reformed calendar in 709. It is expressly stated by Asconius, whose testimony is unanimously accepted, that there was an intercalary month in 702;[3560] and it is admitted by all chronologists that there was no other intercalary month in the seven years between 700 and 708.[3561] There is no doubt that the Kalends of January, 709, corresponded either with the 1st or the 2nd of January, 45 B.C. It is clear, therefore, that when we have ascertained how many days were intercalated in 708, we shall be able, by reckoning backwards, to ascertain the correspondence of any given date in the summer of 700, the year of Caesar’s second expedition, with the Julian calendar,--with a possible error of one day only. This error will be removed if we can ascertain whether the Kalends of January, 709, corresponded with the 1st or the 2nd of January, 45 B.C.

All German scholars who have written upon Roman chronology are agreed that ‘the year of confusion’ contained 445 days, in other words, that 90 days, amounting to four ordinary intercalary months, were intercalated; and they hold that these 90 days were actually composed of three months, namely the Mercedonius, which, in the ordinary course, should in that year have been intercalated immediately after the 23rd of February, and two extraordinary months, amounting to 67 days, which were intercalated between the last day of November and the 1st of December. This view is supported in every detail by Censorinus,[3562] who wrote about 240 A.D. The principal dissentients are De La Nauze, Napoleon the Third and his collaborator, the famous astronomer, Le Verrier, who held that 67 days only were intercalated in 708, and Colonel Stoffel, who, in his Histoire de Jules César--Guerre civile, published in 1887, reaffirmed the same view, but who does not appear to have informed himself of what any of the Germans, except Ideler, had written on the subject. Moreover, the theory of Napoleon, Le Verrier, and Colonel Stoffel is frequently referred to by scholars in terms which imply that they regard the question as still open. The reason which Le Verrier[3563] gives is that in 700 Caesar re-embarked his troops for the return voyage from Britain to Gaul ‘because the equinox was at hand’ (quod aequinoctium suberat[3564]); that the equinox actually took place on the 26th of September of the Julian calendar; and that Caesar informed Cicero on the sixth day before the Kalends of October (which corresponded, on Le Verrier’s theory, with the 21st of September) that he was on the point of bringing back the army.[3565] He remarks that, on the theory of Ideler (who, like all other German scholars, held that 90 days were intercalated in 708), the sixth day before the Kalends of October, 700, would have corresponded with the 30th of August, 54 B.C.;[3566] and he argues that this theory must be wrong because Caesar would not have troubled himself about the approach of the equinox 27 days before it occurred. He also remarks that, although the view that 90 days were intercalated in 708 is supported by Suetonius[3567] and Censorinus,[3568] Dion Cassius[3569] affirmed that 67 days were intercalated, and that Dion expressly added that other writers had asserted that a greater number had been intercalated, but that his own statement was true; and he insists that Dion derived his information from authentic sources.[3570]

It will be shown presently that if Le Verrier was right in his interpretation of Dion’s words, Dion made a mistake. G. F. Unger[3571] holds that he misunderstood the authority from whom he borrowed his statement, and ‘who, like Suetonius, unquestionably regarded the month intercalated in February as an ordinary intercalary month’. Von Göler,[3572] on the other hand, holds that Dion was quite right; that he was referring, not to the ordinary intercalary month, but only to the two extraordinary months intercalated between November and December, which amounted to 67 days; and therefore that his statement tallies with that of Censorinus. Be this, however, as it may, it is absolutely certain that more than 67 days were intercalated in 708. For on the 16th of May, 705, Cicero wrote from Cumae, ‘At present the equinox is delaying us, which has been very stormy’ (Nunc quidem aequinoctium nos moratur, quod valde perturbatum erat[3573]). Now, on Le Verrier’s theory, the 16th of May, 705, corresponded with the 16th of April, 49 B.C. of the Julian calendar; but the equinox occurred on the 24th of March. In order to dispose of this difficulty, Le Verrier is obliged to have recourse to flagrant special pleading: ‘l’équinoxe,’ he says, ‘était passé depuis 21 jours,[3574] et les troubles atmosphériques pouvaient durer encore. Était-ce d’ailleurs autre chose qu’un prétexte pour Cicéron’?[3575] A very thin pretext, one would say,--a pretext which a man of Cicero’s intelligence was hardly likely to resort to.[3576]

But this is not the only proof of the unsoundness of Le Verrier’s theory. The argument which he bases upon Caesar’s words, quod aequinoctium suberat, shows that he completely misunderstands both Caesar’s narrative and the letter of Cicero to which he appeals. Cicero writes, ‘On the 24th of October I received letters from my brother Quintus and from Caesar, dated from the nearest coasts of Britain on the 25th[3577] of September ... they were on the point of bringing back the army’ (a Quinto fratre et a Caesare accept a. d. IX Kal. Nov. litteras, datas a litoribus Britanniae proximis a. d. VI Kal. Octobr. ... exercitum [e] Britannia reportabant[3578]). The letters to which Cicero refers were written, according to Napoleon and Le Verrier, on the 21st of September of the Julian calendar; and, they triumphantly remark, this tallies with Caesar’s statement, that he hurried on his departure from Britain ‘because the equinox was at hand’.[3579] But there is one fact which they overlook. Caesar had a large number of hostages and prisoners, and some of his ships had been lost. He was therefore obliged to transport the hostages, prisoners, and troops to Gaul in two successive trips. Only a few of the ships which made the first trip ever returned to Britain, almost all the rest having been driven back to Gaul by adverse winds. Caesar tells us that he waited for these ships ‘a considerable time (aliquamdiu) in vain’. Then, and not till then, he made the second trip, being obliged to crowd the soldiers into the few ships that he had, and not daring to wait any longer, ‘because the equinox was at hand.’[3580] The letters to which Cicero refers were evidently written before the first trip; for neither Caesar nor Quintus had written to Cicero for a long time;[3581] and they would naturally have dispatched their letters by one of the ships which made the first voyage. It is clear, therefore, that the letters in question were written, not on the 21st of September of the reformed calendar, but ‘a considerable time’ before that day. In fine, if Le Verrier and Napoleon were right, the letters would have been written not from Britain but from Gaul, and Cicero would have written not reportabant (‘they were on the point of bringing back’) but reportaverant (‘they had brought back’[3582]).

Another fact, which Napoleon, Le Verrier, and Colonel Stoffel appear to have overlooked, alone proves that not 67 days only but 90 days were intercalated in that year. The word nundinae is familiar to many readers of Cicero’s letters. To quote the authors of the article NUNDINAE in Smith’s Dictionary of Greek and Roman Antiquities,[3583] ‘the Romans had a system of eight-day weeks, which, like our seven-day weeks, ran on from one month to another and from one year to another without breaking.’ Every eighth day was a market-day, and was called nundinae. Thus, if the 1st of January was a market-day, the next was the 9th, the next the 17th, and so on. We learn from Dion Cassius[3584] that the Kalends of January, 702, was a market-day; the same writer says that, in order to prevent the Kalends of January, 714, from falling on a market-day--a coincidence which was regarded as ill-omened--a day was intercalated extraordinarily in 713;[3585] and it follows that, if there had been no intercalation in 713, the number of days that elapsed from the Kalends of January, 702, to the last day of December, 713, would have been a multiple of 8. Now, on the theory of Napoleon the Third, Le Verrier, and Colonel Stoffel this number would have been 4,401; on the theory that the year 708 contained 445 days, 4,424. The latter number is divisible by 8; the former is not.

There has never been any question but that the number of days intercalated in 708 was either 67 or 90; and the former number has been proved to be wrong.

IV. In order to obtain an absolutely firm foundation for the chronology of Caesar’s second expedition, one more question remains to be answered. In this country it is generally taken for granted that the Kalends of January, 709, the year in which Caesar’s reform of the calendar came into operation, corresponded with the 1st of January, 45 B.C. Various German scholars have, however, attempted to prove that the Kalends of January, 709, fell on the 2nd of January of the Julian calendar.

Let us first see what there is to be learned from the ancient writers.

Pliny[3586] says that when the error in the execution of Caesar’s reform was discovered, it was corrected by the omission of intercalary days during twelve successive years.

Solinus[3587] tells us that Caesar’s reform was vitiated by the pontiffs; for, whereas it had been enjoined that a day should be intercalated on the completion of the fourth year, they made the intercalation at the beginning of the fourth year, not at the end. Thus, Solinus continues, twelve days were intercalated in thirty-six years, whereas only nine ought to have been intercalated.

Suetonius[3588] says that the calendar, as reformed by Caesar, was thrown into confusion by ‘negligence’, and rectified by Augustus; also that Caesar made the calendar year consist of 365 days, so as to bring it into harmony with the solar year, and, abolishing the intercalary month, ordered that one day should be intercalated every fourth year.[3589]

According to Censorinus,[3590] Caesar directed that, in order to compensate for the quarter of a day by which the solar exceeded the calendar year, one day should be intercalated at the end of every quadriennial cycle, after the Terminalia [that is to say, after the 23rd of February].

According to Macrobius,[3591] Caesar directed that one day should be intercalated every fourth year. Macrobius then makes substantially the same charge against the pontiffs as Solinus, and goes on to say that, after the error for which they were responsible had continued for thirty-six years, Augustus corrected it by ordering that twelve years should pass without any intercalation.

Now it is absolutely certain that of the first five years during which the reformed calendar was in force, namely 709, 710, 711, 712, and 713, not one only but two contained an intercalary day.[3592] For, as we have already seen,[3593] Dion Cassius[3594] states that the Kalends of January, 702, was a market-day, and also that, in order to prevent the Kalends of January, 714, from falling on a market-day, a day was intercalated extraordinarily (παρὰ τὰ καθεστηκότα[3595]) in 713;[3596] and, as I had occasion to remark before, it follows that, if there had been no intercalation in 713, the number of days that elapsed from the Kalends of January, 702, to the last day of December, 713, would have been a multiple of 8. This would have been the case if one of the four years 709, 710, 711, and 712 had contained an intercalary day, but not otherwise.[3597] Which was it?

The year 711 may be set aside at once: nobody has ever argued that it was a Leap Year; and no reason can be given to show that it was.

1. Wilhelm Soltau,[3598] the author of a valuable work on Roman chronology, maintains that 709 was the first Leap Year of the reformed calendar. He argues that Matzat’s theory, according to which an intercalation occurred in 710, must be wrong, because it is inconceivable that Caesar, who was then alive, should have allowed an intercalation to take place in the second year of his calendar, in defiance of his own edict, that the intercalation should be made every fourth year (quarto quoque anno intercalandum esse). The theory that the first intercalation was in 712 is, Soltau continues, based only upon the statements, derived from the same source, of Solinus and Macrobius. If, says Soltau, they are right, the Leap Years, before Augustus rectified the error which had been made in carrying out Caesar’s intentions, would have been 712, 715, 718, 721, 724, 727, 730, 733, 736, 739, 742, and 745. But Dion’s statement, that 713 was an intercalary year, proves that this cannot have been the case. The series must, therefore, have been 709, 713, 716, 719 ... 743. Soltau holds that the twelve years during which intercalation was, by the order of Augustus, suspended, lasted from 745 to 756; that Augustus disapproved of Caesar’s (assumed) anticipatory method of reckoning, that is to say, of his having intercalated a day in the first year of his calendar instead of after the conclusion of the fourth; that accordingly the next intercalation, which would naturally have taken place in 757, was omitted; and that the first intercalation after the reform of Augustus occurred in 761. Thus, like other writers[3599] who differ from him on points of detail, Soltau identifies the Kalends of January, 709, with the 2nd of January, 45 B.C.

It will presently be shown that the statements of Macrobius and Solinus do not necessarily lead to the conclusion which Soltau condemns. Meanwhile I may remark that Soltau’s theory is irreconcilable with the very passage in Dion Cassius to which he refers.[3600] Dion says that a day was intercalated extraordinarily (παρὰ τὰ καθεστηκότα) in 713, in order to prevent New Year’s Day in 714 from falling on a market-day,[3601] and that subsequently an intercalary day was struck out.[3602] It is therefore obvious that the intercalation of 713 took place earlier than had been contemplated; and consequently that the previous intercalation must have occurred later than 709; for if, as Soltau maintains, the previous intercalation had taken place in 709, the intercalation of 713 took place at the proper time. If, on the other hand, Dion’s words, παρὰ τὰ καθεστηκότα, mean ‘contrary to the regulations erroneously attributed to Caesar by the pontiffs’, that is to say, contrary to the triennial cycle which they themselves followed, the intercalation, on Soltau’s theory, took place a year too late; for, if the first intercalation had occurred in 709, the object which Dion mentions could have been attained by intercalating in 712.

2. Let us now examine the theory of Matzat,[3603] namely, that the first intercalation took place in 710. This writer believes that Caesar’s reason for intercalating in 710 was to prevent the Kalends of January in the following year from falling on a market-day.[3604] He holds that Dion’s words, παρα τα καθεστηκοτα mean ‘contrary to the actual regulations of Caesar’; and accordingly he believes that those regulations were at the time understood. He maintains, however, that, after 713, the pontiffs intercalated every three years,--namely in 716, 719, 722 ... 743; but he insists that they did this simply for the same reason which had prompted the intercalation in 713, namely to prevent the Kalends of January in each following year from falling on a market-day. The statement of Dion, that, in order to compensate for the day extraordinarily intercalated in 713, another intercalary day was omitted, he takes to mean that the next intercalation, which, on his theory, ought to have occurred in 714, was left out. Finally, he believes that the three superfluous days which had accumulated during the twelve triennial cycles were compensated for by the omission of all intercalations in the years 745-756; and that the first intercalation under the reform of Augustus occurred in 757. On this theory the Kalends of January, 709, corresponded with the 1st of January, 45 B. C.

I have already mentioned the objection which Soltau has brought against Matzat’s theory;[3605] but that objection is inconclusive. It is not true that if Caesar had allowed an intercalation to take place in 710, he would have done so ‘in defiance of his own edict, that the intercalation should be made every fourth year (quarto quoque anno intercalaretur)’. If, according to his scheme, the next intercalation was to take place in 714, the next in 718, and so on, the intercalation would still be made every fourth year. Provided it took place every four years, what difference would it make whether it took place first in 709, 710, 711, 712, or 713? Holzapfel[3606] blames Matzat for disregarding the testimony of Solinus and Macrobius. But Matzat does not disregard their testimony: he simply refuses to admit that they make any definite statement as to the year in which the first intercalation of the Julian calendar occurred. The only statement which would appear to support Holzapfel’s criticism is contained in the words of Solinus, that, ‘whereas it had been enjoined that they [the pontiffs] should intercalate one day in the fourth year, and this ordinance ought to have been carried out on the completion of the fourth year ... they intercalated at the beginning of the fourth year, not at the end’ (nam cum praeceptum esset, anno quarto ut intercalarent unum diem, et oporteret confecto quarto anno id observari ... illi incipiente quarto intercalarunt, non desinente). If by ‘the beginning of the fourth year’ Solinus meant the fourth year of the Julian calendar, that is to say, 712, and if he had original authority for his statement, then Holzapfel is right. But observe the looseness with which Solinus expresses himself. Immediately after saying that the intercalation ought to have taken place ‘in the fourth year’, he says that it ought to have taken place ‘on the completion of the fourth year’. To state the facts correctly required extraordinary precision and nicety of expression; and this requirement he failed to satisfy. His meaning may have been that, in whatever year of the Julian calendar the first intercalation took place, the next ought, by Caesar’s ordinance, to have taken place four years later, and so on. If it be objected that I have suggested an arbitrary interpretation of his meaning, I reply that this interpretation is dictated by the passage in which Dion Cassius states that the intercalation of 713 was ‘contrary to the regulations’ (παρὰ τὰ καθεστηκότα). Holzapfel[3607] says that these words, taken by themselves, may mean one of two things. They may mean that the intercalation of 713 was contrary to the actual regulations of Caesar; or they may mean that it was contrary to the regulations adopted by the pontiffs in misunderstanding or in contravention of Caesar’s regulations. The question, says Holzapfel, can only be settled by other evidence; and the only other evidence is that of Solinus and Macrobius, which shows that the pontiffs misunderstood Caesar’s regulations. As a matter of fact, their evidence does not show this, unless misunderstanding is connoted by the words vitium and error. Matzat[3608] contends that such a misunderstanding would have been impossible, for Caesar must have made his intentions clear. Holzapfel replies that Caesar would no doubt have done so if he had foreseen his own imminent death; but, as he certainly intended, in his capacity as Pontifex Maximus, to superintend the execution of his own arrangements, and thus establish the rule of intercalation which he contemplated, the regulation ut quarto quoque anno intercalaretur might seem sufficient. But Dion Cassius, if his testimony may be accepted, settles the question. Immediately after saying that a day was extraordinarily intercalated in 713, he adds that ‘of course an intercalary day was in turn omitted, in order that the calendar might be brought into harmony with Caesar’s intentions’ (καὶ δῆλον ὅτι [ἡμέρα ἐμβόλιμος] ἀνθυφῃρέθη αὖθις, ὅπως ὁ χρόνος κατὰ τὰ τῷ Καίσαρι τῷ προτέρῳ δόξαντα συμβῇ). Now these words, as Matzat[3609] unanswerably argues, prove that by παρὰ τὰ καθεστηκότα Dion meant ‘contrary to the regulations’ actually made by Caesar.[3610] Holzapfel,[3611] however, tries to explain away Dion’s remark by the argument that the authority whom he followed may have been a contemporary who shared the misconception of the pontiffs; and this I cannot gainsay. Moreover, although I have argued that the words of Solinus may be interpreted in a sense different from that which Holzapfel ascribes to them, I admit that the conclusion which they suggest is that Caesar intended to make his first intercalation in 713, and that the pontiffs made it in 712.

Secondly, it is expressly stated by Macrobius[3612] that the Kalends of January in the year of the Lepidianus tumultus fell on a market-day; and if, as Holzapfel maintains, this statement refers to the Lepidus who, as one of the Triumvirs, revived Sulla’s policy of proscription in 711, it proves that no intercalation occurred either in 710 or in 709.[3613] Matzat, however, maintains that the words Lepidianus tumultus designate the outburst of Lepidus in 676 (78 B. C.).[3614] Unger,[3615] who agrees with Holzapfel, maintains that the official recognition of the superstitious dread with which the Roman populace contemplated the coincidence of the Kalends of January with a market-day, was due to the acts of Lepidus in 711. But, replies Matzat,[3616] for this ‘official recognition’ the approbation of the Pontifex Maximus was necessary. Now in 713, as in 711, the Pontifex Maximus was Lepidus himself; and, according to Macrobius, it was the Lepidianus tumultus which strengthened the popular belief that whenever the Kalends of January fell upon a market-day, the whole year would be darkened by ill-omened events. If, then, says Matzat, we are to believe Unger, Lepidus described the deeds which he had himself done in 711, and by which the joint supremacy of himself and the other two Triumvirs had been established, as a tumultus,--the most horrible events of a year full of horrors! This argument is clever, but I think that it is hardly fair. Lepidus was not obliged to describe anything. Assuming that the Kalends of January, 711, had fallen on a market-day, it is surely intelligible that he should have recognized the wisdom of allaying superstitious fears, even though they had been roused by his own acts, when he could do so by the simple expedient of intercalating a day in 713. I agree with Unger that there does not appear to have been anything very alarming in the affair of 677, even though Lucan[3617] describes it as truces Lepidi motus; and, judging the question without bias on its own merits, I can only conclude that the tumultus Lepidianus was most probably the outbreak of 711. If so, the first intercalary year of the Julian calendar cannot have been either 709 or 710, but must have been 712.

Thirdly, Holzapfel[3618] points out that, if Matzat is right, the intercalary cycle introduced by Augustus did not correspond with that of Caesar. For after the reform of Augustus the intercalary years were odd years, 761, 765, 769, and so on; while Caesar’s first intercalary year is supposed by Matzat to have been 710. Or, if we reckon the quadriennial cycles contemplated by Caesar from the year 709, the intercalations, according to his regulation, would, on Matzat’s theory, occur in the second, those made by Augustus in the first year of each successive cycle.

All this is perfectly true: but what does it matter? The one really important point, namely, that the intercalation should take place every four years, was duly secured by Augustus. Whether it took place in the first, the second, the third, or the fourth year of the cycle, mattered not a jot. Holzapfel’s objection is purely academic.

Fourthly, says Holzapfel, if, as Matzat maintains, Caesar’s only reason for intercalating in 710 was to prevent the Kalends of January in the following year from falling on a market-day, it is difficult to believe that Caesar should not have foreseen that for the same reason it would be necessary to intercalate in 713, 716, and so on, that is to say, every three years; in other words, that it would be impossible to carry out the arrangement which he had himself made.[3619]

This is certainly a reasonable objection, and Matzat has not, so far as I know, attempted to remove it: but it is perhaps conceivable that a man so busy as Caesar should have failed to look far ahead.

Holzapfel[3620] argues, further, that Matzat’s theory, according to which the calendar, as reformed by Augustus, was inaugurated on the 1st of January, 757 (A.D. 4), rests upon the assumption that the first day of the intercalary cycle coincided with the first day of the civil year, whereas it was really the day after the Terminalia, that is to say, the sixth day before the Kalends of March. This, he maintains, is proved (a) by the place which Caesar gave to his intercalary day; (b) by the fact that the two months intercalated between November and December of 708 were called respectively mensis intercalaris prior and mensis intercalaris posterior, and also by the fact that, according to Dion Cassius, the number of days intercalated in that year was only 67, whereas the number of days intercalated in the civil year 708 was 90; (c) by the fact that, according to Macrobius and Solinus, Caesar ordained that the intercalary day [which followed the 23rd of February] should be inserted at the end of the fourth and before the fifth year of the Julian calendar.[3621]

Matzat[3622] summarily replies to these arguments. Referring to Macrobius,[3623] he observes (a) that the place which Caesar gave to the intercalary day was identical with the place which the intercalary day, whenever it occurred, had occupied before his reform; and (b) that the two intercalary months known as mensis intercalaris prior and mensis intercalaris posterior were added to the year 708 in order that the calendar year 709 might begin on the Kalends of January, and thus coincide with the consular year. If it be asked why they were called prior and posterior although another intercalary month had preceded them, the answer is easy: the other month ought in any case to have been intercalated in that year, whereas the prior and posterior were extraordinarily intercalated. The passages in Macrobius and Solinus on which Holzapfel relies have been already explained; and it has been shown that they do not necessarily bear the meaning which he ascribes to them. One fact alone appears to me to dispose of his contention, that the Julian calendar did not come into operation until the sixth day before the Kalends of March, 709: if it did not, why did January in that year contain 31 days, whereas in every previous year it had contained only 29?

Holzapfel also invokes the support of Böckh,[3624] who remarked that it would have been unnatural for a reformer to correct the error caused by the difference of a quarter of a day between the civil and the solar year until the error required correction. The conclusion appears to Holzapfel inevitable that Caesar intended to make his first intercalation as soon as, and not before, the error should have amounted to one day, that is to say, in 713. Matzat,[3625] on the contrary, maintains that logically the proper place for the intercalary day would have been immediately after the second year of the quadriennial cycle. But he does not believe that Caesar cared for such academic considerations. He undoubtedly fixed the place of the intercalary day in the year not on astronomical grounds, but according to usage. Why, then, asks Matzat, should it be considered improbable that he fixed the place of the intercalary year in the quadriennial cycle on the same principle?

3. Holzapfel[3626] holds, as we have just seen, that the Caesarian cycle began on the day after the Terminalia of 709, that is to say, on the sixth day before the Kalends of March; that Caesar intended that the first intercalation should take place in 713; that the pontiffs, misunderstanding his directions, made the first intercalation at the beginning of the fourth year, that is to say, in 712; that, as Dion says, a day was extraordinarily intercalated in 713; that, to compensate for this extraordinary intercalation, a day was omitted in 714, which accordingly comprised 364 days only; that the pontiffs thenceforth intercalated every three years, namely in 715, 718, 721 ... 745; that, to compensate for the three superfluous days which had been intercalated in consequence of the misunderstanding of Caesar’s regulations, the intercalations which ought to have occurred in 749, 753, and 757, were omitted; and that the first intercalation after the reform of Augustus took place in 761.

This theory, as the reader will have already seen, cannot stand unless the evidence of Dion Cassius is to be rejected. Indeed it cannot stand even then. Holzapfel is not justified in assuming that in order to compensate for the extraordinary intercalation of 713, a day was omitted in 714; for, on his own theory, 714 was an ordinary year. Matzat[3627] points out that, in the passage in which Dion[3628] describes the omission of a day to compensate for the extraordinary intercalation--καὶ δῆλον ὅτι ἀνθυφῃρέθη αὖθις, ὅπως ὁ χρόνος κατὰ τὰ τῷ Καίσαρι τῷ προτέρῳ δόξαντα συμβῇ--the words ἡμέρα ἐμβόλιμος (an intercalary day) must necessarily be supplied, as the subject of the verb ἀνθυφῃρέθη, from the preceding sentence. Holzapfel retorts that his view does not involve the assumption of a change of subject. ‘One can very well translate,’ he says, ‘“an intercalary day was inserted, and self-evidently in turn omitted”’ (Man kann sehr wohl übersetzen: ‘es wurde ein Schalttag eingelegt und selbstverständlich wiederum in Abzug gebracht’[3629]). This is not a satisfactory answer; for, on Holzapfel’s own showing, the omitted day was not an intercalary day. There is no evidence that a day was ever withdrawn from an ordinary year in the Roman calendar; and, as Matzat[3630] points out, the best proof that such a proceeding would have been regarded as out of the question is supplied by the procedure of Augustus. Instead of correcting the error of the pontiffs by withdrawing three days from one ordinary year, he omitted three intercalary days in three intercalary years, thus taking twelve years to accomplish a reform which, according to modern notions, might have been accomplished in one. If Dion’s words are interpreted in their natural sense, they evidently mean that the next intercalary day which would have occurred, according to Caesar’s regulations, was omitted. Thenceforth, accordingly, if the first intercalation occurred in 710, the intercalary years were 716, 719, 722 ... 743. Or if, as Holzapfel insists, Dion’s words, παρὰ τὰ καθεστηκότα, mean ‘contrary to the regulations as erroneously interpreted by the pontiffs’, and if, as he also insists, the year in which they first intercalated was 712, then the next year in which they would naturally have intercalated was 715: the extraordinary intercalation of 713 must have been compensated for by the omission of an intercalary day in 715; and the following series of intercalary years must have been 718, 721, 724 ... 745.

Again, Holzapfel’s theory compels him to disregard silently the testimony of Solinus, on whose authority he lays such stress. Solinus[3631] says that twelve days were intercalated in the first thirty-six years of the Julian calendar: according to Holzapfel, thirteen were intercalated.

Lastly, if we accept Holzapfel’s view, that the first day of the Julian calendar was the sixth day before the Kalends of March, 709, we find that a day was intercalated immediately after the end of the fourth year of the cycle, namely, after the Terminalia of 713. But Holzapfel assures us that, according to Macrobius and Solinus, this was not the case.

But I am not arguing against Holzapfel’s theory as regards the first intercalary year of the Julian calendar; and what appears to tell most strongly in favour of it, besides the probability that the Lepidianus tumultus occurred in 711,[3632] is the statement of Macrobius,[3633] that Augustus enacted that the intercalation should take place ‘every fifth year’ (quinto quoque anno), that is to say, according to our reckoning, every four years. These words seem to imply that the pontiffs had actually misunderstood Caesar’s regulation. On Matzat’s theory, however, the pontiffs who intercalated παρὰ τὰ καθεστηκότα in 713 deliberately set that regulation aside in order to avoid the dreaded coincidence of the Kalends of January with a market-day. But, says Matzat, every three years this troublesome necessity recurred; and thus ultimately, as he suggests, the erroneous view might prevail that Caesar had himself intended to intercalate every three years ([anno] quarto non peracto sed incipiente[3634]). But what right have we to assume that after 713 the pontiffs took any account of the nundinal superstition? At all events, if Holzapfel is right in maintaining that the first intercalary year of the reformed calendar was 712, there can be no doubt that the subsequent intercalary years were 713, 718 ... 745; and it is therefore impossible for him to reconcile his view, that the first intercalation under the reform of Augustus occurred in 761, with the statement that Augustus allowed twelve years to pass without any intercalation. Augustus’s first intercalation undoubtedly took place in 757; for in that year, if Caesar’s regulation had been observed, the twelfth intercalation would have occurred.

Opinions may differ as to whether Matzat’s theory or the modification of Holzapfel’s which I have just suggested is the more probable. As, according to the latter, the series of intercalary years must have been 712, 713, 718, 721, 724, 727, 730, 733, 736, 739, 742, 745, it implies that after 713 the pontiffs thought it safe to disregard the nundinal superstition. So far my suggestion may be objectionable. On the other hand, it fits in with all the statements of the ancient writers, except that one remark of Dion Cassius which, as Holzapfel suggests, he may have made on erroneous information;[3635] and particularly it fits in, as no other series which has been suggested does, with the statements that twelve days were intercalated in the first thirty-six years of the Julian calendar,[3636] and that Augustus allowed the next twelve years to pass without any intercalation. However, the difference between Matzat’s theory and mine (which is purely tentative) is unimportant; for they agree in the main point,--that the Kalends of January, 709, corresponded with the 1st of January, 45 B.C.

V. We have now gained the knowledge which will enable us to investigate the chronology of Caesar’s second invasion of Britain. We have ascertained that the Kalends of January, 709, fell on the 1st of January, 45 B.C.; that 90 days were intercalated in 708, which accordingly consisted of 445 days; that a month of 23 days was intercalated in 702, which accordingly comprised 378 days; and that 701, 703, 704, 705, 706, and 707 were ordinary years, each comprising 355 days. It follows that the last day of 700, the year in which Caesar made his second expedition to Britain, corresponded with the 30th of November, 54 B.C., and that the sixth day before the Kalends of October, the day on which he wrote to tell Cicero that he was on the point of bringing back his army from Britain to Gaul, corresponded with the 29th of August. From these data it will be easy to ascertain the correspondence of any date in the year 700 which we find in our authorities with the Julian calendar.

After quitting Cisalpine Gaul, Caesar returned to his army, which had wintered in the country of the Belgae. He made a tour of inspection, visiting the various camps, which were of course in the immediate neighbourhood of the yards where the legionaries had been building the ships for his intended expedition, that is to say, at the Portus Itius (Boulogne), and probably on the estuaries of the Canche, the Authie, the Somme, and the Seine.[3637] After ordering all the ships to assemble at the Portus Itius, he started with four legions in light marching order and 800 cavalry for the country of the Treveri, which, roughly speaking, comprised the greater part of the province of Luxembourg, the Grand Duchy of Luxembourg, and the southern part of Rhenish Prussia.[3638] Two chiefs of this people, named Cingetorix and Indutiomarus, were struggling for supremacy. Cingetorix at once presented himself before Caesar, and promised fidelity. Indutiomarus collected levies, and prepared to fight. Many of the leading men, however, came into Caesar’s camp and made terms for themselves. Indutiomarus found that he had miscalculated his strength, and hastened to excuse himself. Caesar, who had no time to spare, contented himself with taking hostages for his good behaviour and returned to the Portus Itius. About 25 days after his arrival, having meantime been detained by contrary winds, he sailed for Britain.[3639]

The exact date of Caesar’s departure from Cisalpine Gaul is uncertain. On the 2nd of June, that is to say, the 9th of May of the Julian calendar, Cicero received at Rome a letter from his brother, written at Placentia;[3640] and on the following day he received another letter from his brother, written at Blandeno, a town near Placentia, the exact site of which is not known,[3641] and also a letter from Caesar, written apparently at the same place.[3642] Napoleon, erroneously maintaining that the second and the third of these letters were received by Cicero on the 5th, instead of the 3rd, of June, and assuming (‘pour trouver le temps voulu’, as he naïvely remarks) that, in consequence of accidental delays, they took 13 days to reach Rome, concludes that Caesar quitted Blandeno on the 23rd of May, which he identifies with the 22nd[3643] of the same month, but which really corresponded with the 29th of April of the Julian calendar. T. Bergk, on the contrary, maintains that if there had been any delay in the transmission of the letters, Cicero would have mentioned it in his reply; and he supposes that Caesar and Quintus Cicero started on their journey for Transalpine Gaul on the 2nd of June, that is to say, on the 9th of May of the Julian calendar.[3644] All that can be safely said is that, if the letter-carrier travelled at the usual rate, namely between 40 and 50 Roman miles a day,[3645] the two letters were written about the 30th of April or the 1st of May of the Julian calendar; and therefore we have no right to assume that Caesar quitted Blandeno before the former day.

It is possible, as we shall presently see, to fix the date of Caesar’s arrival at the Portus Itius and of his voyage to Britain within a day or two; but it is plainly impossible to make any satisfactory calculation of the dates of his movements from the time when he left Blandeno to the time when he left the country of the Treveri and marched for the Portus Itius; and the minute computations of Napoleon and others are simply elaborate trifling.[3646] For, although Caesar’s average rate of travelling may be estimated approximately, we do not know how far he penetrated into the extensive country of the Treveri; and it is waste of time to guess how long he stayed there. This much only can be said with certainty:--he did not let the grass grow under his feet.[3647] For he left Blandeno about the 30th of April: he arrived, as we shall presently see,[3648] finally at the Portus Itius about the 11th of June; and in those 43 days he travelled from Blandeno across the Alps and across Gaul to the English Channel; moved along the coast to various points between Boulogne and the mouth of the Seine; and marched from Boulogne to the neighbourhood of Sedan, or further--at least 180 miles--and back again.

On the 27th of July, that is to say, on the 2nd of July of the Julian calendar, Cicero wrote to Atticus, ‘Judging from my brother Quintus’s letters, I imagine that by this time he is in Britain’ (ex Q. fratris litteris suspicor iam eum esse in Britannia[3649]). It would be very rash, however, to infer from this that Caesar landed in Britain before, or even as early as, the 2nd of July; for, as we have seen, his embarkation was delayed by the long continuance of adverse winds. The first letter which announced the arrival of the expeditionary force in Britain was referred to by Cicero in a letter to Quintus, in which he says, ‘How I rejoiced at your letter from Britain! I was nervous about the sea and the coast of that island’ (O iucundas mihi tuas de Britannia literas! Timebam Oceanum, timebam litus insulae[3650]). This letter is undated; but it must have been written some time after the one which Cicero wrote to Atticus on the 27th of July; for in the letter to Atticus we find the words, ‘I have undertaken to defend Messius.... After that I have to prepare myself for Drusus, and then for Scaurus’ (Messius defendebatur a nobis ... Deinde me expedio ad Drusum, inde ad Scaurum[3651]); while in the letter to Quintus Cicero wrote, ‘The day I write this Drusus has been acquitted.... The comitia have been put off to September. Scaurus’s trial will take place immediately’ (Quo die haec scripsi, Drusus erat ... absolutus ... Comitia in mensem Septembrem reiecta sunt. Scauri indicium statim exercebitur[3652]). Asconius[3653] tells us that the last day of Scaurus’s trial was the 2nd of September; and Cicero’s remark that ‘the comitia have been put off to September’ makes it evident that he wrote in August; while from his saying that ‘Scaurus’s trial will take place immediately’ we should naturally infer that when he wrote the 2nd of September was not far off. Letters from Britain generally reached Rome in about 27 days;[3654] and accordingly we may conclude that the letter in which Quintus Cicero announced his arrival in Britain was written about the end of July; that is to say, about the 6th of July of the Julian calendar.[3655] Now Caesar says that the tide [in the Straits of Dover] turned westward soon after daybreak on the morning of his arrival in Britain;[3656] and this statement proves that he landed either about the time of full moon or about the time of new moon. There was a full moon on the 21st of July, 54 B.C.; and the previous new moon occurred on the 7th of July. Napoleon[3657] insists that the landing must have taken place on the day of full moon, arguing that without moonlight Caesar could not have undertaken the march which he made on the night following his arrival. But, as we have already seen, Napoleon argues on the erroneous assumption that Caesar intercalated only 67 days in the year 708, and accordingly he fixes all the dates of the unreformed calendar which occur in Cicero’s letters of 700 twenty-three days too late. His argument that Caesar could not have made a night march except by the light of the moon is worthless. It must be remembered that, in the first half of July, there is no real night over any part of the British Isles; and no one familiar with the records of night marches would deny that Caesar could have marched on a clear night at that time of the year. Long[3658] says that ‘of course he had also the moon on the night on which he sailed from the Gallic coast’. There is no ‘of course’ in the matter. Caesar sailed from Gaul to Britain in 55 B.C. after midnight, on the 26th or 27th of August, that is to say either five or four days before the full moon;[3659] and the moon set soon after midnight both of August 25-26 and of August 26-27. Moreover, according to Napoleon himself, Caesar sailed from Britain to Gaul in 55 B.C. on the 12th of September,[3660] that is to say, on a moonless night: as new moon occurred on the 14th, and he did not sail until after midnight, he could not have had the benefit of moonlight unless he had deferred his voyage until the date of the equinox; and I am assured by Captain Iron, the harbour-master of Dover, that on a fine night, especially in July, there would not have been the least difficulty in sailing without a moon. As a matter of fact, William the Conqueror sailed to England on a dark night. ‘The moon,’ says Mr Freeman,[3661] ‘was hidden and the heavens were clouded over. The Duke therefore ordered every ship to bear a light.... The ships were to keep as near together as might be, and to follow closely after the beacon-light of his own ship.’ If, then, we decide that Caesar landed in Britain on the day of new moon, the 7th of July, 54 B.C., we shall not be more than one day wrong; but to fix the date with absolute precision is impossible.[3662]

As Caesar landed about the 7th of July, it follows that he had reached the Portus Itius, where he was delayed about 25 days, about the 11th of June.

On the day after he landed Caesar encountered a British force 12 miles from his camp on the coast. On the following day, while his troops were pursuing the fugitives, he was recalled to the coast by the news that a large number of his ships had been damaged by a storm. He then proceeded to construct a naval camp, and, as soon as it was finished, returned to the point from which he had started.[3663] As the construction of the camp occupied ‘about ten days’ (circiter dies X), we shall not be far wrong if we say that it was finished 12 days after the landing, that is to say, about the 19th of July of the Julian calendar.

We now come to a date the significance of which has hardly been appreciated. At the close of one of his letters to Quintus, Cicero writes, ‘Caesar wrote me a letter from Britain on the 1st of September... in which, to prevent my wondering at not getting one from you, he tells me that you were not with him when he reached the coast’ (Ex Britannia Caesar ad me K. Septembr. dedit litteras ... quibus, ne admirer, quod a te nullas acceperim, scribit se sine te fuisse, cum ad mare accesserit[3664]). This passage proves that Caesar had returned from the interior of Britain to the coast on or before the 1st of September, that is to say, the 5th of August of the Julian calendar. But we have already seen that he did not quit the coast after the construction of his naval camp until about the 19th of July, and that on the 29th of August he was still in Britain, ‘on the point of bringing back the army,’ and did not sail for Gaul till several days later.[3665] We have to decide, then, between two alternatives. Either Caesar had finished the campaign against Cassivellaunus by the 5th of August of the Julian calendar, and thereafter remained on the coast until the 29th of August, when he was able to announce that he was ‘on the point of bringing back the army’; or he made a hurried temporary visit to the coast, the object of which remains unexplained. The latter view is not supported by the Commentaries. Caesar’s narrative certainly leaves the impression that, immediately after the completion of his naval camp, he resumed the military operations which had been interrupted by the shipwreck, and did not again return to the coast until the time came for him to prepare for his voyage to Gaul. He tells us that Cassivellaunus, after the failure of the attack which had been made by his orders upon the naval camp, and in consequence of the reverses which he had suffered, sued for peace; that he ordered Cassivellaunus to furnish hostages;[3666] and that, ‘on receiving the hostages, he led back the army to the sea, where he found the ships repaired’ (Obsidibus acceptis exercitum reducit ad mare, naves invenit refectas[3667]). ‘When they were launched,’ he continues, ‘he determined to take the army back in two trips’ (his deductis ... duobus commeatibus exercitum reportare instituit[3668]). Certainly there is not a word in this to support the view, which is advocated by Vogel, that the visit to the coast which Caesar made on the 1st of September (the 5th of August of the Julian calendar) was purely temporary, and took place before he began to march towards the country of Cassivellaunus; and the motive which Vogel[3669] suggests for the visit--that Caesar wished to inspect the camp once more and to give the necessary instructions before marching against Cassivellaunus--is hardly adequate, unless we grant Vogel’s assumption, that when Caesar returned to the sea the campaign had advanced no further than the stage which he describes in the 17th chapter of his Fifth Book, where three legions under Trebonius inflicted a decisive defeat on the enemy; in other words, that Caesar returned from a point within a day’s march of the sea. But this affair, according to the Commentaries, took place on the very day after Caesar quitted his naval camp in order to resume the campaign; whereas he returned to the sea, on Vogel’s own showing, about 17 days after he quitted the camp. If, on the other hand, we adopt Napoleon’s view[3670]--that Caesar, after he returned to the sea on the 1st of September, remained there until he sailed for Gaul--how are we to account for the 24 days which elapsed before he wrote to tell Cicero that he was on the point of bringing back the army? What was he doing all that time? Napoleon,[3671] indeed, maintains that, on the day on which he wrote this last letter, he actually sailed for Gaul, and Bergk that he had already arrived in Gaul; but it has already been shown that both these assumptions are untenable. Moreover, in order to account for Caesar’s having arrived at the sea so early, Napoleon and Bergk are forced to strain the words of the Commentaries (exercitum reducit ad mare),--to assume that Caesar hurried on in advance of his army, and that it did not reach the coast until several days later.[3672]

Although the problem cannot be definitively solved, I have no doubt but that Vogel’s solution is wrong. Not only does it give the lie to Caesar’s narrative, but it requires us to believe that Caesar had failed for about 17 days to make any headway against the Britons, and had been held in check by them within a dozen miles of the sea. Yet Caesar states that, in consequence of the defeat which Cassivellaunus suffered at the hands of Trebonius on the day after he left the newly-constructed naval camp, the British infantry levies dispersed; and in the same breath he goes on to say that he forthwith marched for the country of Cassivellaunus.[3673] If, as Vogel implies, this statement had been false, surely Quintus Cicero would have informed his brother, in one of the five letters which he wrote before the 1st of September, of the real state of affairs! On the other hand, I find it difficult to believe that Caesar used the words exercitum reducit ad mare loosely, and that he remained on the coast from the 1st to the 25th of September, without once writing to Cicero between those two dates,[3674] and then remained several days longer before embarking. This view, indeed, would compel us to assume that he left Trebonius or one of his other generals to carry on the negotiations with Cassivellaunus which are described in the 22nd chapter of his Fifth Book.

Napoleon, it is true, believes that Caesar did not leave his army and hurry on in advance of it to his naval camp until the negotiations were completed and he had received his hostages from Cassivellaunus.[3675] But on this assumption what motive could he have had, first, for hurrying on in advance of his army, and, secondly, for delaying its re-embarkation until after the lapse of several weeks? I am inclined, therefore, to believe that he made a hurried temporary visit to his naval camp, escorted probably by a small column, and then, having accomplished his purpose, returned to the main army in order to conduct or to complete the negotiations with Cassivellaunus. The motive of his visit may have been connected with the attack which the Kentish chieftains made upon the naval camp.[3676]

The date of Caesar’s return to Gaul can only be given approximately. We have seen that on the 25th of September (the 29th of August of the Julian calendar) he wrote to Cicero, saying that he was on the point of bringing back the army.[3677] Vogel[3678], remarking that, after he reached the coast, his ships had to be launched and loaded, and that he did not sail with the second detachment of troops until he had waited a long time in vain for the return of the ships which had carried the first, concludes that he did not return to Gaul until about 20 days after he wrote to Cicero, that is to say, not until the 15th of October. This date corresponds with the 17th of September of the Julian calendar; and Vogel maintains that it agrees with Caesar’s statement of the reason which led him to hurry on his return, namely, that the equinox was at hand. But it is impossible to estimate, from Caesar’s statement, how many days he waited for the return of his ships. Let us examine the attempts which have been made to gain a clue from Cicero’s correspondence.

Vogel[3679] points out that Cicero,[3680] in a letter written just after the 23rd of November--the 26th of October of the Julian calendar--referred to two letters which he had received from Quintus, and also to a third, which Quintus had handed to Labienus for transmission the day before he dispatched the earlier of the other two, but which had not yet arrived. Now Labienus had remained in Gaul during the invasion of Britain.[3681] It is clear, therefore, that these three letters were not written until after Quintus had returned to Gaul. On the other hand, they would seem to have been the first letters which Quintus wrote to his brother after his return. For Cicero, in the letter in which he referred to them, said, ‘where your Nervii dwell and how far off, I have no idea’ (Ubi enim isti sint Nervii et quam longe absint, nescio[3682]). The Nervii were the tribe in whose country Quintus, with his legion, was to pass the winter.[3683] Evidently Quintus, when he wrote the two letters which Cicero received, had not yet reached the country of the Nervii; for otherwise he could not, on the previous day, have been with Labienus, who was to winter in the country of another tribe. Probably, as Vogel concludes, he wrote from Samarobriva, or Amiens, where Caesar had his head quarters and where the arrangements for the distribution of the legions were made.[3684] Again, in the letter which has been already quoted, Cicero wrote to Quintus, ‘Pray be careful to let me know to whom I am to give the letter which I shall then send you,--to Caesar’s letter-carriers, for him to forward it direct to you, or to those of Labienus?’ (Tu velim cures ut sciam quibus nos dare oporteat eas quas ad te deinde litteras mittemus, Caesarisne tabellariis, ut is ad te protinus mittat, an Labieni[3685]). Vogel remarks that ‘this question is only intelligible on the hypothesis that Quintus was only just beginning to take up his quarters in Gaul’ at the time when he wrote the two letters which Cicero had just received. Now a letter would have required about 25 days for transmission from Samarobriva to Rome;[3686] and accordingly the letters to which Cicero referred, assuming that he replied to them promptly, would have been written about the end of October. Vogel, who thinks that they must have been written within a fortnight after Quintus and Caesar returned to Gaul, infers that they cannot have returned earlier than the 15th of October, the 17th of September of the Julian calendar. His reasoning is ingenious; but unfortunately we do not know exactly how soon after the receipt of his brother’s second letter Cicero wrote, or how many days intervened between the arrival of the first and of the second.

On the whole, it appears to me that all we can say for certain regarding the date of Caesar’s return is this. It cannot be fixed earlier than several days after the 29th of August of the Julian calendar,--the date of the letter in which he informed Cicero that he was on the point of bringing back the army. Bearing in mind that it occurred when ‘the equinox was at hand’, we may place it about the middle of September.

TOPOGRAPHICAL NOTES

On April 25, 1902, I observed more carefully than I had ever done before the coast between Sandgate and West Hythe. To speak of the hills between Sandgate and Hythe as angusti montes is sheer nonsense. Caesar would never have attempted to force a passage inland at any point between Lympne and Sandgate;[3687] nor would the Britons have abandoned these loca superiora, which lay ready to hand. There are, indeed, depressions in the line of hills--(1) just west of Sandgate railway station, (2) nearly opposite the Seabrook (now Imperial) Hotel, and (3) west of Hythe, just west of the point where the road diverges from the military canal; but if Caesar had attempted to force these gaps, he would have found himself entangled in the hills behind. 27.4.02.

* * * * *

Quite recently I explored the easternmost and the westernmost of the three valleys which partially break the continuity of the hills behind Hythe. Neither would have been [reasonably] practicable for an invading army [in the conditions of ancient warfare]. The road leading through the former, which branches off from the road [running from Sandgate] to Hythe, is level for the first 120 yards, and then ascends rapidly for a short distance. Then it is tolerably level until about 100 yards before one gets to the cross-road which turns off to the left, when it ascends rapidly for a long way. A column moving along it would have been exposed to attack from the hills on either side, and particularly on the west. 3.9.03.

* * * * *

On the return voyage from Boulogne [September 5, 1903] I most carefully scrutinized the whole coast-line between Sandgate and the Foreland, as I had often done before on land. Caesar’s description of the angusti montes is applicable only to Shakespeare’s Cliff and the cliffs which extend from the Castle Hill to the neighbourhood of Kingsdown. It is not applicable even to the imposing heights which bound East Wear Bay, because, although they might fairly be called angusti montes, the missiles of which Caesar speaks could only have been thrown on to the beach from the precipitous but low chalk cliffs which form the lowest part of these heights; whereas he plainly means that the missiles would [or rather, could] have been thrown by the enemy who were standing in omnibus collibus; and if he anchored off East Wear Bay, the colles were a quarter of a mile or more from the sea. The low chalk cliffs of East Wear Bay would never have been called montes, although they are the lowest part of a range of montes. The notion that the cliffs between East Wear Bay and Folkestone Harbour, or the cliffs on which the ‘Leas’ stand, as seen from a vessel half a mile from the shore, would have been called angusti montes or montes at all is simply ludicrous. No! It is absolutely certain that Caesar’s angusti montes were the cliffs of Dover,--the cliffs between [and including] the Castle Hill and the Foreland. And as for Airy’s theory, how could cliffs ‘ten to thirty feet high’ have been called angusti montes by an observer standing on the deck of a ship five nautical miles away?[3688]

[Cicero’s description (Att., iv, 16, § 7) of the cliffs which ‘walled in the approaches to the island’--mirificis molibus--is applicable only to the heights behind East Wear Bay and the cliffs of Dover.]

* * * * *

The following notes were jotted down on September 15, 1902, on the deck of a steamer running from Dover to Deal:--

Six colles at present visible off Dover Harbour, i.e. on east of old [pre-Roman and Roman] harbour: one on left, not counting Shakespeare’s Cliff.

Eight from off the Foreland. Nothing could be more appropriate than the expressions in omnibus collibus expositas [hostium copias] and montes angusti. In the various ‘dips’ the hostium copiae would have been very conspicuous.

* * * * *

Behind the low rampart between Kingsdown and Walmer Castle the ground rises. Caesar might have encamped [in his first expedition] on this rising ground or on the rising ground which extends behind the castle towards the church.

Just after passing Deal Castle one turns to the left down Gilford Road, and, after walking about 300 yards, passes under a railway bridge, and then, very gradually ascending for a few paces, walks along a path, which crosses fields. It seemed to me just possible that C. might have encamped on these fields, but very--to the last degree--unlikely.

The only suitable camping-grounds that I can see anywhere are those mentioned above. 16.9.02.

* * * * *

Walked over the cliffs from St. Margaret’s to Walmer.... After walking for a short distance along the edge of this [natural] rampart [south of Walmer Castle], I struck inland along a fence, and came to a path which traverses a rolling chalk plateau, and runs exactly in a straight line with the tower of St. Mary’s Church, Walmer. There is splendid camping-ground on this plateau. The camp would have commanded the approaches from every point of the compass; and the descent from the plateau on the west or landward side is steep. Beyond the valley which bounds the plateau on this side the ground rises again fairly rapidly. The plateau extends northward to a point just south of Walmer Castle, which it commands. If Caesar encamped on the plateau, the camp could be discovered by excavation: but if he encamped on the gently rising ground north-west of the plateau, the camp could not be discovered; for this ground is covered by buildings. (I assume of course that he did not encamp on the high ground west of the plateau or on the ground on which the windmill stands; for both sites are too far from the sea.) 17.9.02.

ADDENDA

PAGE 122. ‘There was certainly a Copper Age ... Ireland.’ Professor Gowland (Journ. Anthr. Inst., xxxvi, 1906, pp. 26-7) deprecates the use of the expression ‘Copper Age’, remarking that ‘the so-called Copper Age possesses no characteristics which are not common to the Neolithic Age, except the imitations and limited use of stone forms in metal’, &c. The question seems to be purely verbal.

PAGE 140. ‘It is worthy of remark ... not pure.’ The proportion of lead in Scottish bronze implements appears to have been remarkable; but Professor Gowland (Journ. Anthr. Inst., xxxvi, 1906, p. 30) observes that lead is ‘found in small quantities in nearly all bronze implements’. See, however, J. Evans, Anc. Bronze Implements, p. 417.

PAGE 148. ‘All the open ones ... sand.’ Professor Gowland (Journ. Anthr. Inst., xxxvi, 1906, p. 36) affirms that ‘moulds of sand or loam were undoubtedly of later times [in the Bronze Age], as there are considerable mechanical difficulties in preparing them’.

PAGE 194, note 3. To the list of counties in which drinking-cups have been found must now be added Kincardineshire (Proc. Soc. Ant. Scot., xl, 1906, pp. 304-6).

PAGE 205, note 4. To the list of papers on cup- and ring-markings may be added Proc. Soc. Ant. Scot., xl, 1906, pp. 318-27.

PAGE 208. ‘Stone circles ... Kincardineshire.’ A circle in Stirlingshire is described in Proc. Soc. Ant. Scot., xl, 1906, pp. 301-4.

PAGE 211, note 2. To the list of papers on Scottish stone circles may be added Proc. Soc. Ant. Scot., xl, 1906, pp. 164-206, 245-54.

PAGE 221. ‘They told him ... ankle-bones.’ This statement rests upon the reading ἀστραγάλων in Diodorus Siculus, v, 22, § 2. Professor Ridgeway (Folk-Lore, i, 1890, p. 83, n. 1) suggests that the true reading may be ἀστραβῶν (saddles).

PAGE 237. ‘Coral ... later period.’ M. S. Reinach (Rev. arch., 4^e sér., vi, 1906, pp. 309-10) argues that the development of enamelling in Gaul was due to the growing dearness of coral, the price of which rose because large quantities were bought for exportation to India after the conquests of Alexander the Great. This view leaves unexplained the continued use of coral in Britain long after it had fallen into disuse in Gaul.

PAGE 259, note 3. In a recent article on vitrified forts (Proc. Soc. Ant. Scot., xl, 1906, pp. 136-50) Lieut.-Col. A. B. McHardy suggests that some of them may have been built in the time of the Vikings, and argues that the vitrifaction was probably never intentional, because (1) ‘in various forts ... the loose stones below the vitrification (sic) are supported by ordinary masonry, which apparently might have been carried up the whole way had the builders so desired,’ and (2) the greatest amount of vitrifaction is generally found ‘where a strong parapet is least needed’. He also thinks that the forts were used for signalling, and that the vitrifaction was produced by smouldering beacon fires.

PAGE 288, note 1. Canon Greenwell, in a paper to which I have already referred (p. 676, n. 6), states that in the ‘Danes’ Graves’ ‘by far the larger number [of bodies] were laid on the left side, and ... about one-third had the head pointing to north-east. It does not seem, however,’ he adds, ‘that the dead were laid with the face turned towards the sun.’ He also observes that ‘the presence of charcoal in close proximity to an unburnt body, so universal a feature in the graves of the Bronze Age [on the Yorkshire Wolds] was not noticed except in a few cases, at the Danes’ Graves, and in these it may only have been there accidentally’.

PAGE 379, note 3. I omitted to mention Professor Karl Pearson’s method of estimating stature. Dr. Beddoe, who has criticized it in Journ. Roy. Inst. Cornwall, xv, 1902, pp. 163-71, thinks that it ‘probably underestimates the stature of tall men’, and observes that in one case, where the actual height of living men had been ascertained, Professor Pearson’s calculations erred considerably more than his own or M. Manouvrier’s.

PAGE 408, note 4. See also Biometrika, iii, 1904, pp. 243-4.

PAGE 490, note 5. Further evidence of tin mining in the Scilly Islands will be found in Memoirs Geol. Survey,--The Geology of the Isles of Scilly, 1906, pp. 10, 11.

PAGES 504-5. ‘Is he not aware ... distinguished.’ C. Müller, in his edition of Ptolemy’s Geography (p. 106), argues that Pliny’s Vectis cannot be the Vectis which was the Isle of Wight, because he mentions it among the islands between England and Ireland. But if this argument is sound and Pliny did not blunder, it remains true that there is no evidence in Pliny or any other writer for identifying Ictis with Vectis.

PAGE 527, note 9. Mr. Spurrell has proved that a slight subsidence has taken place in the London district since the Roman occupation. See also p. 566, supra.

PAGES 581-3. ‘There is one passage ... anchorage.’ The argument in this paragraph depends upon the assumption that the ships which were ‘carried back’ (referrentur) in 55 B.C. to the port from which they had started were laid to, and that they could not lie within less than four points and a half of the wind, or, if, with Falconer and James Smith, we allow six points for leeway, within less than two and a half. It may not perhaps be impertinent to explain that when a ship is laid to her head is brought as close to the wind as possible without losing way, under little sail. In these circumstances the helm is often lashed, and the ship then has a tendency to yaw, that is, to fall off the wind and come up again alternately. An experienced seaman with whom I have recently discussed the matter believes, however, that it might have been practicable for the captains of the ships in question to adopt the plan of ‘head reaching’, that is, working to windward under comparatively low sail but still enough to enable the steersmen to keep the ships under control, to luff up to dangerous seas, and to make considerable headway. As he remarked, the difference between lying to, when the helm is not lashed, and ‘head reaching’ is simply one of degree: in ‘head reaching’ the vessel carries a little more sail; and, judging from illustrations of ancient ships, he thinks that Caesar’s would have been able, when ‘head reaching’, to lie within two points of the wind. Allowing five points for leeway, he holds that they could have made good a course within seven points; and he insists that an ancient ship with its sails properly trimmed could have gone as near the wind as a modern lugger!

This, I need hardly say, is rank heresy, and I do not believe that any one who is conversant with the literature of ancient navigation will accept it. Besides, we do not know whether Gallic ships were rigged exactly like those of the Mediterranean. James Smith (The Voyage and Shipwreck of St. Paul, 1880, pp. 75, 127, 215) may or may not have been right in maintaining that an ancient ship could not make good a course in moderate weather within less than about seven points of the wind; but numerous passages prove that ancient ships, if they could work to windward, could not do so without difficulty. If my heretical friend is right, why were Caesar’s cavalry transports windbound in the ulterior portus, less than eight miles from the port which their captains wished to make, and why were they obliged by contrary winds to put back, even in moderate weather, after they first started from the ulterior portus? M. Jules Vars, indeed (L’art nautique dans l’antiquité, 1887, pp. 181 ff.), apparently doubts whether ancient ships could work to windward at all; but a passage which James Smith quotes from one of Cicero’s letters (Fam., xiv, 5, § 1)--cum sane adversis ventis usi essemus tardeque et incommode navigassemus--and another in Livy, which I am obliged to quote at second hand--hinc atque hinc ad ventos obliqua transferre vela--suggest, if they do not prove, that they could. I hope to discuss the question soon in another place. Meanwhile it is enough to say that if Caesar’s cavalry transports could not work to windward, they could neither ‘head-reach’ nor lie to; and those which were carried back to their port of departure must have fetched it by sailing with the wind nearly abeam. But those which were ‘driven down in great peril to the lower and more westerly part of the island’ (ad inferiorem partem insulae quae est propius solis occasum magno suo cum periculo deicerentur) were evidently scudding; and while the word referrentur (‘were carried back’), which is strongly contrasted with deicerentur, suggests that the others were laid to, Caesar’s words show that they were comparatively in little danger, though, if they had not been brought close to the wind, they would evidently have run as great a risk as the others. A distinguished admiral suggests to me that they did sail with the wind nearly abeam, and disposes of the question of danger by assuming that Caesar exaggerated its force. But, with due deference, I would ask whether, since it was strong enough to drive ashore and wreck ships which were riding at anchor, and since the ships that ran before it were ‘in great peril’, ships sailing with it nearly abeam would not have been in peril equally great. It is universally admitted that ancient ships could sail with the wind abeam: it seems hardly credible that the discovery should not have followed that Gallic ships, which were not shallow, could also make headway within less than eight points of the wind; and I therefore believe that, however rare, from one cause or another, tacking may have been, ships were occasionally laid to in bad weather.

Anyhow, my argument remains unshaken. If the ships which were carried back to the ulterior portus neither lay to nor ‘head-reached’, it is self-evident that they could not have fetched Sangatte, and therefore that the ulterior portus was Ambleteuse: if they sailed with the wind abeam, they could not have fetched any port east of Wissant: if they were laid to, they could not have fetched Sangatte; nor could they have done so even if they ‘head-reached’ unless they possessed a capacity for working to windward with which few modern ships and no ancient ship can be credited. Indeed they could not have done so even then; for it would evidently have been impossible for them to beat across the Channel in a single tide: the flood (see p. 583) would hardly have helped them; and the ebb, prolonged and strengthened by the north-easterly gale, would have carried them far out of their course.

The foregoing remarks apply of course to other passages in which I have mentioned the cavalry transports. By no means could they have returned either from near Hythe or Lympne to Ambleteuse or from near Pevensey to the mouth of the Authie.

PAGE 680. ‘Therefore, unless ... Bekesbourne.’ Mr. George Barrow of the Geological Survey, whom I have consulted, thinks that there is no reason to suppose that the relative level of water and banks in the Little Stour above Littlebourne was appreciably different in 54 B.C. from what it is now.

PAGE 698. ‘The claim ... better grounds.’ Mr. F. H. Baring (Eng. Hist. Rev., October, 1907, pp. 726-8) argues that Caesar crossed the Thames by ‘an undoubted ford just above tidewater at Hampton’, which was mentioned by Lord Lumley in 1685. He argues that, owing to the tides, Brentford would only have been available for ‘five or six hours out of twelve’; that ‘this important fact would surely have been mentioned’; and that the stakes which have been found at Brentford are fifteen inches round and therefore too large. I am not sure that he is referring to the stakes described by Mr. Montagu Sharpe, whom he does not mention; but surely the stakes would not have answered their purpose worse for being strong. I do not believe that Caesar would have mentioned the presence of tides at Brentford if they had not affected his passage; but, as we have seen (p. 696), Mr. Spurrell has argued that the tides did not reach Brentford. If there was a ford at Hampton before the days of locks and weirs, there was another, if the evidence of the name Halliford is trustworthy, at or near Coway Stakes. Are we entitled to assume that either existed in the time of Caesar? The reader will have understood that I do not pin my faith to Mr. Sharpe’s theory: I only think that it is less feebly supported than any other.

PAGE 712, note 2. W. Sternkopf (Hermes, xl, 1905, p. 37) rejects Boot’s emendation, proximis, and conjectures that Cicero wrote proxime, meaning that the letter in question was the most recent of several which had arrived together.

INDEX

[Only such references to authors are included as seemed likely to be useful.]

Abbeville, 4.

Abercromby, J., 6, 168 n. 1, 182 n. 2, 185 n. 2, 192 n. 1.

Aberdeenshire, 81, 151, 194 n. 3, 361 note; stone circles of, 208-10; brachycephalic skeletons found in, 426; ethnological observations in, 442.

Abnoba, 272.

Abydos, 123 n. 1.

Acton, 40 n. 2.

Addedomaros, 363 n. 6.

Adriatic, 243.

Adzes, 75.

Aedui, 331-3.

Aegean Sea, 9.

Africa, 5, 9, 30, 66-7, 82, 123, 197, 205, 211, 251 n. 1, 263 n. 2.

Africus, 555.

Agricola, 226.

Agricultural implements, 76, 89. See Ploughs.

Agriculture, in Neolithic Age, 89-91; in Bronze Age, 151-2; in Early Iron Age, 252-4.

Ain, 296, 410-1.

Airy, Sir G., 558-63, 616-8, 645-8, 652, 655, 659.

Aisne, 110 n. 1.

Akerman, R. Y., 5.

Alderbury, 27.

Aldington Knoll, 639-40.

Allen, J. Romilly, 6, 149 n. 5, 232 n. 2.

‘Alpine’ race, 127-8, 437, 455.

Alprech, 306, 570, 572-3.

Altars, 274, 280; of Paris, 276, 279 note; of Sarrebourg, 281.

Amber, 131; cup, 159 n. 1; ornaments, 162-3, 167-70, 183, 469; trade, 218, 357; amber coast visited by Pytheas, 220, 223-4.

Ambidexterity (alleged), 82 n. 5.

Ambleteuse, cavalry transports sailed from, in 55 B.C., 313, 318-9, 587-8, 593, 616, 639; not identical with Portus Itius, 552, 563-4.

Ambresbury Banks, 137 n. 4.

America, 4, 31, 51, 78, 93-4, 125, 205.

Amesbury, 189.

Amiens, 4.

Ammianus Marcellinus, 273.

Amminus, 370 note.

Amulets, 92-3, 116, 205, 260.

Ancalites, 346, 700.

Ancestor-worship, 116, 283-4.

Ancyra, 363-4, 368.

Andate, 297.

Anderson, J., 7, 135 note.

Andred (Anderida), 501, 507, 614.

Anglesey, 107, 139, 160, 194 n. 3; hut-circles in, 154; Druids, 292.

Animals, worship of, 55, 284; domestication, 55-7; interment, 114, 203, 288; sacrifice, 202-3, 288.

Animism, 50-1.

Antedrigus, 370.

Anthropology, 8, 234; methods of, 376-9, 457 n. 4.

‘Anthropomorphic ape,’ 48; designs in Late Celtic art, 236.

Anthropomorphism, 286.

Anvils, 79.

Appach, F. H., 314 n. 2, 639-44.

Appledore, 533-4, 538-43, 545-6, 549-50, 552, 604, 638, 640, 643. See Rother.

Apollo, 275, 280.

Aquae Calidae, 232 n. 3.

Aquitaine, 40.

Aquitani, 304.

Arbois de Jubainville, H. d’, 228, 281, 291 n. 2, 292 note, 407-8, 410-2, 415 note, 416 n. 6, 421-3, 446-7.

Arbor Low, 476 n. 3.

Archaeologia, 3.

Archaeology, caution needed in, 7; as an aid to ethnological inquiry, 375-6.

Archers, 313, 316, 331, 346, 698.

Arctic animals, 20-1, 40; plants, 15, 19.

Arctic Circle, 225-6.

Argentocoxos, 423, 450.

Argyllshire, 102, 107-8, 185 n. 5, 194 n. 3, 205, 208.

Ariovistus, 300-1.

Armorica, 501, 507.

Arran, 115 n. 8, 208; cairns, 108; skulls, 394.

Arras, interments, 234 n. 1; torque, 241 n. 1; skeleton, 436 note; chariot-burial, 676.

Arreton Down, 145, 182 n. 5.

Arrow-heads, 3, 80-2; bronze, 132 note.

Artabri, 484-5, 487-8, 492.

Artillery, in Caesar’s army, 313, 347.

Artists, palaeolithic, 35, 47, 83.

Aruntas, 53, 199, 463.

Aryans, 153, 272, 275.

Asia, 4, 125, 205; Asia Minor, 31.

Astronomy, 10; astronomical theory of Stonehenge, 216-7, 472-6, 480-1; of other stone circles, 481-2.

Atecotti, 448.

Atkinson, J. W., 5.

Atrebates, 232, 235, 347, 366, 451-2, 454, 695; of Gaul, 300, 309, 364-6.

Atrius, 336, 338, 686.

Atticus, 329, 333, 350.

Augusta, 255, 704.

Augustus, 306, 357, 361, 363, 366-7, 369.

Aulus Plautius, 362 n. 7, 371.

Aurochs (urus), 68, 88.

Aurunculeius Cotta, 314, 324.

Austen, Godwin, 4.

Australia, 49, 51, 58, 72, 116, 197, 199, 206, 377, 463.

Austria, 199.

Authie, 306, 327, 558, 613.

Avebury, 209, 213, 240.

Avebury, Lord, 45. See Lubbock.

Avienus, Festus, 218, 490-2, 512.

Avon (Hampshire), 25, 27.

Awls, bone, 42; stone, 80; bronze, 140, 182.

Axe-hammers, 78.

Axes, stone, 3, 73, 131; bronze, 144; engravings of, 205; iron, 253.

Aylesford, 5; bucket, 237; pottery, 242-4; interments, 287-8; Caesar may have crossed Medway at, 344.

Babylonia, 125.

Badbury Rings, 134, 137.

Balbus, 328.

Balearic islanders, 111.

Baltic, 14.

Banffshire, 194 n. 3; stone circles of, 208-9.

Baoussé-Roussé, 34, 382.

Barbed arrow-heads, 80-1, 477 note.

Bards, 266.

Barham, 679-82.

Barrows, 3, 6, 8, 11. See Long barrows, Round barrows.

Barton Mere, 154.

Basket-work, 197.

Basques, 94, 399-400, 455; Basque language, 407, 420.

Bateman, T., 5, 6 n. 1.

Bath, 207.

Bâtons de commandement, 48 n. 1.

Batten Promontory, 397.

Battersea, 158.

Battle-axes, stone, 78, 131; bronze, 132 note, 145.

Baumes-Chaudes, 93, 401.

Beads, lignite and shale, 92; amber, 163, 167, 469; glass, 162 n. 1, 167-8, 170, 183, 241, 357, 469; jet, 167.

Beansale, 260 n. 1.

Bear, brown, 30, 68; grizzly, 30, 68.

Beavers, 68.

Beddoe, J., 8, 385-6, 393-6, 425-6, 434, 437, 440-1, 445, 448.

Bedford, 40 n. 2.

Beech, 600, 661-2.

Beechey, Sir F. W., 607, 608 n. 3, 612 n. 6.

Beehive huts, 65, 155, 262.

Beer, 224, 261.

Bekesbourne, 679-80.

Belerium, 221, 499, 502 n. 8, 507.

Belgae, 232-6; not identical with ‘round-heads’ of Bronze Age, 429-30; date of their invasion of Britain, 233-4, 434, 445-6, 456, 459-60; theory that they were Goidels, 449-54; British tribe of that name, 232, 454, 460; of Gaul, 250, 300-1.

‘Belgic ditches,’ 11.

Belgium, 30, 38, 82.

Belisama, 281.

Bell barrows, 175, 176 n. 1, 185 n. 3, 470.

‘Bell-pits,’ 516-7.

Bellovaci, 365.

Beltout, 97.

Bere Regis, 162.

Bericus, 370-1.

Berkshire, 36, 101, 130, 134, 138, 153 n. 9, 194 n. 3, 235, 240, 251, 267 n. 3, 347.

Bertrand, A., 9, 438 n. 3.

Berwickshire, 194 n. 3, 262.

Bexhill, 619.

Bibracte, 237, 283, 286 n. 2, 584-5.

Bibroci, 346, 700.

Bigbury, 256-7; stormed by 7th legion (?), 337, 685.

Bill-hooks, 253.

Birdlip, 435 n. 1.

Bits, horses’, 152, 238, 264-5.

Blancnez, Cape, 306, 565.

Blandeno, 329, 727.

Blashenwell Farm, 63.

Bleasdale, 179.

‘Blue-stones,’ 214, 470, 479-80.

Boadicea, 269, 296-7, 347.

Boars, 68, 88, 202, 357, 407; worshipped, 284.

Boduni, 362 n. 7.

Bognor, 19.

Boissier, G., 277.

Bokerly Dyke, 260 n. 1.

Bol, 451 n. 2.

Bone implements, 5, 42, 82, 158, 160.

Bonnington, 604, 639-40, 642.

Bononia, 591 n. 1.

Borreby, 430, 441.

Bos longifrons, 88.

Boudicca. See Boadicea.

Boulder clay, 15, 18, 20, 22-3.

Boule, Monsieur M., 14 n. 1, 29, 30 n. 3.

Boulogne, 306, 552-3, 557, 572-7, 579, 584; identical with Portus Itius, 585-95, 643, 652, 658.

Bournemouth, 25, 32.

Bowl barrows, 175, 176 n. 1, 185 n. 1, 470.

Bracelets, 158, 163, 165; Late Celtic, 241.

Brachycephalic invaders. See Round-headed invaders.

Brandon, 23, 44. 69-70.

Brassempouy, 198, 383 n. 2.

Breeches, 264-5.

Breithyn, 460.

Brentford, Caesar may have crossed Thames at, 344, 661, 697-8, 741.

Bretain, 460-1.

Bridge, 680-1.

Brigantes, 269, 359, 360 n. 2.

Brigantia, 280.

Brigit, 280.

Bristol Channel, 20, 62, 224.

Britain, materials for ‘prehistory’ of, 1-12; continental during part of Pleistocene Period, 19-22; origins of civilization neolithic, 63-4; geography of neolithic, 64; comparative backwardness of bronze culture, 171; political and social conditions of, compared with those of Gaul, 270-1; ties between Britons and Gauls, 299-300; how Britons were affected by Caesar’s campaigns in Gaul, 300; Caesar’s invasions of, 301-54; history of, between Caesar’s departure and A.D. 43, 355-72; Roman conquest of, 372.

Britanni (Gallic tribe), 235, 459-60.

Britanni, the name, 459-61.

British Association, 4.

British Isles identified with Cassiterides, 491, 493-4, 497-8.

British Museum, 9, 12 n. 1, 70, 158, 217 n. 1.

Brittani, 459-60.

Brittany, 19, 65, 109, 111, 155 n. 1, 171, 185, 194, 200, 212.

Brittones, 459-60.

Brixham Cave, 37.

Broad Down, 159 n. 1.

Broca, P., 9, 93, 376-7, 381, 430, 437-9.

Brochs, 7, 262, 464.

Bronze Age, 6-7, 121-230; period of its commencement, 126-7; no apparent interval between, and Roman occupation at Cranborne Chase, 267; origin of bronze, 9, 124-6; of British bronze culture, 126; bronze implements used for centuries in Britain before Iron Age, 123-4; social organization of invaders, 128-9; their settlements, 129-31; coins, 249, 358; implements and weapons, 140-9; hoards, 149-50; ornaments, 158, 161-7; backwardness of culture in Britain, 171. See Transition.

Brooches, 183 n. 6, 232, 240, 264.

Brutus, Decimus, 304.

Bruyos, islands of, wrongly identified with Cassiterides, 488.

Bryce, T. H., 7, 435 n. 1.

Brythonic, 421-3.

Brythons, 228-9, 232-6, 409-12; date of their invasion of Britain, 233-4, 445-6; religion, 274; did they hold aloof from Druidism? 290-1; not identical with early round-headed invaders, 428-40; ethnology of, 444-53, 456, 458-61.

Buckets, 237, 246.

Buckinghamshire, 36, 153, 194 n. 3.

Buckland, W., 4.

‘Bulb of percussion,’ 44.

Bulford, 183.

Bull, wild, 48.

‘Bull-roarers,’ 49.

Bulverhythe, 604, 621 n. 3.

Burghead, 259.

Burial customs, 8; neolithic, 110-2, 114-5; of Bronze Age, 173, 184-91, 200-5; of Early Iron Age, 286-8.

Bury St. Edmunds skull, 33, 380, 455.

Bute, 94 note, 110, 185.

Buttons, 161-2, 189; gold, 131, 162; stone, bone, wood, and jet, 161; with V-shaped holes, 161-2; bronze and ivory, 162; in interments, 189.

Caddington, 25, 39, 44, 60.

Cae Gwyn, 23.

Caermarthenshire, 35.

Caesar, Gaius Julius, his limited knowledge of Britain, 1-2; mentions no Gallic invaders of Britain except Belgae, 232-3; wrote to Q. Cicero in Greek characters, 266; on Britons of the interior, 267; on social and political conditions of Gaul and Britain, 268-71; on Gallic religion, 274-8, 281-2, 285; on Druids, 289-90, 292-6, 298 n. 6; how his Gallic campaigns affected Britons, 300; first invasion of Britain, 301-25; landing-place, 309-12, 315-6, 595-665; disembarkation, 316-7, 673; sites of his camps, 317, 335, 673-4; credibility of his narrative, 319, 349 n. 3, 666-72; forces passage of Stour near Canterbury and storms a British fort, 337, 678-85; his earlier operations in 54 B.C., 337-8, 685-8; second combat with the Britons, 339-41, 688-91; crosses the Thames, 345-6, 692-9; storms Cassivellaunus’s stronghold, 347, 699-702; results of his invasions, 355-72; his reform of the calendar, 707-26. See Table of Contents.

‘Caesar’s Camp’ (Folkestone), 134 n. 12.

Cairns, 67. See Chambered barrows, Horned cairns, Round barrows.

Caithness, chambered cairns of, 102, 106; brochs, 262; skulls, 394.

Calais, not Portus Itius, 306, 552, 565; coast between, and Somme, 517-8.

Caledonians, 361 note, 416-8.

Calendar, of Coligny, 296; Julian, 707-26.

Caligula, 369.

Callernish, 207, 209, 215 n. 3.

Calleva, 255, 366; inscription found at, 451-2.

Calver, E. K., 607-10.

Cambrian Archaeological Association, 5.

Cambridgeshire, 36, 194 n. 3.

Camden, W., 2, 12, 213, 345.

Camps, 7, 11, 156-7. See Hill-forts.

Camulodunum, 255, 278, 299 n. 5, 359, 362.

Camulogenus, 342.

Camulus, 278, 454 n. 4.

Canche, 306, 324, 327, 518, 586.

Cann, 267 n. 2.

Cannibalism, 113-4, 268 n. 1.

Canoes, 61.

Canterbury, 253, 336-7, 339, 344, 660, 682-5.

Cantii, 255, 366, 454.

Cantium, 663.

Cantonal organization, 233 note.

Caractacus. See Caratacus.

Caratacus, 370-1.

Cardiganshire, 139.

Carl’s Wark, 135.

Carn Brea, 134, 258.

Carnarvonshire, 66 n. 3, 184, 194 n. 3, 205, 295 n. 1.

Carse of Stirling, 62.

Cartailhac, E., 9.

Carthaginians, Atlantic trade of, 220, 491, 496 n. 5, 512; trade inherited by Veneti, 303.

Cartismandua, 269.

Cassi, 346, 700-1.

Cassiobury, 701.

Cassiterides, 483-98, 514.

Cassiteros, 494.

Cassivellaunus, his opposition to Trinovantes, 300, 327; elected commander-in-chief against Caesar, 339; his military operations, 341-7, 669; sues for peace, 349-50, 670-1; distance of his country from the sea, 692-3, 704; site of his stronghold, 699-702. See also 731-3.

Castle Law, 258-9.

Cattedown Cave, 380, 455.

Catti, 360.

Cattle, 57 n. 2, 88, 151, 354, 357, 406.

Catuvellauni, 232, 235, 299 n. 5, 354, 361-2, 701, 704; oppose Trinovantes, 300, 309; Commius tries to check, 365; their territory, 454.

Cauldrons, 158.

Cavalry, Gallic, raised by Caesar, 301, 313, 331; Caesar feels want of, in 55 B.C., 317, 320, 324, 635-6; operations of British, in 55 B.C., 316, 321; in 54 B.C., 336, 339-40, 342-3; Caesar’s, fail to reach Britain in 55 B.C., 318-9; service rendered by Commius’s, 323; operations of Caesar’s, in 54 B.C., 335-7, 339-46, 616, 686-91, 698-9; how British fought in conjunction with chariots, 674, 676-7, 688-91; British depicted on coins, 689. See Transports.

‘Cave-men,’ 22, 38, 383-5.

Caversham, 32.

Caves, exploration of, 4-5; frescoes in Pyrenaean, 35 n. 3; implements found in, 36, 42; inhabited in Palaeolithic Age, 37, 46; implements compared with those of drift, 38-40, 383-5; neolithic relics in, 67; inhabited and used as sepulchres in Neolithic Age, 100; and in Bronze Age, 178; still inhabited in Early Iron Age, 261; neolithic skeletons found in, 393-6. See Brixham, Creswell Crags, Heathery Burn, Kent’s Cavern, Rains Cave, &c.

Cavities under barrows, 202 n. 3.

Cefn, 395.

Celtac, 430, 436, 438-9, 444.

Celtic languages, 8, 127-8, 433. See Brythonic, Goidelic.

‘Celticans,’ 413, 445.

Celts, not identical with earlier round-headed invaders, 127-8, 428-40; date of their invasion of Britain, 128, 229, 233-4, 432-3, 445-52, 455; ethnology of, 433-40, 444-54, 456-8. See Brythons, Goidels, ‘P’ Celts, ‘Q’ Celts.

Celts, stone, 69, 75-7; bronze, 140-1, 144; socketed, 126, 141, 144, 148-9; celts in interments, 181 n. 3, 183; as amulets, 205; bronze socketed celt associated with silver coins, 267 n. 2.

Cemeteries, of Bronze Age, 178-9; of Early Iron Age, 287-8. See ‘Danes’ Graves’, Scorborough Park.

Cenimagni, 346-7, 700.

Cenotaphs, 180-1, 212.

Cephalic index, 396-8.

Cereals, 64, 89, 407. See Corn.

Cerne Giant, 206 n. 5.

Cernunnos, 282 n. 2, 284.

Chalk, used for manure, 253, 515-7; chalk downs, prehistoric settlements on, 68, 130; cultivated, 90, 660 n. 4.

Chambered barrows, compared with dolmens, 65-6; chambered long barrows, 101-6, 173-4; round, 101-2, 107-8, 173; chambered cairns of South-Western Scotland, 108-9; interments in chambers, 112, 173. See Bleasdale, Horned cairns, Ronsay, Wor Barrow.

Chambers, subterranean, 87.

Chancelade, 35, 382, 385 n. 1, 455.

Chanctonbury Rings, 97.

Channel, English, 16, 19 n. 1, 32, 300, 303, 311, 334, 372.

Channel Islands, 19, 102, 194, 196, 222, 501.

Chariots, 312, 328; oppose Caesar’s landing, 316, 597, 642-3; operations of, in 54 B.C., 339-45, 352-3, 689-92, 699; general account of, 674-7.

Chellean implements, 39 n. 1, 41 n. 3.

Cherbury, 138.

Cheshire, 16, 208, 235, 252, 279.

Chevron, 149, 162, 197-9, 236.

Chiefs, barrows erected in honour of, 100-1, 114, 117, 177-9. See Kings.

Chifflet, J., 2.

Chilham, 682-3.

Chinese may have been the first to use bronze, 125.

Chipping of tools, 44, 73-5.

Chisels, stone, 75, 77; bronze, 140-1, 144.

Chishull, E., 363.

Christison, J., 7, 87 n. 1, 259 n. 3.

Chronology, palaeolithic, 31-2; neolithic, 32 n. 1, 62-4; of the Bronze Age, 126-7, 432-3; of round barrows, 181-4, 476 n. 1; of Caesar’s invasions of Britain, 706-7, 726-35. See Calendar.

Chrysoister, 262 note.

Church, British, 372.

Church Hole, 45 note.

Churingas, 199.

Cicero, M. Tullius, his friendship with Diviciacus, 275; on Deiotarus, 297; his correspondence about Caesar’s invasions of Britain, 327-9, 333, 335, 348-50, 707, 728, 731-5; in relation to credibility of Caesar’s narrative, 666-8.

Cicero, Quintus, 266; with Caesar in Britain, 328-9; his correspondence, 333, 335, 667-8, 728, 731, 734-5.

Cicht, 414.

Cinerary urns, 180, 182 n. 5, 185-7, 190, 193-4, 288, 467; Late Celtic, 242-3, 288.

Cingetorix, 330, 370.

Circles, concentric, on shields, 146; on bronze celts and spear-heads, 149; on an incense cup, 183; symbolical of sun-worship, 198-200, 206-7; on cists and rocks, 205-6; concentric stone circles, 177; megalithic, 209.

Cisalpine Gaul, 326-7, 726-7.

Cissbury, 7; flint mines, 69, 71; date of fort, 98; entrances, 138 n. 1; fort described, 256; implements of palaeolithic type at, 387.

Cists, 8, 173, 177, 179, 186-7, 205.

Civil jurisdiction exercised by Druids, 293-4.

Civitates, 269.

Clans, 52-5, 118, 128, 269, 271. See Intertribal war.

Classis Britannica, 545, 589 n. 1, 664.

Claudius Caesar, 239, 249, 357-9, 371, 589.

Claverdon, 260 n. 1.

Cleveland, 184-6.

‘Cliff-castles,’ 135, 136 n. 1.

Climate in Palaeolithic Age, 25 n. 3.

Clothing, of palaeolithic man, 47; neolithic, 91; in Bronze Age, 156, 160-2; corpses interred in clothes, 189; Late Celtic clothing, 264-5.

Clwyd, Vale of, 23.

Clyde, 62.

Coast, between Calais and the Somme, 517-8; of Kent in Caesar’s time, 518-52.

Cock-fighting, 264.

Coffey, G., 169 n. 2, 170 n. 1, 172 n. 2, 198 n. 3.

Coffins, 159 n. 1, 173.

Coins, British, 5, 11; Dumnonii had none, 222, 501; probably first struck in Britain by Belgae, 233-4; general account of British, 248-50; inscribed, show knowledge of writing, 266; religious significance of, 273; as materials for British history, 357-70; found near Sandown Castle and at Stonar, 520-1; and at Deal, 524; chariots depicted on, 675; also mounted warriors, 689.

Colchester, 255; inscription, 415. See Camulodunum.

Coles, F. T., 7, 476.

Coligny, calendar of, 296.

Collignon, R., 377, 399-400, 431 n. 1, 439 n. 1.

Columba, 421.

Commentaries, Caesar’s, 2, 10, 267, 298 n. 6, 339, 342, 348, 659; credibility of, 666-72; certain passages in, discussed:-- quod inde ... traiectus (iv, 31, § 3), 554, 571; tertia fere vigilia (iv, 21, § 3), 615; montibus angustis (iv, 23, § 3), 597 n. 2, 613-4, 641-2, 652; ad horam nonam ... exspectavit (iv, 23, § 4), 597, 648-9; aliae ad inferiorem ... deicerentur (iv, 28, § 2), 636; quo ex portu ... continenti (v, 2, § 3), 558, 562, 619; sub sinistra conspexit (v, 8, § 2), 575-7, 616, 643, 655; aestus commutationem secutus (v, 8, § 3), 657; summa tranquillitate (v, 23, § 6), 617-9. See Mollis.

Commius, opposes Cassivellaunus, 300; sent by Caesar to Britain, 309, 597; arrested, 318; service performed by his cavalry, 323; accompanies Caesar to Britain (54 B.C.), 333; negotiates for Cassivellaunus, 349, 670-1; his dynasty, 361; rebels against Caesar, 364-5; his conquests in Britain, 365-6, 371.

Comparative method, 8.

Concubines, 270 n. 2.

‘Continuance Theory,’ 295.

Contorted drift, 25, 39, 60.

Contracted position, 110-1, 187, 287.

Cooking, 46, 89, 155.

Copper Age, 121-2, 739; mining, 139, 251; metallurgy, 139-40; coins, 249, 358, 362; imported into Britain (?), 252 n. 1; stain of, at Stonehenge, 470.

Coracles, 247.

Coral, 237, 240, 265, 739.

Corbilo, 221, 500-1, 507-8, 512.

Cores, 149.

Corinium, 255.

Coritani, 235, 390-1, 452.

Corn, not grown in Palaeolithic Age, 46; ear of, sculptured in cave at Lourdes, 57; grown on chalk downs, 90, 660 n. 4; cultivated in Bronze Age, 151-2, 224; how threshed, 224; cultivated largely in Early Iron Age, 253-4, 260; remains of, at Hunsbury, 267; reaped by 7th legion, 321; supplied by Trinovantes, 346; exported in Strabo’s time, 357.

Cornavii, 235.

Cornwall, 20, 101-2, 115 n. 8, 130, 134, 140, 152, 154, 174, 205, 239, 250, 359-60; identified with Cassiterides, 493, 497; cairns, 107 n. 8, 108; hill-forts, 135; mining, 139, 249, 502-3 n. 8, 509-11; huts, 153-5, 261; poor in relics, 168, 200; interments, 184; peculiar pottery, 195 note; megalithic monuments, 208, 209 note, 211 n. 1; trade, 218, 501-2, 504-7, 509-11.

Corus, 555 n. 2.

Cosmas Indicopleustes, 226.

Côte-d’Or, 436.

Cotentin, 304.

Cotswold Hills, 176, 210.

Councils, 270, 330, 332.

Couvade, 94-5, 407 n. 6.

Coway Stakes, 693-7.

Cowlam, 183 n. 6, 231.

Cow’s tooth, 203.

Cranborne Chase, 103 n. 2, 105, 138, 175 n. 4, 177 n. 2, 179, 201 note, 267. See Bronze Age, Handley Down, Handley Hill Camp, Martin Down Camp, Rushmore, South Lodge Camp, Wor Barrow, Woodcuts.

Cranbrook Castle, 135 note.

Cranial forms, 378.

Craniology, 3. See Cephalic index, Cranial forms, Ethnology.

Crannogs, 263 n. 2, 463.

Crassus (the triumvir), 334.

Crassus, Publius (consul of 95 B.C.), 494.

Crassus, Publius (son of the triumvir), 303-4, 485, 488-9, 493-7.

Crayford, 42, 44.

Credibility of Caesar’s narrative, 349 n. 3, 666-72.

Cremation in Neolithic Age, 110, 401; in Bronze Age, 184-90, 229; religious significance of, different from that of inhumation (?) 204-5, 465-6; practised by Belgae, 234 n. 1, 286; Sergi’s views on, 400-1.

Creswell Crags, 35, 37, 39, 47, 382. See Church Hole, Robin Hood Cave.

Crete, 5.

Criminal jurisdiction, 293-4.

Crinan, 177 n. 4.

Crocodiles in London Clay, 14.

Croham Hurst, 85.

Croll, James, 31.

Cro-Magnon, 382, 397 n. 8.

Cromer, 23.

Cromlechs, 65 n. 3.

Croydon, 85, 87.

Cruthni, 411-2, 418-9.

Cumberland, 101, 194 n. 3, 205, 247 n. 3.

Cumbrian mountains, 14.

Cunnington, W., 3, 186 n. 4, 188 n. 6, 469 n. 7, 480.

Cunobeline, 361-3, 366, 368-70, 372.

Cup-markings, 205-6.

Currency-bars, 250-1.

Curvilinear decoration, 236-9.

Cyprus, 198 n. 3.

Daggers, 82, 131, 182; flint, 80; bone, 82; bronze, 131, 140, 145, 147; decoration of, 149; with handles of amber and ivory, 162; in interments, 182, 201; carved dagger-handles (?) 35; Late Celtic sheath, 239.

Dalmatia, 367, 442.

Dananns, 391 n. 5.

‘Danes’ Graves,’ 288 n. 1, 435 n. 1, 739.

Danube, 170.

D’Anville, J. B. B., 2, 634. See Stadia.

Darent, 25.

Dartmoor, 6, 96; hut-circles, 154-6; kistvaens, 189 n. 1; stone rows, 208, 212; ‘blue-stones’ fetched from (?), 214; tin, 503 note.

Darwin, Sir G., 606.

Daventry, 110 n. 12.

David, Félicien, 275.

Davis, J. B., 5, 8, 394.

Dawkins, W. Boyd, 5, 23 n. 7, 40 n. 2, 45 note, 84 n. 2, 125 n. 2, 129 n. 2, 383-90, 400 n. 3, 406-7, 429, 465-6.

Dea Arduinna, 272.

Deae matres, 283.

Deal, amber found at, 168; reconnaissance of, by Volusenus, 311; Caesar’s landing at, 316; ancient topography of, 519, 521-4; Caesar drifts to latitude of, 575; theory that Caesar landed at, in 55 B.C., 604-5, 626-7, 632, 641 n. 1, 644-62, 664-5; distance between, and Great Stour, 682; castle, 521-3, 525, 604, 664.

Deer, 68, 88, 91, 113.

Deer-horn implements, 74, 158, 160. See Picks.

Deiotarus, 297.

Demeter, 210 n. 5.

Denbighshire, 35, 184, 188 n. 2, 194 n. 3.

Dene-holes, 253, 515-7.

Denmark, 66, 77, 89, 105 n. 2, 127, 144, 172, 194.

Derbyshire, 5, 35, 37, 68, 83, 102, 110 n. 5, 129-30, 135, 165, 167, 175-6, 194 n. 3, 208, 235; chambered cairns of, 108; poor in ornaments, 169; interments in, 184, 185 n. 3, 188 n. 2.

Derwent Moor, 108.

Desjardins, A. E. E., 562 n. 2, 573 n. 2, 583, 586, 589-90, 591 n. 1.

Deux-Sèvres, 410-1.

Devonshire, 20, 35, 117, 130, 133, 135, 155, 162, 174, 194 n. 3, 208, 359-60; cists in, 179; interments, 184.

Devonshire Association, 6.

Dew-ponds, 138.

Dewlish, 27 n. 5.

Dilston Park, 192 n. 1.

Diodorus Siculus, 1, 111, 219, 223, 265, 294, 358.

Diogenes Laertius, 289, 291 n. 2.

Dion Cassius, 1, 360 n. 4, 628, 630-1, 650-1.

Dis Pater, 281-3, 296.

Disk barrows, 175-6, 183 n. 1, 189, 208; near Stonehenge, 469; near Rollright Stones, 470.

Disks, bronze, 158; gold, 167; symbolical of sun worship, 207.

Diviciacus, 299-300, 454 n. 4.

Diviciacus (the Druid), 275, 298 n. 6, 332.

Division of labour, 83.

Dobuni, 255, 362, 370.

Dogger Bank, 22.

Dogs, not domesticated in Palaeolithic Age, 46; used for hunting and eaten in Neolithic Age, 88; interment of, 114.

Dolmens, 8; evidence of, as to origin of later neolithic invaders of Britain, 65-7, 402-5.

Dolo, 644 n. 2.

Domestic animals, introduced into Britain by neolithic immigrants, 64, 88-9; largely eaten in Bronze Age, 151; bones of, in barrows, 202; European origin of, 406-7.

Domestication of animals, 55-7.

Domnoveros, 360 n. 2.

Dorchester (Oxfordshire), 246.

Dorsetshire, 27 n. 5, 32 n. 2, 96, 101, 116, 130, 133-4, 151, 156, 160, 162, 177 n. 2, 178, 194 n. 3, 205, 208, 242, 251; absence of chambered barrows in, 102; bowl and disk barrows, 175-6; interments, 184, 187-8, 190, 203; incense-cups rare, 196.

Dover, 336, 604, 616, 641; Straits of, in Ice Age, 16, 22; time of high tide at, on Aug. 26-8, 55 B.C., 609-10; inscribed tiles found at, 664.

Dover Castle, 32, 311, 315, 530 note, 653.

Dover cliffs, 315, 627, 642; erosion of, 528-30; Airy denies that Caesar anchored off, 631-4, 652.

Dover harbour, 310, 314, 530-1, 597 n. 2; Caesar hoped to land in, 314, 645-6, 651-2.

Downs, The, 526-7, 665 n. 3.

Dowris, 152.

Dress. See Clothing.

Driffield, 240. See ‘Danes’ Graves.’

Drinking-cups, 6, 97 n. 2, 109 n. 2, 179, 191-2, 194-5, 288, 443, 467.

Druidesses (?), 296-7.

Druids, opposed to anthropomorphism, 83 n. 3, 286; dolmens erected by (?), 114-5; wore ornaments, 163; may have worshipped in stone circles, 210; wrote in Greek characters, 265; political power of, 271, 298 n. 6; influence on Celtic religion, 276, 298; general account of, 289-98.

‘Drums,’ 199.

Dryades, 297 note.

Dubnoreix, 331, 360 n. 2.

Dubnovellaunus, 363, 366-9.

Duck Marsh, 547-8.

Duelling, 261.

Du Fresne, C., 2.

Dumbuck, 463-4.

Dumnonii, 359 n. 12, 447, 501; of Scotland, 447-8.

Dumnorix, 269, 331-3, 358, 360 n. 2.

Dumnovellaunus, 360 n. 2. See Dubnovellaunus.

Dunbuie, 463-4.

Dungeness, 536, 632, 654.

Durham, 101, 130, 133 note, 194 n. 3, 267 n. 1, 359.

Durotriges, 133, 359.

Durovernum, 336. See Canterbury.

Dwellings, exploration of, needed, 11; palaeolithic, 46; neolithic, 84-7; of Bronze Age, 153-60; of the rich in Early Iron Age, 254; of Early Iron Age generally, 256-7, 259-64. See Beehive huts, Caves, ‘Earth-houses,’ Hut-circles, Hut-clusters, Lake-dwellings, Mound-dwellings, ‘Picts’ houses,’ Weems.

Dymchurch, 532, 534-6, 538, 551.

Each End, 519.

Early Iron Age, 5, 7, 9, 230-300. See Iron, ‘Late Celtic’ Period.

Earrings, 165-6.

‘Earth-houses,’ 7. 391.

Earthworks, 7, 11, 98 n. 3, 157 n. 2. See Camps, Flamborough Head, Hill-forts.

East Ham skull, 395, 397.

East Runton, 23.

East Wear Bay, 310, 532, 635, 646, 650-1.

Eceni, 235. See Iceni.

Eggardun, 84 n. 2, 96, 153.

Egypt, 9, 82, 122-5, 171 n. 3.

Ehenside Tarn, 76.

Elephant, 23 n. 7, 48, 386 n. 5; straight-tusked, 37, 40; elephas meridionalis, 27 n. 5.

Elevation of land, in Pleistocene Period, 15-7, 19-22; in Neolithic Age, 20, 62.

Elginshire, 165, 170 n. 1, 194 n. 3.

Elks, 30, 68.

Elton, 178.

Ely, Isle of, 254.

Emporiae, 248 n. 2.

Enamelling, 237-8, 246, 265.

Eoliths, 25-30, 46; eolithic man (?), 379-80.

Epaticcus, 361-2.

Epona, 278-9, 284 n. 5.

Eppillus, 365-6, 452 n. 3.

Equinox, 323, 350; chronological arguments based upon mention of, 709 n. 2, 710-3, 734.

Eratosthenes, 219, 223.

Esk, 134.

Eskimos, 42, 48 n. 1, 102, 388-9.

Essex, 36, 137 n. 4, 235, 242, 253-4, 346, 363 n. 6.

Esus, 276, 278.

Ethnography, 8-9; ethnographic census, 375 n. 1.

Ethnology, 10, 375-458. See Celts, ‘Iberian’ race, Long Barrow race, Mediterranean race, Picts, Pygmies, Round-headed invaders, Skeletons.

Etruria, 158.

Evans, A. J., 5, 9, 171 n. 3, 233, 288, 468-70, 476-8.

Evans, Sir J., 4-5, 7 n. 1, 21 n. 5, 23, 24 note, 26, 28 n. 2, 32 n. 2, 40 n. 2, 59, 74, 126, 232 n. 3, 360, 370 n. 4, 386-8, 432.

Excavation, 6-8, 11, 96, 134 n. 12.

Exogamy, 53.

Extended position, 110 n. 11, 188, 287.

Eyford, 114.

‘Fabricators,’ 74.

Fairlight, 310.

‘Fairy-darts,’ 3.

Falkirk, 62.

Falster, 127, 430, 441.

Fens, 77-8.

Fernacre, 211 n. 1.

Fernworthy. 212 n. 2.

Fetichism, 116.

Fibulae. See Brooches.

Fifeshire, 174, 186, 190 n. 9, 194 n. 3.

Finns, 441.

Fir-Bolg, 451 n. 2.

Fire, how produced in prehistoric times, 46, 80; worshipped, 282, 462.

Fish-hooks, 151.

Fisherton, 84 n. 2.

Five Wells, 108.

Flagon, bronze, 246.

Flakes, flint, 41, 44, 73-4, 79-80, 82; in interments, 115; chips in cinerary urns, 204, 288.

Flamborough Head, earthworks near, 157 n. 2, 441.

Flint quarries, 7, 73. See Cissbury, Grime’s Graves.

Flintshire, 162, 252.

Flower, Sir W., 4.

Folkestone, 531-2, 650, 652; harbour, 310, 634-5, 645-6; theory that Caesar landed at, 604.

Folk-lore, 8-10, 94, 375; of Rollright Stones, 213.

Folkton Wold, 199, 464.

Fontenay, 401.

Food-vessels, 191, 193-4, 196, 467; food-vessel (?) found in Stonehenge, 467.

Fordwich, 660, 683-4.

Forest Bed, 23.

Forests, submerged, 20, 62-3; obstruct settlers, 68, 130; difficulty of clearing, 90-1; gradually cleared, 151, 253; worshipped, 272; Forest of Dean, 251. See Andred, Wealden Forest.

Forfarshire, 167, 194 n. 3.

Forgers, 358.

Forth, 62.

Fowls, 55, 264.

Fox, bones of, in urn, 203.

France, 9, 28, 35, 47, 56, 63, 65 n. 3, 82, 96, 105 n. 2, 109-10, 113, 115, 119, 141, 144, 151, 168, 171, 205, 211-2, 259.

Franks, Sir A. W., 6.

Frazer, J. G., 58 n. 2, 277, 463.

Frere, John, 3-4.

Friar’s Heel, 214, 472, 474-6.

Funeral feasts, 113-4, 202.

Fyvie, 81.

Gades, 485, 497, 511-2.

Gaelic, 227, 229.

‘Galates,’ 438 n. 3.

Galicia, headlands of, identified with Cassiterides, 489.

Galleys, in war with Veneti, 302-5; in Caesar’s first invasion of Britain, 313, 316-8; wrecked, 320. See also 335, 338, 554-5, 596-9, 662.

Galli, senses in which Caesar used the word, 290, 438 n. 3.

Games, 264.

Gard, 200.

Garrowby Hill, 181 n. 1.

Garson, J. G., 8, 379 n. 2, 382 n. 2, 385, 394-6, 408.

Garton-on-the-Wolds, 385 n. 7.

Gaul, 1, 65, 67, 128, 148, 218; British bronze culture indebted to, 126; bronze imported from, 144; trade with, 172 (see Merchants); Gallic coins in Britain, 250; how Britons were affected by Caesar’s campaigns in, 300.

Gav’r Inis, 171, 200.

Geese, Britons forbidden to eat, 55; goose interred in a long barrow, 114.

Geikie, Sir A., 31.

Gelderland, 194.

Geminus, 219, 225.

Geoffrey of Monmouth, 362, 369.

Geologists, 8.

Germanicus, 369.

Germany, 60-1, 66, 68, 105 n. 2, 110, 115, 140, 145, 194, 205, 211; German influence on British bronze culture, 126; cavalry, 354, 690.

Gesoriacum, 306, 583-4, 586, 588 n. 1, 589-91.

Glamorganshire, 35, 37, 134, 168, 194 n. 3, 203, 208, 426.

Glass beads, in Bronze Age, 162 n. 1, 167-8, 170, 183; in Early Iron Age, 241, 357; in barrows near Stonehenge, 469.

Glastonbury, 7, 241-2, 251, 263-4; skeletons, 435 n. 1.

Gloucestershire, 8, 92, 101, 106, 110 n. 10, 112, 114, 184, 208, 251.

Gloves, 47, 388.

Glutton, 20, 47.

Goats, 71, 88, 113, 406.

Goidelic Celts, 234, 291, 409-14; not identical with early round-headed invaders, 195, 428-40; had they invaded Britain before Caesar’s time? 228-9, 445-52, 455; religion of, 274, 281; language, 420-4, 444-53, 455-6, 458; inscriptions, 446-7, 449, 451-2.

Gold, worn and used for decoration by wealthy in Bronze Age, 131, 162-3, 165; associated with tin, 139; sources of, 168, 249; exported from Ireland, 168; comparative abundance of relics in Scotland, 168; and in Wiltshire, 168-70, 469; ornaments in Scottish cemeteries, 179; rare in Early Iron Age, 241; coins, 248-50, 358; ornaments imported in Strabo’s time, 357.

Golden Bough, 10.

Goodwin Sands, 312, 334, 525-8, 655-9.

Gop, 178.

Gouges, stone, 75, 77-8; bronze, 140-1, 144.

Gower, 107.

Gowland, W., 6, 140 n. 1, 251, 739.

Grannos, 280.

Gravels, 4, 21 n. 5, 25, 30, 36. See Alderbury, Eoliths, High-level drift, Plateau gravels.

Graves, moundless, 178-9, 208, 232 n. 2, 287.

Gray’s Inn Lane, 4.

Great Orme’s Head, 66 n. 3.

Great Pyramid, 123 n. 1.

Great Stour, Caesar defeats Britons on banks of, 336, 353, 660-2, 664, 678-9, 682-5. See also 339, 687, Chilham, Fordwich, Sturry, Thanington, Wye.

Greeks, 204, 217-8; Greek letters, 266.

Greenhill, 186.

Greenland, 17, 94.

Greenwell, W., 5, 26, 103 n. 2, 109 n. 2, 181 n. 3, 184 n. 6, 185 n. 5, 231, 232 n. 2, 409, 434.

Grenelle race, 405, 431 n. 2, 438 n. 3, 443, 455-6, 458.

Grime’s Graves, 69, 471.

Grimspound, 155.

Grimthorpe, 436 note.

Grinding of stone implements, 73.

Grisnez, Cape, 2, 32, 306, 319, 565-6; erosion of, 528; identical with Itian promontory (?), 570, 572-4.

Grove Ferry, 684.

Grovehurst, 85.

Guest, E., 11, 565-6, 569, 571-4, 577-8, 605, 694-8, 704.

Gutuater, 293 n. 2.

Haddon, A. C., 8, 431 n. 2, 432 n. 5.

Hagbourne Hill, bronze and iron objects associated at, 153 n. 9, 267; Late Celtic pins, 240.

Halberds, 145.

Halliford, 345, 696, 698.

Hallstatt, 8 note, 124; glass beads, 168, 241; bronze objects of Hallstatt period, 182 n. 2; amber necklaces, 183, 469; chronology of period, hardly represented in Britain by iron objects, 229, 231; connexion of Late Celtic culture with, 236.

Hamitic dialects, 67, 405-6.

Ham Marshes, 63, 395.

Hammeldon Down, 162.

Hammersmith, 240.

Hammer-stones, 42, 44, 73-5, 79.

Hammers, stone, 78, 149; bronze, 149.

Hampshire, 25, 32, 36, 85, 130, 138, 194 n. 3, 232 n. 3, 235, 251, 347, 365.

Hamy, E. T., 9, 385-6, 436.

Handles, of palaeolithic tools, 44; of neolithic, 76-8; of bronze celts, 141; of daggers, 145, 147; of Late Celtic swords, 238.

Handley Down, 144 n. 10; moundless interments, 178.

Handley Hill Camp, 156.

Hanging Grimston, 85.

Hangman’s Wood, 515-7.

Hares, Britons forbidden to eat, 55; used in divination, 297.

Harpoons, 37, 42, 63.

Harrison, B., 25, 28.

Hartland, S., 58 n. 2, 204 note, 416.

Hastings, 82 n. 6, 530.

Hastings (the pirate), 540 note, 638-9.

Hatchets, stone, 69-70, 74-7; bronze, 144.

Haverfield, F. J., 7, 139 n. 3, 250 n. 8, 254, 298 n. 6, 347 note, 371 n. 3, 422 n. 2, 483, 509-11.

Hawkins, E., 5.

Hayes Common, 85-6.

Hearn, Lafcadio, 275.

Heathery Burn Cave, 157-60; skeletons, 444; pottery, 467.

Hebrides, 191, 198, 262.

Heller, H. J., 553, 557 n. 4, 564, 569, 574-8, 592-3, 636, 643 n. 3, 647 n. 1, 648 n. 4.

Helmets, 238.

Helvetii, 300-1, 332, 342, 432, 439.

Hercules, 362, 513 n. 3.

Herefordshire Beacon, 259 n. 3.

Herne Bay, 37.

Herodian, 1, 418-9.

Hertfordshire, 36, 235.

Hervé, G., 9, 400 n. 3, 431 n. 2.

‘Hiatus’ (?), 11, 13, 59-61, 385-90.

Hicks, H., 23.

Highfield, 84 n. 2.

High-level drift, 24, 27, 30-2, 37, 39.

Highlands, Scottish, 14.

Hill-forts, of Neolithic Age, 95-6; of Bronze Age, 132-9; of Early Iron Age, 255-9, 337.

Himilco, 218, 490, 512.

Hinduism, 275 note.

Hinks, A., 217 n. 1., 472-3.

Hipparchus, 219.

Hippopotamus, 30, 40 n. 2, 48.

Hissarlik, 191, 198 note, 199-200.

Hitcham, 153.

Hitchin, 20 n. 3.

‘Hoar Rock in the Wood,’ 504.

Hoards, 126, 140, 149-50, 181 n. 3, 252 n. 1; of gold ornaments, 165.

Hoare, Sir R. C., 3, 113, 159 n. 1, 162, 189-90, 215, 288 n. 1.

Hod Hill, 7, 240, 251, 255, 260.

Holderness, 87, 130, 154, 263.

Holed stones, 8, 115-6, 403; at Aylesford, 288.

Holland, 119.

Hollingbury, 134.

Holyhead, 154.

Homer, 190, 218.

Homme Mort, 401.

Homo primigenius (?), 34

Homo sapiens, 14, 34.

Horace, 84, 367.

Horned cairns, 106-7, 394, 403.

Horses, 47; semi-domestication of, in Palaeolithic Age, 56; rare in long barrows, 113 and n. 8; eaten in Britain, 113, 152; small size of British, 152, 343, 676; well-bred, imported by Gauls, 342.

Horse-trappings, 152, 251, 264-5.

Hove, 159 n. 1.

Hoxne, 3, 20 n. 3, 22.

Hudan Fleot, 541, 542 n. 1.

Human sacrifice, in Neolithic Age, 112-3; in Bronze Age, 203-4; by Druids, 288, 293, 297-8.

Hünebedden, 403.

Hungary, 122, 126, 170.

Hunsbury, 7, 94 note, 137 n. 4, 238, 253, 259-60, 267.

Hunting, 46, 88, 151.

‘Hurlers, The,’ 481-2.

Hurleyford, 345.

Hurst, 604; theory that Caesar landed at, 638-9, 646-7; theory that Caesar landed between, and Kennardington, 639-44.

Hurstbourne, 84 n. 2.

Hut-circles, 6-7, 154-6, 263-4.

Hut-clusters, 262 note.

Huxley, T. H., 8, 34, 377, 381, 396-7, 430.

Hydrography, 10.

Hyena, 30, 37, 40, 48.

Hythe, 310, 532-3, 836-40, 547-9, 604-5; theory that Caesar landed at, 622-37, 640-2, 650, 652, 655-7, 659-61; Hythe harbour, useless to Caesar, 310, 632-3; conformation of, 532, 536, 547-9, 622-4; as described by Appach, 535 n. 5, 641.

‘Iberian’ race, 65, 398-401, 406-7, 455-6.

Ice Age, 14-25, 30-1.

Iceland, 223, 225.

Iceni, 235, 269, 347, 358, 370.

Icht, 413.

Ictis, 221-3, 246, 413, 499-500, 511; not Thanet, 500-1; nor the Isle of Wight, 501-7; must be identified with St. Michael’s Mount, 502-7.

Ightham, 25.

Ilkley, 207.

Illyricum, 303, 327.

Immortality of the soul, 50, 117, 201-4, 274-6.

Incense-cups, 183, 191, 194, 196, 467, 477.

India, 31, 65, 82, 123, 177, 211.

Indo-European language, 127, 433; ‘Indo-European p,’ 410, 450-2.

Indutiomarus, 330, 333, 370, 667.

Infanticide, 128.

Infants, interment of, 128; cremation of, 190.

Inhumation, in Palaeolithic Age at Mentone, 49, 204, 389, 460; in Neolithic Age in Britain, 110-2; in Bronze Age, 173, 184-9; religious significance of, different from that of cremation (?), 204-5, 465-6; on Arras Farm, 234 n. 1; in Early Iron Age, 286-8; Sergi’s views on, 400.

Inscriptions, 10; as materials for history of Celtic religion, 273-4, 277-82; at Colchester, 415-6; ‘Pictish,’ 420-1, 423; Goidelic, 446-7, 449, 451-2.

Interglacial period (?), 15-6, 18, 20, 23.

Intertribal war, 68, 81, 95, 101, 119, 129, 131-2, 268-9.

Inverness, 17.

Inverness-shire, 102, 115 n. 8, 194 n. 3, 208.

Ipswich, 42.

Ireland, in Ice Age, 16; dolmens in, 66, 403, 405; pygmy flints, 82; chambered cairns, 108; well-worship, 116; Copper Age, 122; spearheads, 148; cauldrons, 158; gold, 168; spiral ornament, 171; only two drinking-cups found in, 195; food-vessels and incense-cups frequent in, 196; disks, 207; ‘bluestones,’ 214; position of, indicated by Pytheas, 223, 352, and Caesar, 352; vitrified forts, 259; crannogs, 263 n. 2; cannibalism, 268 n. 1; skulls of Long Barrow type found in, 398; ‘Pictish succession’ in, 415; no ‘characteristic’ Round Barrow skulls found in, 432; Goidelic invaders of, 446, 449; Irish invaders of Britain, 448; Ireland colonized by Belgae (?), 451 n. 2. See Dowris, Lough Crew, New Grange, and 108, 205, 432.

Irish Sea, in Ice Age, 16.

Iron implements not used in British Bronze Age, 123-4; date of their introduction into Britain, 126, 231-2; probably introduced by Brythons, 232; iron pyrites, 46, 80; mines, 230, 251, 260, 357. See Axes, Bill-hooks, Currency bars, Hallstatt, Mirrors, Noreia, Ploughs, Sickles, Swords.

Ischalis, 232 n. 3.

Islay, 108.

Italy, 9, 21, 211; probably a Copper Age in, 122; British bronze culture connected with, 126.

Itchen, 32.

Ἴτιον, 556, 570-1, 577 8.

Itius. See Portus Itius.

Jade, 69 n. 1.

Japan, 66, 478.

Javelin-heads, 80.

Jet, 161, 167.

Jewellery, origin of, 92; of Bronze Age, 163-70; sham, 167; Late Celtic, 241.

Julia, 348.

Julian calendar, 707-26.

Jullian, C., 226 n. 3, 273 n. 1, 277, 278 n. 2, 286 n. 6, 292 note, 298 n. 6.

‘Julliberrie’s Grave,’ 683 n. 6.

Jupiter, 275-6, 279-83, 285.

Kabyles, 416.

Κασσίτερος, 433, 453, 494.

Keiss, 396.

Kellythorpe, 162, 189.

Kennardington, 639-40, 643.

Kent, 5, 30, 33, 36, 67, 96 n. 3, 101, 130, 133, 194 n. 3, 235, 242, 254; poor in ornaments, 170; prototype of British coins frequent in, 249; dene-holes, 253; survival of old sepulchral rites, 268; kings of, attack Caesar’s naval camp, 346. See Aylesford, Caesar, Coast, Commius, Cunobeline, Deal, Dover, Great Stour, Landing in Britain, North Foreland, South Foreland, Walmer, &c.

Kent’s Cavern, 4, 37, 42, 244.

Khasis, 214, 478.

Kilmartin, 205.

Kilts, 265.

Kimmeridge shale beads, 92; cups, 159 n. 1; bangles, 241.

‘Kimris,’ 438-9.

Kincardineshire, 208-9, 739.

Kings, 269-71; of Kent, attack Caesar’s naval camp in 54 B.C., 353, 360, 669-70.

Kingsdown, 311, 334, 523, 651, 674.

Kingship, origin of, 58, 100-1, 117.

Kingston-on-Thames, 696-7.

Kingston (on the Little Stour), 679-80.

Kinross-shire, 194 n. 3.

Kirkcudbright, 135.

Kirkwall, 268 n. 1.

Kistvaens, 65 n. 4, 96, 189 n. 1.

Kit’s Coty House, 66 n. 3.

Knives, of flint, 75; of bronze, 182.

Krapina, 381.

Kymri, 449-51.

Laberius Durus, 340, 683 n. 6.

Labienus, sent to country of Treveri, 304; punishes Morini, 324, 593; left in command in Gaul (54 B.C.), 334; Caesar applies to, for shipwrights, 338; warns Caesar of danger in Gaul, 349; builds ships, 350, 584; orders assassination of Commius, 365.

Lake barrow, 167, 469.

Lake District, 16.

Lake-dwellings, 7; of Holderness, 87, 153-4; of Switzerland, 87, 157. See Barton Mere, Crannogs, Glastonbury, Munro.

Lakes glacial origin of, 15; worshipped, 116, 272.

La Madelaine, 46, 99, 383.

Lamiis tribus, 272.

Lamps, 70, 258.

Lancashire, 82 n. 6, 171, 179, 205, 267.

Lance-heads, 82.

Landing in Britain, place of Caesar’s, 309-12, 315-6, 595-665.

Land’s End, 135; Pytheas lands near, 221; skull found near, 396; Ictis located off, by Müllenhoff, 502 n. 8.

Lang, A., 49 n. 7, 52 n. 6, 53 n. 4, 198 note, 206 n. 4, 463-4.

Langbank, 463-4.

Language, 48; neolithic, 67, 405-6. See Brythonic, Celtic, Goidelic, Philology.

Lapps, 102.

Largie, 109 n. 2.

‘Late Celtic’ Period, 5-6; art, 9, 84 n. 2, 236-46, 372. See Early Iron Age.

La Tène, cauldrons, 158 n. 2; culture named after settlement of, 236, 241; swords, 238; brooches, 240.

Lathe, 159 n. 1. See Potter’s wheel.

Latin, known by some of the Britons, 266, 368, 372.

Laugerie Basse, 35, 382.

Lea, 60, 347, 702.

Lead, in bronze, 140, 739; leaden celts, 148-9, 252; lead mines, 252.

Leaf-shaped arrow-heads, 80-1; swords, 146; spears, 148.

Ledbury, 396-7.

Lee way, 326, 582, 613, 625, 634, 656, 659, 740.

Legions, 301; 10th, 313, 316; 7th, 313, 321, 337, 343, 636, 677.

Lepidianus tumultus, 719-21, 725.

Lewes, 7, 256. See Mount Caburn.

Lewin, T., 533, 535-6, 542, 546-9, 563, 587, 607-11, 622-38, 648, 649 n. 1, 650-9, 701-3.

Lewis, 207-8, 262.

Liane, 306, 314, 324, 331, 571, 594; ancient depth and extension of estuary, 586-7.

Life, duration of in Neolithic Age, 91; in Bronze Age, 152-3.

Ligurian coast, 61; language, 296 n. 4, 408.

Limen, 535, 538-43, 545.

Lincolnshire, 35, 82 n. 6, 130, 194 n. 3; round barrows in, 187.

Lions, 30, 48.

Littlebourne, 680, 682.

Little Stour, 660, 679-82, 685.

Littleton Drew, 105.

Livy, 285.

Llandebie, 395.

Llandudno, 139.

Llangorse, 263 n. 2.

Loch Etive, 107.

Lockyer, Sir N., 216-7, 472-6, 480-2.

Loire, tin shipped from Ictis to, 221, 223, 246, 500-1, 505-9; ships built in, and lent by tribes near, for war with Veneti, 302-4; Strabo on passage from, to Britain, 577.

Lomea, 526.

Londinium, 255; name of, does not appear on any British coin, 359; was it pre-Roman? 703-5.

London, palaeolithic implements found in, 39; topography of, in Caesar’s time, 255; road from, to Durovernum, 336; Cassivellaunus’s stronghold wrongly located at, 701-2.

Long, G., 567-9, 571, 577-8, 590-3, 603 note.

Long Barrow race, ethnology of, 64-7, 393-407, 455-6, 458.

Long barrows, 101-3, 173; orientation of, 103; construction, 103-5. See Chambered barrows.

Long Hole Cave, 37.

Lord’s Down, 188-9.

Lorthet, 99.

Los Murciélagos, 100.

Lossio Veda, 415.

Lough Crew, 206.

Lourdes, 57.

Lower Greensand, 26-7.

Lozenge pattern, 197, 198, n. 3, 239.

Lozère, 35.

Lubbock, Sir J., 4. See Avebury, Lord.

Lucan, 279, 285 n. 8, 355, 628, 630.

Lucretius, 50, 124 n. 5.

Lug, 277.

Lugotorix, 348.

Lugudunum, 277, 283.

Lunettes, 165.

Lutcombe Castle, 134.

Lyall, Sir A., 10, 204 note, 275, 277

Lydd, 535, 536 n. 1, 538, 543.

Lydney, 280-1.

Lyell, Sir C., 4, 45, 223.

Lympne, 309-10, 532-3, 538-45, 547, 551-2; theory that Caesar landed at, 622-37, 642, 650-2, 656.

Mabinogion, 274.

Mabon. See Maponus.

Macbain, A., 420-3.

MacEnery, J., 4.

Magi, 297 n. 3.

Magic, 47-8, 57-8, 92 n. 6, 117; connexion of, with religion, 58, 461-2.

Maiden Bower Camp, 97.

Maiden Castle, 134, 137.

Mainland, 226.

Malden, H. E., 539 n. 7, 556, 607, 615 n. 1, 619-20, 638-9, 646-7.

Malvern Hills, 68.

Mammoth, 4, 21, 30, 35, 40; teeth of, found in situ in peat, 386 n. 5.

Man, Isle of, 180, 205. See Monapia.

Mandubracius, 327, 333, 339, 361, 700.

Manonvrier, 9, 379 n. 2.

Μάντεις, 297 note.

Mantes, 29.

Maoris, 77.

Maponus, 280.

Mark Antony, 365.

Marliano, R. de, 2.

Marlborough bucket, 237, 246.

Marne, cremation in, in Neolithic Age, 110 n. 1; ‘owl-heads’, 200; Caesar builds ships on river (?), 327; skeletons of Early Iron Age, 436; chariot-burials, 676.

Marriage, 128, 269-70.

Marrow, 46-7.

Mars, 275, 277-9, 282-3.

Marshes (?) near Caesar’s landing-place, 628, 630-2, 653-4.

Martial, 368.

Martin Down Camp, 156-7, 467.

Mas d’Azil, 48, 49 n. 7, 61; painted pebbles, 99, 263, 464.

Massilia, British trade with, 172, 218, 499-501, 507-8; Pytheas calculates latitude of, 219; date of foundation, 511; Massilians introduce coins into Gaul, 248; import tin from Spain (?), 495-6.

Matlock, 252.

Matriarchy, 52 n. 6, 94-5, 351; among the Picts, 414-7.

Mediterranean, 200, 218, 231, 307, 326; ‘Mediterranean race,’ 65, 398, 400-1, 406-7, 455.

Mêdûm, 125.

Medway, 25; crossed by Caesar, 344.

Megalithic monuments, 5-6, 177. See Menhirs, Stone circles, Stone rows.

Menapii, 302, 314, 324.

Mendip Hills, 252.

Menhirs, 114, 208, 285.

Mentone, 34, 49, 204, 382.

Merchants, of Gaul, 1, 307-8, 310, 331.

Mercury, 274, 277-8, 282-3, 285.

Merionethshire, 171, 184, 205, 242.

‘Mesolithic’ implements, 59, 387, 388 n. 1.

Mesopotamia, 122, 125 n. 1.

Metallurgy, 121, 139-40.

Metempsychosis, 293, 295-6.

Mexico, 125.

Mice, bones of, in urn, 203.

Mictis, 500. See Ictis.

Midacritus, 485, 514.

Middlesex, 36, 235.

Midlands, 14, 16.

Midnight sun, 225-6.

Midsummer festivals, 280, 475.

Military history, 10, 352, 595-6.

Millfield, 85.

Minerva, 275, 280-2.

Mining, of flint, 69-71; of copper, 139 n. 1, 502-3 n. 8; of iron, 231, 251; of lead, 252. See Tin.

Mirrors, 239-40, 264.

Mistletoe, 298.

Mitchell, Sir A., 6, 203 n. 4, 248.

Moel Tryfaen, 17 note.

Mold, 162.

Mole, bones of, in urn, 203.

Mollis (B. G., v, 9, § 1), 628-30.

Mommsen, Th., 355-6, 715 n. 6.

Monapia, 450-1.

Monarchy. See Kings.

Mongoloid tribes, 51.

Monmouthshire, 194 n. 3, 360.

Montelius, O., 9, 123 n. 1, 402, 432, 476 n. 1.

Monzie, 153.

Moon, worshipped, 116, 282; influence of, on tides, announced by Pytheas and Posidonius, 219, 319, 336; year computed by revolutions of, in ancient calendars, 296, 475, 707; full moon of Aug. 31, 55 B.C., 319, 600-3; exact time of, 610, 665-6; new moon of Jan. 2, 45 B.C., 722 n. 2; Caesar sailed for Britain in 54 B.C. about day of new moon, 728-9.

Moravia, skeletons found at, 34, 381; interment practised in, in Palaeolithic Age, 49.

Morayshire, 165.

Morbihan, 110 n. 1, 205, 302-4.

Moredun, 435 n. 1.

Morini, 133, 302; Caesar’s campaign against, 305-6; envoys from, proffer submission, 133, 312; Cotta, 314; attack Romans, punished, 324, 593-4; shortest passage to Britain from their country, 554, 571, 596, 619; Caesar sails from their country, 558-63; Strabo on passage to Britain from, 555-6, 577-9; portus Morinorum Britannicus, 589; induced by Britons to mediate with Caesar (?), 672.

Mortars, 79.

Mortillet, G. de, 38, 39 n. 1.

Mortimer, J. R., 5, 101 n. 3, 157 n. 2, 174 n. 1, 177 n. 6, 181 n. 1, 186 n. 4, 195 n. 5, 197 n. 1, 393 n. 4, 425 n. 4.

Moulds, 148-9.

‘Mound dwellings,’ 390-2.

Mountains worshipped, 272.

Mount Caburn, 256.

Mountfield, 165.

Mousterian implements, 40-1, 384.

Moustier, Le, cave of, 40.

Muir n’ Icht, 572.

Müller, S., 9, 402.

Mullers, 79, 90.

Munro, R., 107 n. 2, 263 n. 2, 463-5.

Muskham, 396-7.

Mycenaeans, 204.

Myres, J. L., 9, 378 n. 6, 402.

Napoleon III, 602 n. 5, 603.

Narbo, trade with, 499-500, 508.

Naval camp, constructed by Caesar in 54 B.C., 338, 686-7; attacked by Kentish kings, 346-8, 661; Caesar’s unexplained visit to, 348-9, 669, 672, 731-3; site of, 673-4.

Neanderthal skull, 33-4, 380-2; race, 385, 397, 455.

Necklaces, 47; amber, 163, 167, 169; jet, 167.

Needles, of bone, 37, 42.

Needles, The, 32.

Neolithic Age, 11, 62-120; early immigrants of British, 62-3; British civilization originated in, 63-4; later invaders of, 64-7; ethnology of inhabitants, 64-7, 393-409; settlements, 67-9; implements, 71-83; specialization of industries, 83; implements used after introduction of bronze, 132. See, Dwellings, Agriculture, Clothing, Cookery, Hill-forts, Religion, Transition, &c.

Nervii, 330, 342, 352, 734.

Nether Swell, 105.

New Grange, 170, 200, 478.

Nicholson, E. W. B., 410-1, 419-23, 449-53.

Nights, shortness of, 225-6, 351.

Nile, 30.

Nilsson, S., 102, 479.

Nobles, 271.

Nodons, 281.

Noreia, 231.

Norfolk, 18-20, 36, 85, 162, 194 n. 3, 253, 263 n. 2, 347.

Normanton, 162.

North Bavant, 97.

North Downs, 26, 662.

North Foreland, 309, 575-7, 657 n. 3.

North Sea, 14.

Northamptonshire, 36, 110 n. 11, 137 n. 4, 238, 251, 361.

Northumberland, 8, 133 n. 1, 154, 161, 167, 171, 179, 194 n. 3, 205, 208, 359; hut-circles, 154-5; cremation, 184; drinking-cups, 192.

Norway, 126. See Scandinavia.

Novantae, 447-8.

Nundinae, 713 ff.

Oban, 62, 394.

Oestrymnides, 491.

Offa’s Dyke, 260 n. 1.

Oldbury, 46, 134.

Old Sarum, 259 n. 3, 481.

Ons, 483 n. 3, 487-8, 497.

Ordovices, 233 note.

Orientation of long barrows and chambered cairns, 103; of skeletons in round barrows, 188; of stone circles, 210-1, 481-2; of Stonehenge, 216-7, 472-6, 480-1; of skeletons in interments of Early Iron Age, 287-8, 739. See Hurlers.

Orkney, 87 n. 1, 97 n. 2, 109, 198, 208; chambered cairns in, 102, 408; holed stones, 115 n. 8; amber necklace, 169; barrows, 175; brochs, 262. See Ronsay, Stromness, Unstan.

Ormiegill, 106.

Ornament, on pottery, origin of, 89, 198 n. 3; on bronze weapons, 149; on pottery of the Bronze Age, 197-200; curvilinear, 236-9.

Ornaments, 92. See Jewellery.

Ortels, A., 2.

Osismii, 221.

Οὐάτεις, 297 note.

Ouse, 48.

Oval barrows, 105 n. 2, 108.

Oxfordshire, 36, 84 n. 2, 101, 183 n. 1, 194 n. 3, 208, 235, 239-40, 246, 248 n. 2.

‘P’ Celts, 227-8, 409-10.

Palaeolithic Age, 4, 13-61; chronology of, 8, 31-2; relation of palaeolithic man to Ice Age, 22-5; environment of palaeolithic man in Britain, 30-1; whence did he come?, 30-1; skeletons, 33-5, 380-3; races, 34-5, 383-5; artists, 35; range of hunters in Britain, 35-6; implements, 3-4, 24, 38-42; where implements have been found, 36-7; ‘Palaeolithic Floor,’ 39; workshops, 42-4; culture of inhabitants in Britain, 45-9; religion, 49-51; did palaeolithic man leave descendants in Britain?, 59-61, 385-90.

Palestine, 30.

Palms in London Clay, 14.

Palstaves, 141, 144; of Scandinavia, 172.

Papa Westray, Holm of, 102.

Paris, altars of, 276, 279 note.

Parisi, 235, 360 n. 2, 450-1.

Park Cwm, 107.

Pasturage, 88, 150-1.

Paviland Cave, 168, 397 n. 8.

Peanfahel, 421-2.

Pebbles, painted, 49; in brochs, 262, 464.

Peebles-shire, 257 n. 5.

Peik-, 414.

Pengelly, W., 223.

Pennine Range, 68.

Pennocrucium, 450.

Pentagram, 295.

Pentland Firth, 224.

Pen-y-Gaer, 257.

Peristaliths, 105, 208 n. 2.

Perthes, B. de, 4.

Perthi-Chwareu, 55, 395.

Perthshire, 153, 194 n. 3, 208, 361 note.

Peru, 125.

Pestles, 75.

Petrie, W. Flinders, 9, 111 n. 3, 402, 479.

Pevensey, 558; theory that Caesar landed at, 604-5, 611-21.

Peytrel, gold, 131, 163.

Philip of Macedon, 248.

Philology, 8; as an aid to ethnological inquiry, 229, 375-6; Celtic philologists differ on fundamentals, 453.

Phoenicians, 172, 219 n. 4, 221, 479, 489-91, 493-5. 497-8. 511-4. See Cassiterides, Tin.

Phrygians, 514.

Picks, deer-horn, 69, 71; at Stonehenge, 215, 470-1.

Pict (the name), 412-4, 419.

Pictones, 419.

Picts, 351, 391 n. 5; the ‘Pictish question’, 409-24, 456; Pictish inscriptions, 420-3.

‘Picts’ houses,’ 102 n. 4, 261, 391.

Piette, E., 9, 99.

Pigs, 88, 407; interment of, in barrows, 203.

Pilgrim’s Way, 247, 256, 337.

Pins, of Bronze Age, 161; Late Celtic, 240.

Pit-dwellings, 84-7, 153, 261.

Pits, in Hunsbury, Mount Caburn, and Worlebury forts, 256, 260.

Pitt-Rivers, A., 6-7, 71, 84 n. 2, 97 n. 7, 123 note, 136, 138, 144 n. 10, 175 n. 4, 176 n. 1, 179, 197, 201 note, 202 n. 3, 212, 215, 256-7, 267, 441.

Placard Cave, 99.

Placentia, 329.

Plas Newydd, 107.

Plateau gravels, 25-8, 36.

Pleistocene Period, 4, 11, 14, 27. See Ice Age.

Pliny, 219, 224, 296, 592-3.

Ploughs, 152 n. 2, 253.

Plutarch, 628, 630-1.

Plymouth, 33. See Cattedown Cave.

Polished stone implements, 73.

Polyandry, 351, 414-7.

Polybius, 219-20, 226 n. 3, 285.

Polytheism, 51, 276, 282.

Pomponius Mela, 1, 295 n. 1.

Pont Newydd, 40 n. 2.

Port Erin, 180.

Portsdown Hill, 20.

Portugal, 194, 263 n. 2; Portuguese neolithic chambers, 87. See Los Murciélagos.

Portus Itius, Caesar sails from, to Britain, 306-7, 312, 327, 330; dispatch vessels ply between, and Britain, 348; question of its site, 552-95. See Ambleteuse, Boulogne, Calais, Somme, Wissant.

Portus Lemanis, 533, 538-41, 543-9, 551-2, 622.

Portus Ritupis, 519.

Posidonius, on tides, 219 n. 4, 319; on Gallic banquets, 261; on tin trade, 484, 499; no evidence that he visited Britain, 499 n. 2.

Potter’s wheel, 191, 242.

Pottery, not made in Palaeolithic Age, 46; at Hurstbourne and Highfield, 84 n. 2; neolithic, 89, 96-7, 108, 109 n. 2; domestic, of Bronze Age, 159, 467; of Early Iron Age, 244; sepulchral, of Bronze Age, 191-9, 467; Late Celtic pottery, 242-4, 288; ‘Samian,’ 372; potsherds in barrows, 113-4, 203-4; at Stonehenge, 469 n. 7. See Cinerary urns, Drinking-cups, Food-vessels, Incense-cups.

Prah Sands, 36 note.

Prasutagus, 358; the name, 450, 452.

Prayer, 117, 290, 297.

Prehistoric ages, indefiniteness of, 72.

Prehistoric Britain, how our knowledge of it has been obtained, 1-12.

Prehistoric Room (British Museum), 9, 70, 217 n. 1.

Prestwich, Sir J., 4, 23, 26-7.

Pretani, 411-3, 459.

Pretanic island, 227-8, 411-3, 459.

Πρετανικαὶ νῆσοι, 411, 459-61.

Promontory, rounded by Caesar in 55 B.C., 600, 650.

Property, private, in land, 252; British women might own property, 269-70.

Prydein, 411, 413, 418-9.

Ptolemy, 235, 255, 422-3.

Puttenham, 137.

Pygmies, 390-3.

Pygmy flints, 82-3.

Pyrenees, palaeolithic artists of, 35 n. 3; cave-dwellers, 47; pottery of dolmens, 109; misplaced by Strabo, 488.

Pythagoras, 294-5.

Pytheas, 152; his voyage, 217-26; his scientific and geographical work, 218-9, 221, 223, 351-2; an authority on British ethnology, 227-9, 411-2, 445-6, 459; discredited in Caesar’s time, 307; on tides, 219, 319; misunderstood by Professor Ridgeway, 495; Diodorus Siculus ultimately derived description of Ictis from, 499. See Ictis, Moon, Thule.

‘Q’ Celts, 227-8, 409-10.

Qicti, 414.

Qrtanic, 228.

Qrtanoi, 412, 452, 459.

Quaternary Period, 14.

Quatrefages, A. de, 9, 385.

Querns, 253, 262, 264, 361 note.

Quiberon Bay, 304.

Quintus Cicero. See Cicero.

Rains Cave, 178.

Raised beach, 62.

Ramsgate, 519, 575, 577.

Ranke, J., 9, 439-40.

Rapier-shaped swords, 147.

Razors, 158, 160.

Read, C. H., 7, 134 n. 12, 182 n. 5, 217 n. 1, 430 note, 431, 505-6.

Reculver, 36-7, 336.

Red hair, 440.

Regni, 366, 617.

Reid, C., 19, 23, 27, 28 n. 2, 36 n. 1, 40 n. 2, 61 n. 3, 222, 503-7.

Reinach, S., 9, 47, 57 n. 5, 83 n. 3, 121 n. 1, 125 n. 4, 171 n. 3, 201 n. 3, 210, 277, 279, 292 note, 405 n. 7, 406 n. 6, 461-2, 493-4, 513-4.

Reindeer, 40, 68.

Religion, 10; may have been a motive of palaeolithic art, 48; and of geometrical decoration, 199 note; religion of palaeolithic man, 49-51; of neolithic man, 114-8; in Bronze Age, 200-7; Celtic, 271-90, 297-8; the birthday of religion, 461-3. See Altars, Animism, Anthropomorphism, ‘Continuance theory,’ Druidism, Magic, Metempsychosis, Retribution, Sun worship, Temples, Totemism.

Reliquiae Diluvianae, 4.

Remi, 299 n. 5, 454 n. 4.

Retribution, religious doctrine of, 296.

Rhee Wall, 535, 538-40, 542, 548; built by Romans, 549-52.

Rhine, some of the brachycephalic invaders come from, to Britain, 128, 443; Strabo on passage from, to Britain, 577.

Rhinoceros, woolly, 20; big-nosed, 40.

Rhoda, 248 n. 2.

Rhonddha valley, 134.

Rhône, tin exported from Britain to mouth of, 222, 499, 513.

Rhosdigre, 395.

Rhys, Sir J., 228, 290, 291 nn. 1-2, 351, 367 n. 9, 390-2, 405 n. 8, 409-24, 429-30, 433 n. 4, 446-9, 453-4, 459-61, 500, 508-9.

Richborough, 336, 641; topography, 519-20; distance from Gesoriacum, 591-2; theory that Caesar landed at, 604, 663-4; that he encamped on, 674.

Ridgeway Hill, 203.

Ridgeway, W., 9, 188 n. 2, 204, 221 n. 3, 495, 500-1, 507-9, 569-70, 619-21.

Rillaton, 195 note.

Rings, 165, 167, 183 n. 1.

Rivers, worshipped, 116, 272.

River-bed skulls, 8, 396-7.

River-drift, 22, 36; ‘river-drift men,’ 38, 383-5. See Caves, High-level drift.

Riviera, 35, 67. See Baoussé-Roussé, Mentone.

Robertsbridge, 616, 678.

Robin Hood Cave, 35, 45 note.

Rochester, 344.

Rodmarton, 105, 401.

Rolleston, G., 8, 112, 377-8, 406, 425 n. 4, 426, 432, 441, 662 n. 3.

Rollright Stones, 210, 470.

Rome, thanksgiving service at, for Caesar’s first invasion of Britain, 325; Roman troops versus British chariots, 341-3; growth of Roman influence in Britain, 356-8, 362-3, 368-72; flight of British princes to Rome, 366.

Romney, 535, 537-9, 540 n. 2, 543, 547, 550.

Romney Marsh, 310; ancient geography of, 532-52, 622-3, 638, 640; Maistre Wace anticipated modern view that Caesar landed on, 644.

Ross-shire, 115 n. 8, 168, 194 n. 3.

Rother, ancient course of, 533, 537-43, 552; Airy holds that Caesar defeated Britons on in 54 B.C., 616.

Rouge, 80, 264.

Round barrows and cairns, 6, 8, 107-8, 119-20, 173-6; first erected in Neolithic Age, 119-20, 408-9; ditches, banks, and stone circles belonging to, 175-7, 207-8; not erected only in memory of chiefs, 177-9; cenotaphs, 180-1; chronology of round barrows, 181-4, 476 n. 1; round barrows of Early Iron Age, 287. See Bell barrows, Bowl barrows, Disk barrows.

Round-headed invaders of Britain, 127-8, 424-44; begin to arrive in Neolithic Age, 119, 127, 408-9; of ‘characteristic’ type, 425-8, 444, 455; of short stature (‘Alpine’ type), 426-8, 455; earlier round-headed invaders not Celtic, 428-40.

Roundway Hill, 109 n. 3.

‘Row Grave’ skulls, 443 n. 5.

Royal Archaeological Institute, 5.

Royal Irish Academy, 5.

Royal Society of Antiquaries of Ireland, 5.

Rudstone, 201 n. 3.

Rufina, 368.

Rushmore, round barrows at, 201 note, 202 n. 3, 212 n. 2; did not contain bronze, 215.

Rutupiae, distance from Gesoriacum, 591; theory that Caesar landed at, 663-4. See Richborough.

Rye, 604.

Sabinus, 304, 314, 324.

Sabre-toothed tiger, 37.

Sacrifice, 118. See Animals, Human sacrifice.

Salisbury Plain, forts on, 133, 137 n. 4; sarsens, 214; Salisbury Spire, 217, 481.

Salmon, Ph., 9, 382 n. 2.

Sanson, N., 2.

Sandgate, coast between, and Dover, 531-2; in connexion with question of Caesar’s landing-place, 627, 638, 640-2, 646-7, 651.

Sandown Castle, 312, 323; coins found near, 520-1; coast between, and Walmer Castle, 521-5.

Sandtun, 539, 541, 542 n. 1.

Sandwich, 311, 323; Caesar lands near in 54 B.C., 335-6, 664-5; coast between, and Sandown Castle, 519-20; decay of port, 526; theory that Caesar landed at, in 55 B.C., 604-5, 660-4. See also 651, 657-8, 673-4, 683.

Sangatte, 306; not Caesar’s ulterior portus, 581-3, 585, 619, 639, 740-1.

Sarrebourg, altar at, 281.

Sarsens, 214-5, 470-1, 479-80.

Saturn, 282.

Saws, flint, 41, 79, 132; bronze, 132 n. 2, 140; many flint saws in one barrow, 201 n. 3; iron saws, 253.

Scabbards, 147; Late Celtic, 237, 239.

Scaliger, J., 2.

Scandinavia, 9, 14, 66-7, 77, 102, 115, 168, 185, 195 note, 205-6, 211, 218, 404, 441; trade, 170-1; superiority of bronze culture, 172. See Thule.

Schneider, R., 557, 589.

Scilly Islands, chambered barrows in, 102; people of, traded by barter, 359; identified with Cassiterides, 486, 490-3, 497-8, 513.

Scorborough Park, 435 n. 1.

Scotland, in Ice Age, 16, 21, 24; subsidence of, 62; dolmens, 66, 403 n. 5; axe-hammers, 79; barbed arrow-heads, 81; chambered cairns, 101, 107; Bronze Age began late, 132; lead in bronze, 140; cauldrons, 158; ornaments, 163, 165, 167-9; cairns, 174; cemeteries, 178-9; interments, 185, 200; drinking-cups, 192, 195 note; stone circles, 207-8, 476; only one interment of Early Iron Age, 232 n. 2; Late Celtic pins, 240; vitrified forts, 259 n. 3; dwellings and brochs, 261-2; crannogs, 263 n. 2; no coins struck in, 359. See also 109, 129-30, 133, 141, 205, the counties, Ethnology, &c.

Scrapers, palaeolithic, 41; neolithic, 79; on Dartmoor, 156.

Sculptor, neolithic, 70.

Sculptured stones, 8, 177, 183, 205-7.

Scythians, use of breeches borrowed from, 265.

Seaford, hill-fort at, 98 note, 136, 137 n. 1.

Secondary interments, 112, 173, 186 n. 4, 188-9.

Segontiaci, 346, 361, 700.

Seine, 327; Strabo on passage from, to Britain, 577.

Selgovae, 448.

Selsea, 19.

Senotigirnios, 360 n. 2.

Sepulchral pottery, of Bronze Age, 191-9, 467; Late Celtic, 242-3, 288.

Sergi, G., 9, 377-8, 398, 400-2, 404 n. 6, 406.

Sevenoaks, 25-6, 254.

Sewing, 47.

Shakespeare, 204.

Shakespeare’s Cliff, 310, 532.

Sheep, 88, 151, 357, 406.

Shells, 11, 16-7; shell-fish eaten, 63, 157.

Shetland, 67, 129 n. 4, 225-6, 262. See Thule.

Shields, 145-6; Scandinavian, 172; Late Celtic, 237, 244-5.

Ships, figured on Scandinavian rocks, 171 n. 3; British, 246-7; of the Veneti, 304; Caesar’s, wrecked, 319-20, 338. See Galleys, Transports.

Shorncliffe, 532, 536, 622-4, 651.

Shropshire, 208, 359.

Sibbald, Sir R., 4.

Siberia, 60-1.

Sickles, stone, 80; bronze, 144-5; iron, 253.

Sidbury Hill, 216, 472-3, 481.

Silbury Hill, 180-1.

Silchester, 255; inscription found at, 410, 451.

Silura, 359 n. 12.

Silures, 281, 359 n. 12, 398.

Silver coins, 249, 358, 362; bronze celt found with, 267 n. 2.

Simulacra, 285.

Sion type, 429-30.

Siret, MM., 9.

Sitting posture, 188 n. 2.

Skeletons, 3; palaeolithic, 33-5, 380-3; neolithic, 64, 393-8; of late Neolithic Age and Bronze Age, 127-8, 424-8; of Heathery Burn Cave, 159-60, 444; of Early Iron Age, 234, 434-6.

Skin, vessels of, 42; clothing, 47, 91, 156, 161, 267.

Skulls, 11. See Skeletons.

Sling-bullets, 264, 268.

Slingers, in Caesar’s army, 313, 331, 346, 698.

Small Down Camp, 134.

Small Downs, 526, 665 n. 3.

Smertullos, 279 note.

Smith, C. Roach, 5.

Smith, W. Robertson, 203 n. 4, 252 n. 4, 277.

Smith, Worthington G., 44.

Smyth, Admiral W. H., 608.

Societies, archaeological, formation of, 5.

Society of Antiquaries, 3.

Society of Antiquaries of Scotland, 5.

Socketed weapons, 141, 144-5, 148-9; very rare in interments of Bronze Age, 181-4.

Solent, 32.

Solinus, 359.

Solutré, 39, 41, 383-4.

Solway Moss, 76.

Somaliland, 31; did Mediterranean race originate in? 406.

Somersetshire, 5, 101, 133-4, 175, 194 n. 3, 208, 232 n. 3, 250-1, 359-60.

Somme, 4, 302-3, 306, 327; coast between, and Calais, 517-8; estuary of, not Portus Itius, 558-63, 617, 621.

Somme Bionne, 238, 243.

South Downs, entrenchments on, 96, 98; settlements on, 130.

South Foreland, 2, 223, 311, 315, 319, 334, 575 n. 4, 582, 627, 642-3, 650, 653, 659, 662-3; ancient configuration of cliffs, 528-30; Caesar drifted past, in 54 B.C., 616, 620, 655-6.

South Lodge Camp, 156-7.

Spain, 9, 21, 65, 82, 111, 171, 194, 200, 211, 220-1, 263 n. 2, 327; Copper Age, 122; trade, 172; tin obtained from, 512.

Spear-heads, palaeolithic (?), 41; neolithic, 80, 147; bronze spears, 131, 145-8; socketed spear-heads not found in interments of Bronze Age, 181, 182 nn. 2, 5, 184.

Spettisbury, 251.

Spindle-whorls, 91-2, 156, 160, 264.

Spiral ornament, 170-1, 200, 239.

Spy, 33-4, 381.

St. Albans, 255, 347, 701.

St. David’s Head, fort on, 258.

St. George’s Hill, 694-5.

St. Keverne, 435 n. 1.

St. Leonards, 614, 617.

St. Margaret’s Bay, 311; movements of shingle at, 523, 529 n. 3.

St. Michael’s Mount, 222-3, 502-3, 506-7, 513. See Ictis.

St. Valéry-sur-Somme, 243.

Stadia, 591-2.

Staffordshire, 82 n. 6, 129, 167, 194 n. 3, 203, 205, 208, 252.

Stalagmite, 59, 222, 386-7, 504.

Standard-bearer of 10th legion, 316-7.

Standards, military, 284.

Standlake, 84 n. 2, 183 n. 1.

Stannon, 211 n. 1.

Stanton Drew, 481-2.

Stars, worshipped, 116.

Statius, 114 n. 8.

Statues, 274, 279, 283-6; statuettes at Brassempouy, 383 n. 2.

Stature, relative, of sexes in Neolithic Age, 91; in Bronze Age, 152; methods of estimating, 378-9, 740.

Steatopygous race, 383 n. 2.

Stokes, Whitley, 421-2, 450, 453.

Stonar, 519-20, 522 n. 6, 524 n. 2.

Stone Ages, 4.

Stone balls, 170.

Stone circles, 2 n. 1, 207-17, 285, 476-9; within or enclosing barrows, 176-7; solar temple theory, 210-1, 216-7, 478-9; astral temples (?), 481-2.

Stonehenge, 5-6; barrows near, 113-4, 167, 169-70, 174-5; avenue, 209; outlying stone, 210; original form and construction, 213-4, 479-80; date, 215-7, 468-77; purpose of builders, 477-81. See Evans, A. J., Hinks, Lockyer, Webb.

Stone implements, purposely broken in barrows, 115. See Flakes, Neolithic Age, &c.

Stone rows, 208.

Stone Street, 543-4, 548-9.

Stones, engraved, 205-6.

Stoney Littleton, 105.

Stour (Hampshire), 25.

Stour. See Great Stour and Little Stour.

Strabo, 1, 219, 223-5, 234, 261, 302, 357-8, 368-9, 558, 569-71, 577-9.

Stromness, 87 n. 1.

Stukeley, W., 2, 12, 174, 210.

Sturry, 336, 660, 683-5.

Stutfall Castle, 336, 544-7, 622, 638-9.

Submerged forests. See Forests.

Submergence, in Ice Age, 16-7; in Neolithic Age, 62-4.

Subsidence, in Neolithic Age, 64; of SE. Britain and NE. Gaul since Roman times, 527, 566, 740.

Sucellos, 281 n. 8.

Suessiones, 299, 454 n. 4.

Suetonius, 1, 363.

Suffolk, 3. 22-3, 36, 69, 153, 194 n. 3, 263 n. 2, 347; chisels, 77; tankard, 241.

Sulpicius Rufus, 314.

Sunbury, 696.

Sunken Kirk, 212 n. 2.

Sun-worship, 116, 206-7, 210-1, 216-7, 280, 282, 472-6, 478, 480. See Concentric rings, Disks, Stone circles.

Surrey, 36, 82 n. 5, 97, 130, 137, 254, 362, 365.

Sussex, in Pleistocene Period, 19, 25; terrace cultivation, 253. See also 36, 98 note, 130, 134, 165, 176, 194 n. 3, 232 n. 3, 256.

Sutherlandshire, 150, 194 n. 3, 361 note.

Swanscombe skull, 33, 380.

Swastika, 199 note, 207, 244.

Sweden, 77, 126. See Scandinavia.

Switzerland, 205. See Lake-dwellings.

Swords, 131, 145-7, 172; not found in interments of Bronze Age, 181-2, 184; Late Celtic, 238-9.

Syracuse, 248 n. 2.

Syria, 66, 211, 213.

Taboos, 54, 118, 201 n. 3.

Tacitus, 1, 161, 239, 249, 268, 286, 292, 355, 358, 375, 398-9, 415, 418 n. 1.

Taddington, 108.

Taexali, 448.

Tamesi, 399 n. 1, 453.

Tanarus, 279.

Tanged blades, 145, 147, 182 n. 5.

Tankards, 241-2.

Taplow, 147.

Taranis, 278-9, 281. See Tanarus.

Tarvos Trigaranus. 278, 284.

Tasciovanus, 361, 365, 368.

Tasmanians, 31, 44, 49 n. 5, 462.

Tattooing, 418-20.

Teddington, 696-7.

Teeth, of neolithic population compared with those of Bronze Age, 90, 152.

Temperate animals, 20-1, 383.

Temples, 284-6. See Stone Circles, Stonehenge.

Τενάγη, 631.

Terminalia, 722-3.

Tertiary man, 13-4; deposits, 27.

Test, river, 32.

Textile fabrics, 89. See Clothing.

Thames, southern limit of glacial movement, 18, 36 n. 1; implements found in drift, 22, 23 n. 7, 24, 26, 42; level of, in Palaeolithic Age, 30, 32; implements and weapons found in bed of, 124, 147, 158, 238-9, 244; forded by Caesar, 345-6, 692-9; his march to, 660-1.

Thanet, not to be identified with Ictis, 222, 500-2; ancient configuration of, 519.

Thanington, 336, 683-5.

Thule, 223-6, 367; confounded by Pliny with (M)ictis, 499 n. 5, 505.

Thurnam, J. T., 8, 97 n. 2, 102 n. 4, 112-3, 181 n. 2, 393-5, 401-2, 426 n. 5, 427, 429-30, 434.

Tiberius, 369.

Tidal currents, 10, 311, 315, 334-5, 595-6, 599; question of, in connexion with Caesar’s invasions of Britain, 605-11, 612-3, 620-1, 625, 634, 638, 640, 641 n. 1, 645 n. 3, 647-9, 655-9. See Airy, Darwin.

Tides, Posidonius and Pytheas on, 219 n. 4, 319; spring and neap, 601-2; times of high tide at Dover on Aug. 26-8, 55 B.C., 610.

Tim...., 363-4.

Timaeus, 273, 351.

Timagenes, 294.

Tin, 121-2, 125; early mining of, in Cornwall, 139, 502-3 n. 8; smelting of, 140; coins, 249; produced in the Cassiterides, 483-6, 488-98; British trade in, 218, 220-3, 251, 307, 358, 483-514; had trade temporarily ceased in Caesar’s time? 509; or before A.D. 50? 509-11; did Phoenicians trade for, with Britain? 511-4. See Cassiterides, cassiteros, Ictis, κασσίτερος.

Tincommius, 364-9.

Titurius. See Sabinus.

Togodumnus, 370-1, 453.

Tongs, 157-8.

Torquay, 4, 37, 244.

Torques, of Bronze Age, 163; Late Celtic, 93, 241.

Totemism, 51-7, 284.

Toutates, 278-9, 281-3.

Towyn-y-Capel, 396.

Trackways, 247-8, 344.

Trade, in Neolithic Age, 71; in Bronze Age, 167-72; Late Celtic, 246-8, 269; of Veneti, 302; trade stimulated by Caesar’s invasions, 308, 356-7, 371; British, described by Strabo, 358. See Tin.

Transition between Palaeolithic Age and Neolithic Age, 13, 63; from Neolithic to Bronze Age, 71-2, 120, 131-2, 139-40; from Bronze to Early Iron Age, 230, 267. See Hiatus.

Transmigration of souls, 293. See Metempsychosis.

Transports, used by Caesar in 55 B.C., 313-20, 324, 554, 596; constructed by him in 55-4 B.C., 326, 331, 555, 599; constructed by Labienus, 350, 584; transports which conveyed Caesar’s cavalry in 55 B.C., 313-4, 318-9, 554-6, 597-8; could not have returned in gale of Aug. 30 from near Pevensey to Authie, 558, 613; or from near Walmer or Hythe to Sangatte, 581-3; or from near Pevensey to Sangatte, 618-9; or from near Hythe or Lympne to Ambleteuse, 624-5; or from near Hurst to Sangatte or Ambleteuse, 639; or from near Bonnington to Ambleteuse, 643; some did return from near Walmer to Ambleteuse, 319, 588, 651; day on which they sailed from Gaul, 601.

Trebatius, 328.

Trebonius accompanies Caesar to Britain in 54 B.C., 334; defeats Britons, 341, 353, 692.

Trees worshipped, 272.

Trelan Bahow, 239.

Trent, skull found in, 63, 396.

Trepanning, 79, 93, 260.

Treveri, 330, 667-8.

Triads, 274.

Tribute, British tribes ordered by Caesar to pay, 350, 356, 670-1; not levied by Augustus, 368.

Trinovantes, 235, 299 n. 5; furnish grain to Caesar, 254, 339, 343, 346; opposed to Catuvellauni, 300, 309, 327, 361; Caesar enters their territory, 346-7, 702; subdued by Cunobeline, 362; by Dubnovellaunus, 363, 366.

Tropical animals, 20-1, 40.

Triskele, 242.

Trumpets, 172, 317.

Turner, J. M. W., 213.

Turner, Sir W., 8, 109 note, 375 n. 1, 377.

Tweezers, 160, 264.

Tyddyn Bleiddyn, 395.

Tylor, Prof. E. B., 10, 49 n. 5, 50, 277, 295 n. 1, 461, 463, 675.

Ty Mawr, 154, 155 n. 1.

Uley, 106-7.

Ulterior portus, 554, 556.

Unstan, 97 n. 2.

Uphill, 395.

Upper Greensand, 27.

Upper Swell, 106, 110 n. 11, 112.

Upton Lovel, 163.

Uxellon, 448.

Uxisama, 221.

V-shaped holes, 161, 162 n. 1.

Vacomagi, 448.

Vada, 631.

Varro, 286.

Vectis, 504-5, 740.

Veneti, 223; Caesar’s campaign against, 302-5; Strabo’s explanation of their hostility to Caesar, 308; their trade, 506, 508-9.

Venta, 232 n. 3.

Vepogen, 415-7.

Vercingetorix, 353, 364-5

Vergil, 367.

Verica, 365-6.

Verulamium, 255; perhaps the stronghold of Cassivellaunus, 347, 701-2; its mint, 359, 361-2

Vesta, 286.

Vigo, islands near, identified with Cassiterides, 487-9, 494-6.

Virchow, R., 9.

Vitrified forts, 259 n. 3, 739.

Volisios, 360 n. 2.

Volusenus reconnoitres British coast, 308-10, 554, 596-7; Caesar acts upon his report, 315; attempts to assassinate Commius, 365; various topographical inferences from Caesar’s account of his reconnaissance, 613, 627, 639-40, 645-6, 651-2.

Votadini, 235.

Vulcan, 276.

Wagons, 152, 221, 247, 505.

Walbrook, 255, 703.

Wales, 5, 14, 24, 65 n. 3, 66, 86 n. 4, 101-2, 124, 129, 133, 135, 174, 208, 233, 259 n. 3, 263 n. 2, 359; iron tools rare in, in pre-Roman times, 266. See the counties.

Walmer, 253, 316, 321, 323, 519-21, 576, 604, 626-7, 674; theory that Caesar landed at (and Deal) in 55 B.C., 644-62; castle, 311, 521-5, 604, 653, 664, 674; church, 311, 673-4. See Deal.

Walton-on-Thames, 693, 695.

Wandsworth, 239.

Wansdyke, 260 n. 1.

War Ditches, 454.

War, internecine, in Britain, 95, 129, 131, 133, 268-9, 339.

Warminster Downs, 90 n. 2.

Warne, C., 5.

Warren, The, 532.

Warwickshire, 131, 260 n. 1.

Water Eaton, 240.

Water supply, 138-9, 256.

Watling Street, 344, 704-5.

Weald, 26, 357.

Wealden Forest, 98, 253, 310, 615 n. 1. See Andred.

Wealth, 167, 269, 357.

Weaving. See Spindle whorls.

Webb, E. J., 474-6, 482.

Weems, 153. 391.

Wellesley, Sir A., 645-6.

Well-worship, 116, 272, 283.

Wendover, 701.

West Furze, 87.

West Hythe, 253-4; ancient topography of, 539, 541, 544-5, 547-9, 622-3, 636, 639.

West Hythe Oaks, 533, 535-6, 542, 545-9, 552, 624.

West Indies, 76.

West Kennet, 97 n. 2, 103, 105, 403.

Westmorland, 101, 194 n. 3, 205. 212 n. 2, 359.

West Wickham, 81, 89 n. 8.

Weybourne, pit-dwellings at, 85; theory that Caesar landed at, 604.

Wheel, god of the, 279-80.

Whetstones, 75.

Whit Tor, 96, 134, 155.

Wight, Isle of, 19, 27 n. 4, 32, 67, 145, 251; wrongly identified with Ictis, 222-3, 501-7.

William the Conqueror, 314, 563, 614.

Wilson, D., 7.

Wiltshire, 27, 97, 109 n. 2, 130, 156, 160, 182 n. 5, 194 n. 3, 208, 215, 217, 232 n. 3, 250, 256, 287, 288 n. 1; long barrows, 101-2, 105; population subdued by Bronze Age invaders, 129; ornaments, 162-3, 165, 167-70, 172; round barrows, 175-8; interments, 184, 187-91, 196, 203; coins, 359, 362.

Wind shifted in Caesar’s first voyage to Britain, 314, 626-7; dropped in second voyage, 334, 576 n. 1; great influence of winds on tides, 595, 602, 608.

Windsor, 48.

Winkelbury, 169 n. 9, 256.

Winterbourne Stoke, 162 n. 1.

Wissant, 306-7, 552-3, 555 n. 2, 557-8; not Portus Itius, 565-85, 588 n. 5, 589, 619, 652.

Witham, shield found in, 237, 284.

Wolds. See Yorkshire.

Wolseley, Lord, on warfare with savages, 354; on range of vision, 612 n. 3; on fords, 694.

Wolves, 68, 98.

Women, lot of, in Neolithic Age, 91; in Bronze Age, 152; in Early Iron Age, 269-70.

Woodcuts, 84 n. 2, 138, 261 n. 3.

Wooden tools, 42.

Woodnesborough, 519.

Woodwork, Late Celtic, 241-2.

Wookey Hole, 37.

Wor Barrow, 103 n. 2, 105, 111 n. 3, 179.

Worcestershire, 131, 134, 251, 360.

Workshops, 7; palaeolithic, 42; neolithic, 70, 85.

Worlebury, 253, 255-6, 434 n. 6.

Worth, 335, 660, 673-4.

Wright, Dr. W., 385 n. 7, 427-8.

Wrist-guards, 82, 162.

Writing, primitive, 99; in Early iron Age, 265-6.

Wye, 633, 660-1, 678.

Yarhouse, 106.

Yarnbury, 137 n. 4.

Ynys Prydein, 411.

Yorkshire, 24, 35, 68, 101, 171 n. 3, 174 n. 5, 181 n. 1, 194 n. 3, 201 n. 3, 205, 231-2, 234 n. 1, 235, 241, 250, 264, 267, 288 n. 1; chisels from Wolds, 77; survival of couvade in, 95; cremation prevalent in, in Neolithic Age, 110; cannibalism (?) in barrows, 113; implements deposited in graves, 115 n. 4; evidence from skeletons of commingling of races, 129, 427; why Wolds were thickly peopled, 130; rarity of bronze relics in Wolds barrows, 132, 181 n. 3, 183 n. 6; large oxen reared on Wolds, 151; poverty of people of Wolds, 156; ornaments, 161, 165, 167, 169 n. 7; barrows, 177; interments, 184, 187-9, 191, 200.

Zeeland, 207.

Zeus, 217, 285.

Zimmer, Prof., 405 n. 8, 409, 411, 413-6.

Zoomorphic designs, 237.

Oxford: Printed at the Clarendon Press by HORACE HART, M.A.

FOOTNOTES:

[1] Dr. Joseph Anderson (Scotland in Pagan Times,--the Bronze and Stone Ages, 1886, p. 135) has pointed out that the subject of stone circles was first treated in a scientific spirit in 1692--long before Stukeley wrote--by Prof. Garden of Aberdeen (Archaeologia, i, 1770, pp. 312-9).

[2] See A. Pitt-Rivers, Excavations in Cranborne Chase, iv, 1898, p. 18 (preface), and cf. Archaeol. Cambr., 6th ser., vi, 1906, p. 72.

[3] xiii, 1800, pp. 204-5.

[4] Sir A. Mitchell, The Past in the Present, 1880, pp. 155-7; Sir J. Evans, Ancient Stone Implements ... of Great Britain, 2nd ed., 1897, pp. 56-61, 65, 362-8; Rev. arch., 4^{e} sér. 1, vii, 1906, pp. 239-59.

[5] Sir J. Evans, Anc. Stone Implements, 1897, p. 363.

[6] None of these excavators, however, was so thorough as Pitt-Rivers; and Bateman was often careless (see H. St. G. Gray’s Index to ‘Excavations in Cranborne Chase’, 1905, p. xvi). But it must be remembered that to do such work properly, not only skill and perseverance are needed, but also money.

[7] Pitt-Rivers remarks (Excavations in Cranborne Chase, iii, 1892, pp. x, 254) that a rampart almost always yields something, for example, pottery, which throws light on the period of its construction; while his experience shows the importance of ‘digging the whole of a camp over, down to the undisturbed soil’ (ib., iv, 14 [preface]). He tells us (ib., p. 4) that when he was excavating the earthwork called South Lodge Camp, ‘in the first three sections little or nothing was found, which shows what very false conceptions are liable to be formed by merely digging one or two sections in a camp.’ See also vol. iii, pp. xi, 13; vol. iv, pp. 46-8, 138, 144, 187; Trans. Epping Forest ... Naturalists’ Field Club, ii, 1882, pp. 59-60; and Report of ... the Brit. Association, 1904, pp. 691-700.

[8] The Past in the Present, 1880.

[9] The following passage from Sir John Evans’s Ancient Bronze Implements ... of Great Britain and Ireland, 1880, pp. 25-6, is instructive:--‘In company with Sir John Lubbock I was engaged in opening a grave [at Hallstatt] in which we had come to an interment of the Early Iron Age, accompanied by a socketed celt and spear-heads of iron, when amidst the bones I caught sight of a thin metallic disc of a yellowish colour which looked like a coin. Up to that time no coin had ever been found in any one of the many hundred graves which had been examined, and I eagerly picked up this disc. It proved to be a “sechser”, or six-kreutzer piece, with the date 1826, which by some means had worked its way down among the crevices in the stony ground.... Had this coin been of Roman date it might have afforded an argument for bringing down the date of the Hallstatt cemetery some centuries in the chronological scale. As it is, it affords a wholesome caution against drawing important inferences from the mere collocation of objects when there is any possibility of the apparent association being only due to accident.’

[10] See Mr. H. Balfour’s interesting introduction to Pitt-Rivers’s Evolution of Culture, 1906, p. xiv.

[11] Guide to the Antiquities of the Stone Age (Brit. Museum), 1902, p. 76. See also Journ. Anthr. Inst., xxxiii, 1903, p. 18.

[12] Vict. Hist. of ... Sussex, i, 1905, p. 22.

[13] Proc. Dorset Nat. Hist. and Antiq. Field Club, xxi, 1900, p. 75.

[14] See A. Pitt-Rivers, Excavations in Cranborne Chase, iii, p. xii; iv, p. 28.

[15] Ib., iii, p. xiii.

[16] Mr. N.W. Thomas (Man, v, 1905, No. 25, pp. 47-8) points out that ‘some five years ago the Berlin collections [of ethnographical objects] from British possessions were seven times as large as those in our national museum, and since then this disproportion has not been decreased’, the reason being that ‘the men whom the nation pays to perform certain duties [in national expeditions] are permitted to retain the objects collected in the performance of those duties’. But the nation is to blame as much as the Government.

[17] See pp. 61, 385-90, infra.

[18] Lafcadio Hearn, Kokoro, p. 290.

[19] ‘Comme paléontologiste,’ says M. Marcellin Boule, ‘je crois fermement à l’existence de l’Homme tertiaire: je ne doute pas qu’on trouvera un jour ses traces’ (L’Anthropologie, xvi, 1905, p. 267).

[20] Sir A. Geikie, Text-book of Geology, 4th ed., ii, 1903, pp. 1224-5, 1231.

[21] Sir A. Ramsay’s theory (Physical Geol. and Geogr. of Great Britain, 6th ed., 1894, p. 269), that the basins of the Scottish and Cumbrian lakes were scooped out of the rocks by glaciers, was held by no British geologist a few years ago, except in a modified form. See A. J. Jukes-Browne, Student’s Handbook of Phys. Geol., 2nd ed., 1892, pp. 159, 624, 629-30; T. G. Bonney, Ice-Work, Present and Past, 1896, pp. 80-94; and Sir A. Geikie, Text-book of Geol., 1903, i, 552; ii, 1323-4, 1385-6. Prof. W. M. Davis of Harvard has, however, recently produced fresh evidence ‘in favour of the excavating power of glaciers’ (Trans. Roy. Soc., Edinburgh, xl, part ii, 1902, p. 457); and Ramsay’s theory is ‘in no wise extinct’ (Quart. Journ. Geol. Soc., lxii, 1906, p. 166); but Prof. E. J. Garwood has recently investigated the Alpine lakes near Airolo, and holds (ib., p. 190) that, with a few possible exceptions, they ‘do not seem to be due to ice-erosion’.

[22] For instance, H. B. Woodward, Geol. of England and Wales, 2nd ed., 1887, pp. 475-512; J. Prestwich, Geology, ii, 1888, pp. 453-4, 469; A. J. Jukes-Browne, The Building of the Brit. Isles, 1888, pp. 281, 289, 294-6; Sir A. C. Ramsay, Phys. Geol. and Geogr. of Great Britain, 6th ed., 1894, pp. 229, 238, 242-3, 246-8, 252, 259, 263, 276; T. G. Bonney, Ice-Work, Present and Past, 1896, pp. 121, 277; and Sir A. Geikie, Text-book of Geology, 1903, i, 169; ii, 1302-32.

[23] Mr. T. Mellard Reade, in an interesting paper (Nat. Science, iii, 1893, pp. 423-35) has argued against the view that these shells were carried up the hill of Moel Tryfaen by a glacier. See also, in support of the theory of a period of extensive submergence, Geol. Mag., 1893, pp. 35-7, 104-7; 1896, pp. 488-92; 1897, pp. 229-33.

[24] H. Carvill Lewis, Papers and Notes, &c., 1894, pp. 375-6; Report of ... the Brit. Association, 1893 (1894), pp. 483-514; Nature, Aug. 16, 1906, p. 399; Quart. Journ. Geol. Soc., lxii, 1906, pp. 33, 39. Mr. T. F. Jamieson, the author of the last-named paper, suggests that the submergence may have been confined to the northern part of Scotland.

[25] Nat. Science, iv, 1894, p. 472. Cf. _Trans. Roy. Soc. Edinburgh_, xl, part i, 1904, p. 82.

[26] Clement Reid, Origin of the Brit. Flora, 1899, pp. 39-40.

[27] Vict. Hist. of ... Hampshire, i, 253; Vict. Hist. of ... Somerset, i, 176.

[28] Vict. Hist. of ... Sussex, i, 25-6. See p. 19, infra.

[29] Geol. Mag., 1895, pp. 63-4.

[30] C. Reid, Origin of the Brit. Flora, p. 38.

[31] See Sir A. Geikie’s Text-book of Geology, ii, 1903, p. 1313. Cf. Nature, Aug. 16, 1906, pp. 388-9, 399.

[32] See Sir A. Geikie’s Text-book of Geology, ii, 1903, p. 1313.

[33] Proc. Geologists’ Association, ix, 1887, pp. 111-2.

[34] Quart. Journ. Geol. Soc., liv, 1898, pp. 197-227, especially p. 209; Trans. Roy. Soc. Edinburgh, xl, part i, 1904, p. 83; Vict. Hist. of ... Durham, i, 24. Professor Bonney, however (Quart. Journ. Geol. Soc., lxii, 1906, pp. 491-2, 498), remains unconvinced.

[35] See Mr. G. W. Lamplugh’s opening address, delivered in Section C of the British Association (Nature, Aug. 16, 1906, pp. 387-400). Mr. T. F. Jamieson, in a valuable and interesting paper (Quart. Journ. Geol. Soc., lxii, 1906, p. 23), observes that ‘although we have some evidence of more than one recurrence of an ice-sheet in [Aberdeenshire] ... no evidence has hitherto been obtained of warm intervals, further than that which may be inferred from the melting away of the vast mass of ice which preceded and followed the deposition of the Red Clay and the shell-bed at Clava and elsewhere. It must have taken a great deal of heat to melt these enormous masses.’

[36] Is it certain that an elevation of seventy feet would not have been enough to unite Britain with the Continent? For thousands of years the scour of the tides must have been deepening the Channel. [On April 11, 1906, I submitted to Mr. Clement Reid, in a conversation which I had with him at the Geological Museum, the gist of the argument by which I endeavour to show (pp. 20-2, infra) that during some part of the Palaeolithic Age Britain must have been continental. He virtually admitted its force, remarking that an elevation of seventy feet would have enabled animals to cross from Gaul to Britain, as the scour of the tides had doubtless deepened the Channel.]

[37] Origin of the Brit. Flora, pp. 37, 38.

[38] Ib., p. 41; Quart. Journ. Geol. Soc., xlviii, 1892, pp. 344-61. Cf. Vict. Hist. of ... Hampshire, i, 32.

[39] Origin of the Brit. Flora, p. 39. See also Memoirs Geol. Survey--The Geology of the Country around Cromer, 1882, p. 90.

[40] Vict. Hist. of ... Hampshire, i, 33; _Vict. Hist. of ... Sussex_, i, 22.

[41] Vict. Hist. of ... Hampshire, i, 33-4; Origin of the Brit. Flora, pp. 42-3.

[42] Report of the Brit. Association, 1896, pp. 410-11. Cf. Vict. Hist. of ... Sussex, i, 25-6. Mr. Reid has, however, concluded, from an examination of the palaeolithic deposits at Hitchin (Proc. Roy. Soc., lxi, 1897, pp. 40-9, and especially p. 46), as well as at Hoxne, that before the time of the palaeolithic inhabitants of those districts the land had again sunk.

[43] Origin of the Brit. Flora, p. 46. Cf. A. J. Jukes-Browne, The Building of the Brit. Isles, pp. 291-2, 302.

[44] A. R. Wallace, Island Life, 1880, pp. 315-17; J. Prestwich, Geology, ii, 1888, pp. 523-5. See p. 62, infra.

[45] W. Boyd Dawkins, Cave-Hunting, 1874, p. 124; _Quart. Journ. Geol. Soc._, xxxv, 1879, pp. 139-42; liv, 1898, pp. xcv-xcvi.

[46] Mr. Reid tells us that there was no vegetation except dwarf birches and willows and other Arctic plants. See Nat. Science, i, 1892, pp. 430, 432; C. Reid, Origin of the Brit. Flora, pp. 40, 42; Vict. Hist. of ... Hampshire, i, 32-3.

[47] This fact seems to have escaped the notice of a writer who argues (Nat. Science, iii, 1893, pp. 261-6) that ‘England was not restocked by a land connexion from the Continent after glaciation’, and affirms (p. 266) that ‘almost the only evidence of a post-glacial connexion with the Continent is the supposed necessity of such to account for our present fauna and flora’. The mammoth fed upon coniferous trees, fragments of the wood of which have been found in the crevices of its teeth (A. S. Woodward, Outlines of Vertebrate Palaeontology, 1898, p. 306).

[48] Quart. Journ. Geol. Soc., xxxv, 1879, pp. 142-3.

[49] Sir Henry Howorth has written a series of articles (Geol. Mag., 1892, pp. 250-8, 396-405; Nat. Science, xii, 1898, pp. 261-70), in which he claims to have proved that ‘in no instance, so far as we know, does the drift actually underlie any land surface containing the remains of the mammoth and of its contemporaries.’ Translated into the language of geologists to whom the glacial period is not a nightmare, this is tantamount to an assertion that the mammoth was neither postglacial, nor interglacial, nor glacial, but preglacial. Mr. A. J. Jukes-Browne (Geol. Mag., 1892, p. 575) replies that ‘gravels containing mammoth remains occur in many other valleys [besides that of the Great Ouse], which are generally considered to have been eroded out of a widespread mantle of Glacial Drift’, and that ‘this conclusion is not shaken by anything which Sir H. Howorth has written’; while Sir John Evans (Anc. Stone Implements, 1879, p. 701) remarks that ‘in some cases, as at Fisherton, the worked flints have been found below the remains of mammoth’. Since the gravel at Hoxne, in which bones of the mammoth and of extinct animals contemporary with the mammoth were found (Sir J. Evans, Anc. Stone Implements, 1897, pp. 573-5; C. Reid, Origin of the Brit. Flora, p. 77; Guide to the Ant. of the Stone Age [Brit. Museum], p. 20), was shown by the committee who excavated it (Report of ... the Brit. Association, 1896, pp. 400-11) to be later than the latest glaciation of the district, Sir Henry naturally discredited their report; and he did so by declaring (Nat. Science, xii, 1898, p 266) that the members of the committee were ‘already committed ... to the view that the implement-bearing deposit at Hoxne was newer than the Drift. This,’ he continued, ‘was not very promising.... It was, in fact, indecent.’

[50] Cf. Quart. Journ. Geol. Soc., xxiii, 1867, p. 107, with Geol. Mag., 1878, p. 98, and Boyd Dawkins, Early Man in Britain, 1880, p. 149; and see also the same writer’s Cave-Hunting, 1874, p. 362. Mr. H. B. Woodward states (Vict. Hist. of ... Norfolk, i, 23) that ‘the Dogger Bank is a remnant of old Pleistocene deposits; as Mr. Reid suggests [Memoirs Geol. Survey,--The Geology of the Country around Cromer, 1882, p. 122], a re-extension of the old Rhine estuary’. I confess that I do not understand how Mr. Reid would reconcile this suggestion with his belief that the Channel was formed in the earliest part of the Ice Age.

[51] Vict. Hist. of ... Hampshire, i, 34.

[52] Memoirs Geol. Survey,--The Geology of the Country around Cromer, p. 122; Vict. Hist. of ... Norfolk, i, 23.

[53] Sir J. Evans, Anc. Stone Implements, 1897, pp. 577, 683, 685, 697.

[54] See p. 21, n. 5, supra, and cf. Journ. Anthr. Inst., xxxv, 1905, pp. 308, 310.

[55] Journ. Anthr. Inst., xxix, 1899, pp. 333-4; Proc. Liverpool Geol. Soc., ix, 1901, pp. 18-19.

[56] Memoirs Geol. Survey,--On the Manufacture of Gun-Flints, 1879, p. 68. Cf. S. H. Miller and S. B. J. Skertchly, The Fenland, 1878, pp. 548-51.

[57] Anc. Stone Implements, 1897, p. 568. See, however, Proc. Geologists’ Association, ix, 1887, p. 126.

[58] Quart. Journ. Geol. Soc., xliii, 1887, p. 117; liv, 1898, pp. lxxxvi-lxxxix.

[59] Ib., lx, 1904, pp. 132-3.

[60] Professor Boyd Dawkins (Quart. Journ. Geol. Soc., xxiii, 1867, pp. 91-109, and especially pp. 106-9) has argued that the Lower Brick-Earths in the Thames Valley, under which implements have been found, were preglacial,--locally, I presume. He observed that not one of ‘the Post-glacial Arctic mammalia’, namely, the glutton, lemming, marmot, musk-sheep, elk, and reindeer, is represented in this deposit, and maintains that, on the other hand, the presence of Elephas priscus and the big-nosed rhinoceros (Rhinoceros megarhinus) ‘indicates the affinity of the [Brick-Earth] group to the Praeglacial deposits of Norfolk’, &c. Prestwich, however (ib., xxviii, 1872, p. 445), differed from the professor; and Sir John Evans (ib., p. 446) remarked that if the brick-earth were preglacial ‘there would be a great difficulty in accounting for the presence of the high beds at Shackleton and Highbury, as these, though in a valley confessedly excavated by the river, and regarded as of more recent age than the lower beds, would yet be at a far higher level’.

[61] Prof. P. F. Kendall maintains (ib., lx, 1904, p. 132) that even the Hoxne implements ‘were of very late Glacial, perhaps the very latest Glacial Age,’--not, as I understand, of the Hoxne district, but of Britain as a whole. Cf. Man, iii, 1903, No. 31, p. 59.

[62] Proc. Geologists’ Association, ix. 1887, p. 129; J. Prestwich. Controverted Questions, p. 45.

[63] See Geol. Mag., 1894, p. 79.

[64] Sir J. Evans, Anc. Stone Implements, 1897, p. 580.

[65] Quart. Journ. Geol. Soc., xxviii, 1872, p. 435; xxxv, 1879, pp. 142-3.

[66] Memoirs Geol. Survey,--The Geology of the Country around Bournemouth, 1898, p. 10; C. Reid, Origin of the Brit. Flora, pp. 44-5; Vict. Hist. of ... Hants, i, 35; Vict. Hist. of ... Sussex, i, 22.

[67] Worthington G. Smith, Man, the Primeval Savage, 1894, pp. 170, 173, 191, 217-8. Cf. H. B. Woodward, Geol. of England and Wales, 1887, pp. 510-12, and Sir J. Evans, Anc. Stone Implements, 1897, p. 698.

[68] The descriptions, based upon the remains of fauna in caves, that have been given of the climate of Southern Gaul in successive periods of the Palaeolithic Age, however true they may be, do not apply in Britain. The little that is known of our climate suggests to Mr. Clement Reid ‘extremes with sharp alternations of cold, drought, and sudden floods’ (Man, iii, 1903, No. 29, p. 56, with which cf. Proc. Roy. Soc., lxi, 1897, p. 46).

[69] Journ. Anthr. Inst., xxiii, 1894, p. 275; B. Harrison, Outline of the Hist. of the Eol. Flint Implements, 1904, pp. 9-10.

[70] Nineteenth Century, April, 1895, p. 623.

[71] Journ. Anthr. Inst., xxi, 1892, p. 272.

[72] Controverted Questions, p. 77.

[73] Ib., p. 78. See also Quart. Journ. Geol. Soc., liv, 1898, p. 298.

[74] The Rev. R. Ashington Bullen (Nat. Science, xii, 1898, p. 107) says that he has found ‘worked flints of the plateau types’ in ‘valley gravels’, but that they were ‘derived specimens’. See also ib., pp. 111-16. On the other hand, see p. 27, infra.

[75] Nat. Science, v, 1894, pp. 269, 271-2; Nineteenth Century, April, 1895, p. 626.

[76] J. Prestwich, Controverted Questions, 1895, p. 54. Cf. Quart. Journ. Geol. Soc., xlv, 1889, p. 295, and Journ. Vict. Inst., xxxiii, 1901, p. 223.

[77] Quart. Journ. Geol. Soc., xxviii, 1872, pp. 39-40; Sir J. Evans, Anc. Stone Implements, 1897, p. 632; Memoirs Geol. Survey,--The Geology of the Country around Ringwood, 1902, pp. 36, 39.

[78] As Mr. Clement Reid points out (ib., pp. 36-7), Prestwich’s implement ‘was not found in place, but picked up among fallen material.... The Alderbury gravel,’ he remarks, ‘judging from its less elevation above the river, is probably newer than the supposed Palaeolithic gravel north of Redlynch; yet it yields implements of more ancient type.’ See also Journ. Anthr. Inst., xxxv, 1905, p. 354, where Mr. H. Warren affirms that eoliths are associated with palaeolithic implements in the drift of High Down, Isle of Wight; and cf. Man, v, 1905, No. 80, p. 146.

[79] Geol. Mag., 1903, pp. 105-6. Mr. Reid thinks that the beds in which these flints have been found are not necessarily of Pliocene date, as they may have been remaniés. Eoliths are said to have been unearthed from gravels at Dewlish in Dorsetshire side by side with the bones of the extinct elephant known as Elephas meridionalis, whose remains have never yet been met with in this island except in preglacial beds (Quart. Journ. Geol. Soc., xliv, 1888, pp. 318-24; lxi, 1905, pp. 35-8; Journ. Vict. Inst., xxxiii, 1901, pp. 212-3); but these flints were so battered that Mr. Reid, who accepts many eoliths as genuine tools and regards them as ‘bad palaeoliths’, was obliged to reject them (Memoirs Geol. Survey,--The Geology of the Country around Ringwood, p. 36).

[80] Quart. Journ. Geol. Soc., liv, 1898, pp. 291, 293-4.

[81] Association franç. pour l’avancement des sc., 1903, 1^{re} partie, pp. 246-7; Nature, lxxii, 1905, pp. 438-9; Journ. Anthr. Inst., xxxv, 1905, p. 261. Are any of these flints identical in form with the characteristic Kentish specimens?

Sir John Evans, who is unable to accept the authenticity of any eoliths, nevertheless believes, or did believe in 1897 (Anc. Stone Implements, pp. 608-9), that the palaeolithic implements which have been found on the plateau belonged to a time when the ‘continuous slope now extending from the neighbourhood of the Thames to the summit of the chalk escarpment’ was ‘continued southward ... over a part of what is now the Lower Greensand area, if not, indeed, into that of the Weald’. In other words he believed that the palaeoliths were as old as the eoliths, and therefore that the question of the authenticity of the latter was unimportant. It is, however, now generally recognized that this view was based upon a misconception. Mr. Harrison (Outline of the Hist. of the Eol. Flint Implements, p. 17) states that ‘palaeoliths and eoliths have been found together only on the surface and never in the drifts in situ’. Cf. J. Prestwich, Controverted Questions, p. 64.

Mr. Clement Reid (Vict. Hist. of ... Hampshire, i, 34) sees no reason for believing that any of the Kentish eoliths are older than palaeolithic implements in general. In a conversation which I had with him on April 11, 1906, he remarked that the patches of drift in which the eoliths had been found were generally dominated by higher ground, and that he could find no evidence that the flints had been washed down from the Weald. Eoliths have, however, been found in a pit at Terry’s Lodge ‘on the summit of the escarpment at a height of 770 feet’ (Journ. Anthr. Inst., xxxv, 1905, p. 300. Cf. Essex Naturalist, xiii, 1904, p. 332).

[82] L’Anthropologie, xvi, 1905, pp. 257-67.

[83] Man, v, 1905, No. 102, p. 179. Cf. No. 92, p. 165.

[84] Ib., No. 91, p. 165.

[85] Ib., No. 103, pp. 180-83. Cf. Journ. Anthr. Inst., xxxv, 1905, p. 363, fig. 7.

[86] Ib., p. 361.

[87] Mr, Hazzledine Warren (Journ. Anthr. Inst., xxxv, 1905, p. 358) goes so far as to affirm that the mineral condition of some of Mr. Harrison’s eoliths ‘shows that they ... are ... clearly as late as the neolithic age’. There is a bibliography of eoliths in Geol. Mag., 1903, pp. 108-10, to which may be added, besides the works quoted in this chapter, Rev. de l’École d’anthr., xiv, 1904, pp. 240-6.

[88] See p. 32, infra. It must of course be remembered that this description applies only to one part of the Palaeolithic Age: palaeolithic man was still here when the Thames had cut out its valley to its present depth.

[89] See p. 40, n. 2, infra.

[90] M. Boule (L’Anthr. xiv, 1903, p. 533) regards the question of the existence of a palaeolithic age in Egypt as unsettled; but, as Mr. H. R. Hall observes (Man, v, 1905, No. 19, p. 34), ‘German investigators ... have no doubt whatever that the Pitt-Rivers flints from Thebes and those of palaeolithic type from the Wâdî esh-Shêkh and elsewhere are in reality palaeolithic.’ See also A. Pitt-Rivers, Excavations in Cranborne Chase, iv. 9-11 (preface), and Association franç. pour l’avancement des sc., 1903, 2^{e} partie, p. 860. Palaeolithic implements are also said to have been found in Patagonia (L’Anthr., xvii, 1906, p. 255).

[91] Sir John Evans in Report of ... the Brit. Association, 1897 (1898), p. 14. Cf. Anc. Stone Implements, 1897, pp. 528-30, 650-54. In regard to palaeolithic remains in America see also Congrès internat. d’anthr. et d’archéol. préhist., 1900 (1902), p. 191.

[92] Journ. Anthr. Inst., xxiii, 1894, p. 147. Cf. A. Pitt-Rivers, Evolution of Culture, 1906, p. xvi.

[93] Climate and Time, 1885, pp. 327-8.

[94] Geol. Mag., 1868, pp. 249-54.

[95] J. Prestwich, Controverted Questions, pp. 22, 42; Sir J. Evans, Anc. Stone Implements, 1897, pp. 705-7.

[96] Geol. Mag., 1895, pp. 3-13, 55-65; A. Geikie, Text-book of Geology, 1903, pp. 1326-7. See also Nature, lii, 1895, p. 594; liii, 1895-6, pp. 29, 196, 220, 269, 295, 317, 340, 388, 460; and Quart. Journ. Geol. Soc., lviii, 1902, pp. 37-45.

[97] Sir J. Evans, Anc. Stone Implements, 1897, pp. 592, 708-9.

[98] Clement Reid in Vict. Hist. of ... Hampshire, i, 34; Archaeologia, lix, part ii, 1906, p. 286, and _Memoirs Geol. Survey,--The Geology of the Country around Ringwood_, pp. 31-2.

Sir John Evans (Anc. Stone Implements, 1897, pp. 634-5, 690-93) has argued that the implementiferous gravel which caps the cliff at Bournemouth was deposited by the Solent river; but Mr. Clement Reid thinks it ‘very doubtful whether it was a deposit formed by ordinary river action’ (Vict. Hist. of ... Hampshire, i, 34); and (ib., pp. 27-8, 34) he is inclined to believe that the continuity of the Isle of Wight with Hampshire and Dorsetshire was already interrupted in late Pliocene times, though the Solent may perhaps have been merely an estuary and not a strait even in the time of the so-called interglacial estuarine deposits. See p. 20, supra.

[99] Dr. A. J. Evans (Report of ... the Brit. Association, 1904 [1906], p. 721) calculates that the earliest settlement at Knossos in Crete (which was neolithic) is about 12,000 years old; but he assumes that in the western court of the palace ‘the average rate of deposit was fairly continuous’. Prof. Montelius (L’Anthr., xvii, 1906, p. 137) argues from the stratigraphy of finds at Susa that the beginning of the Neolithic Age in the East may be dated about 18,000 B.C. But even if these calculations could be established, it would still remain doubtful whether our Palaeolithic Age was not partly contemporary with a neolithic civilization in more genial climates. Probably it was (ib., p. 164). Against the theory which would minimize the antiquity of the Palaeolithic Age, see ib., xv, 1904, p. 66, and Quart. Journ. Geol. Soc., xliii, 1887, p. 410, and in favour of it L’Anthr., xvii, 1906, p. 27, n. 1.

[100] Trans. Devon. Association, xix, 1887, pp. 419-37. Neither of the skulls could be removed intact, but one was photographed (ib., p. 433). The forehead recedes, but not excessively: the supraciliary ridge is strong, but not abnormally developed.

[101] See p. 380, infra.

[102] L’Anthr., xvii, 1906, pp. 70-3. See pp. 380-1, infra.

[103] See pp. 380-1, infra.

[104] See p. 381, infra.

[105] See pp. 382-3, infra.

[106] See pp. 382-3, infra.

[107] Mr. C. H. Read (Guide to the Ant. of the Stone Age [Brit. Museum], pl. 1 and p. 49) has no doubt that they were dagger-handles; but the abbé Breuil (L’Anthr., xvi, 1905, p. 632) affirms that it is ‘démontré qu’aucune des sculptures dont on a voulu faire des manches de poignard n’ont eu ce rôle’.

[108] E. Lartet and H. Christy, Reliquiae Aquitanicae, passim; L’Anthr., v, 1894, pp. 129-46; vi, 1895, p. 143; xiv, 1903, pp. 295-315; xv, 1904, pp. 129-76, 625-44. Among the palaeolithic artists were not only carvers and engravers but also draughtsmen and even painters. On the walls of caves in the Spanish Pyrenees are many-coloured frescoes, depicting animals as well as objects the meaning of which is still unknown. See L’Anthr., xv, 1904, p. 629; xvi, 1905, pp. 437, 442; Rev. de l’École d’anthr., xiv, 1904, pp. 320-5; xv, 1905, pp. 150-5; and Man, vi, 1906, No. 63, p. 96.

[109] Quart. Journ. Geol. Soc., xxxiii, 1877, p. 582.

[110] Sir J. Evans, Anc. Stone Implements, 1897, p. 581; Guide to the Ant. of the Stone-Age (Brit. Museum), p. 6. Mr. Clement Reid (Quart. Journ. Geol. Soc., lx, 1904, pp. 106-12), has described ‘a probable Palaeolithic Floor [or old land surface] at Prah Sands, Cornwall’; but in the discussion which followed the reading of the paper he admitted that he ‘would not like to speak confidently as to any one of the stones being an implement’.

[111] See Nat. Science, iii, 1893, p. 369; Vict. Hist. of ... Hampshire, i, 35; Mem. Geol. Survey,--The Geology of the Country around Ringwood, 1902, p. 48; Man, iii, 1903, No. 29, p. 56; and Quart. Journ. Geol. Soc., lx, 1904, p. 130.

Mr. Clement Reid, as those who are familiar with his writings must have seen, does not believe that many of the deposits classed as river-drift (see Evans’s Anc. Stone Implements, 1897, pp. 662-709, especially 679) deserve that title. In a conversation which I had with him on April 11, 1906, he remarked that he could see no reason to suppose that palaeolithic man ‘was an aquatic animal’; that much of the so-called river-drift would probably be found, under minute examination, not to be due to fluviatile action; and that the geology of the Thames Valley, which in the Glacial Epoch was on the edge of the ice, presented great difficulties. See, however, Mr. H. B. Woodward’s article in Vict. Hist. of ... Buckingham, i, 22.

[112] See, however, Guide to the Ant. of the Stone Age (Brit. Museum), p. 3.

[113] Sir J. Evans, Anc. Stone Implements, 1897, pp. 613-7.

[114] Sir J. Evans, Anc. Stone Implements, 1897, pp. 479, 488-525; Phil. Trans., clxiii, 1874, pp. 553-70; Quart. Journ. Geol. Soc., xxxii, 1876, pp. 240-58; xxxiii, 1877, pp. 579-612; xxxv, 1879, pp. 724-35.

[115] Sir J. Evans, Anc. Stone Implements, 1897, p. 485.

[116] Ib., pp. 474-5.

[117] See pp. 383-5, infra.

[118] Sir J. Evans, Anc. Stone Implements, 1897, pp. 475, 483-5, 528, 530, 575-6: Quart. Journ. Geol. Soc., lx, 1904, p. 132. In de Mortillet’s classification the oldest type was the Chellean (called after Chelles in the department of Seine-et-Marne): then followed successively the types represented in the cave of Le Moustier, at Solutré, and in the cave of La Madelaine. Dr. M. Hoernes (Der diluviale Mensch in Europa, 1903, pp. 21, 63, 185-6, &c.) combines the Chellean and Mousterian periods. (See also Rev. mensuelle de l’École d’anthr., v, 1895, p. 407, and L’Anthr., xv, 1904, pp. 27, 196-8).

[119] Quart. Journ. Geol. Soc., xxxii, 1876, pp. 252-3; Sir J. Evans, Anc. Stone Implements, 1897, pp. 522-3.

[120] Worthington G. Smith, Man, the Primeval Savage, pp. 215, 220.

[121] Worthington G. Smith, Man, the Primeval Savage, pp. 60-89, 96-175. Cf. L’Anthr., xvi, 1905, p. 27.

[122] Prof. Boyd Dawkins (Early Man in Britain, p. 192) says that, except at Pont Newydd, ‘the association of traces of man with the remains of hippopotamus has, as yet, not been observed in any bone caves either in this country or on the Continent’. Sir John Evans, who does not mention such remains in his notice of Pont Newydd (Anc. Stone Implements, 1897, p. 521), records their discovery, without associated implements, in the ‘mid-terrace gravels’ near Acton (ib., p. 591), and in gravels of the same character as those which yield implements, near Bedford (p. 533) and at Folkestone (p. 621). Evidently (pp. 699-700, with which cf. Boyd Dawkins and W. Ayshford, Brit. Pleistocene Mammalia, 1866, p. xxviii) he has no doubt that the hippopotamus was contemporary in Britain with palaeolithic man; but Mr. Clement Reid, in a conversation which I had with him on April 11, 1906, questioned whether its bones had ever been found together with implements.

[123] Sir J. Evans, Anc. Stone Implements, 1897, pp. 528, 533, 536, 591; M. Hoernes, Der diluviale Mensch in Europa. 1903, p. 13. Readers who are interested in the question which is raised by the discoveries of Arctic in association with tropical mammalian remains should consult Quart. Journ. Geol. Soc., xxviii, 1872, pp. 426-43; xxxv, 1879, p. 142; W. Boyd Dawkins, Early Man in Britain, pp. 113-14; E. Piette, La France préhist. par M. Cartailhac, 1890, pp. 5-6; Nat. Science, i, 1892, p. 432; iii, 1893, pp. 262-3; Lord Avebury, Prehist. Times, 6th ed., 1900, p. 290; Guide to the Ant. of the Stone Age (Brit. Museum), pp. 36-7; and R. Lydekker, Mostly Mammals, p. 269. See also, in regard to the contrast between the intermingling of tropical and Arctic animals in Britain and Northern Gaul and their succession in South-Western Gaul, M. Hoernes, op. cit., p. 193, and L’Anthr., xiii, 1902, pp. 305, 317.

[124] Ib., xv, 1904, pp. 57-8; xvi, 1905, p. 67.

[125] See p. 384, infra.

[126] Man, vi, 1906, No. 63, p. 94. Chellean implements have been found at Le Moustier, evidently in situ, in the second layer from the top, among those of the Madelaine period.

[127] Report of ... the Brit. Association, 1903, pp. 804-5.

[128] Report of ... the Brit. Association, 1904 (1905), p. 726; Journ. Anthr. Inst., xxxiv, 1904, p. 308.

[129] Sir J. Evans, Anc. Stone Implements, 1897, pp. 504-7, 512-7, 523, 565-6, 581, 640-9, 655-6; Worthington G. Smith, Man, the Primeval Savage, pp. 110-11, 121, 248-9.

[130] Archaeol. Journal, xxxvii, 1880, pp. 294-9.

[131] Worthington G. Smith, Man, the Primeval Savage, pp. 113-4, 116, 142-3, 165.

[132] Ib., pp. 262-7; Vict. Hist. of ... Hertford, i, 224; Sir J. Evans, Anc. Stone Implements, 1897, p. 274. Cf. J. Prestwich, Controverted Questions, pp. 76-7.

[133] Journ. Anthr. Inst., xxiii, 1894, p. 145.

[134] Sir J. Evans, Anc. Stone Implements, 1897, pp. 645, 656; Worthington G. Smith, Man, the Primeval Savage, p. 222; Vict. Hist of ... Hertford, i. 224. Professor Boyd Dawkins (Early Man in Britain, pp. 183-4) affirms that certain implements found in the upper cave-earth of Church Hole and the Robin Hood Cave at Creswell Crags ‘had obviously been let into a handle ... by which the edge of one side had been protected, while the other was worn away by use’; and in Nature (May 22, 1902, p. 77) it is stated that a palaeolithic implement, recently discovered near Ipswich, ‘shows signs of having been worked for hafting.’

[135] Sir J. Evans, Anc. Stone Implements, 1897, pp. 645, 655.

[136] Prehist. Times, 1900, p. 332.

[137] See Journ. Roy. United Service Inst., xii, 1868, pp. 408-9.

[138] Antiquity of Man, 4th ed., 1873, p. 422.

[139] Sir J. Evans, Anc. Stone Implements, 1897, pp. 486, 657-8. Traces of corn have been found in French palaeolithic caves, though there is no evidence that it was cultivated. See Congrès internal. d’anthr. et d’archéol. préhist., 1900 (1902), p. 408.

[140] E. Lartet and H. Christy, Reliquiae Aquitanicae, B. 1, pl. ii, fig. 5. Prof. Boyd Hawkins (Early Man in Britain, p. 214) has no doubt about the subject of the drawing: I confess that I am not so certain.

[141] ‘It is doubtful,’ says H. E. Schoolcraft (Indian Tribes of the United States, i, 1851, p. 433), ‘whether an area of fifty thousand acres, left in the forest state, is more than sufficient to sustain by the chase a single hunter.’ One may be allowed, however, to suspect an exaggeration in this estimate; otherwise how could the communities who dwelled at Caddington and Crayford (see pp. 39, 42-4, supra) have escaped starvation? See also A. Lang, The Secret of the Totem, 1905, pp. 6-7, 88-9, 151-2.

[142] See Mr. Lewis Abbott’s paper in J. Salmon’s Guide to Sevenoaks, 1905, pp. 120-1. Cf. Archaeol. Journal, xxxix, 1882, p. 17.

[143] Life of Sir J. Prestwich, 1899, p. 376.

[144] Sir J. Evans, Anc. Stone Implements, p. 501.

[145] W. Boyd Dawkins, Early Man in Britain, p. 211. I must admit that I feel doubtful whether the illustration in Reliquiae Aquitanicae which Professor Dawkins reproduces really represents gloves.

[146] Sir J. Evans, Anc. Stone Implements, 1897, p. 657.

[147] L’Anthr., xiv, 1903, pp. 258, 263-5; Comptes rendus ... de l’Acad. des Sciences, 1903, pp. 1536-7.

[148] E. Lartet and H. Christy, Reliquiae Aquitanicae, p. 209.

[149] L’Anthr., xv, 1904, p. 174.

[150] Ib., v, 1894, p. 146.

[151] Bull. et mém. de la Soc. d’anthr., 5^e sér. iii, 1902, p. 771. It is remarkable that Ezekiel (viii. 10-11), speaking of seventy of ‘the ancients of the house of Israel’ who were worshipping in a court, says that he saw therein ‘every form of creeping things, and abominable beasts ... pourtrayed upon the wall round about’. These were ‘unclean’ animals, which were not to be eaten. Cf. A. Lang, Custom and Myth, 2nd ed., 1885, p. 115.

M. Reinach also insists (L’Anthr., xiv, 1903, pp. 264-5) that the so-called sceptres, or bátons de commandement--engraved and perforated instruments of reindeer-horn--which have been found in French palaeolithic caves, were used in magical ceremonies; whereas it has been proved by Dr. O. Schoetensack (ib., xii, 1901, pp. 140-4) that they were merely dress-fasteners similar to those which are used by the Eskimos.

[152] See A. Lang, Custom and Myth, 1885, pp. 294, 296.

[153] L’Anthr., xvii, 1906, p. 293.

[154] See p. 34, supra.

[155] L’Anthr., xvi, 1905, p. 395. Cf. p. 321 of the same volume, and vol. iv, 1893, p. 550.

[156] Journ. Anthr. Inst., xxi, 1892, p. 297; xxiii, 1894, pp. 147, 151.

[157] L’Anthr., xvii, 1906, pp. 293-4. The Tasmanians ‘placed weapons near the grave for the dead friend’s soul to use’ (E. B. Tylor in Ency. Brit., xxv, 1902, p. 467). Cf. pp. 200-2, infra.

[158] See pp. 262-3, 464, infra.

[159] Mr. Andrew Lang (Man, iv, 1904, No. 22, p. 37), remarking that in the cave of Mas d’Azil, in the department of the Ariège, there has been found a pendeloque of bone which exactly resembles some Australian ‘bull-roarers’ (L’Anthr., xiv, 1903, pp. 655-60), infers that ‘palaeolithic and neolithic man ... probably had such religious ideas as among savages are attached to bull-roarers’. There is an interesting chapter on bull-roarers (which in this country are more familiar to schoolboys than to scholars) in Mr. Lang’s Custom and Myth, 1885, pp. 29-44.

[160] See E. B. Tylor, Prim. Culture, 4th ed., 1903, i, 417-24.

[161] Ib., p. 424.

[162] See pp. 461-3, infra.

[163] It is said that totemism exists in New Guinea (Man, v, 1905, No. 2) and on the Gold Coast (Journ. Anthr. Inst., xxxvi, 1906, pp. 178-88).

[164] See A. Lang, The Secret of the Totem, pp. 2, 66.

[165] See W. Robertson Smith, The Religion of the Semites, 1901, p. 355.

[166] E. B. Tylor, Prim. Culture, ii, 1903, p. 237.

[167] See Journ. Anthr. Inst., xxviii, 1899, p. 146.

[168] Ib., p. 147.

[169] A. Lang, The Secret of the Totem, pp. 22-3. I confess that I cannot understand why descent should have been reckoned in the female line if, as Mr. Lang apparently holds, the master of each little primitive group was the only sire in that group. [I am glad to find that Dr. W. H. D. Rouse (Folk-Lore, xvii, 1906, p. 25) has argued in the same sense.]

[170] E. B. Tylor, Prim. Culture, ii, 1903, p. 236; B. Spencer and F. J. Gillen, Native Tribes of Central Australia, 1899, pp. 73, 121; Man, iv, 1904, No. 93, p. 143. See also No. 98, p. 150.

[171] L’Anthr., xiii, 1902, pp. 665-7.

[172] A. Lang, The Secret of the Totem, pp. 59-89, especially 66, 68, 70, 72-4, 89. See also pp. 7-8 in regard to the complex organization of Australian tribes. The Aruntas (ib., pp. 17-18) do not inherit their totems, which are ‘determined by local accident’.

[173] Ib., p. 29. Mr. Lang conjectures (p. 114) that the master of a small group, actuated by sexual jealousy, ‘expelled all his adult sons as they came to puberty.’ Such a group, he remarks, would have been ‘necessarily exogamous in practice’, and then (ib., p. 143) would have come the rule, ‘No marriage within the local group.’ But would it have been to the interest of the master to expel sons who were useful? What would have become of them? Would not the same sexual jealousy that ex hypothesi prompted their expulsion have prevented the master of any other group from receiving them? And if the master was killed in hunting after he had expelled his sons, what became of the other members of the group?

In connexion with Mr. Lang’s book, see Man, vi, 1906, No. 17, pp. 27-8, No. 34, pp. 51-4, No. 87, p. 131, and No. 112, p. 182.

[174] A. Lang, The Secret of the Totem, pp. 116, 127-8, 153. Cf. Lord Avebury’s Origin of Civilisation, 1902, p. 275.

[175] A. Lang, The Secret of the Totem, p. 125.

[176] Archaeol. Rev., iii, 1889, p. 220.

[177] Ib., p. 227.

[178] Notes and Queries, 3rd ser., iv, 1863, pp. 82, 158. These passages, which are referred to by Mr. Gomme in vol. iii of Archaeol. Rev., do not support the statement in the text about the geese of Great Crosby, for which he is responsible.

[179] Archaeol. Rev., iii, 1889, p. 355.

[180] B. G., v, 12, § 6.

[181] M. S. Reinach’s explanation of this passage (Rev. celt., xxi, 1900, p. 275) was anticipated by Elton (Origins of English Hist., 2nd ed., 1890, p. 288).

[182] Journ. Anthr. Inst., xxviii, 1899, pp. 141, 143-4, 148. M. Reinach seems to make this assumption when he says (see preceding note) that we are justified in affirming that ‘chez certaines tribus au moins de la Bretagne, le lièvre, l’oie et la poule étaient des animaux sacrés, c’est-à-dire, des totems’. I am glad to find that M. Camille Jullian (Rev. des études anc., iv, 1902, p. 274) also rejects M. Reinach’s guess; but he continues, ‘Je suis, du reste, convaincu, avec M. Reinach, que les Celtes ont connu le totémisme ... par exemple, si bran(n)os signifie ... “le corbeau”, une tribu gauloise avait pris cet oiseau pour totem ... Aulerci Brannovices,’ &c. On the much surer evidence of such names as Bull, Lamb, Herring, Roach, and many others, M. Jullian might conclude that ‘les Anglais du vingtième siècle connaissent le totémisme’. It is perhaps reasonable to conjecture that the name Brannovices may point to a remote age when the ancestors of the historic Celts had totems: but it is quite certain that the Celts of whom M. Jullian is thinking knew nothing about totemism; and the superstitions which forbade the Britons to eat hares, geese, and fowls, may have been absolutely unconnected with totemism. See Lord Avebury’s Origin of Civilisation, 1902, p. 19. Miss Eleanor Hull (Folk-Lore, xii, 1901, p. 49) observes that ‘there is one example of what appears to be a true totemistic idea in those [Irish] stories.... It is in Cúchulainn’s prohibition to eat the flesh of a hound because it was his namesake.’

[183] W. Boyd Dawkins, Cave-Hunting, p. 165.

[184] Notably Dr. F. B. Jevons (Folk-Lore, x, 1899, pp. 374-5) and M. S. Reinach (Rev. celt., xxi, 1900, pp. 283, 299, 305).

[185] See W. Ridgeway, Origin ... of the Thoroughbred Horse, 1905, pp. 90-1, 479, and L’Anthr., xvii, 1906, pp. 27-53, especially 27, 29, and figs. 1 and 1_a_. If these illustrations, which purport to reproduce late palaeolithic engravings of horses, are accurate, they unquestionably depict halters, though M. Zaborowski (Association franç. pour l’avancement des sc., 32^e sess., 1903, 2^{e} partie, p. 849) thinks that they only represent lassoes.

[186] See a very interesting review [by Mr. Andrew Lang?] in the Athenæum, April 22, 1905, pp. 502-3, of M. Reinach’s Cultes, mythes et religions, and also papers on the domestication of animals in the numbers for April 29 (p. 533), May 6 (p. 565), and May 13 (p. 597).

[187] It is impossible to tell whether in Ancient Britain oxen were at any time regarded as sacred, as they apparently were among the early Phoenicians, the Libyans, the ancestors of the Greeks, and other primitive peoples, their flesh being never eaten except in sacrificial feasts, partaken of by the whole clan. See W. Robertson Smith, The Religion of the Semites, 1901, pp. 296-311.

[188] L’Anthr., xiv, 1903, pp. 355-7. Cf. Congrès internat. d’anthr. et d’archéol. préhist., 1900 (1902), pp. 408-9.

[189] Journ. Anthr. Inst., xxviii, 1899, p. 148. Cf. Man, v, 1905, No. 2, p. 6.

[190] Chambers’s Ency., vi, 1901, p. 795. M. S. Reinach (L’Anthr. xvi, 1905, p. 660) regards magic as ‘la mère de toutes les vraies sciences’.

[191] J. G. Frazer, Early Hist. of the Kingship, 1905, pp. 37-9, 43-4, 77-8, &c.

[192] See Man, vi, 1906, No. 40, p. 62, No. 112, p. 189 (for a criticism of Dr. Frazer’s ‘oil-and-water theory’ of magic and religion), and Mr. Sidney Hartland’s most interesting presidential address to the Anthropological Section of the British Association (Times, Aug. 7, 1906, p. 11, cols. 4-6).

[193] Folk-Lore, xv, 1904, pp. 159-60.

[194] L’Anthr., xvi, 1905, pp. 574-5.

[195] Ib., p. 660.

[196] Man, v, 1905, No. 10, pp. 18-19.

[197] J. G. Frazer, Early Hist. of the Kingship, passim.

[198] See Man, vi, 1906, No. 29, p. 46.

[199] See A. Lang, Custom and Myth, 1885, p. 237.

[200] Ib., p. 242; J. G. Frazer, Early Hist. of the Kingship, pp. 2, 3, 36-7.

[201] See pp. 387-8, infra.

[202] Sir J. Evans, Anc. Stone Implements, 1897, p. 704.

[203] See p. 20, supra.

[204] See p. 398, infra.

[205] Vict. Hist. of ... Somerset, i, 178.

[206] See pp. 382, 389, n. 6, infra.

[207] See p. 389, infra. M. L. Siret (L’Anthr., xvii, 1906, p. 127) says the same for Spain.

[208] See pp. 385-90, infra. ‘The ... transition,’ says Mr. Clement Reid (Origin of the Brit. Flora, p. 45), ‘from the Palaeolithic to the Neolithic is, unfortunately, one of the most obscure, and I can only suggest that the break is more apparent than real, and that one follows the other in close succession.’ See also p. 93 of the same book.

[209] See Mr. Clement Reid’s chapter in Vict. Hist. of ... Hants, i, 35-6.

[210] See A. J. Jukes-Browne, The Building of the Brit. Isles, p. 300. Mr. Clement Reid (Origin of the Brit. Flora, p. 46) states that in the early part of the Neolithic Age ‘the land stood ... some 60 or 70 feet above its present level’. Cf. p. 20, supra.

[211] Journ. Ethn. Soc., N. S., ii, 1870, pp. 141-5; J. Prestwich, Geology, ii, 523-4; A. J. Jukes-Browne, _The Building of the Brit. Isles_, pp. 300-2; Clement Reid, Origin of the Brit. Flora, p. 46.

[212] Nature, Jan. 6, 1898, p. 235; Archaeol. Journal, lv, 1898, p. 271.

[213] Ib., p. 272.

[214] Archaeol. Journal, lv, 1898, p. 270; Proc. Soc. Ant. Scot., xxix, 1895, pp. 223-7, 431-2, 438; R. Munro, Prehist. Problems, 1897, p. 72.

[215] See L’Anthr., vii, 1896, pp. 319-24, and M. Hoernes, Der diluviale Mensch in Europa, p. 185.

[216] Proc. Dorset Nat. Hist. and Ant. Field Club, xvii, 1896, pp. 67-75.

[217] See pp. 395-7, infra.

[218] See Vict. Hist. of ... Hants, i, 256.

[219] Ib., p. 37; Vict. Hist. of ... Sussex, i, 25.

[220] See Vict. Hist. of ... Hertford, i, 229, and Archaeol. Journal, lv, 1898, p. 285. Dr. A. H. Keane’s extravagant estimates of the length of the Neolithic Age in Europe, which vary between the limits of ‘scarcely less than 60,000 years’ (Ethnology, 2nd ed., 1896, p. 55) and ‘over 100,000 years’ (ib., p. 116), are based upon obsolete calculations of the chronology of the Glacial Period. See pp. 31-2, supra.

[221] See pp. 126-7, infra.

[222] See Rev. de l’École d’anthr., xv, 1905, pp. 408-14, especially p. 412.

[223] See pp. 398-407, infra.

[224] See pp. 427-8, 433, 443, infra.

[225] This use of the word ‘dolmen’, which obtains in France, although megalithic chambers enclosed in tumuli are there sometimes called by the same name (Archaeol. Cambr., 5th ser., xvii, 1900, p. 221), is becoming common in this country; but in Wales dolmens are still known as cromlechs, a name which in France is applied only to stone circles.

[226] See Archaeologia, xlii, 1869, p. 214. The kistvaens of Dartmoor are really small dolmens.

[227] J. Anderson, Scotland in Pagan Times,--the Bronze and Stone Ages, 1886, p. 232; W. C. Borlase, Dolmens of Ireland, ii, 1897, pp. 461-2.

[228] Archaeol. Cambr., 5th ser., xvii, 1900, p. 222; W. C. Borlase, Dolmens of Ireland, ii, 446.

[229] Ib., p. 426; Proc. Soc. Ant., 2nd ser., v, 1870-3, pp. 367-70; viii, 1879-81, pp. 287-9; Dict. des sc. anthr., 1883, pp. 388, 1078; Archaeol. Cambr., 5th ser., xvii, 1900, p. 221; B.C. A. Windle, Remains of the Prehist. Age, pp. 174-7; Rev. de l’École d’anthr., xiv, 1904, pp. 259-62. In the eighteenth century the famous Kentish dolmen called Kit’s Coty House was still partly enclosed within a sepulchral mound. The Rev. W. C. Lukis, in a letter to Mr. George Payne (Collectanea Cantiana, 1893, p. 127), says, ‘I have a letter written ... in 1723 by one Hercules Ayleway [in which] ... Kit’s Coty is represented as being partly in a long barrow.’ See also Borlase, op. cit., iii, 752-3.

Mr. A. L. Lewis (Man, vii, 1907, No. 26, p. 38) says that a dolmen on Great Orme’s Head shows ‘that there certainly were dolmens that were never buried, but were intended to be “free-standing”’.

[230] W. C. Borlase, Dolmens of Ireland, ii, 424-6, 612-3.

[231] In Pembrokeshire, Glamorganshire, Merionethshire, Carnarvonshire, and Anglesey. Elsewhere they are almost entirely wanting, perhaps owing to the lack of suitable stones (Archaeol. Cambr., 6th ser., iv, 1904, p. 199).

[232] See pp. 402-5, infra.

[233] See pp. 405-6, infra.

[234] See p. 382, infra.

[235] See pp. 101-2, infra.

[236] B. G., vi, 28.

[237] Journ. Brit. Archaeol. Association, xxxiv, 1878, p. 351; W. Boyd Dawkins, Early Man in Britain, pp. 257-62, 484; R. Munro, The Lake-Dwellings of Europe, 1890, p. 488; Archaeologia, lv, 1897, pp. 130-1, 158.

[238] See Journ. Anthr. Inst., v, 1876, p. 359; the topographical index in Sir J. Evans’s Anc. Stone Implements, 1897; and the Victoria County Histories of Berks. (i, 276), Hants (i, 257), Lancs. (i, 212), Northampton (i, 139), Sussex (i, 311, 313, 470), and Worcester (i, 180).

[239] Sir J. Evans, Ancient Stone Implements, 1897, pp. 65-6, 104-6, 107-9, 129-30, 213, &c. Implements of jade and jadeite, which are common in the lake-dwellings (F. Keller, Lake Dwellings of Switzerland [trans. J. E. Lee], i, 1878, pp. 72, 195-6, 215-6), are very rare in Britain (Evans, op. cit., p. 109), and were doubtless imported, as jade apparently does not exist in situ in Europe, except in Silesia and Styria (Journ. Anthr. Inst., x, 1881, p. 359; xx, 1890-1, pp. 332-42, especially 334 and 338; Report of ... the Brit. Association, 1890, p. 971; L’Anthr., iv, 1893, p. 555).

[240] Similar lamps have been found in neolithic caves in France (Ass. franç. pour l’avancement des sc., 32^{e} session, 1903, 2^{e} partie, pp. 896-900), and are still used in China (Proc. Soc. Ant. Scot., xxii, 1888, p. 81). Cf. p. 258, infra.

[241] Journ. Ethn. Soc., N. S., ii, 1870, p. 430.

[242] Ib., p. 427.

[243] Ib., pp. 419-39; S. B. J. Skertchly, Memoirs of the Geol. Survey,--On the Manufacture of Gun-Flints, pp. 39-41, 71, 74.

[244] Journ. Ethn. Soc., N. S., ii, 1870, p. 439.

[245] Journ. Anthr. Inst., v, 1876, pp. 368-74. See p. 98, infra.

[246] Sir J. Evans, Anc. Stone Implements, 1897, pp. 80, 85-6; Guide to the Ant. of the Stone Age (Brit. Museum), p. 79.

[247] See p. 214, infra.

[248] Journ. Anthr. Inst., v, 1876, pp. 357-62, 382, 479; Archaeologia, xlii, 1869, pp. 59-60, 66, 68-9, 73-4; xlv, 1880, pp. 337-8, 340-7; Guide to the Ant. of the Stone Age (Brit. Museum), pp. 69-70; Vict. Hist. of ... Sussex, i, 315.

[249] See J. Anderson, Scotland in Pagan Times,--the Bronze and Stone Ages, p. 306.

[250] Journ. Anthr. Inst., xxxii, 1902, p. 395.

[251] See pp. 131-2, 230, infra; E. B. Tylor, Early Hist. of Mankind, 2nd ed., 1870, p. 194; A. Lang, Custom and Myth, 1885, p. 11; O. Schrader, Prehist. Ant. of the Aryan Peoples, 1890, p. 234; and Journ. Anthr. Inst., xxxiii, 1903, pp. 246-58.

[252] pp. 78-81, infra.

[253] Nature, Jan. 13, 1898, pp. 257-8.

[254] Traces of polishing are said to have been found on French implements of late palaeolithic age (Ass. franç. pour l’avancement des sc., 13^{e} sess., 1884, 1^{re} part., p. 212; L’Anthr., iv, 1893, p. 550).

[255] Sir J. Evans, Anc. Stone Implements, 1897, pp. 73, 85-6, and Guide to the Ant. of the Stone Age (Brit. Museum), pp. 69-70. Much depended upon the nature of the material. Certain hard stones, for instance granite and diorite, were necessarily ground and polished. See L’Anthr., iv, 1893, p. 550, and Guide to the Ant. of the Stone Age (Brit. Museum), p. 69.

[256] Sir J. Evans, Anc. Stone Implements, 1897, p. 276.

[257] Ib., pp. 28-9, 31.

[258] Sir J. Evans, Anc. Stone Implements, 1897, pp. 14-37, 43. Cf. 15th Ann. Report American Bureau of Ethn., 1893-4 (1897), p. 25.

[259] Sir J. Evans, Anc. Stone Implements, 1897, pp. 37-43, 412, 414-6; Journ. Anthr. Inst., xxxiii, 1903, p. 47.

[260] J. A. H. Murray, New Eng. Dict., ii, 215. Cf. Sir J. Evans, Anc. Stone Implements, 1897, p. 55.

[261] Sir J. Evans, Anc. Stone Implements, 1897, pp. 66, 107.

[262] Ib., pp. 71, 172, 205; Proc. Suffolk Inst. of Archaeology, xi, 1903, p. 329.

[263] Sir J. Evans, Anc. Stone Implements, 1897, pp. 129, 235. See also E. B. Tylor, Early Hist. of Mankind, 1870, pp. 205-6; Prim. Culture, 1903, i, 65.

[264] Sir J. Evans, Anc. Stone Implements, 1897, pp. 136, 171.

[265] J. Anderson, Scotland in Pagan Times,--The Bronze and Stone Ages, pp. 353-4.

[266] Archaeologia, xliv, 1873, pp. 281-3.

[267] F. Keller, The Lake Dwellings of Switzerland, i, 1878, pp. 21-2, 38, 57, 90, &c.

[268] An axe-hammer has, however, been found in the Liverpool Docks, scored with a groove, along which a withy was perhaps twisted to serve as a handle (Sir J. Evans, Anc. Stone Implements, 1897, pp. 168-9; Vict. Hist. of ... Lancs, i, 218).

[269] M. Hippolyte Müller (L’Anthr., xiv, 1903, pp. 424-6) has cut down numerous trees with flint axes, which were uninjured by the experiments. Two of the trees were felled in thirteen and fourteen minutes respectively.

[270] Sir J. Evans, Anc. Stone Implements, 1897, pp. 79, 171-2, 195-6.

[271] Ib., pp. 175-6.

[272] Sir J. Evans, Anc. Stone Implements, 1897, p. 178.

[273] Journ. Anthr. Inst., iv, 1875, p. 403.

[274] Sir J. Evans, Anc. Stone Implements, 1897, pp. 183-4.

[275] Sir J. Evans, Anc. Stone Implements, 1897, p. 215; L’Anthr., iv, 1893, p. 489.

[276] See Archaeologia, xliii, 1871, p. 409.

[277] J. Anderson, Scotland in Pagan Times,--the Bronze and Stone Ages, p. 246; Archaeol. Cambr., 6th ser., iii, 1903, p. 234; Proc. Soc. Ant. Scot., xxxviii, 1904, p. 355.

[278] Sir J. Evans, Anc. Stone Implements, 1897, pp. 183-4, 195, 215, 231.

[279] Ib., pp. 238-9, 245, 247-8, 250-2.

[280] Ib., pp. 275-6, 289. See pp. 92-3, infra.

[281] See Anthr. Rev., iii, 1865 (Journ. Anthr. Soc., p. lxvi); Archaeologia, xlii, 1869, pp. 229-30; and Anc. Stone Implements, 1897, p. 294. Dr. R. Munro. (Prehist. Problems, pp. 325-30, 359) would refer all the British flint saws to the Bronze Age, on the ground that ‘bronze saws have never yet been found in the British Isles’: but this statement is inaccurate (see p. 132, infra); and, as we have seen (p. 41, supra), serrated palaeolithic flints have been unearthed in a gravel-pit at Swanscombe.

[282] Sir J. Evans, Anc. Stone Implements, 1897, pp. 299-300, 311-2.

[283] Ib., pp. 312-9; Mem. Geol. Survey,--On the Manufacture of Gun Flints, p. 39.

[284] Sir J. Evans, Anc. Stone Implements, 1897, p. 321.

[285] Ib., pp. 326-32, 356-7.

[286] Sir J. Evans, Anc. Stone Implements, 1897, pp. 360-1, 369-98; Man, vii, 1907, No. 25, p. 37; No. 37, p. 56.

[287] See Journ. Anthr. Inst., xxxiii, 1903, p. 54.

[288] Sir J. Evans, Anc. Stone Implements, p. 377.

[289] J. Anderson, Scotland in Pagan Times,--the Bronze and Stone Ages, p. 272.

[290] Journ. Anthr. Inst., xxix, 1899, p. 131.

[291] Proc. Soc. Ant. Scot., xi, 1876, p. 509.

[292] Sir J. Evans, Anc. Stone Implements, 1897, p. 361.

[293] Ib., pp. 428-30; Proc. Soc. Ant. Scot., xxxviii, 1904, pp. 355, 361.

[294] Sir J. Evans, Anc. Stone Implements, pp. 431, 433-4.

[295] Worthington G. Smith, Man, the Primeval Savage, p. 304. Messrs. W. Johnson and W. Wright (Neol. Man in N.-E. Surrey, 1903, pp. 49, 169), who have been diligent in collecting tools from North-Eastern Surrey, remark that ‘some were fitted for use in the right hand, and others for the left’, and conclude that the people who used them were ambidextrous. But surely the more natural conclusion would be that some were left-handed!

[296] On the moors near Sheffield and in East Lancashire, in Staffordshire and Lincolnshire, and at Hastings.

[297] Donegal.

[298] Sir J. Evans, Anc. Stone Implements, 1897, pp. 324-5; Proc. Roy. Irish Acad., 3rd ser., vi, 1900-2, pp. 362-3; Reliquary, N. S., vii, 1901, pp. 123-6; Proc. Soc. Ant. Scot., xxxv, 1901, pp. 98-101; Man, ii, 1902, No. 15, pp. 18-22.

[299] See Journ. Anthr. Inst., xxix, 1899, p. 136.

[300] See L’Anthr., v, 1894, pp. 20-1, 146, and Guide to the Ant. of the Stone Age (Brit. Museum), pp. 77-8. M. Salomon Reinach (Rev. celt., xiii, 1892, pp. 193-9) attributes the absence of such artistic remains in France to the influence of Druids.

[301] Carm., ii, 15, 10-20.

[302] Professor B.C. A. Windle (Remains of the Prehist. Age, p. 257) affirms that excavations in the stronghold of Eggardun in Dorsetshire have proved that ‘pit-dwellings were in use in the Pre-metallic period’; and he remarks (ib., p. 258) that there was ‘no trace of any metallic object in the pits examined by Stevens at Hurstbourne or in those at Standlake’ in Oxfordshire. It is shown on p. 97, infra, that there is no sufficient reason for referring Eggardun to the Neolithic Age: bronze was discovered at Standlake (Archaeologia, xxxvii, 1857, p. 368), which, according to Pitt-Rivers (Excavations in Cranborne Chase, i, 20-1), bore such an ‘exact resemblance to the [Romano-British] Woodcuts village’, that, in his judgement, ‘further excavations would have proved it to have been of the Roman or Late-Celtic period’: Romano-British pottery was found in the pits at Hurstbourne; and Dr. Stevens himself (Parochial Hist. of St. Mary Bourne, 1888, p. 34) only claimed that ‘the flint implements ... establish that the site, if not the dwellings, was occupied by the people of the Neolithic Age’. Professor Boyd Dawkins (Vict. Hist. of ... Hants, i, 262) rightly refers the dwellings to the Iron Age.

Bone weaving-combs, which were found in pit-dwellings at Highfield, near Fisherton in Wiltshire, evidently belonged, like the querns with which they were associated (J. Stevens, Parochial Hist. of St. Mary Bourne, p. 25), to the Early Iron Age (cf. Sir J. Evans, Ancient Stone Implements, 1897, p. 251, and Reliquary, N. S., vii, 1901, p. 115); and although Professor Boyd Dawkins (Early Man, &c., p. 268) pleads that the pottery, which was ‘ornamented with incised curves’, was ‘not turned in the lathe’, that does not prove that it was made in the Neolithic or even the Bronze Age (see p. 244, infra); while the ‘curves’ suggest that it was Late Celtic.

[303] Journ. Anthr. Inst., xi, 1882, pp. 472-3.

[304] Archaeol. Cant., xiii, 1880, pp. 122-6.

[305] Journ. Anthr. Inst., xxix, 1899, p. 128; Vict. Hist. of ... Surrey, i, 237.

[306] Proc. Soc. Ant., 2nd ser., xii, 1887-9, pp. 258-63; xvii, 1897-9, pp. 216-21; Journ. Anthr. Inst., xxix, 1899, pp. 124, 127, 134.

[307] Norfolk Archaeology, iii, 1852, pp. 232-6; _Journ. Anthr. Inst._, xxix, 1899, p. 127; Vict. Hist. of ... Surrey, i, 236. Cf. Archaeol. Cambr., 6th ser., iv, 1904, p. 200.

[308] Vict. Hist. of ... Hants, i, 258-9.

[309] Journ. Anthr. Inst., xxix, 1899, pp. 127, 139.

[310] Ib., p. 140. Stone mounds have been discovered in South Wales by Messrs. T. C. Cantrill and O. T. Jones, who regard them as ‘probably the remains of prehistoric hearths or cooking-places’ (Archaeol. Cambr., 6th ser., vi, 1906, p. 17); but, as they ‘range in diameter from 6 feet or so to as much as 50 feet’ (ib., p. 19), I would suggest that the cooks must have been of Brobdingnagian stature.

For descriptions of other pit-dwellings which may perhaps be of neolithic age, see G. Young, Hist. of Whitby, ii, 1817, pp. 666-83; T. Bateman, Vestiges of the Ant. of Derbyshire, 1848, p. 126; Journ. Brit. Archaeol. Association, xi, 1855, pp. 305-13; Anthr. Rev., v, 1867, p. 253; and Archaeologia, xlii, 1869, pp. 223-4.

[311] Proc. Soc. Ant. Scot., xxxvii, 1903, pp. 370-415.

[312] A chambered mound in Stromness, Orkney, which was not a sepulchre but a dwelling, has also been recently excavated, and contained a stone implement (Proc. Soc. Ant. Scot., xxxvii, 1903, pp. 352-9). Mr. Christison (ib., xxxviii, 1904, pp. 5-6) tentatively refers it to the Neolithic Age, while the discoverer more cautiously says that it must have been built in ‘a remote period, not ... because the implement is made of stone, but because the type is an ancient one’.

[313] Mr. George Clinch. See his article in Surrey Archaeol. Collections, xvii, 1902, pp. 181-3.

[314] Matériaux pour l’hist. ... de l’homme, 3^e sér., ii, 1885, pp. 1-18.

[315] R. Munro, The Lake-Dwellings of Europe, 1890, pp. 470-1, 489. Cf. Journ. Roy. Soc. Ant. Ireland, 5th ser., x, 1900 (1901), pp. 208, 235, and Guide to the Ant. of the Bronze Age (Brit. Museum), pp. 142-3. See p. 154, infra.

[316] Mem. Anthr. Soc., iii, 1870, p. 76; W. Greenwell, Brit. Barrows, 1877, p. 742.

[317] Journ. Brit. Archaeol. Ass., N. S., v, 1899, p. 285.

[318] W. Greenwell, Brit. Barrows, p. 742.

[319] Journ. Ethn. Soc., ii, 1870, p. 431.

[320] W. Greenwell, Brit. Barrows, pp. 740-1. Cf. p. 151, infra.

[321] Archaeologia, lv, 1897, pp. 132-3, 150; Archaeol. Journal, liv, 1897, p. 379; R. Lydekker, Mostly Mammals, pp. 52, 299.

[322] W. Greenwell, Brit. Barrows, pp. 743-4; F. Keller, Lake Dwellings of Switzerland, i, 1878, p. 479.

[323] Journ. Ethn. Soc., ii, 1870, p. 431.

[324] Journ. Anthr. Inst., xxix, 1899, pp. 135-6; J. R. Mortimer, Forty Years’ Researches, p. lxx. Cf. E. B. Tylor, Early Hist. of Mankind, 1870, pp. 262-70, and Ency. Brit., xxv, 1902, p. 467.

[325] Archaeol. Review, i, 1888, p. 6.

[326] Congrès internat. d’anthr. et d’arch. prehist., 1900 (1902), p. 407.

[327] F. Keller, Lake Dwellings, &c., i, 1878, pp. 518-36.

[328] Sir J. Evans, Anc. Stone Implements, 1897, p. 257.

[329] A stone which appears to have been used as a grain-crusher was found in the neolithic village at West Wickham (Journ. Anthr. Inst., xxix, 1899, p. 133).

[330] See p. 76, supra.

[331] W. Greenwell, Brit. Barrows, pp. 744-5.

[332] Archaeologia, xix, 1821, p. 48; Mem. Anthr. Soc., i, 1865, p. 144; Anthr. Rev., iii, 1865 (Journ. Anthr. Soc., p. lxvii); W. Greenwell, Brit. Barrows, pp. 701-5; Brit. Med. Journal, 1903, pp. 809-10; Vict. Hist. of ... Derby, i, 168. It must be admitted that in some cases the teeth of neolithic skulls are as much worn as those of the Bronze Age.

Messrs. W. Johnson and W. Wright (Neol. Man in N.-E. Surrey, 1903, pp. 53-4), referring to Science Gossip, July, 1901, p. 36, affirm that under the tartar which covered the teeth of a skeleton in a neolithic barrow on Warminster Downs were found particles of quartzite,--‘apparently the rubbings from the mortar in which the corn was ground.’ But the writer of the article in Science Gossip states that bronze was found in the barrow.

[333] See O. Schrader, Prehist. Ant. of the Aryan Peoples, p. 286, and Journ. Roy. United Service Inst., xiii, 1870, p. 518.

[334] See Journ. Anthr. Inst., iii, 1874, pp. 35-6, and Mr. Clement Reid’s article in Vict. Hist. of ... Sussex, i, 9-10.

[335] Journ. Anthr. Inst., v, 1876, pp. 121, 478. See also p. 152, infra.

[336] W. Greenwell, Brit. Barrows, pp. 659-60, 704.

[337] Nature, Nov. 22, 1894, p. 92.

[338] F. Keller, Lake Dwellings of Switzerland, i, 1878, pp. 44, 46, 56, 63-4, 67, 69, 505-17, &c.

[339] W. Greenwell, Brit. Barrows, pp. 744-5.

[340] Sir J. Evans, Anc. Stone Implements, 1897, pp. 436-9.

[341] Ib., p. 465.

[342] See p. 80, supra, and Anc. Stone Implements, 1897, p. 312.

[343] J. Anderson, Scotland in Pagan Times,--the Bronze and Stone Ages, p. 240.

[344] W. Greenwell, Brit. Barrows, pp. 519-20, 543.

[345] ‘The fact is,’ says Professor Ridgeway (Man, iii, 1903, No. 97, pp. 171-2), ‘that mankind was led to wear such objects by magic rather than by aesthetic considerations ... the use of all the objects still employed in modern jewellery has primarily arisen from the magical powers attributed to them, by which they were thought to protect the wearer.’ M. Salomon Reinach’s review (L’Anthr., xiv, 1903, pp. 711-12) of Professor Ridgeway’s article is worth reading.

[346] A. Pitt-Rivers, Excavations in Cranborne Chase, ii, 179; Folk-Lore, xii, 1901, p. 175.

[347] See Lord Avebury’s Origin of Civilisation, 1902, pp. 54-8.

[348] The primitive method was apparently scraping (L’Anthr., viii, 1897, p. 204). M. Hippolyte Müller (ib., xiv, 1903, pp. 430-4) has performed the operation on four skulls by scraping with a flint implement; and he concludes that this method was adopted in the case of living patients. It appears (ib., p. 434) that the distinguished anthropologist, M. Capitan, has been impelled by scientific ardour to experiment ‘sur plusieurs chiens vivants’. What will happen if the Anti-Vivisection Society hears of this?

[349] Bull. de la Soc. d’anthr. de Paris, 3^e sér., iv, 1881, p. 107; vi, 1883, pp. 318-9; Journ. Anthr. Inst., xi, 1882, pp. 9, 12-4, 16; xvii, 1888, pp. 101, 106; Proc. Soc. Ant. Scot., xxvi, 1892, pp. 5, 8, 14-5, 17-8, 21, 28, 30-2; Sir J. Evans, Anc. Stone Implements, 1897, p. 289; R. Munro, Prehist. Problems, pp. 191-232; E. B. Tylor, Prim. Culture, i, 1903, p. 295; Bull. et mém. de la Soc. d’anthr., 5^e sér., v, 1904, pp. 67-73; Guide to the Ant. of the Early Iron Age (Brit. Museum), p. 58. See also Man, v, 1905, No. 27, p. 49. A perforated skull which was found in an interment in Bute may perhaps show that the practice of trepanning existed in this country as early as the Bronze Age; but a physician who has examined the perforation believes that it was not produced by trepanning (Proc. Soc. Ant. Scot., xxxviii, 1904, pp. 67-8). Artificially perforated skulls were found just outside the Late Celtic fortified village of Hunsbury near Northampton. See A. Pitt-Rivers, Excavations in Cranborne Chase, vol. iii, unnumbered page following pl. ccxxviii.

[350] Diodorus Siculus, v, 14, § 2.

[351] Apollonius Rhodius, ii, 1011-4.

[352] Strabo, iii, 4, § 17.

[353] See Lord Avebury’s Origin of Civilisation, 1902, pp. 25, 157-68, and L’Anthr., v, 1894, pp. 352-7. According to Dr. G. A. Wilken (ib., p. 356), certain communities in the Indian archipelago who practised the couvade in 1894 were in a state of transition from matriarchy to father-right. Cf. Man, vi, 1906, No. 74, p. 112.

[354] Rev. celt., vii, 1886, p. 227.

[355] Academy, xxv, 1884, p. 112.

[356] E. B. Tylor, Early Hist. of Mankind, 1870, pp. 293-304; Prim. Culture, 1903, i, 84.

[357] See, however, pp. 117-8, infra.

[358] See Trans. Epping Forest ... Field Club, ii, 1882, p. 60; iii, 1884, p. 228.

[359] Bull. et mém. de la Soc. d’anthr. de Paris, i, 1900, p. 53.

[360] Journ. Brit. Archaeol. Ass., N. S., vii, 1901, p. 17.

[361] See p. 156, infra.

[362] Proc. Dorset Nat. Hist. and Ant. Field Club, xxii, 1901, pp. 28-42; Proc. Soc. Ant., 2nd ser., xviii, 1901, pp. 258-62.

[363] Archaeologia, xlii, 1869, pp. 195, 232.

[364] It should be noted that pottery of the types that characterized the Bronze Age was still manufactured in the Early Iron Age. See p. 244, infra. Fragments of pottery were found in the long barrow of West Kennet in Wiltshire; but Thurnam (Archaeologia, xlii, 1869, p. 231) regards it as ‘very doubtful whether they belong to the people by whom the chamber was erected’. They seem to have been portions of ‘food-vessels’, which belong to the Bronze Age (see p. 191, infra); and Pitt-Rivers (Excavations in Cranborne Chase, iv, 1898, pp. 100, 163) says that they ‘must probably have got in subsequently to the construction of the barrow’. See also ib., pp. 147 and 162 (fig. 8); and, for examples of round-bottomed domestic pottery which have been found both in long barrows and in certain round barrows that may have been erected in the Stone Age, see Brit. Barrows, pp. 488-9, 509, and J. R. Mortimer, Forty Years’ Researches, p. lxviii.

The Scottish chambered cairns have yielded more pottery than the English barrows. Most of the vessels lacked decoration; but some were ornamented either with cord patterns or by impressions of the potter’s finger-tips and nails, or with vertical flutings; while a chambered cairn at Unstan, in Orkney, contained a vessel with triangular ornament of a kind which, as we shall see (pp. 197-8, infra), was characteristic of the Bronze Age. It may, however, have been manufactured at a time when bronze was coming into use in Southern Britain. See Proc. Soc. Ant. Scot., xix, 1885, pp. 346-8, and J. Anderson, Scotland in Pagan Times,--the Bronze and Stone Ages, pp. 240, 248-9, 252, 272-3, 294-7.

Some of the ‘drinking-cups’ which have been exhumed from round barrows doubtless belong to the end of the Neolithic Age (see pp. 192-3, infra); and a curious vessel, which Bateman (Vestiges of the Ant. of Derbyshire, 1848, p. 43) described as ‘a small drinking or incense cup of novel and unprecedented shape’, was found in a round barrow the neolithic age of which is certain. See Man, vi, 1906, No. 44, pp. 70-1.

[365] Archaeologia, xlii, 1869, p. 44.

[366] Ib., pp. 32-3. See also Vict. Hist. of ... Sussex, i, 458, 461, 463.

[367] Worthington G. Smith, Man, the Primeval Savage, pp. 319-20; Vict. Hist. of ... Bedford, i, 160.

[368] Vict. Hist. of ... Surrey, i, 237-8.

[369] Gen. Pitt-Rivers (Journ. Anthr. Inst., vi, 1877, pp. 359-60), after remarking that in the region between Seaford and Beachy Head ‘the débris of flint manufacture is so far abundant on the surface ... that the fact of finding flint flakes in the interior of these entrenchments [Seaford and Beltout] is no proof whatever of their being of the age of these entrenchments’, goes on to say that ‘this does not apply to other parts of the Downs of Sussex and elsewhere. There, worked flints are found in patches here and there; but considerable distances may be traversed without coming to these patches, and the fact of finding them in unusual numbers in the insides of these earthworks remains to testify to the probability of their having been used by the inhabitants of them.’

[370] Journ. Anthr. Inst., v, 1876, pp. 383-4. Mr. George Clinch (Vict. Hist. of ... Sussex, i, 316), referring to certain implements which he found himself at Cissbury, says, ‘The position of the flint flakes and chips upon [the side of one of the mounds] proves that the earthworks were completed during the Neolithic Age’.

[371] L’Anthr., xiv, 1903, pp. 444-62, and especially 450-2. See also A. Bertrand, Archéol. celt. et gaul., 1889, p. 105, n. 1, and Congrès internat. d’anthr. et d’arch. préhist., 1900 (1902), pp. 430-1.

[372] Vict. Hist. of ... Sussex, i, 469-70. The authors of an interesting article in the Cornhill Magazine (May, 1906, pp. 611-2), which, however, contains some unverifiable statements, assert that earthworks of a peculiar form, ‘usually at the base of a hill on the edge of a plain,’ were designed as a protection against wolves.

[373] L’Anthr., xv, 1904, p. 159.

[374] Ib., pp. 162, 165; Man, iv, 1904, No. 22, p. 37; Fort. Rev., Oct., 1904, pp. 635-9. M. Piette, however (L’Anthr., xvi, 1905, pp. 6-7), holds that certain symbols which he himself discovered in the cave of Gourdan form a real inscription.

[375] L’Anthr., xv, 1904, p. 162.

[376] Gongora y Martinez, Antiguëdades prehist. de Andalucia, 1868, p. 40, fig. 24. Cf. Fort. Rev., Oct., 1904, p. 643. See also in regard to primitive writing Journ. Anthr. Inst., xxix, 1899, pp. 204-6, Rev. arch., 4^e sér., i, 1903, pp. 231-2, and Man, iv, 1904, No. 22, p. 37.

[377] Archaeol. Cambr., 4th ser., iii, 1872, p. 25; W. Boyd Dawkins, Cave Hunting, pp. 155-9; Quart. Journ. Geol. Soc., lx, 1904, pp. 335-48.

[378] See Mem. Anthr. Soc., i, 1865, p. 133, and W. Greenwell, Brit. Barrows, pp. 1-2.

[379] See p. 58, supra.

[380] See Folk-Lore, xii, 1901, pp. 28-9.

[381] Archaeologia, xlii, 1869, p. 109. Mr. J. R. Mortimer (Forty Years’ Researches, &c., 1905, p. lxxxi) is ‘slightly inclined to consider that the long barrows [of the Yorkshire Wolds] ... are more recent than the greater number of the round ones’; but the only reason which he gives for this singular opinion, namely, that he has frequently found both long and round skulls in the round barrows of the same district, has no weight against the facts which have led all other investigators to regard the long as earlier than the round barrows. See p. 393, infra. Not only has no metal ever been found with a primary interment in a long barrow, but sepulchral pottery is also wanting. See Man, v, 1905, No. 86, p. 159. If the contents of certain long barrows ‘do not show any features of interest differing from those found in [some] round barrows’ (Forty Years’ Researches, p. xix), that only suggests that long barrows were still made for some time after the first interment in a round barrow took place. See W. Greenwell, Brit. Barrows, pp. 509, 556.

[382] J. B. Davis and J. Thurnam, Crania Britannica, ii, 1865, pl. 33, p. 2; Archaeologia, xlii, 1869, pp. 170, 176, 202, 206-7; Mem. Anthr. Soc., iii, 1870, p. 41; W. Greenwell, Brit. Barrows, pp. 479, 484-511, 550-6; B.C. A. Windle, Remains of the Prehist. Age, pp. 155, 157, 159-63, 166-71; Vict. Hist. of ... Durham, i, 200, 207. A few long barrows are said to exist in Lancashire, but it is doubtful whether they can really be classed as such (Vict. Hist. of ... Lancs., i, 211).

[383] W. Boyd Dawkins, Cave Hunting, p. 162; J. Anderson, Scotland in Pagan Times,--the Bronze and Stone Ages, pp. 232-67; Proc. Soc. Ant. Scot., xxxvi, 1902, pp. 39-42.

[384] Archaeologia, xlii, 1869, pp. 206-7; W. C. Lukis, _Prehist. Stone Monuments of the Brit. Isles,--Cornwall_, 1885, p. 13; J. Anderson, op. cit., pp. 268-303; _Journ. Brit. Archaeol. Association_, N. S., vi, 1900, p. 7; Journ. Anthr. Inst., xxxii, 1902, p. 404.

[385] Archaeologia, xlii, 1869, pp. 206-7; Mem. Anthr. Soc., iii, 1870, p. 41.

[386] See S. Nilsson, Primitive Inhabitants of Scandinavia, 1868, pp. 124-58; Journ. Ethn. Soc., N. S., ii, 1870, p. 448; W. Greenwell, Brit. Barrows, p. 536; Archaeol. Review, ii, 1889, p. 314; E. Cartailhac, La France préhist., 1889, p. 195; and A. H. Keane, Ethnology, 1896, p. 126, note. The ‘Picts’ houses’, whose resemblance to chambered tumuli, according to Thurnam (Archaeologia, xlii, 1869, pp. 223-4), is such that ‘in particular instances it has been doubted whether the structure ... was a dwelling or a tomb’, belong to a much later period than that of the tumuli. See pp. 261 and 391, infra.

[387] Journ. Anthr. Inst., v, 1876, pp. 130-1; W. Greenwell, Brit. Barrows, p. 536, n. 2.

[388] Canon Greenwell (ib., p. 485, n. 1) says that in North Gloucestershire ‘the rule of the primary interment having been made at the larger end of the mound by no means holds good in all cases’. See also p. 504. In the Wor Barrow on Cranborne Chase the primary interments lay south of the centre of an oblong enclosure, which is described on p. 106, infra (A. Pitt-Rivers, Excavations in Cranborne Chase, iv, 20-1 [preface]).

[389] Archaeologia, xlii, 1869, pp. 181, 208-9; Mem. Anthr. Soc., iii, 1870, p. 41; W. Greenwell, Brit. Barrows, pp. 484, 487-8, 491, 497, 501, 505, 509, 511, 513, 515, 521, 524; Dict. des. sc. anthr., 1883, p. 387; J. Anderson, Scotland in Pagan Times,--the Bronze and Stone Ages, pp. 264-5; E. Cartailhac, La France préhist., 1889, p. 183; Proc. Soc. Ant., 2nd ser., xv, 1893-5, p. 404; W. C. Borlase, Dolmens of Ireland, ii, 489-90.

[390] Archaeologia, xlii, 1869, pp. 172, 208; J. B. Davis and J. Thurnam, Crania Britannica, ii, pl. 50, pp. 1-2.

[391] J. B. Davis and J. Thurnam, Crania Britannica, ii, pl. 5, p. 2; Archaeologia, xlii, 1869, p. 212, pl. xiv.

[392] Ib., pp. 172-3, 209; W. Greenwell, Brit. Barrows, p. 544. See p. 177, infra.

[393] See p. 65, supra.

[394] Oval neolithic barrows, which were not only fenced by peristaliths, but also had ellipses of stone on the surface, and which, like the West Kennet barrow, were each surmounted by a dolmen, exist in Northern Germany, west of the Vistula (L’Anthr., iv, 1893, p. 487), in Denmark (A. Bertrand, Archéol. celt. et gaul., 1889, pp. 163-4), and in France (ib., p. 166). Cf. Archaeologia, xlii, 1869, p. 165 (pl. xii, figs. 3 and 4), 211, note b.

[395] J. B. Davis and J. Thurnam, Crania Britannica, ii, pl. 24, pp. 2-3, pl. 5, p. 2; Archaeologia, xlii, 1869, pp. 209-21; W. Greenwell, Brit. Barrows, p. 544; J. Anderson, Scotland in Pagan Times,--the Bronze and Stone Ages, p. 232.

[396] Excavations in Cranborne Chase, iv, pp. 20-1 (preface).

[397] Journ. Anthr. Inst., v, 1876, p. 153, fig. 1, 165, fig. 1.

[398] W. Greenwell, Brit. Barrows, p. 536; J. Anderson, Scotland in Pagan Times,--the Bronze and Stone Ages, pp. 232, 266-7; Proc. Soc. Ant. Scot., xxxvi, 1902, pp. 39-42.

[399] J. Anderson, op. cit., p. 300.

[400] See, however, p. 108, and p. 109, n. 2, infra. According to Dr. R. Munro (Prehist. Scotland, 1899, p. 325), ‘Although many [Scottish] graves have been examined which contained ... stone and nothing of bronze, it does not follow that they were earlier than others in which bronze articles were found. It seems to me’, he continues, ‘that the vast majority of the sepulchral memorials hitherto explored within the Scottish area date from the introduction of bronze’.

The evidence that the Scottish chambered tombs belonged to the Scottish Stone Age is precisely the same as that which is almost unanimously accepted for the English long barrows. Since we find that not a single article of bronze has ever been found with a primary interment in a Scottish chambered cairn, while bronze is abundant in the short cists and unchambered cairns of the same country; that the skeletons in chambered cairns belong to the same stock as the people who built the long barrows (see pp. 393-4, infra); and that the chambered cairns and the chambered long barrows are structurally akin, we may infer that the former, like the latter, belonged to the Stone Age. That, however, some of them may have been built after bronze had been introduced into Southern Britain is not improbable.

[401] J. Anderson, Scotland in Pagan Times,--the Bronze and Stone Ages, pp. 249-50, 258, 264, 272, 274.

[402] Journ. Ethn. Soc., N. S., ii, 1870, pp. 416-9.

[403] Archaeol. Journal, xxviii, 1871, pp. 85-96.

[404] An oval cairn, however, at Pawton in East Cornwall (W. C. Lukis, The Prehist. Stone Monuments of the Brit. Isles,--Cornwall, 1885, p. 11) contains a cist, apparently contemporary with its erection, and is therefore presumably later than the chambered round barrows. It has been suggested (Vict. Hist. of ... Derby, i, 175-6) that the oval form of some barrows may be due to the addition, following secondary interments, of new material.

[405] Ib., pp. 166-9.

[406] T. Bateman, Ten Years’ Diggings, &c., 1861, pp. 253-4. The Derwent Moor barrow was opened in 1780, when the art of excavation was in its infancy; and the urn in question may have belonged to a secondary interment.

[407] Report of ... the Brit. Association, 1888 (1889), pp. 289-316. In regard to the chambered round barrows of Derbyshire, see also Journ. Brit. Archaeol. Association, N. S., vi, 1900, p. 7, and cf. W. Greenwell, Brit. Barrows, pp. 447-52.

[408] Archaeologia, xlix, 1885, pp. 189-92, 194-7; W. C. Borlase, Dolmens of Ireland, i, 145, 149, ii, 418, 441-2, 445-6, 448, 451, 462. See also Borlase’s Nenia Cornubiae, 1872, p. 3, and Vict. Hist. of ... Cornwall, i, 358.

[409] Proc. Soc. Ant. Scot., xxxvi, 1902, pp. 74-181: xxxvii, 1903, pp. 36-67; xxxviii, 1904, pp. 17-81; Journ. Anthr. Inst., xxxii, 1902, pp. 398-406. These cairns have no passages of entrance. Their outline was commonly rectangular, the ground-plan being defined by flagstones, arranged at one end in a semi-circle, the space within which led to a low portal that gave access to the chamber. The latter consisted of two sections, one above the other, of which the lower was built of large lateral slabs, covered by flagstones, and divided by other slabs into compartments, while the upper was formed of small flags laid horizontally.

A chambered cairn of abnormal form in the island of Ronsay, Orkney, which has been described by Sir William Turner (Proc. Soc. Ant. Scot., xxxvii, 1903, pp. 73-82), ‘consisted of a central part and four recesses’ (ib., pp. 74-5, fig. 1); and on its roof were cremation cists ‘quite different in character from the short cists so frequently found in Scotland’ (ib., p. 79).

[410] Journ. Anthr. Inst., xxxii, 1902, pp. 398, 405; _Proc. Soc. Ant. Scot._, xxxviii, 1904, p. 78.

[411] Ib., xxxvi, 1902, pp. 154-5. The chronology of the peculiar chambered cairns of South-Western Scotland is somewhat perplexing. On the one hand the structure of the cairns is presumptive evidence that they were built in neolithic times (J. Anderson, Scotland in Pagan Times,--the Bronze and Stone Ages, pp. 271-2; Proc. Soc. Ant. Scot., xxxvi, 1902, p. 136); the successive interments which were made in them were characteristic of the same period (ib., p. 134); the pottery which they contained is almost all of neolithic type (ib., pp. 165-7; xxxviii, 1904, pp. 78-9); and the presence of drinking-cups does not necessarily point to a later date (p. 193, infra). On the other hand a perforated stone hammer, which was found in one of the chambers (ib., xxxvi, 1902, p. 100), belongs to a class of implements which in this country were generally post-neolithic (p. 78, supra); an elegant bowl which was obtained by Canon Greenwell in a cairn on Largie Farm, near Crinan (Proc. Soc. Ant. Scot., xxxvi, 1902, pp. 165-7), although neolithic in form, is somewhat similar to a food-vessel figured in Brit. Barrows, p. 88, fig. 73, which, like it, is ornamented with vertical flutings; and one of the drinking-cups deposited in the same cairn is ‘almost identical in size, shape, and ornamentation’ with a specimen that was associated with a bronze dagger in a barrow on Roundway Hill, Wiltshire (Proc. Soc. Ant. Scot., vi, 1867, pp. 345, n. 1, 347). I conclude that these cairns were locally of neolithic age, but that the influence of the later culture had made itself felt in the district,--in short, that they belonged to a period of transition.

[412] M. de Baye exaggerates when he says (L’archéol. préhist., 1888, p. 108) that in France inhumation in the Neolithic Age was almost universal. M. E. Cartailhac (Matériaux pour l’hist. ... de l’homme, xxii, 1888, pp. 1-2, 4, 6-7; La France préhist., 1889, pp. 270-6) gives numerous instances of incineration in neolithic tombs in the departments of the Aisne, the Marne, the Morbihan, &c.

[413] W. C. Borlase, Dolmens of Ireland, ii, 520.

[414] Archaeologia, xlii, 1869, pp. 191-2, 224-6; Journ. Anthr. Inst., v, 1876, p. 129.

[415] Ib.; Guide to the Ant. of the Bronze Age (Brit. Museum), p. 53.

[416] Proc. Soc. Ant. Scot., xxxviii, 1904, p. 81. Cremation was common in the Neolithic Age in Derbyshire (Report of ... the Brit. Association, 1888 [1890], p. 316).

[417] See, however, p. 187, infra.

[418] See Journ. Anthr. Inst., v, 1876, pp. 130-1.

[419] W. Greenwell, Brit. Barrows, pp. 690-1; J. Anderson, Scotland in Pagan Times,--the Bronze and Stone Ages, pp. 244-5, 250, 274, 293, 300; Proc. Soc. Ant. Scot., xxxviii, 1904, p. 42.

[420] W. Greenwell, Brit. Barrows, p. 506; Guide to the Ant. of the Stone Age (Brit. Museum), p. 72.

[421] In a long barrow at Upper Swell in Gloucestershire (W. Greenwell, Brit. Barrows, pp. 524, 526, 536) the primary interment was in a true grave. No similar interment, so far as I know, has been found in any long barrow.

[422] Archaeologia, xlii, 1869, pp. 189, 224-5. Sir J. Evans (Archaeol. Journal, xxxv, 1878, p. 266) gives an instance of a skeleton of neolithic age found in the extended position near Daventry in Northamptonshire.

[423] Archaeologia, xlii, 1869, p. 189; W. Greenwell, _Brit. Barrows_, pp. 23-4.

[424] Archaeologia, xv, 1806, p. 339; xix, 1821, pp. 43-4; xlii, 1869, pp. 184-5.

[425] Ib., pp. 190-1; W. Greenwell, Brit. Barrows, p. 504; A. Pitt-Rivers, Excavations in Cranborne Chase, iv, 20-1 (preface); Guide to the Ant. of the Stone Age (Brit. Museum), pp. 72-3. In the Wor Barrow on Cranborne Chase (see p. 105, supra) Pitt-Rivers discovered six skeletons ‘huddled together beneath a small mound’, which was within and distinct from the monument itself. Three of them were crouched, and the rest ‘put in with them as bones, with the long-bones laid out by the side of the skull’. Referring to a discovery made in Egypt by Professor Flinders Petrie, Pitt-Rivers observes that the bones of a skeleton of the Fifth Dynasty (about 3500 B.C.) ‘had been cut up and put in a box, with an effigy of the deceased by the side of it. Something of this sort’, he continues, ‘may have occurred here’.

[426] v, 18, § 2.

[427] Guide to the Ant. of the Stone Age (Brit. Museum), p. 73.

[428] Archaeologia, xlii, 1869, pp. 184-5; W. Greenwell, Brit. Barrows, pp. 500-1.

[429] Journ. Anthr. Inst., v, 1876, pp. 134-8. Even in cremation deposits the bones are often imperfect and disconnected; and, previous to cremation, the bodies must have been stored in an ossuary (W. Greenwell, Brit. Barrows, p. 547).

[430] Ib., pp. 527, 533-4, 547-8

[431] Archaeologia, xlii, 1869, pp. 185, 191-2, 222, 227; _Mem. Anthr. Soc._, iii, 1870, p. 76.

[432] W. Greenwell, Brit. Barrows, pp. 686-93. See also Canon Greenwell’s remarks on pp. 544-5. Although he agrees in the main with Rolleston, he admits the probability that Thurnam ‘found signs of violent breakage upon a few skulls’.

[433] Archaeologia, xxxviii, 1860, p. 421.

[434] Gen. Pothier, Les tumulus du plateau de Ger, 1900, pp. 30-1.

[435] See the references in Greenwell’s Brit. Barrows, p. 685.

[436] See E. B. Tylor, Prim. Culture, i, 1903, pp. 458-67.

[437] Archaeologia, xlii, 1869, pp. 188, 222. Cf. Bull. de la Soc. d’anthr. de Paris, 2^e sér., ii, 1867, pp. 326-32; J. R. Mortimer, Forty Years’ Researches, p. xxiv, and W. Greenwell, Brit. Barrows, p. 15; and, for a valuable caution against forming hasty inferences as to cannibalism, Journ. Derby. Archaeol. and Nat. Hist. Soc., xv, 1893, p. 162.

[438] See p. 268, n. 1, infra.

[439] W. Greenwell, Brit. Barrows, pp. 486, 499, 544.

[440] ‘Remains of the horse,’ says Lord Avebury (Prehist. Times, 6th ed., 1900, p. 160, with which cf. p. 152, n. 5, infra), ‘are very rare in English barrows, and I know no well authenticated case of their occurrence in a long barrow’. See, however, Archaeologia, xlii, 1869, pp. 228-9. In Gaul at all events in the Neolithic Age horses abounded (Association franç. pour l’avancement des sc., 32^e sess., 1903, 2^e part., p. 851).

[441] Archaeologia, xlii, 1869, pp. 182-3, 227-8, 237-8, 241.

[442] See E. B. Tylor, Early Hist. of Mankind, 1870, p. 131.

[443] Trans. Ethn. Soc., N. S., iii, 1865, p. 317.

[444] W. Greenwell, Brit. Barrows, p. 735.

[445] Archaeologia, xlii, 1869, p. 183.

[446] B. G., v, 12, § 6.

[447] See W. Robertson Smith, The Religion of the Semites, 1901, p. 350.

[448] Primus in orbe deos fecit timor (Statius, Theb., iii, 661). Cf. E. B. Tylor, Prim. Culture, ii, 1903, p. 230.

[449] Ib., pp. 28, 111, 113.

[450] Acad. des inscr. et belles-lettres,--comptes-rendus des séances de l’année 1892, 4^e sér., xx, 6-7.

[451] See pp. 289-98, infra.

[452] L’Anthr., iv, 1893, p. 488; Archaeologia, xlii, 1869, p. 232.

[453] Ib., pp. 193, 229, 232; W. Greenwell, Brit. Barrows, p. 543.

[454] See Trans. Ethn. Soc., iii, 1865, p. 316, and cf. W. Greenwell, Brit. Barrows, pp. 57-60, 120-1, whose remarks, though they apply to the round barrows of the Yorkshire Wolds, are relevant.

[455] Journ. Anthr. Inst., vi, 1877, p. 500 (with which cf. p. 293); W. Boyd Dawkins, Early Man in Britain, p. 287. Cf. J. R. Mortimer, Forty Years’ Researches, p. 1.

[456] See pp. 201-2, infra.

[457] See Lord Avebury’s Origin of Civilisation, 1902, pp. 35, 301. Professor Boyd Dawkins misunderstands the custom.

[458] See pp. 288, 403, infra; Archaeol. Cambr., 3rd ser., x, 1864, pp. 292, 296, 298; Archaeologia, xlii, 1869, pp. 216-7; and W.C. Borlase, Dolmens of Ireland, i, 68, 75, 95, 174. Numerous holed stones exist in the vicinity of barrows and stone circles, which were probably erected in the Bronze Age, in Cornwall, Ross-shire, Inverness-shire, the Orkneys, and the island of Arran; but their significance is unknown. There are holed megaliths in Britain (one of which has in recent times been used for curing weakly children, whose mothers passed them through it) that do not belong to dolmens (W. C. Lukis, Prehist. Stone Monuments of the Brit. Isles,--Cornwall, p. 17).

[459] E. B. Tylor, Prim. Culture, i, 1903, p. 454. Cf. Rev. des études anc., vii, 1905, pp. 31-2, and p. 288, infra.

[460] See E. B. Tylor, Prim. Culture, 1903, i, 477, ii, 144-5, 159, 242-3.

[461] Ib., i, 120.

[462] Ib., ii, 209-14; G. L. Gomme, Ethnology in Folk-lore, 1892, pp. 78-9; Lord Avebury, Prehist. Times, 1900, pp. 207-8. See also W. Robertson Smith, The Religion of the Semites, 1901, p. 182.

[463] E. B. Tylor, Prim. Culture, i, 1903, pp. 108-10; ii, 209.

[464] Ib., i, 357-8. Cf. A. Lang, Custom and Myth, 1885, pp. 124, 131, 137, 142.

[465] E. B. Tylor, Prim. Culture, ii, 1903, pp. 186, 248-9, 255.

[466] Ib., p. 364. Cf. Journ. Anthr. Inst., xxxiv, 1904, p. 264, and Mr. R. R. Marett’s interesting article, ‘From Spell to Prayer’ (Folk-Lore, xv, 1904, pp. 132-65).

[467] L’Anthr., xv, 1904, p. 120; A. Lang, The Clyde Mystery, 1905, pp. 82, 89, 99.

[468] E. B. Tylor, Prim. Culture, ii, 1903, p. 22.

[469] E. B. Tylor, Prim. Culture, ii, 1903, p. 185.

[470] See W. Robertson Smith, The Religion of the Semites, 1901, p. 152; Lord Avebury, Origin of Civilisation, 1902, pp. 466-70; L’Anthr., xvi, 1905, p. 660; and A. Lang, The Secret of the Totem, p. 2.

[471] See W. Robertson Smith, The Religion of the Semites, 1901, pp. 38, 41, 92, 119, &c.

[472] Ib., pp. 253-6, 263.

[473] Journ. Anthr. Inst., xxviii, 1899, p. 145; E. B. Tylor, Prim. Culture, ii, 1903, p. 370.

[474] ‘... die Phantasie, welche wie aller Poesie so auch aller Historie Mutter ist’ (Th. Mommsen, Röm. Gesch., v, 1885, p. 5).

[475] See pp. 408-9, infra.

[476] W. C. Borlase, Dolmens of Ireland, ii, 516-7, 549, 563; Rev. de l’École d’anthr., xv, 1905, pp. 213-4. The rarity of long barrows may partly be explained by supposing that a certain proportion of the others belonged to the late Neolithic Age. That some did is certain. See Proc. Soc. Ant. Scot., xxxvi, 1902, pp. 159, 171, and pp. 408-9, infra.

[477] M. Salomon Reinach (L’Anthr., xvi, 1905, p. 659) characteristically remarks in regard to the discovery of metals that ‘On l’explique ordinairement par une succession de hasards heureux, en oubliant que l’humanité primitive, n’ayant aucune idée de l’utilisation industrielle des métaux, ne pouvait en arriver là du premier coup.... Aujourd’hui toute la métallurgie primitive me semble un chapitre de l’histoire des religions.... On a soumis ces métaux [gold and tin] à l’action du feu, au cours d’opérations magiques; ainsi naquit l’idée de traiter de même les minerais de cuivre ... et d’en dégager le métal brillant qui ressemble à l’or ... L’alchimie primitive, absolument étrangère à toute application industrielle, chercha à manier des substances divines par l’action du feu, à opérer ... des hiérogamies analogues à celle qui conduisit les agriculteurs à la découverte de la griffe. L’alliage du bronze fut un des résultats de leurs efforts.’

That smiths were sometimes regarded with superstitious awe by those who did not share their secrets (O. Schrader, Prehist. Ant. of the Aryan Peoples, pp. 165-8); that metallurgy was connected at various points with religion;--so much may be granted. But to say that ‘primitive alchemy’ (if it existed) had no industrial application is simply to make an unverifiable and improbable assertion. The discovery that ores could be smelted must have been accidental. Why should not the ‘alchemist’, however superstitious he may have been, have thereupon conceived the idea of turning gold to account for the manufacture of ornaments, or copper for that of axes?

[478] Proc. Soc. Ant., 2nd ser., xvi, 1895-7, p. 333.

[479] See Man, ii, 1902, No. 19, p. 29.

[480] Journ. Anthr. Inst., xxxi, 1901, pp. 278-9; Man, iv, 1904, No. 5, pp. 13-4.

[481] Rev. mens. de l’École d’anthr., iii, 1893, pp. 227-9; Report of ... the Brit. Association, 1896, p. 911.

[482] See L’Anthr., iv, 1893, p. 559; xvi, 1905, p. 198.

[483] Congrès internat. d’anthr. et d’archéol. préhist., 1900 (1902), p. 340; Vict. Hist. of ... Hertford, i, 232; C. H. Read, Guide to the Ant. of the Bronze Age (Brit. Museum), 1904, pp. 5-6, 27, 101, 111, 130. M. P. du Chatellier (Rev. de l’École d’anthr., xiii, 1903, pp. 169-72) thinks that there was a copper age in Brittany, but admits that he cannot settle the question. Professor O. Montelius affirms (Man, v, 1905, No. 7, p. 13) that ‘copper had been used there [in Britain] for a long time’ before bronze; but British archaeologists do not bow to his authority. Pitt-Rivers (Journ. Roy. United Service Inst., xiii, 1870, p. 520) remarked that ‘it is not surprising that on the first discovery of the advantages of [adding tin to copper] ... all the old implements of copper, wherever procurable, should have been taken to the melting-pot for conversion into bronze, and we should thus be left with such scanty evidence of the existence of an age of copper’. Still we have sufficient evidence for Ireland, and not for Britain. Professor Gowland (Journ. Anthr. Inst., xxxvi, 1906, p. 23) attributes the scarcity of copper celts in England to ‘the occurrence of mixed copper and tin ores in Cornwall.’

[484] Man, iii, 1903, No. 8, pp. 147-9. Professor Montelius (ib., v, 1905, No. 7, pp. 13-4) denies that iron was used by the Egyptians before the fifteenth century B.C., and insists that the lump of iron oxide which was found at Abydos in association with copper implements of the Sixth Dynasty, or about 3200 B.C., ‘does not prove the use of iron, only the existence of that metal’. The professor doubtless wrote ‘existence’ by a slip for ‘knowledge’. But, as Mr. H. R. Hall points out (ib., No. 40, pp. 69-71), he ignores the discovery in the Great Pyramid of a piece of worked iron, which is now in the Third Egyptian Room of the British Museum (Case K 29, No. 2433), and to which a date about 300 years older ‘is assigned on good prima facie grounds’; and Mr. Hall reasonably asks whether the discovery of the lump of iron oxide does not corroborate the other. It is unlikely that an unworked lump of iron would have been deposited along with copper tools; and we may fairly suppose that the lump is the remains of an iron tool.

[485] Archaeologia, lvi, 1899, pp. 302-3; L. Beck, Die Gesch. des Eisens, i, 1884, pp. 593-6. As far as I can see, all that is proved by the instances which Beck has collected is--what we know already--that stone implements continued in use after the Iron Age had begun. In regard to the discovery, mentioned on p. 595, which Worsaee made in a stone chamber, may not graves of this kind have been built here and there after the Neolithic Age? See pp. 108, 109, n. 2, supra.

[486] See Man, v, 1905, No. 7, p. 13.

[487] Journ. Anthr. Inst., xxx, 1900, p. 16.

[488] J. Evans, Anc. Bronze Implements, &c., p. 25. See also p. 95.

[489] Ib., p. 23.

[490] See O. Schrader, Prehist. Ant. of the Aryan Peoples, pp. 194, 203-4, 242; Guide to the Ant. of the Bronze Age (Brit. Museum), pp. 3-4; Rev. mens. de l’École d’anthr., iii, 1893, pp. 105-6, 118, 120-2; Daremberg and Saglio, Dict. des ant. grecques et rom., ii, 1075; and Pauly’s Real-Encyclopädie, v, 1905, col. 2143. Classical scholars will remember that Lucretius (v, 1286), in his powerful description of prehistoric times, affirmed that bronze was used before iron.

[491] See O. Schrader, Prehist. Ant. of the Aryan Peoples, pp. 192-3, Reallexicon der indogermanischen Altertumskunde, 1901, pp. 200-1; and Guide to the Ant. of the Bronze Age (Brit. Museum), pp. 9-11. Dr. Schrader argues that as there are no special names for bronze in the languages of any of the ancient bronze-using peoples except the inhabitants of Mesopotamia, in whose tongue bronze, as distinct from copper (urudu), was designated by the word zabar, they must have been the inventors of bronze. See p. 494, infra.

[492] Professor Boyd Dawkins (Early Man in Britain, pp. 408-9) has compiled tables which show that the percentage of tin in British bronze implements varies between 5·09 and 18·31, and in French between 1·50 and 21·5. He concludes (ib., p. 410) that ‘the uniformity of the composition of the cutting implements of the Bronze Age implies that the art of compounding tin with copper was discovered in one place, from which the knowledge of it spread over ... Europe and Asia, and the greater part of the Americas. Had it spread from separate centres, this uniformity would have been impossible.’ The uniformity which subsists between 5·09 and 18·31, and between 1·50 and 21·5 is remarkable.

[493] See Sir John Evans’s Anc. Bronze Implements, p. 420; Rev. d’anthr., 3^e sér., iii. 1888, pp. 209-10; and Lord Avebury’s Prehist. Times, 1900, pp. 53-7.

[494] L’Anthr., iv, 1893, pp. 548, 561-2, 566. M. Salomon Reinach (ib., iii, 1892, p. 280) has gone so far as to suggest that ‘les origines mêmes de la métallurgie du bronze’ should be sought in Western Europe.

[495] See Journ. Roy. United Service Inst., xiii, 1870, p. 539, and J. Evans, Anc. Bronze Implements, p. 419.

[496] Ib., p. 476.

[497] Ib., pp. 476-7. See p. 144, n. 5, infra.

[498] J. Evans, Anc. Bronze Implements, pp. 420, 477; Association franç. pour l’avancement des sc., 2^e partie, 1903, p. 931; L’Anthr., xvi, 1905, p. 168.

[499] J. Evans, Anc. Bronze Implements, pp. 470-2. See pp. 231-4, infra.

[500] J. Evans, Anc. Bronze Implements, pp. 472-3.

[501] Journ. Anthr. Inst., xxix, 1899, pp. 308-10; Man, v, 1905, No. 7, p. 13.

[502] See pp. 33-4, supra.

[503] See pp. 424-44, infra.

[504] W. Greenwell, Brit. Barrows, pp. 111-2.

[505] Ib., pp. 118-20.

[506] Archaeologia, xliii, 1871, p. 313; Report of ... the Brit. Association, 1888 (1889), p. 316. Cf. O. Schrader, Prehist. Ant. of the Aryan Peoples, p. 389.

[507] W. Greenwell, Brit. Barrows, pp. 127, 711-2; _Vict. Hist. of ... Derby_, i, 180, 190, n. 1. See pp. 427-8, infra.

[508] See pp. 131-2, infra. Professor Boyd Dawkins (Early Man in Britain, p. 342) asserts that ‘bronze weapons ensured victory [to the brachycephalic immigrants] over enemies armed with the old weapons of stone’. On page 344 he remarks that ‘while the chiefs and the rich possessed bronze implements and weapons, the poorer classes would naturally continue to use those of stone’, &c. How could bronze weapons have decided battles if only ‘the chiefs and the rich’ wielded them?

[509] Vict. Hist. of ... Lancs, i, 212, 239.

[510] Since the publication of Sir John Evans’s work, bronze weapons have been found in the Orkney and Shetland Islands (Proc. Soc. Ant. Scot., xxi, 1887, pp. 340-2).

[511] See p. 133, n. 1, infra.

[512] See the Topographical Index in Sir J. Evans’s Anc. Bronze Implements; Vict. Hist. of ... Northampton, i, 142; Vict. Hist. of ... Nottingham, i, 289; Journ. Anthr. Inst., xxxii, 1902, p. 386; and Archaeol. Journal, lxiv, 1904, pp. 310-2. It must not, however, be supposed that mere statistics of finds are necessarily valid evidence. See Proc. Soc. Ant. Scot., xxxviii, 1904, pp. 496-7.

[513] W. Greenwell, Brit. Barrows, pp. 118, 133-5.

[514] Vict. Hist. of ... Worcester, i, 183-4. Cf. J. Evans, Anc. Bronze Implements, pp. 81, 88, 129, 368.

[515] W. Greenwell, Brit. Barrows, pp. 159-60; J. Evans, Anc. Bronze Implements, p. 318, Anc. Stone Implements, 1897, p. 195; Archaeologia, xliii, 1871, pp. 409-12; lii, 1890, pp. 60-1; liv, 1895, p. 105; Archaeol. Cambr., 6th ser., ii, 1902, pp. 60-1. Flint arrow-heads are also found with burials of the Bronze Age in France (J. Anderson, Scotland in Pagan Times,--the Bronze and Stone Ages, p. 171, n. 1); and Sir J. Evans (Anc. Bronze Implements, p. 318) says that ‘many of the bronze arrow-heads found on the Continent appear to belong to the Early Iron Age’. He admits, however, that some very small spear-heads, so called, ‘may possibly have served to point arrows’. Bronze battle-axes are unknown in Northern France as well as in Britain (ib., pp. 161-2).

[516] Ib., pp. 19-20, 41, 51, 165, 189-90, 224-5, 256, 480, 487; Anc. Stone Implements, 1897, p. 143; Trans. Ethn. Soc., N. S., iii, 1865, p. 313; Archaeologia, xliii, 1871, pp. 412-3, 435-6, 438; W. Greenwell, Brit. Barrows, pp. 38, 43, 360; A. Pitt-Rivers, Excavations in Cranborne Chase, iv, 11, 17; _Trans. Devon Association_, xxxiv, 1902, p. 128; Vict. Hist. of ... Lancs, i, 218.

[517] No British bronze saws are mentioned in Sir J. Evans’s work, which was published in 1880; but see Proc. Soc. Ant., 2nd ser., xi, 1885-7, p. 12; The Naturalist, 1903, pp. 206-7; and J. R. Mortimer, Forty Years’ Researches, p. 182.

[518] J. Evans, Anc. Bronze Implements, p. 173.

[519] Proc. Soc. Ant. Scot., xxxviii, 1904, p. 78.

[520] See B. C. A. Windle, Remains of the Prehist. Age, &c., pp. 232-47. Durham, strange to say, has hardly any, though they abound in Northumberland (W. Greenwell, Brit. Barrows, pp. 378, 440; Journ. Brit. Archaeol. Association, N. S., xi, 1905, p. 168). In highly cultivated districts many have of course been destroyed.

[521] Archaeol. Journal, xxix, 1872, p. 160, n. 1; xxxii, 1875, p. 292.

[522] See A. Pitt Rivers, Excavations in Cranborne Chase, iii, 5, 7-8.

[523] J. Evans, Anc. Bronze Implements, p. 250.

[524] Trans. Devon. Association, xxxi, 1899, pp. 146-55; Trans. Hon. Soc. Cymmrodorion, 1898-9 (1900), p. 19.

[525] Archaeol. Cambr., 6th ser., ii, 1902, pp. 252-60.

[526] Proc. Somerset. Archaeol. and Nat. Hist. Soc., 1, 1905, part ii, pp. 32-49.

[527] 42nd Annual Report Roy. Inst. Cornwall, 1860, pp. 17-43; Journ. Roy. Inst. Cornwall, xiii, 1895 (1896), pp. 98-9; Trans. Hon. Soc. Cymmrodorion, 1898-9 (1900), p. 19; Trans. Roy. Irish Acad., xxxi, 1896-1902, p. 633.

[528] Ib., pp. 618-9, 623; Journ. Roy. Soc. Ant. Ireland, xxxv, 1905, pp. 244-5.

[529] Vict. Hist. of ... Worcester, i, 182.

[530] G. Payne, Collectanea Cantiana, 1893, pp. 176-7.

[531] Vict. Hist. of ... Sussex, i, 471.

[532] Vict. Hist. of ... Berks, i, 261.

[533] Journ. Brit. Archaeol. Association, xxviii, 1872, pp. 40, 42; Man, iv, 1904, No. 105, pp. 161-2. Mr. Cunnington (Proc. Dorset. Nat. Hist. and Ant. Field Club, xxiv, 1903, pp. xxxiv-xxxviii) has wasted much labour in endeavouring to prove that the Maiden Castle was constructed by the Romans. In 1882 and following years he excavated in the eastern division of the fort and found remains of a Roman building, which proves merely that the fort was occupied in Roman times.

[534] I am glad to find that I have the support of Mr. C. H. Read (Guide to the Ant. of the Bronze Age, p. 78) and Mr. Reginald Smith (Guide to the Ant. of the Early Iron Age, 1905, p. 122), who unhesitatingly refer the Dorsetshire hill-forts in general to ‘the Bronze, and possibly, in some cases, the Neolithic period’. Mr. H. St. G. Gray strains Pitt-Rivers’s doctrine when he argues (Index to Excavations in Cranborne Chase, 1905, p. xix; Proc. Somerset. Archaeol. and Nat. Hist. Soc., 3rd ser., ix, 1903 [1905], p. 28) that, without excavation, it is idle to express any opinion as to the age of a camp. He says that ‘Caesar’s Camp’ at Folkestone ‘was always considered to be pre-Roman before Lane-Fox excavated it and proved it to be of Norman construction’ This remark is incorrect (see Excavations in Cranborne Chase, iii, p. xi), and, even if it were true, would only prove that less was known about the principles of construction of prehistoric British camps in former days than now. I cannot conceive how anybody could on a priori grounds suppose ‘Caesar’s Camp’ to be pre-Roman, even if he had only seen the plan on the 25-inch O. S. map. Of course I freely admit that, without excavation, it would be generally (though not always) idle to express any opinion as to the particular prehistoric epoch to which a fort belonged.

Dr. Joseph Anderson (Scotland in Pagan Times,--the Bronze and Stone Ages, p. 139, with which cf. D. Christison, Early Fortifications in Scotland, 1898, pp. 350-3, 380-1) observes that no one Scottish fort ‘can be assigned with certainty to the Bronze Age’; but it is nevertheless morally certain that many did then exist.

Mr. R. Burnard (Vict. Hist. of ... Devon, i, 366) remarks that it is unsafe to infer from the fact that Cranbrook Castle, in the valley of the Teign, contained pottery of Bronze Age type, that it belonged to that period; for such pottery was also used in the Early Iron Age. Certainly it was; and so also were bronze implements (see pp. 266-7, infra); but, as a rule, the discovery of Bronze Age pottery, or bronze implements, unaccompanied by objects of the Early Iron Age, is enough to raise a presumption, which in most instances would be correct, that the site was occupied in the Bronze Age.

[535] Scheme for recording Ancient Defensive Earthworks and Fortified Enclosures, 1903, pp. 2-3, 6 (published by the Congress of Archaeological Societies). My classification differs slightly in form, but not in substance, from that adopted by the Congress.

[536] Journ. Derbyshire Archaeol. and Nat. Hist. Soc., xxv, 1903, pp. 175-80.

[537] Archaeologia, xliv, 1873, p. 424; xlix, 1885, p. 181; Journ. Roy. Soc. Ant. Ireland, xxxv, 1905, p. 244; Archaeol. Cambr., 3rd ser., xi, 1865, pp. 77-81. Many of the ‘cliff-castles’ probably do not belong to the Bronze Age (see Vict. Hist. of ... Cornwall, i, 451-2, 458-9).

[538] Journ. Anthr. Inst., vi, 1877, pp. 288-9. See also Archaeologia, xlvi, 1881, p. 458.

[539] Journ. Anthr. Inst., vi, 1877, pp. 288-9. The same feature exists in the camp at Seaford.

[540] See p. 98, supra.

[541] See Archaeol. Journal, xxii, 1865, p. 354.

[542] E.g. Ambresbury Banks in Essex, Yarnbury on Salisbury Plain, and Hunsbury near Northampton. See R. C. Hoare, Anc. Wilts, i, 1812, pp. 89-90; Trans. Epping Forest ... Field Club, ii, 1882, pp. 55-68; and Journ. Brit. Archaeol. Association, N. S., vii, 1901, p. 23.

[543] See pp. 259-60, infra.

[544] Mr. I. Chalkley Gould (Trans. Essex Archaeol. Soc., N. S., viii, 1903, p. 139, with which cf. Journ. Derby. Archaeol. and Nat. Hist. Soc., xxiv, 1902, p. 29) says that ‘the early constructors fixed on the highest points, and ... adopted a system of tortuous and involved entrances’; and that in later times engineers ‘no longer depended on involved tortuous entrances’. There is some truth in this; but forts which were at all events occupied in the Early Iron Age stood ‘on the highest points’, and the entrances of Cissbury (Archaeologia, xlv, 1880, p. 338, pl. xxvi), which was probably erected in the Neolithic Age (pp. 97-8, supra), and of many other hill-forts were not tortuous.

[545] See, for instance, Trans. Devon. Association, xxxi, 1899, p. 151, xxxiii, 1901, pp. 129-38; and Archaeol. Cambr., 6th ser., ii, 1902, pp. 252-60.

[546] See Vict. Hist. of ... Derby, i, 364, and pp. 257-8, infra.

[547] Proc. Geologists’ Association, 1887-8 (1889), pp. 376-7; Papers Hants Field Club, iii, 1896, part ii, p. 175; W. Johnson and W. Wright, Neol. Man in N.-E. Surrey, 1903, p. 47; Cornhill Mag., May, 1906, p. 612.

[548] Excavations in Cranborne Chase, iv, 19-20.

[549] See my Caesar’s Conquest of Gaul, 1903, pp. 51, 160-1.

[550] See Archaeologia, xlii, 1869, p. 51; A. Pitt-Rivers, Excavations in Cranborne Chase, ii, 238; and Journ. Derby. Archaeol. and Nat. Hist. Soc., xxiii, 1901, pp. 112-3.

[551] Prof. F. J. Haverfield (Eng. Hist. Rev., xix, 1904, p. 746) says that ‘the scanty archaeological evidence hardly seems to justify ... Cornish mining so early as B.C. 800’. But it does prove that bronze implements were made in Britain in the earlier part of the Bronze Age,--considerably earlier than 800 B.C.; and the tin must have been obtained either from Cornwall, or from Dartmoor, or from both. There is no evidence of prehistoric mining in Dartmoor, and there is in Cornwall (pp. 502-3, n. 8, infra).

[552] See Prof. Gowland’s interesting paper in Archaeologia, lvi, 1892, pp. 267-322, especially 268, 284-5, 287, 296.

[553] J. Evans, Anc. Bronze Implements, p. 425. I do not know what the results of later analyses have been. See also Archaeologia, liv, 1895, p. 97, and Addenda, p. 739.

[554] J. Evans, Anc. Bronze Implements, pp. 39, 188, 204-5, 222, 487.

[555] Ib., pp. 48, 69-70, 74, 76, 153.

[556] Ib., pp. 70-3, 160.

[557] Ib., p. 468.

[558] Ib., pp. 107-8.

[559] Ib., pp. 160, 162-3.

[560] Archaeologia, liv, 1895, p. 98.

[561] J. Evans, Anc. Bronze Implements, p. 167. Cf. Archaeol. Journal, xxvi, 1869, pp. 346-50.

[562] J. Evans, Anc. Bronze Implements, pp. 173, 175-6.

[563] Ib., pp. 207-8.

[564] J. Evans, Anc. Bronze Implements, p. 480.

[565] Ib., p. 479; Papers Hants Field Club, iii, 1895, pp. 56-7.

[566] J. Evans, Anc. Bronze Implements, p. 106.

[567] Sir J. Evans (ib., p. 483) observes that ‘our socketed celts appear to have had the cradle of their family in Western Germany’. See also ib., pp. 107-8.

[568] Ib., pp. 84, 108, 483.

[569] Ib., pp. 108, 135.

[570] Ib., pp. 114-5.

[571] Journ. Roy. United Service Inst., xiii, 1870, p. 531; J. Evans, Anc. Bronze Implements, pp. 75, 133.

[572] Pitt-Rivers (Excavations in Cranborne Chase, iv, 106-7), referring to Ancient Bronze Implements, p. 72, remarks that a palstave found on the bottom of the Angle Ditch on Handley Down was probably used in excavating the ditch, the sides of the lower part of which are ‘scored all along by vertical grooves’, some of which ‘coincide with the width of the flat side of the palstave’.

[573] J. Evans, Anc. Bronze Implements, pp. 194, 197-200, 202; J. Anderson, Scotland in Pagan Times,--the Bronze and Stone Ages, p. 202; Proc. Soc. Ant., 2nd ser., xv, 1893-5, pp. 358-60.

[574] J. Evans, Anc. Bronze Implements, pp. 204, 222-4.

[575] Ib., pp. 257-60, 473, 480; Archaeologia, xxxvi, 1855, pp. 326-31.

[576] J. Evans, Anc. Bronze Implements, pp. 261-2, 265, 269-70.

[577] Ib., pp. 248-9, 256, 273, 342-3, 354-5.

[578] J. Evans, Anc. Bronze Implements, pp. 343-8, 354-5, 481; Archaeologia, xxvii, 1838, pp. 298-300, liv, 1895, p. 112; J. Anderson, Scotland in Pagan Times,--the Bronze and Stone Ages, p. 227.

[579] Ib., pp. 171, 173; J. Evans, Anc. Bronze Implements, pp. 250-1, 278-81, 297, 301-2, 308, 481; Archaeologia, liv, 1895, p. 112.

[580] Ib., xliii, 1871, pp. 455-6; J. Evans, _Anc. Bronze Implements_, p. 310.

[581] Proc. Soc. Ant., 2nd ser., xix, No. ii, 1902, pp. 287-9.

[582] ‘The openings’ [in one specimen], says Sir J. Evans (Anc. Bronze Implements, p. 332), ‘are about 17 inches from the point. An Irish friend has suggested that they were for the reception of poison, but after the blade had penetrated seventeen inches into the human body such an use of poison would probably be superfluous.’

[583] J. Evans, Anc. Bronze Implements, pp. 341-2; Guide to the Ant. of the Bronze Age (Brit. Museum), pp. 68-9, 81.

[584] J. Evans, Anc. Bronze Implements, pp. 337-8, 341-2.

[585] Ib., p. 315.

[586] Ib., p. 311.

[587] J. Evans, Anc. Bronze Implements, pp. 107, 427-8, 430, 438, 440-1, 445; Proc. Soc. Ant., 2nd ser., xvi, 1895-7, pp. 328-30; xx, 1904-5, p. 259; Proc. Soc. Ant. Scot., xxxviii, 1904, pp. 487-505; Guide to the Ant. of the Bronze Age (Brit. Museum), pp. 72-3.

[588] J. Evans, Anc. Bronze Implements, pp. 186, 451-2. See p. 71, supra.

[589] Ib., pp. 67-8, 177, 179, 181, 451-3.

[590] Daggers with chevron ornament are very rare in the British Isles except in Ireland (Archaeol. Cambr., 6th ser., ii, 1902, p. 221). One or two English specimens are noticed by Sir John Evans (Anc. Bronze Implements, pp. 232, 238).

[591] Ib., pp. 108, 320, 330. Mr. Romilly Allen is mistaken when he says (Archaeol. Cambr., 6th ser., ii, 1902, p. 220) that winged, looped, and socketed celts never have chevron ornament. Several instances are given in Anc. Bronze Implements (pp. 74, 84, 90, 126, 128, 132).

[592] See pp. 181-4, infra.

[593] Anc. Bronze Implements, pp. 457-70. For details of hoards which have been found since the publication of Sir John Evans’s book, see Archaeol. Cambr., 5th ser., i, 1884, pp. 225-7; Proc. Soc. Ant., 2nd ser., xi, 1885-7, pp. 12, 42-51; xv, 1893-5, p. 138; xvi, 1895-7, pp. 96-8, 327-30; xviii, 1900, pp. 285-7; Archaeologia, xlviii, 1888, pp. 106-14; Proc. Soc. Ant. Scot., xxvi, 1892, pp. 182-8; xxxv, 1901, pp. 266-75; Papers Hants Field Club, iii, 1895, pp. 53-66; and Vict. Hist. of ... Surrey, i, 241.

[594] Anc. Bronze Implements, pp. 457-8; Proc. Soc. Ant. Scot., xxxviii, 1904, p. 504.

[595] Some hoards of damaged and broken implements, unaccompanied by copper cakes, may have been formed for barter with a bronze-founder (Vict. Hist. of ... Surrey, i, 240).

[596] See p. 126, supra.

[597] Cf. Archaeologia, xliii, 1871, p. 536, with Greenwell’s Brit. Barrows, p. 740, and Pitt-Rivers’s Excavations in Cranborne Chase, iv, 132.

[598] W. Greenwell, Brit. Barrows, pp. 110-1; Archaeologia, liv, 1895, pp. 110-1; A. Pitt-Rivers, Excavations in Cranborne Chase, iv, 19.

[599] J. Evans, Anc. Bronze Implements, pp. 192, 236.

[600] J. R. Mortimer, Forty Years’ Researches, &c., pp. 111-2. See also Brit. Barrows, p. 114; and Journ. Anthr. Inst., xxxiv, 1904, pp. 392, 396.

[601] See W. Robertson Smith, The Religion of the Semites, 1901, pp. 260-1.

[602] See p. 90, supra.

[603] Oxen drawing a plough are depicted on rock-carvings in Scandinavia (Congrès internat. d’anthr. et d’archéol. préhist., i, 1874 [1877], pp. 454 [fig. 1], 473 [fig. 31]).

[604] W. Greenwell, Brit. Barrows, pp. 220, 262; Archaeologia, liv, 1895, p. 110; A. Pitt-Rivers, Excavations in Cranborne Chase, iv, 127-8.

[605] O. Montelius, Civilisation of Sweden in Heathen Times, 1888, pp. 71-6; Guide to the Ant. of the Bronze Age (Brit. Museum), p. 141.

[606] J. Evans, Anc. Bronze Implements, pp. 399-400, 404. Harness rings have been found, according to Dr. J. Anderson (Scotland in Pagan Times,--the Bronze and Stone Ages, p. 168), in Scottish hoards. Professor W. Ridgeway (The Thoroughbred Horse, p. 92) argues that ‘the use of the horse by man in the British Isles cannot be placed before the end’ of the Bronze Age.

[607] See p. 221, infra.

[608] W. Greenwell, Brit. Barrows, pp. 656-7.

[609] Nature, Nov. 22, 1894, p. 92.

[610] O. Schrader, Prehist. Ant. of the Aryan Peoples, p. 379.

[611] W. Greenwell, Brit. Barrows, pp. 139, 591-2, 599.

[612] See pp. 85-6, supra.

[613] See Guide to the Ant. of the Bronze Age (Brit. Museum), pp. 52-3; and p. 467, infra.

[614] D. Wilson, Prehist. Annals of Scotland, 2nd ed., i, 1863, p. 107.

[615] See p. 87, supra.

[616] See B.C. A. Windle, Remains of the Prehist. Age, &c., p. 279.

[617] Anc. Bronze Implements, p. 486. Bronze spear-heads were associated with objects of the Early Iron Age in a hoard found on Hagbourne Hill in Berkshire, which belonged to a period of transition. See p. 267, infra.

[618] Guide to the Ant. of the Bronze Age (Brit. Museum), p. 143. The argument is no doubt generally sound, and no find, as far as I know, refutes it; but I do not think that it is absolutely conclusive. Stone implements were undoubtedly used in the Early Iron Age; and bronze and iron implements have been found together.

[619] Archaeol. Journal, xxiv, 1867, pp. 229-35; xxvi, 1869, pp. 301-5, 317; xxvii, 1870, pp. 158-9. The workshops may have been used in the Bronze Age; but one, in which iron slag was found, contained Roman coins. Huts similar to those of Ty Mawr have been explored in Brittany (ib., p. 148).

[620] Archaeologia, xlv, 1880, pp. 356-8. See p. 86, supra.

[621] W. C. Lukis, Prehist. Stone Monuments of the Brit. Isles,--Cornwall, pp. 18-9.

[622] W. Greenwell, Brit. Barrows, p. 402; Reliquary, viii, 1902, p. 92.

[623] Vict. Hist. of ... Devon, i, 350, 352.

[624] Ib., p. 349; Trans. Devon. Association, xxxiv, 1902, p. 161.

[625] Ib., p. 163.

[626] Ib., pp. 161-3.

[627] Vict. Hist. of ... Devon, i, 354.

[628] Trans. Devon. Association, xxxi, 1899, p. 148.

[629] Ib., xxxiv, 1902, p. 160. Cf. Archaeologia, xlv, 1880, pp. 362-3.

[630] Trans. Devon. Association, xxxiv, 1902, p. 160.

[631] Reliquary, N. S., viii, 1902, p. 91.

[632] Trans. Devon. Association, xxxiv, 1902, p. 163. Bronze has been found, though rarely, in graves on Dartmoor (Ib., p. 130; Reliquary, N. S., vii, 1901. p. 95).

[633] Ib., p. 92.

[634] A. Pitt-Rivers, Excavations in Cranborne Chase, iv, 46-8.

[635] Ib., pp. 14, 15-6, 20 (preface), 185-90, and pl. 306. See also Journ. Anthr. Inst., xxxiv, 1904, pp. 387-97. Mr. J. R. Mortimer (Forty Years’ Researches, pp. 365, 369) describes an ‘extensive labyrinth of entrenchments’, which ‘traversed the high grounds of the [Yorkshire] Wolds in every direction, forming a network ... connecting hill to hill and valley to valley’. He states (p. 379) that several round barrows have been mutilated by these entrenchments ‘in a manner which shows that the latter are the more recent’ [but does not show that they are later than the Bronze Age], but that they are ‘at least for the most part pre-Roman, being in several instances crossed by what are believed to be portions of ... Roman roads’; and he concludes, disagreeing with Pitt-Rivers (see p. 441, n. 2, infra), that they were intended to protect cattle against robbers. In regard to the entrenchments near Flamborough Head which Pitt Rivers excavated, I prefer his guidance.

[636] W. Greenwell, Brit. Barrows, pp. 107-8; J. Evans, Anc. Bronze Implements, pp. 179, 185; Archaeologia, liv, 1895, pp. 87-114. Canon Greenwell (ib., p. 103) conjectures that the disks, which have analogues in France, Switzerland, and Italy, may have been worn as ornaments upon the breast; but their use is uncertain. Cf. Anc. Bronze Implements, pp. 401-3.

[637] Ib., pp. 408-14; J. Anderson, Scotland in Pagan Times,--the Bronze and Stone Ages, p. 205; Guide to the Ant. of the Bronze Age (Brit. Museum), pp. 28-30, 84. Cauldrons with ring-handles and rounded bottoms have been found in Ireland and in various parts of Scotland, but, according to Dr. J. Anderson (Proc. Soc. Ant. Scot., xix, 1885, pp. 313-5; xxxix, 1905, pp. 14-20) and Mr. C. H. Read (Guide, &c., p. 30), not on the Continent. Dr. R. Munro, however (Lake Dwellings of Europe, 1890, p. 290), affirms that they have been found in the famous settlement of La Tène. They belonged indeed to the very latest period of the Bronze Age, if not to the Early Iron Age (Anc. Bronze Implements, p. 410). Only one cauldron of the flat-bottomed type has been found in Scotland (Proc. Soc. Ant. Scot., xxii, 1888, p. 36). The British cauldrons of this kind differ in details from continental examples, especially in their handles, which are rings, and may be of native manufacture. Anyhow, the Heathery Burn Cave cauldron had been mended with a degree of skill which shows that British workmen knew how to rivet plates together.

[638] Professor Boyd Dawkins (Early Man in Britain, pp. 360-2) remarks that if a well-known amber cup, which was found at Hove, was of British workmanship, it proves that the use of the lathe was known in Britain in the Bronze Age. Sir John Evans (Anc. Stone Implements, 1897, pp. 445-50), like the professor, thinks that the cup may have been imported; but he points out that cups made of shale, which were certainly turned in a lathe, and were most probably of British manufacture, have been found in round barrows on Broad Down near Honiton. Sir R. C. Hoare (Anc. Wilts, i, 122-3) found in the trunk of a tree inside a bowl barrow along with a skeleton an urn which he described as ‘different both in shape and colour to any we have ever found in the British sepulchres’, and which appeared to him to have been turned in a lathe. Still the statement in the text is, generally speaking, true both of the British Isles and of Northern and Western Europe.

[639] Trans. Devon. Association, xxxiv, 1902, p. 125. See also p. 467, infra.

[640] Archaeologia, xliii, 1871, pp. 338-43, and pl. xxix; liv, 1895, p. 110; W. Greenwell, Brit. Barrows, pp. 11, 106-8; Trans. Devon. Association, xxxiv, 1902, p. 125; Proc. Somerset. Archaeol. and Nat. Hist. Soc., 1, 1905, part ii, p. 42; J. R. Mortimer, Forty Years’ Researches, pp. lxvii, 9, 82.

[641] See pp. 395-7, infra.

[642] Archaeologia, liv, 1895, pp. 112-4.

[643] Ib., pp. 94, 108.

[644] Sir J. Evans, Anc. Stone Implements, 1897, pp. 438-9; Anc. Bronze Implements, p. 383.

[645] Sir A. Mitchell, The Past in the Present, pp. 5-6, 12.

[646] Archaeologia, liv, 1895, pp. 108-9.

[647] Ib., xliii, 1871, p. 440; J. Evans, Anc. Bronze Implements, pp. 191-2; A. Pitt-Rivers, Excavations in Cranborne Chase, iv, 162-3.

[648] ‘So called’, says Pitt-Rivers (ib., i, 66), ‘because some of them are found with an edge as sharp as a penknife’. The thought of shaving with a bronze razor is not pleasant; but the negroes of Tanganyika still use razors of this metal (L’Anthr., xiv, 1903, pp. 667-75; xv, 1904, p. 116); and everybody knows that the Flamen Dialis might only shave with a bronze knife. See J. G. Frazer, The Golden Bough, i, 1900, p. 242.

[649] Archaeologia, xliii, 1871, p. 451; liv, 1895, p. 99; J. Evans, Anc. Bronze Implements, pp. 217-21, 480.

[650] See p. 156, supra. Cf. W. Greenwell, Brit. Barrows, pp. 118, 142.

[651] Ib., p. 411.

[652] See p. 189, infra.

[653] Germania, 17.

[654] Archaeol. Journal, ix, 1852, p. 8; Archaeologia, xliii, 1871, pp. 432-4; liv, 1895, pp. 101, 107; W. Greenwell, Brit. Barrows, pp. 15, 31, 33; J. Evans, Anc. Bronze Implements, pp. 366-73; J. Anderson, Scotland in Pagan Times,--the Bronze and Stone Ages, p. 168.

[655] W. Greenwell, Brit. Barrows, pp. 32-3, 54-6; Archaeologia, xliii, 1871, pp. 510-2, 519-22; J. R. Mortimer, Forty Years’ Researches, p. xli.

[656] Sir J. Evans, Anc. Stone Implements, 1897, pp. 452-5; Anc. Bronze Implements, pp. 400-1; Proc. Soc. Ant. Scot., xxxvi, 1902, pp. 473-4, 477, 480-5. One button with a V-shaped perforation, found in a barrow at Winterbourne Stoke, Wiltshire, was associated with glass beads (ib., p. 474), which (p. 183, infra) appear to have belonged to a comparatively advanced period of the Bronze Age. Cf. Proc. Soc. Ant. Scot., xxxvi, 1902, p. 481.

[657] R. C. Hoare, Anc. Wilts, i. 99, pl. x; Archaeologia, xliii, 1871, pp. 524-5.

[658] Ib., pp. 490-1, 502-3; J. R. Mortimer, Forty Years’ Researches, pp. li, 92.

[659] Archaeologia, xliii, 1871, p. 492.

[660] W. Greenwell, Brit. Barrows, p. 55, n. 1.

[661] Trans. Devon. Association, v, 1872, pp. 554-5 and pl. ii.

[662] Anc. Wilts, i, 202, pl. xxvii, 2. See also Archaeologia, xliii, 1871, p. 459, and Anc. Bronze Implements, pp. 51, 232.

[663] Archaeologia, xxvi, 1836, pp. 422-31; Archaeol. Journal, lviii, 1901, p. 324; Proc. Soc. Ant., 2nd ser., xviii, 1899-1901, pp. 223-4.

[664] Archaeologia, xliii, 1871, p. 501. Cf. p. 502, note c.

[665] See Anc. Bronze Implements, pp. 395, 481, 487.

[666] Ib., p. 481.

[667] Ib., p. 395.

[668] Archaeologia, liv, 1895, p. 102. Sir John Evans (Anc. Bronze Implements, p. 402) does not accept them as armlets; but cf. Guide to the Ant. of the Bronze Age (Brit. Museum), p. 85.

[669] J. Evans, Anc. Bronze Implements, pp. 76, 90, 96, 375-9. Cf. W. Greenwell, Brit. Barrows, pp. 55, n. 1, 436. Bronze torques of all these patterns have also been collected. Funicular torques are unknown in Scotland.

[670] J. Evans, Anc. Bronze Implements, pp. 283, 381-7; Archaeologia, xliii, 1871, pp. 489-90, 528; J. Anderson, Scotland in Pagan Times,--the Bronze and Stone Ages, pp. 94-5, 168, 217; A. Pitt-Rivers, Excavations in Cranborne Chase, iv, 6.

[671] Proc. Soc. Ant., 2nd ser., ii, 1861-4, pp. 247-8; J. Anderson, Scotland in Pagan Times,--the Bronze and Stone Ages, p. 224, n. 1; Mr. G. Clinch (Vict. Hist. of ... Sussex, i, 320) thinks that the Mountfield hoard probably belonged to the Late Celtic period.

[672] J. Anderson, Scotland in Pagan Times,--the Bronze and Stone Ages, pp. 217, 220-1.

[673] Archaeol. Journal, xxiv, 1867, pp. 197, 201; J. Evans, _Anc. Bronze Implements_, p. 42; J. Anderson, _Scotland in Pagan Times,--the Bronze and Stone Ages_, p. 222; Rev. celt., xxi, 1900, pp. 166-73; Vict. Hist. of ... Cornwall, i, 356. Cf. L’Anthr., xvii, 1906, p. 135.

[674] See Proc. Soc. Ant. Scot., xxxv, 1901, p. 263.

[675] Archaeologia, xliii, 1871, p. 470; J. Evans, Anc. Bronze Implements, pp. 390-1.

[676] Ib., pp. 391-3; Archaeologia, xliii, 1871, pp. 471, 531; W. Greenwell, Brit. Barrows, pp. 117, 223, 324; J. R. Mortimer, Forty Years’ Researches, pp. xlvi, 218. A pair of ornaments, which may have been ear-rings, have been found in a barrow in Wiltshire. Cf. Guide to the Ant. of the Bronze Age (Brit. Museum), p. 94, with Anc. Bronze Implements, p. 393. Ear-rings of the Bronze Age are equally rare in France (Proc. Soc. Ant. Scot., xxxv, 1901, pp. 267, 273).

[677] J. Evans, Anc. Bronze Implements, pp. 392-3

[678] Archaeologia, xliii, 1871, pp. 514-5, 522; J. Anderson, Scotland in Pagan Times,--the Bronze and Stone Ages, pp. 51-2; Proc. Soc. Ant. Scot., xxxv, 1901, pp. 270-1; Archaeol. Cambr., 6th ser., ii, 1902, p. 209.

[679] See p. 469, infra.

[680] Archaeologia, xliii, 1871, pp. 494-5, 500, 504, 526-7.

[681] Proc. Soc. Ant. Scot., xxvi, 1892, p. 183.

[682] J. Evans, Anc. Bronze Implements, pp. 394, 485; L’Anthr., xvi, 1905, pp. 173-5. Mr. L. M^cLellan (Proc. Soc. Ant. Scot., xl, 1906, pp. 396-402), unlike Mr. Abercromby (Journ. Anthr. Inst., xxxix, 1905, p. 262), argues that the paste beads were made in Britain.

[683] Archaeologia, xliii, 1871, pp. 492-4.

[684] J. Evans, Anc. Bronze Implements, p. 485.

[685] Journ. Roy. Inst. Cornwall, xvi, 1904, p. 103.

[686] Journ. Roy. Soc. Ant., Ireland, 5th ser., v, 1895, p. 23; Report of ... the Brit. Association, 1896, pp. 912-4; L’Anthr., vii, 1896, pp. 688-9; Rev. celt., xxi, 1900, pp. 166-75; Guide to the Ant. of the Bronze Age (Brit. Museum), pp. 29, 145-6; Man, v, 1905, No. 7, p. 13. For evidences of intercourse between Scotland and Ireland see Proc. Soc. Ant. Scot., vi, 1867, pp. 350-1; W. Greenwell, Brit. Barrows, p. 62; and R. Munro, Prehist. Scotland, p. 290.

[687] Archaeologia, xliii, 1871, pp. 508-9; J. Evans, Anc. Bronze Implements, p. 484; Journ. Roy. Soc. Ant. Ireland, 5th ser., vi, 1896, p. 37, n. 1.

[688] This conjecture, I find, has the support of Mr. Coffey (ib., p. 39).

[689] See p. 357, infra.

[690] Proc. Soc. Ant. Scot., iii, 1860, pp. 183, 195.

[691] Ib., xxvi, 1892, pp. 186-7; W. Greenwell, Brit. Barrows, p. 55.

[692] Journ. Anthr. Inst., xxxv, 1905, p. 261.

[693] W. Greenwell, Brit. Barrows, pp. 54-6; Archaeologia, xliii, 1871, pp. 492, 494-5, 507, 530; lii, 1890, pp. 58-9. A glass bead has been found in one Derbyshire barrow (Proc. Soc. Ant., 2nd ser., xv, 1893-5, p. 425). Gold has only once been found in a barrow on the Yorkshire Wolds (Archaeologia, lii, 58-9); and of 379 interments only 10, of which 2 were Late Celtic, were found there accompanied by ornaments (Brit. Barrows, pp. 51-2); whereas in Wiltshire 64 were found out of 354 (Archaeologia, xliii, 488).

[694] The absence of gold and amber which distinguishes the group of barrows round the great stone circle at Avebury from those associated with Stonehenge (Crania Britannica, ii, pl. 11, p. 5, n. 11) is remarkable. Perhaps it may be due partly to the greater antiquity of the Avebury barrows.

[695] In regard to the poverty of the people who, probably in the Early Iron Age, used the stronghold of Winkelbury Hill, 13 miles WSW. of Salisbury, see A. Pitt-Rivers, Excavations in Cranborne Chase, ii, 236.

[696] Dr. Joseph Anderson assigns these balls, which have been found only in Scotland, to the Iron Age; but Mr. George Coffey (Journ. Roy. Soc. Ant. Ireland, 5th ser., vi, 1896, p. 42) thinks that ‘the general character of these spirals appears to be distinctly Bronze Age, not Late Celtic’. It has been pointed out (Proc. Soc. Ant., 2nd ser., xvi, 1895-7, pp. 408-9) that Mr. Coffey’s theory is ‘strengthened by the fact that stone balls of this class have been found associated with cist burials [of the Bronze Age] ... near Ballater, and ... [in] Elginshire’.

[697] Meath, Louth, Fermanagh, Tyrone, and Donegal.

[698] See Mr. G. Coffey’s articles in Journ. Roy. Soc. Ant. Ireland, 5th ser., iv. 1894, pp. 349-79; v, 1895, pp. 16-29, 195-211; vi, 1896, pp. 34-69, and especially 40-2, 65; Proc. Soc. Ant. Scot., xxi, 1887, p. 144; xxiii, 1889, p. 133; xxix, 1895, pp. 191-4; xxxiii, 1899, pp. 363-4, 368; L’Anthr., vii, 1896, pp. 688-9; xvii, 1906, p. 332, fig. 6; Trans. Cumberland and Westmorland Ant. and Archaeol. Soc., N. S., ii, 1902, pp. 381-2; Romilly Allen, Celtic Art in Pagan and Christian Times, 1904, pp. 50-3; C. H. Read, Guide to the Ant. of the Bronze Age (Brit. Museum), p. 96; and Vict. Hist. of ... Lancs, i, 240. Mr. Read observes (op. cit., p. 103) that some of the Scandinavian bronzes have been shown by analysis to have been imported from Britain.

[699] M. Salomon Reinach (L’Anthr., iv, 1893, pp. 688-90), if I do not misunderstand him, although he admits that the British Isles in the Bronze Age were connected by trade with Scandinavia, thinks it probable that the spiral reached the former by way of Spain. He observes that among Scandinavian rock-sculptures there are no spirals; that certain designs--for example the boat--are common to sepulchral monuments in Brittany and in Ireland; that there are striking points of resemblance between the bronze culture of Ireland and Spain; and that designs which have been found in the East Riding of Yorkshire represent Aegean types which also appear in the departments of the Marne and the Gard (see p. 200, infra). These arguments seem unavailing against those stated in the text, and especially against the almost complete absence of the spiral from Spain, Gaul, and Southern England. If the figure does not appear on Scandinavian rocks, it abounds on Scandinavian weapons and ornaments; and on rocks boats are frequently represented (O. Montelius, Civilisation of Sweden in Heathen Times, 1888, pp. 46, 73-6). Nor is M. Reinach’s reasoning sound when he goes on to argue that we have no right to trace the Bronze Age spiral of Western Europe to an Egyptian source because the same design has been found engraved on mammoths’ tusks in the Pyrenaean cave of Espélungues at Arudy (L’Anthr., xv, 1904, p. 146, fig. 24; xvi, 1905, p. 5, fig. 4). As Dr. Arthur Evans remarks (Report of ... the Brit. Association, 1896, p. 913), ‘the earliest cultural strata of Europe, from the Neolithic period onwards, betray an entire absence of the recurring spiral motive. When we find it later propagating itself as a definite ornamental system in a regular chronological succession throughout an otherwise inter-related European zone, we have every right to trace it to a common source.

[700] See pp. 499-500 and 511-4, infra.

[701] See Journ. Anthr. Inst., xxx, 1900, p. 94, and the Hon. John Abercromby’s article in the same periodical, xxxv, 1905, pp. 256-64, especially 262. Mr. George Coffey (Journ. Roy. Soc. Ant. Ireland, 5th ser., vi, 1896, p. 40), remarking that Britain and Gaul were on a lower plane of civilization in the Bronze Age than Scandinavia, argues that one cause may have been that ‘the sea-way south of the Elbe was possibly closed to Scandinavian enterprise in the Bronze Age’. But the North Sea was not; and apparently there was nothing to prevent Scandinavian traders from landing on our eastern coasts if they had thought it worth while.

[702] See p. 119, supra. Prof. B.C. A. Windle (Remains of the Prehist. Age, pp. 153-73) gives a fairly complete list. It would be superfluous to print references for barrows belonging to counties not mentioned in my text; for full lists are being given in the Victoria County History.

[703] Long barrows may possibly have been erected in remote districts after bronze had been introduced into Southern Britain. See Archaeol. Cambr., 5th ser., viii, 1891, pp. 33-7.

[704] A. Pitt-Rivers, Excavations in Cranborne Chase, iv, 58-9. Cf. Archaeologia, xlii, 1869, p. 196.

[705] See p. 108, supra.

[706] W. Greenwell, Brit. Barrows, pp. 13, 451; Trans. Devon. Association, xxxiv, 1902, p. 111.

[707] Archaeologia, xliii, 1871, pp. 309, 314, 326; W. Greenwell, Brit. Barrows, pp. 12-3.

[708] Ib., pp. 31-2; R. C. Hoare, Anc. Wilts, i, 52, 122-5; J. Hutchins, Hist. and Ant. of Dorset, 3rd ed., i, 1861, p. 100; Archaeologia, xliii, 1871, pp. 314-5; A. Pitt-Rivers, Excavations in Cranborne Chase, i, 4; Proc. Soc. Ant. Scot., xxxix, 1905, pp. 179-81; J. R. Mortimer, Forty Years’ Researches, p. xxvii.

[709] J. R. Mortimer, Forty Years’ Researches, p. xxv.

[710] The same degeneration took place in Gaul (A. Bertrand, Archéol. celt. et gaul., 1889, p. 104; Rev. de l’École d’anthr., xv, 1905, pp. 213-4).

[711] Mr. J. R. Mortimer (Forty Years’ Researches, p. xxv) believes that on the Yorkshire Wolds barrows were occasionally erected over the dwellings in which the dead had lived; but the evidence which he adduces, except in one instance (pp. 182-3), appears to me weak. See pp. 155, 328-9, 336-7.

[712] Stonehenge, 1740, p. 45.

[713] Archaeologia, lii, 1890, p. 63.

[714] Ib., xlix, 1885, p. 183.

[715] J. Anderson, Scotland in Pagan Times,--the Bronze and Stone Ages, pp. 3-4; Trans. Glasgow Archaeol. Soc., N. S., iii, part ii, 1899, p. 499.

[716] Trans. Devon. Association, xxxvii, 1902, p. 106.

[717] Archaeologia, xliii, 1871, p. 309; lii, 1890, p. 63; _Proc. Soc. Ant._, 2nd ser., viii, 1879-81, p. 289.

[718] J. R. Mortimer, Forty Years’ Researches, p. xxi.

[719] Archaeologia, xliii, 1871, pp. 291-2; W. Greenwell, Brit. Barrows, pp. 4-5.

[720] Ib., p. 3; Archaeologia, xliii, 1871, pp. 290-1, 301-4; A. Pitt-Rivers, Excavations in Cranborne Chase, ii, 7-8, 64.

[721] Archaeologia, xliii, 1871, p. 291. It has often been assumed that barrows had no ditches because none were visible; but Pitt-Rivers (Excavations in Cranborne Chase, i, 4) showed that several barrows on his estate were surrounded by ditches ‘of which no trace was seen before excavating’. See also ib., ii, 7-8.

[722] Archaeologia, xliii, 1871, pp. 302-4; W. Greenwell, Brit. Barrows, pp. 3-4. Composite bowl barrows and bell barrows are described in Archaeologia, xliii, 297-300.

[723] Archaeologia, xliii, 1871, pp. 293-5, 303-4; Folk-Lore, vi, 1895, pp. 14-5. Pitt-Rivers (Excavations in Cranborne Chase, iv, 145) suggests that the common form of disk barrows ‘may have arisen through a failure to carry out the original intention’. ‘The first idea,’ he continues, ‘of the mourners ... may probably have been to erect a large monument ... and the ditch in such a case would contain a large area. In the course of a few days, however, the grief may have abated, and laziness supervened, in which case the arrested tumulus would assume the form described. The habit of all primitive peoples ... of lashing themselves up into a frenzy on the occasion of a death, and general excitability upon any uncommon occurrence, followed by a speedy relapse, favours this hypothesis. When, however, a vallum is seen to follow the line of the ditch, this cause cannot be assigned to the particular structure. It may, however, be a form that has become persistent and conventionalized through the cause already mentioned.’ This ingenious theory seems to imply that the motive of laziness only began to operate when disk barrows came into fashion.

The few disk barrows of Derbyshire have no apparent ditches (Vict. Hist. of ... Derby, i, 169). That county also contains barrows constructed differently and of different materials from those which undoubtedly belong to the Bronze Age: their date is uncertain, but may be Romano-British (ib., pp. 186-9; Proc. Soc. Ant., 2nd ser., xv, 1893-5, p. 427).

Mr. G. F. Tregelles (Vict. Hist. of ... Cornwall, i, 358) thinks that in Cornwall the distinctions between conical, bowl, bell, flat, and ring barrows ‘may be little more than differences in height’.

[724] Journal Brit. Archaeol. Association, xviii, 1862, p. 39; Proc. Soc. Ant., 2nd ser., viii, 1879-81, pp. 289, 291-2; x, 1884-5, pp. 305-6; W. C. Lukis, Prehist. Stone Monuments of the Brit. Isles,--Cornwall, p. 6; Archaeologia, lii, 1890, p. 63; Trans. Cumberland and Westmorland Ant. and Archaeol. Soc., N. S., i, 1901, pp. 295-9.

[725] Vict. Hist. of ... Derby, i. 169.

[726] The ditches of the Cranborne Chase barrows in Dorsetshire, just outside the frontier of Wiltshire, are sometimes incomplete (A. Pitt-Rivers, Excavations in Cranborne Chase, iv, 138).

[727] See p. 104, supra.

[728] W. Greenwell, Brit. Barrows, pp. 6-8; Proc. Soc. Ant., 2nd ser., xi, 1885-7, p. 434. Canon Greenwell (Proc. Soc. Ant. Scot., vi, 1867, p. 339), speaking of a cairn near Crinan in Argyllshire, surrounded by ‘a double circle of stones’, which ‘stood from 3 feet to 5 feet apart, except for a space ... where, in both circles, four stones were found placed close together’, says, ‘This is not an unusual feature in circles which enclose burials; in fact it is, in one shape or another, almost universal.... The object seems to be to make the circle incomplete.... When the circle is made of stones placed close together, or is formed of earth, then one or more openings occur in it.’ On the other hand, Mr. W. C. Borlase (Archaeologia, xlix, 1885, p. 183) found that in Cornwall the stone rings enclosing barrows were almost always continuous.

[729] W. Greenwell, Brit. Barrows, p. 8.

[730] A. Pitt-Rivers, Excavations in Cranborne Chase, iv, 138. Mr. J. R. Mortimer (Forty Years’ Researches, p. xxii), referring to the incomplete stone rings or trenches which are found within barrows, suggests that they were intended ‘to mark off ... the sacred spot in which the ceremony and interment were afterwards to be conducted, and that the break in the circle had no other significance than to serve as a place of ingress and egress’.

[731] Ib., p. lxxi; W. Greenwell, Brit. Barrows, p. 112.

[732] Archaeologia, xlii, 1869, pp. 169-71.

[733] Archaeol. Journal, lviii, 1901, pp. 328-31.

[734] Journ. Derby. Archaeol. and Nat. Hist. Soc., xi, 1889, pp. 39-44; xiv, 1892, pp. 244-7, 250; xv, 1893, pp. 161-2.

[735] In certain cases, however, the mound may have been so worn down by denudation as to escape notice.

[736] A. Pitt-Rivers, Excavations in Cranborne Chase, iv, 148.

[737] J. Anderson, Scotland in Pagan Times,--the Bronze and Stone Ages, pp. 28, 37-8; Proc. Soc. Ant. Scot., xx, 1886, pp. 114, 240-1; xxix, 1895, pp. 46-8; xxxv, 1901, pp. 258-66; xxxix, 1905, pp. 189, 528-32.

[738] Anthr. Rev., v, 1867, p. 255.

[739] W. Greenwell, Brit. Barrows, pp. 112-3.

[740] J. Anderson, Scotland in Pagan Times,--the Bronze and Stone Ages, pp. 38, 48-50, 63-4.

[741] Excavations in Cranborne Chase, iv, 22-3 (preface). ‘There is no knowing,’ says Pitt-Rivers, ‘how many of these graves without mounds or ditches may exist in the soil; as they show no mark on the surface, they can only be found accidentally’.

[742] W. Greenwell, Brit. Barrows, p. 113.

[743] Trans. Devon. Association, xxxiv, 1902, p. 111.

[744] Brit. Barrows, p. 402. For other instances of moundless graves see Anthr. Rev., iii, 1865 (Journ. Anthr. Soc., p. lxvii); Journ. Brit. Archaeol. Association, xlv, 1889, pp. 112-22; Proc. Soc. Ant., 2nd ser., xvi, 1895-7, p. 335; W. C. Borlase, Dolmens of Ireland, iii, 1013; Archaeol. Cambr., 6th ser., ii, 1902, p. 28; Wilts Archaeol. and Nat. Hist. Mag., xxxiii, 1904, pp. 410-1; Archeol. Aeliana, 3rd ser., ii, 1906, p. 132; and Vict. Hist. of ... Lancs, i, 245. Mr. J. R. Mortimer, Forty Years’ Researches, p. lxxii, suggests that ‘the great number of small cairns which even yet exist on the uncultivated moors of Yorkshire’ may have ‘belonged to the masses’.

[745] See pp. 105-6, supra.

[746] Trans. Lancs. and Cheshire Ant. Soc., xviii, 1900 (1901), pp. 114-24.

[747] These cists are assigned by the excavators, Dr. W. A. Herdman and Mr. P. M. C. Kermode (Proc. Liverpool Biol. Soc., viii, 1894, pp. 159-72), to the Neolithic Age; but the evidence which they adduce is purely negative. Perhaps the people who built them only had stone tools; but the fact that the interments were in cists and accompanied by cinerary urns proves that they were made after bronze had come into use.

[748] A. Pitt-Rivers, Excavations in Cranborne Chase, ii, 4; Archaeologia, lii, 1890, pp. 24-7, 41, 60; W. C. Borlase, Dolmens of Ireland, ii, 634; Général Pothier, Les tumulus du plateau de Ger, 1900, pp. 28-9. See also Journ. Anthr. Inst., xv, 1886, pp. 95-7.

[749] J. H. F. Brabner, Gazetteer of England and Wales, vi, 31.

[750] W. Greenwell, Brit. Barrows, p. 2, n. 2. Mr. J. R. Mortimer (Forty Years’ Researches, p. xviii, n. ||) remarks that ‘it could hardly be expected that these two small openings would be more likely to find the primary grave ... than two rat holes would be likely to come upon the ashes of a mouse placed under a mound ten feet in diameter’.

[751] Wilts Archaeol. and Nat. Hist. Mag., xxiii, 1887, pp. 245-52. At the foot of Garrow by Hill, on the Yorkshire Wolds, there is another gigantic mound, 50 feet high and 250 feet in diameter, which has not been opened (J. R. Mortimer, Forty Years’ Researches, p. xx).

[752] J. Evans, Anc. Bronze Implements, pp. 134, 273-4, 277, 342; W. Greenwell, Brit. Barrows, pp. 44, 48-9; Archaeologia, xliii, 1871, pp. 442-6. See also Journ. Brit. Archaeol. Association, x, 1855, p. 8. Dr. Thurnam (Archaeologia, xliii, 447 and note b) quotes one instance of the discovery in a round barrow of a socketed celt, which, notwithstanding the doubts expressed by Sir John Evans (Anc. Bronze Implements, p. 134), appears to have been contemporaneous with the interment; and another is mentioned in Anc. Bronze Implements, p. 114. On the other hand, Pitt-Rivers, referring to the spear-head mentioned in Archaeologia, xliii, 447, says, ‘I am informed by Mr. William Cunnington ... that ... it was found by his grandfather ... immediately under the turf near Stonehenge, and not in a barrow’. The reference given by Pitt-Rivers (Excavations in Cranborne Chase, iv, 20) to Archaeologia is incorrect.

[753] See J. Anderson, Scotland in Pagan Times,--the Bronze and Stone Ages, pp. 169-70. Canon Greenwell (Brit. Barrows, pp. 48-50) asserts that knife-daggers and flat axes--the only bronze implements found in Yorkshire barrows--never accompany swords and spear-heads in hoards, and argues that this proves the early date of round barrows in general. If the assertion were true, the fact would prove the early date of those Yorkshire barrows in which daggers and flat axes were found; but the question is whether many other barrows do not belong to later periods of the Bronze Age. Moreover, flat axes have been found twice in hoards,--one with palstaves (J. Evans, Anc. Bronze Implements, p. 464), another with swords (ib., p. 466).

[754] Archaeologia, xliii, 1871, p. 463.

[755] The Hon. John Abercromby (Journ. Anthr. Inst., xxxv, 1905, p. 262) affirms that ‘the few bronze swords, spear-heads, etc., of the Bronze Age and Hallstatt Period, that first filtered in driblets into this country, and were then reproduced with variations by native smiths, were too precious to be laid by for ever in a grave, even at the end of the Bronze Age in Britain’. Is not this begging the question? If small bronze weapons were ‘laid by for ever’ in graves in the earlier period, when bronze was scarce, why should not large ones have been laid by when it was common? And if gold ornaments were not too costly to be sacrificed, why should bronze swords have been deemed so precious?

[756] Anc. Bronze Implements, pp. 473-4.

[757] See p. 288, infra, and Addenda.

[758] See p. 181, n. 2, supra; Archaeologia, xliii, 1871, pp. 447-8; and Guide to the Ant. of the Bronze Age (Brit. Museum), p. 93. Sir John Evans (Anc. Bronze Implements, p. 310), referring to the tanged weapons of Arreton Down type and to certain blades found in the Wiltshire barrows, which (see pp. 145, 147, supra) he thinks ‘may have been the heads of spears rather than the blades of daggers’, remarks that ‘at the period to which they belong the art of making cores must have been known, as the ferrule found at Arreton Down will testify’. This is significant; but on p. 473 he refers the ‘tanged spear-heads or daggers’ to the second period of the Bronze Age,--earlier than that of palstaves and socketed celts.

Mr. C. H. Read (Guide to the Ant. of the Bronze Age, p. 24) observes that cinerary urns ‘possibly represent the period during which swords and spear-heads of bronze were manufactured ... by our population’. It is true that certain urns of overhanging rim type were contemporary with socketed weapons (Journ. Anthr. Inst., xxxv, 1905, pp. 262-4); but the oldest urns were much earlier. Cf. Proc. Soc. Ant., 2nd ser., xviii, 1899-1901, pp. 251-3.

[759] Archaeol. Rev., ii, 1889, p. 323. Cf. Folk-Lore, vi, 1895, pp. 15-7. Dr. Evans remarks in this article that ‘the characteristic form presented by a spiral ring of bronze found in one urn leads one indeed to believe that these flat disk-barrows of Standlake [in Oxfordshire] belong to a time when iron was coming into use’.

[760] Journ. Anthr. Inst., xxxv, 1905, pp. 256-64.

[761] Archaeologia, xliii, 1871, pp. 366, 369.

[762] See p. 149, supra.

[763] See pp. 205-6, infra.

[764] Flat bronze celts were found by Canon Greenwell in two only of the multitudes of barrows which he has explored not only in the northern counties but also in Wiltshire and Berkshire (Archaeologia, lii, 1890, p. 3); while Mr. J. R. Mortimer (Forty Years’ Researches, p. xlvi) never found one with any of the 893 interments which he examined on the Yorkshire Wolds. The canon opened four Late Celtic barrows in the parish of Cowlam, of which he says (Brit. Barrows, p. 212), ‘Had the bodies occurred without the necklace, fibula, or armlets, I should not have hesitated the least about classing these four barrows with the other barrows in the immediate vicinity, which were of the time of stone, or more probably of bronze.’ Is it unreasonable to conclude that a few other barrows which contain no relics of the late Bronze Age may nevertheless belong to that time?

[765] L’Anthr., xvii, 1906, pp. 321-42. Cf. W. Greenwell, _Brit. Barrows_, p. 44, n. 2, and Rev. de l’École d’anthr., xv, 1905, pp. 213, 215.

[766] W. Greenwell, Brit. Barrows, pp. 21, 333.

[767] Ib., p. 21, n. 1.

[768] Ib. See p. 185, n. 3, infra.

[769] W. Greenwell, Brit. Barrows, pp. 19-20; J. R. Mortimer, Forty Years’ Researches, p. xxxiv. Canon Greenwell remarks, however (pp. 28-30), that charcoal was almost always found in contact with unburnt bodies; and he was doubtful whether it was merely the ashes of the fire at which the funeral feast had been cooked, or might be regarded as a sign that the corpses had been passed through fire, just as in baptism aspersion was substituted for immersion. But this would of course imply that cremation on the Wolds was earlier than inhumation. Cf. J. R. Mortimer, Forty Years’ Researches, p. lxxvii.

[770] W. Greenwell, Brit. Barrows, pp. 21 n. l, 445; Archaeologia, xliii, 1871, p. 310; xliv, 1873, p. 426; lii, 1890, pp. 37-8, 43; Report of ... the Brit. Association, 1888 (1889), p. 315; Journ. Anthr. Inst., xxxii, 1902, p. 386. In regard to the few interments by inhumation that have been found in Cornwall see Vict. Hist. of ... Cornwall, i, 362-3, 366.

[771] Trans. Devon. Association, xxxiv, 1902, p. 119.

[772] Archaeol. Scotica, ii, 1822, pp. 76-102; iii, 1831, pp. 40-50; Nature, Jan. 13, 1898, p. 236.

[773] O. Montelius, Sur la chronologie de l’âge du bronze, 1885, p. 3.

[774] Mr. Abercromby in Proc. Soc. Ant. Scot., xxxviii, 1904, p. 364, denies that any drinking-cups were contemporary with cinerary urns; but in Journ. Anthr. Inst., xxxii, 1902, p. 385, he affirms that some were.

[775] Archaeologia, xliii, 1871, p. 390; A. Pitt-Rivers, Excavations in Cranborne Chase, ii, 6; Journ. Anthr. Inst., xxxii, 1902, pp. 375, 381; Proc. Soc. Ant. Scot., xxxviii, 1904, p. 364. Five more drinking-cups have lately been found with burnt bones in two cists in Dilston Park, Northumberland (Archaeol. Aeliana, 3rd ser., ii, 1906, pp. 142-6, 148).

Mr. John Ward (Vict. Hist. of ... Derby, i, 177-8) shows from an examination of the sepulchral pottery of Derbyshire (cf. pp. 191-6, infra) that in those districts in which interments of both kinds are found cremation was, generally speaking, later than inhumation. This conclusion is supported by the fact that in Wiltshire, where cremation on the whole greatly predominates, it occurs only about as often as inhumation in bowl and bell barrows (Archaeologia, xliii, 1871, p. 293).

[776] L’Anthr., xvii, 1906, p. 326.

[777] Proc. Soc. Ant. Scot., xxxviii, 1904, p. 78. Canon Greenwell (Ib., vi, 1867, p. 343, n. 2), speaking of a cairn near Crinan in Argyllshire, which he explored, remarks that ‘in this part of Scotland at all events the earliest interments in the large megalithic chambers are of burnt bodies’.

[778] Proc. Soc. Ant. Scot., vii, 1870, pp. 268-70; xx, 1886, p. 252; xxix, 1895, pp. 191-4; Archaeologia, xxx, 1844, p. 335; xliii, 1871, pp. 450-1; lii, 1890, pp. 25, 64; J. B. Davis and J. Thurnam, Crania Britannica, ii, pl. 1, p. 1; W. Greenwell, Brit. Barrows, pp. 20-1; J. Anderson, Scotland in Pagan Times,--the Bronze and Stone Ages, p. 90; A. Pitt-Rivers, Excavations in Cranborne Chase, ii, 2, 29; iii, 17; Journ. Brit. Archaeol. Association, N. S., vi, 1900, p. 10. Secondary interments by inhumation sometimes succeeded primary interments by cremation (Excavations in Cranborne Chase, iv, 173; Proc. Dorset Nat. Hist. and Ant. Field Club, xvi, 1895, p. 50).

[779] Proc. Soc. Ant. Scot., xxxvi, 1902, pp. 650-1.

[780] Archaeologia, xliii, 1871, pp. 312, 450-1; Journ. Brit. Archaeol. Association, N. S., vi, 1900, p. 10; Guide to the Ant. of the Bronze Age (Brit. Museum), p. 16. The two latter authorities only suggest that in cases like that mentioned in the text the unburnt body was that of the chief.

[781] Archaeologia, xv, 1806, pp. 340-1. The nine skeletons may have belonged to a secondary interment, but Cunnington inferred from the careless manner in which they had been buried that they were ‘slaves or dependents of the great personage below’. Mr. J. R. Mortimer (Forty Years’ Researches, p. xxxii), remarking that ‘in several instances where the body of the chief burial was reduced to ashes the attendants [?] were inhumed’, argues that in some cases cremation, in others inhumation, was considered the more honourable mode of sepulture. Perhaps.

[782] W. Greenwell, Brit. Barrows, pp. 20-1. Cf. J. R. Mortimer, Forty Years’ Researches, pp. xxxii, 60, 318.

[783] Proc. Soc. Ant., 2nd ser., xvi, 1895-7, p. 304.

[784] Archaeologia, xliii, 1871, pp. 313-4.

[785] Ib., lii, 58-9; J. R. Mortimer, Forty Years’ Researches, p. xxxviii. Cf. Guide to the Ant. of the Bronze Age (Brit. Museum), p. 90.

[786] Nature, Jan. 13, 1898, p. 237; J. Anderson, Scotland in Pagan Times,--the Bronze and Stone Ages, pp. 17, 74-5.

[787] Guide to the Ant. of the Bronze Age (Brit. Museum), p. 48.

[788] Archaeologia, xliii, 1871, p. 309; W. Greenwell, Brit. Barrows, p. 12.

[789] W. Greenwell, Brit. Barrows, pp. 22-3; Archaeologia, xliii, 1871, pp. 315-8. Cf. Journ. Brit. Archaeol. Association, N. S., vi, 1900, pp. 8-9.

[790] C. Warne, Celtic Tumuli of Dorset, 1866 (‘Tumuli opened at Various Periods’, pp. 10-1, 72, 76). The instance mentioned on p. 72 is, in my opinion, doubtful, and no certain pre-Roman instances are recorded in the sections entitled ‘Personal Researches’ and ‘Communications from Personal Friends’. Prof. Ridgeway (Early Age of Greece, i, 1901, p. 502), referring to Greenwell’s Brit. Barrows, p. 22, states that ‘in Dorsetshire ... the extended position seems to be the prevalent one’, a remark which illustrates the danger of relying on second-hand evidence.

In a few cases in Derbyshire and elsewhere in which the body has been found sitting the posture was perhaps due to some accident in filling up the grave (Archaeologia, xliii, 1871, pp. 318-20; Journ. Brit. Archaeol. Association, xxxviii, 1882, pp. 109-10). Two skeletons, however, were found sitting, back to back, in a barrow in Denbighshire (Crania Britannica, ii, pl. 23, p. 1).

[791] W. Greenwell, Brit. Barrows, p. 12. In a round barrow on the south of the road between Rochester and Gravesend, and about midway between Chalk Church and the Crown Inn, five skeletons were found in the trench near the bottom (Archaeol. Cant., xxiv, 1900, pp. 86-90).

[792] C. Warne, Celtic Tumuli of Dorset, pp. 46-9. Cf. pp. 36-7. In Derbyshire ‘secondary interments are found in any position, central or otherwise’ (Vict. Hist. of ... Derby, i, 176).

[793] W. Greenwell, Brit. Barrows, pp. 12-3.

[794] Archaeologia, xliii, 1871, pp. 321-3. Cunnington, however, remarks (ib. xv, 1806, p. 343) that ‘on the top of barrows we find the skeletons in every direction’.

[795] W. Greenwell, Brit. Barrows, pp. 25-6; Archaeologia, lii, 1890, pp. 25, 38, 64; Wilts Archaeol. and Nat. Hist. Mag., xxxiii, 1904, pp. 412-3; J. R. Mortimer, Forty Years’ Researches, p. xxxvii; Vict. Hist. of ... Derby, i, 173. All the kistvaens of Dartmoor lie at one end between north and west, at the other between south and east of the corrected compass (Trans. Devon. Association, xxxiii, 1901, pp. 121-2; xxxiv, 1902, p. 164); and the cairns near the Land’s End have ‘an aspect ranging from south-east to south-west’ (Archaeologia, xlix, 1885, p. 182). Cf. Rev. arch., 4^e sér., v, 1905, p. 307.

[796] Archaeologia, xxxiv, 1862, p. 255; xliii, 1871, pp. 314-5; W. Greenwell, Brit. Barrows, pp. 31-2; J. Anderson, Scotland in Pagan Times,--the Bronze and Stone Ages, pp. 74-5; Proc. Soc. Ant. Scot., xxxix, 1905, p. 552; J. R. Mortimer, Forty Years’ Researches, p. xli.

[797] Anc. Wilts, i, 124.

[798] J. R. Mortimer, Forty Years’ Researches, pp. xxxiii, 15-6, 63, 66, 77; A. Pitt-Rivers, Excavations in Cranborne Chase, iv, 66.

[799] Ib., ii, 1-2; E. T. Stevens, Flint Chips, 1870, p. 410; Archaeologia, xliii, 1871, pp. 324-5; W. Greenwell, Brit. Barrows, p. 14; Trans. Devon. Association, xxxiv, 1902, p. 108.

[800] Archaeologia, xliii, 1871, pp. 294, 310, 325-6.

[801] Ib., p. 326. Cf. W. Greenwell, Brit. Barrows, p. 445; Nature, Jan. 13, 1898, p. 236; Journ. Brit. Archaeol. Association, N. S., vi, 1900, pp. 8-9; and Trans. Devon. Association, xxxiv, 1902, pp. 108-9.

[802] Trans. Cumberland and Westmorland Ant. and Archaeol. Soc., N. S., i, 1901, pp. 295-9. Cf. Proc. Soc. Ant. Scot., xxxix, 1905, p. 547.

[803] Ib., xxxvi, 1902, p. 644.

[804] Archaeologia, xliii, 1871, p. 326. On the other hand the urns which Canon Greenwell found in Yorkshire were usually placed upright (Brit. Barrows, p. 14).

[805] Guide to the Ant. of the Bronze Age (Brit. Museum), p. 49.

[806] Iliad, xxiii, 254. Cf. xxiv, 795-6.

[807] Archaeologia, xliii, 1871, p. 326.

[808] J. Anderson, Scotland in Pagan Times,--the Bronze and Stone Ages, pp. 43-7. An urn only ⅞ inch high, which of course could not have been used for containing ashes, has been found in a cairn of the Bronze Age in Fifeshire (Proc. Soc. Ant. Scot., xxxvi, 1902, p. 641).

[809] J. Anderson, op. cit., pp. 51-2, 74-5; W. Greenwell, Brit. Barrows, pp. 62, 72-3, 139, 277, 291, 297.

[810] A. Pitt-Rivers, Excavations in Cranborne Chase, ii, 29. Canon Greenwell (Archaeologia, lii, 1890, pp. 63-4) has described in an interesting paragraph ‘the infinite variety, within certain limits, which is found in connection with the burials of the Bronze Age’.

[811] Archaeologia, xliii, 1871, p. 331; Brit. Barrows, p. 74; J. R. Mortimer, Forty Years’ Researches, pp. lv, lix.

[812] See pp. 442-3, infra.

[813] Sir A. Mitchell, The Past in the Present, 1880, p. 28; Archaeologia, xliii, 1871, p. 334.

[814] See p. 159, n. 1, supra.

[815] W. Greenwell, Brit. Barrows, p. 63. See, however, Archaeologia, xlix, 1885, p. 184.

[816] Archaeologia, xliii, 1871, pp. 391, 396; Journ. Anthr. Inst., xxxii, 1902, pp. 373-97; Proc. Soc. Ant. Scot., xxxviii, 1904, pp. 323-410; xxxix, 1905, pp. 326-44; J. R. Mortimer, Forty Years’ Researches, pp. lxv-lxvii.