Colchester Castle a Roman Building, and the oldest and noblest monument of the Romans in Britain. With illustrative plans and sketches

Front matter

No. I. THE SOUTH FRONT OF COLCHESTER CASTLE. A A Supposed original Ground Line. B B Depth at which foundations have been discovered. The Dome is modem, and the ivy as it appeared in 1855. COLCHESTER CASTLE i ROMAN BUILDING, AND THE OLDEST AND THE NOBLEST MONUMENT OF THE ROMANS IN BRITAIN. With Illustrative Plans and Sketches. I'.Y GEORGE BUCKLER, AUTHOR OF "TWENTY-TWO OF THE CHURCHES OF ESSEX." COLCHESTER : BENHAM AND HARRISON. LONDON: SIMPKIN, MARSHALL AND CO stationers' hall court. M.D.CCC.LXXVI. THIS WORK IS DEDICATED TO THE MEMORY OF THE LATF. REVEREND HENRY JENKINS, Who during a long life devoted much time, and untiring labour, to the Study and Investigation of Eoman Antiquities in this Country. A deep research into the History of Colchester and its Neighbourhood, and repeated explorations in and around the Castle, led him to decide that it could be no other than the Templed Citadel which the Eoman Colonists built at Colonia Camulodunum to their Deified Emperor Claudius.

He also ascer tained, correctly, that it is much larger than the Keep of any Castle, and concluded that its History, as a Eoman Fortress and Temple, is confirmed by its Architectural Details.

Preface

. The late Rev. Henry Jenkins was Rector of Stanway, near Colchester, from 1880 to 1874, and as President of the Colchester Archæological Society he delivered an able and suggestive Lecture before that Society, in 1850, on the Roman origin of Colchester Castle. The Lecture was published in 1853, upon which a controversy sprung up with respect to the origin and consequent age of the building, whether it is Roman or Norman. In 1861 Mr. Jenkins wrote the History of the Castle, and in 1869 he issued a revised Edition. His writings prove his great learning and his knowledge of the Celtic language, especially its British dialects. The difference of a thousand years in the supposed age of a building within easy reach, led me, as an Antiquary, to study its construc tion, and spare time, during nine years, enabled me to make the requisite sketches, and to examine the most important Roman buildings now extant, and those commonly considered to have been built by the Normans, but I had no intention of publishing the result of my enquiry.

Misconceptions concerning Colchester Castle, and Roman and Norman buildings generally, crept into the controversy of 1853. These would have lain dormant, had not some of them been again authoritatively published in The Essex Standard (Aug. 12, 1864), in favour of the Norman theory. The question being thus resuscitated, 1 was induced to contribute to the leading Newspapers in Colchester certain letters, the result of my enquiry, showing the similarity of Colchester Castle to Roman buildings, and its dissimilarity to buildings commonly called Norman Keeps. The present pamphlet is in three parts. Paei I. — The substance of the Letters describing the Architecture of the Castle, and showing its dissimilarity to Norman Keeps, written from Notes and Sketches taken between 1855 and 1864. Part II. — Portions of the Rev. Henry Jenkins's History of the Castle, bearing most strongly on the vexed question of its Roman construction, printed from the Author's Manuscripts of 1872.

(1.) The indentity of Camulodunum with Colchester. (2.) The Architecture of the Castle. (3.) The C.istle, the Temple of Claudius. Part HI. — A note on the White Tower of London G. B. 107, Camberwell New Road, London, S.E., July, 1876. GREENSTEAD No. 2. PLAN OF COLCHESTER. SHOWING THE CASTLE ; THE REMAINS OF THE TUTELA ; THE TWO ROMAN SEWERS WHICH ISSUED FROM THE CASTLE ; THE TOWN WALL ; AND THE SITUATION OF THE TEMENOS OR GLEBE OF THE TEMPLE, WITHIN AND WITHOUT THE TOWN WALL. A Lias Roman Pavement. B B Roman foundations of the Tutela. C Mouth of the Sewer beyond the Town Wall. D Town Wall built over Sewer. The part not etched, had neither streets nor houses upon it down to the time of the Town Map, by Gilbert. (1847.) Part I. Site. The site on which Colchester Castle is built, is not such as would have been chosen by tbe Nor mans for a fortified place if rebance is to be placed upon tbe collective opinions of Engbsb and French archæologists during the last half century; for on those authorities tbe Normans would have selected tbe highest ground, or a site close to tbe river;* and they would have raised a mound for tbe keep.

Construction. Tbe constructive features which at tracted tbe attention of tbe Eev. Henry Jenkins prove to be characteristics of Eoman Architecture, and a careful study of tbe simple but ingenious methods of building practised by tbe skilled and scientific Eoman masons will show tbe superiority of their work over that of tbe Norman masons. Tbe following are leading features of Eoman work manship in Colchester Castle : — Tbe sobd Platform, or Basement, raised considerably higher than tbe adjacent ground in order that tbe Temple might be exalted high above all surrounding objects. Tbe wide and, in a military sense, unprotected Entrance. Tbe close prornixity of the Well to tbe main entrance. Tbe plain and simple manner in which the Adytum and tbe Podium are vaulted, tbe hemispherical vaulting of tbe niches and the vaulting generally. * Durham Castle furnishes a good example of a site for a Norman Castle.

(Jour. Brit. Arch. Assoc. vi., 211.) £ The method of setting-back tbe Vaults from tbe face of tbe walls'" and tbe manner in which tbe arches, cross vaults, and end walls, are inserted under tbe principal vaults. Tbe care bestowed in working-out intersections in tbe vaults, especially those over the entrances of tbe staircases and tbe vaulted chambers. Tbe use of tiles for bridging over putlog boles : tbe observance of this method is conspicuous in Eoman buildings where quoins and bands of tiles form a dominant feature. The bridging over with tiles, tbe flues to receive tbe bars of tbe gateway and tbe entrance to tbe staircase, when drawn back. Tbe small upright panels for inscription letters on tbe South and East fronts. Tbe quoins of stone for one-third of tbe height and of tiles for tbe remaining two-thirds.

Tbe prototypes of tbe foregoing constructive features are to be found in Eoman Temples, but not amongst Norman Keeps. Vaults and vaulted chambers occur in Norman buildings ; but for workmanship and size, they are as unworthy of comparison with those in this Castle as also are the Norman scraps of herringbone brickwork, with tbe twelve hundred and sixty feet of such walling at Colchester. Magnitude. Some misconception may yet prevail as to the size of Colchester Castle in comparison with other buildings which have been used as, and are commonly called Keeps. Of tbe following Keeps, namely, Colchester, White Castle Monmoutb, White Tower London, Norwich, Canterbury, Dover, Eoches ter, Castle Eising, Bamborough, Portcbester, Corse, Bowes, Newcastle, Hedingham, Odiham, Guildford, Carisbrooke, Lincoln, Orford, Penline, Newark, Eicb mond, Conisborough, Peake, Leeds and Goodricb, Colchester is by far tbe largest.

It measures one * This method of setting-back the vaults is conspicuous in the ancient buildings of Eome, and was noticed, in 1822, by Bayley in the White Tower. (Pp. 112, 113.) hundred and fifty-two feet six inches, by one hundred and eleven feet six, exclusive of all projections. As a near approximation to these dimensions White Castle, Monmoutb, was erroneously described in 1853 as a Keep, one hundred and forty-five feet by one hundred and six feet, but it is only an oval area of open ground, enclosed by a wall, around which are six small circular towers, no one of which is distinguishable as a Keep. The next in size to Colchester is the White Tower of London, but it is tbii-ty-six feet shorter and fourteen feet narrower. The Dext is Norwich, fifteen feet shorter and five feet narrower than the White Tower, and therefore Norwich, tbe third in size, is fifty-one feet shorter and nineteen feet narrower than Colchester.

The remainder of the Keeps diminish in size in the order named, tbe smallest being Goodrich, thirty feet square. Altitude. Colchester Castle is the lowest, in pro portion to its size, of all the buildings called Keeps, and as its beigbt is not one-thnd of its square, it cannot with any advantage be equally compared with Keeps which are said to have been built from one to two squares in height. The Eoman system of measuring bundings by squares is not really appbcable to Norman Keeps, nor of any value in comparing those Keeps with the Castle at Colchester. Materials. The materials composing this enormous structure are cement stone, tiles, a hard stone, and mortar. Cement Stone. The cement stone was brought from Harwich, and the thickness of tbe walls is built of it, in tbe ancient manner known as opus incertum, i.e., stones in their natural forms fitted closely together.

In broken places of tbe walls, not weathered by ex posure, tbe stones are unimpaired. Tbe facing stones, called septaria, are cement stones squared and laid in horizontal courses, and a large proportion of them remain entire. Hard Stone. Tbe bard stone is tbe shelly oolite b limestone of Portland, which tbe Eomans obtained in large quantities, and used extensively ; moreover they could readily transport it from Dorsetshire to Essex for the Claudian Temple at Colonia Camulodunum. The oobte limestone of Barnack is of the same geologic formation as that of Portland, and the hardness and specific gravity of tbe two are about equal. That the Early Normans brought stone from Barnack to build a Castle in Essex within the conjectural period of from 1087 to 1120* is improbable, it being doubtful whether the Eomans, during tbe period named, bad sufficient command over tbe intervening country to enable them to convey large quantities of stone from Barnack to Colchester.

The mere suggestion is contrary to tbe general conclusions arrived at by Archæologists, namely, that the Normans resorted to tbe nearest Eoman Station or Villa for building materials when tbe locabty failed to furnish them. Mortar. It bas been said that the mortar is not Eoman ; but tbe very formation of the structure being Eoman, the mortar which binds tbe materials together must necessarily be Eoman. Judging from tbe com position of the mortar, irrespective of the construction of the building, it is Eoman. Tbe composition of the two kinds having long since been decided to be totally different. Here is an independent authority upon the subject. Eoman mortar is composed of lime, pounded tiles, sand and gravel, more or less coarse, and even small pebble stones : these ingredients vary in thenquantities, but usuaUy tbe lime and pounded tiles predominate.

Norman mortar is totally different from tbe Eoman, it contains no pounded tiles and tbe sand is generally in excess, f The mortar with which Colchester Castle was built contains Lime, Pounded Tiles, Sand, Gravel, Pebble Stones, and Water, with no excess of Sand. Tbe * Jour. Brit. Arch. Assoc. xxi., 283. t Ibid ii., 35. mortar used in repairs entered in tbe Great Eoll of the Pipe between A.D. 1130 and 1205,* and in those effected since Wheeley's time, may contain an excess of sand, but if so, tbe fact does not affect the question. Twenty-two of tbe twenty-six Keeps which have been specially examined, were referred to in tbe con troversy as Norman buildings, t Norman Keeps are said to be formed after One model.]; Now inasmuch as of those twenty-two Keeps, ten are oblong ; eight, square ; one a circle ; one an oval ; one a polygon ; seven have one divisional wall, and three, two divi sional walls, it is not easy to comprehend what is meant by one model, but however it may be used in reference to Norman Keeps it is not appbcable to Colchester Castle.

In deabng with the internal area of Colchester, reference has been made to the Keep of Castle Eising on account of its division into three areas. § But it is dissimilar in all respects to Colchester Castle ; and in size, so small (sixty-nine feet by fiftyeight feet) that the West court of Colchester is larger than it and the Keep of Goodrich added together, and consequently the Norman theory with respect to Colchester Castle is not strengthened by comparing it with Castle Eising. Plan. Colchester Castle is an oblong, now divided longitudinally into two parts familiarly known as the West and East courts (No. 11) : the former measures ninety feet by fifty-nine feet six inches, and the latter eighty-eight feet by nineteen feet six. Tbe outer walls are twelve feet four inches in thickness and the divi sional wall seven feet nine inches.

There were two longitudinal walls, but one of them was pulled down in 1683. Tbe trustworthy authorities on this disputed question are confirmed by tangible evidences in the * Jour. Brit. Arch. Assoc. xxi., 280, 281. t On the improbability of all the buildings commonly called Norman having been erected by the Normans, or during the Normau period. See ArcU. Jour., iii., 290, and Jour. Brit. Arch. Assoc, x., 142. Also note *, Page 17. J Arch. Jour, i., 94. § Jour. Brit. Arch. Assoc. xxi., 282 (note). B building. The removal of one of tbe walls led Mr. Jenkins to discover that the wall thus removed bad been an addition, and that tbe West court was, in the first instance, Hypethral, an open space* and the following facts sustain bis reasoning on this point. (1.) The completeness of tbe North-West tower in all its arrangements; its isolation from the rest of the building ; and tbe loss of one of its cameræ privatæ.

(2.) The exclusiveness of the staircase tower for the Southern portion of the Castle. (3.) Inscription panels on the South and East fronts only. (4.) Openings which have been blocked up in tbe North-East and North-West towers. Towers. The five huge Towers at tbe angles of Colchester Castle are unlike anything to be found amongst Normans Keeps. One on tbe West side, forty feet by twenty-nine feet. One at the West end of the South front, eleven feet one inch, by seven feet ten inches. Another at the East end of tbe South front. twenty feet nine inches, by seven feet seven. One at tbe North-west angle, twenty-six feet nine inches, by twenty-four feet nine, and another at the North-east angle, twenty-five feet by twenty-two feet nine inches. Besides the five towers there are buttresses on the North, West and East sides: the largest being eight feet ten inches by eighteen inches, the smallest eight feet one inch by thirteen inches.

None of the Castle Keeps, and not even tbe White Tower have such towers as the before-mentioned, and their position and magnitude render Colchester Castle as remarkable for its towers as it is for its vaults, and in both respects it is unlike Norman Keeps. Tbe largest tower, that on the West side, forty feet by twenty-nine feet, is larger than the Keeps of either of the following Castles, viz. : — Peake, Leeds, and * Jour. Brit. Arch. Assoc. xxi., 282. t Norman Keeps are strengthened along the sides by the usual flat Roman buttress rising from a general plinth and dying into the wall below its summit. The end pilasters of each face unite at and cap the angle and rise a story above the walls to form four angular turrets. (Arch. Jour, i., 95.) Goodrich. It contains the only staircase which led from the ground floor, and the position and size of this staircase are not to be matched in any Castle Keep.

The doorway leading to the stairs is just within the entrance gateway, tbe plan of which does not exhibit the skilful means of protection, the design of which is erroneously attributed by many writers on Castle Keeps to have originated with tbe Normans when constructing their castles in the Eleventh century.* The stairs, seven feet in width, were repaired in the last century, and the inclined, plane at the bottom was made in preference to renewing the lower most steps. The spiral vault is of tiles, laid edgewise in a bed of concrete, which bears the impress of the planks on which it was laid. At ten feet from the bottom tbe fine wrought ashlar ceases, and above this for some distance the walls are of septaria, with bands of tiles six inches apart. Several of the original steps, eight inches in height and built of three courses of tiles, remain tolerably perfect, they are in every respect like tbe steps in tbe galleries of the Cobseum at Eome.

The tiles are sixteen inches by eleven, by two inches and threequarters, and as sound and durable as ever, and highly pobshed by long use. Above the first floor, the walls are not regularly ashlared, but there are bands of tiles in tbe lower portion. Tbe old work ceases at tbe following heights ; the old but not original steps at six feet above the first floor ; the vault at eighteen feet ; and the newel at twenty-two feet ; above these points the staircase was pulled down in 1683, and re-built in tbe last century. Besides the staircase, this tower contains two small chambers, one on eacb floor. One of them on the ground floor is but a Eecess eleven feet by seven, by nine feet two inches in height, (c. No. 18) ; it is vaulted, and tbe semi-circular end made hemispherical; it does not appear to have been shut in with a wall or * The Keep at Penline Castle is the counterpart of a military tower at Argos, I Donaldson's Antiq., Athens, 1830, iv., 23.)

ColcJiester Castle. 12 door to entitle it to tbe name of a Room. Tbe chamber on tbe first floor bas, however, a doorway of Eoman construction (No. 3) ; tbe room is lofty, and measures fifteen feet six inches by ten feet four inches, and sixteen feet in height ; it is spanned by a semi-circular vault of tiles ; there are the remains of two original windows — one is walled up, the other made into a fireplace ; two modern windows have been made by breaking out the wall at the back of two c ameræ privatæ, the vents or descending flues from which are perfect on the outside* The two towers on the South front were originally compact masses of building throughout their height, notwith standing that there is now on the top of one, a cavity capable of hold ing sufficient earth and moisture to si;stain a good-sized tree ; and on the other, slight walls around the outer edges to support the modern No. 3.

DOORWAY INTO VAULTED chamber, now the record room. roof. These towers were apparently designed to act as abutments, and their position shows that they were calculated to resist lateral pressure where it might otherwise have been deficient : tbe Western tower is at the angle of the square of the building, and provides against tbe thrust of a dome over the staircase : the Eastern tower is at the other angle of the square of tbe building, similarly placed This chamber is now the Record Room. and for apparently a similar purpose. Thus far tbe Eev. Henry Jenkins's supposition that there was a dome over the apse as well as over tbe staircase is countenanced, and perhaps no better reason can be assigned for there being towers at these points. A dome over the circle of the apse would be about twice the diameter of a dome over the stafrcase, and the size of the tower to receive tbe thrust is in the same pro portion,* this principle is correctly carried out.

Other bmldings which have been used as Keeps do not analagously furnish so satisfactory a reason for there being towers at these two points, and nothing approximating to them is to be found in Norman Keeps. Tbe towers at the North-West and the North-East angles are alike in one respect, inasmuch as they are both sobd masses of building from tbe basement to the first floor. Their equals in size and solidity are to be met with in Eoman Castles. The North-West tower is a remarkable feature. Mr. Jenkins shrewdly guessed that it was the Watch Towers of the Temple, and the military character be assigned to it is borne out by tbe peculiarity of its arrangements. At tbe first-floor level tbe North face sets back seven feet six inches and forms an outside landing (a) to the Postern (b), the remainder of the landing is on tbe inside (c) ; tbe waUs, the vaults and the paits adjacent are built with great care ; the walls are of stone, and the vaults and groins like those of the entrance and principal exit be- NORTH.

No. 4. PLAN OF N.W. TOWER. * Mr. Jenkins was indebted to the Rev. J. S. Dolby for the suggestion that there was a dome over the apse. Domes are in accordance with Roman architecture. (Collectanœ Antique, v., 236.) f Watch Towers. See Lower's Pevensey, 1853, p. 7. CoUectanœ Antiquæ, v., 236. Donaldson's Antiq. Athens, iv., 23. Josephus War, v. 4, 1. longing to tbe grand staircase. This tower, above tbe first-floor level, is almost wholly occupied by a staircase twelve feet in diameter ; tbe steps, five feet in width, are modernized ; the walls are of stone, in which a harder stone is used as bonding courses ; tbe newel is not so finished a piece of masonry as that of the great staff-case, but tbe spiral vault and tbe windows of both staircases are similar. At tbe foot of the stairs, and partly under them, is a camera privata, (d), the entrance to which is by an arch of Eoman character, built under tbe vault.

The form of the camera privata is complete, and the vent or descending flue from it remains out side in the West wall, (e). Near this vent is another, (f), but there is no access to the second camera privata : a close inspection of tbe wall shows that an opening off the staircase was roughly filled up at a remote period, (g). This alteration, and the loss of tbe camera privata probably ensued upon the West wall being raised as suggested by Mr. Jenkins. This tower was always higher than that at the North-East angle, tbe spiral vault of the staircase parapet. being higher than the The North -East tower contains a chamber on tbe first-floor, thirteen feet two inches by ten feet, and six teen feet in height. On the North side an original win dow is filled up : on tbe East side, two windows have been thrown into one : and, on tbe West side, an opening (r) bas been filled-in.

Tbe entrance is by a vaulted lobby (s) four feet ten inches by three feet. NORTH No. 5. PLAN OF N.E. TOWER In the five towers are two staff-cases, two chambers, one recess, and tbe remains of four cameræ privatæ. It was stated in 1864 that the Norman Keeps of Hedingham and Orford have similar chambers in their towers,* but Colchester Castle is unique in respect of its towers and the accommodation provided in them. The Appendages to the Keeps of Norwich, Dover, Eochester, Castle Eising, f Newcastle, and Hedingham are sometimes called Towers, but they, in no respect resemble tbe towers of Colchester in shape, position, or utility. Well. Tbe situation of tbe Well is remarkable from its being on tbe ground floor, just witffin tbe entrance gateway. Such an exposed position for the supply of water to the Keep of a Castle has no parallel amongst Norman Keeps.

Archæologists have supposed that the Normans placed their wells for security, in the substance of the wall, and led a pipe to the upper storeys of the Keep in order that the garrison should not be without water when they had lost possession of the lower storeys. Tbe want of this precaution mib tates against the theory of Colchester Castle being a Norman Keep, but it strengthens Mr. Jenkins's theory of it being a Eoman Temple, because be says the Well is conveniently situate for the use of tbe Priets, and for the ablution of the Temple. Great Gateway. The mouldings of the Gateway led King, (1776),+ and Britton, (1806), § to say that it was not so ancient as the rest of the building, and Mr. Jenkins says tbe only Norman moulding throughout the Castle, is that over the main portal. Upon carefully examining tffe gateway it is evident that tbe facing of it only bas been partially altered. * Essex Gazette, 16th Sept. t Of eleven hundred and fifteen Norman Castles, said to have been built in Stephen's reign, not enough now remains to show what a Norman Castle really was.

(Harrod's Gleanings (Castle Rising), 1867, p. 44.) Hedingham is a Tower, a Keep, not a Castle. (Vetusta Mon. Majendie, iii., 2.) { Archœologia, iv., 406. § Arch. Antiq. i., p. 2. First. A third only of the stone used in making the alteration is Caen stone, the rest being hard stone taken from the fabric. Second. The - line of the masonry bounding the alteration (No. 1) shows that it was built subsequentlyto the septaria and tile walls. Third. The No. 6. MOULDINGS GREAT GATEWAY. Norman mouldings are made to fit an older arcff. Fourth. The ends of the mouldings are let into the side wall of tbe tower (a), and tbe chisel marks on the tower show that the masonry was faultily cut (because it was in situ) for tbe mouldings. Fifth. The mortar joints (b c d) under the capitals and abacus are imperfectly made and clumsily wedged up.

Sixth. Tbe septaria and tiles of the South front are rudely broken off. It is tbe only part of the Castle which has been made ornamental, and it is remarkable that after the wear and tear of many centuries, the difference between the fabric and tbe alteration should be distinctly visible. The width of the gateway is seven feet seven inches in the clear, and nine feet fourinches in the rebate. The rebate for tbe massive wooden gates is halfway through the wall : in front of these gates was the portcullis, tbe grooves for which remain. The whole gateway is of hard stone, with the exception of the semicircular arch inside, which is of tiles. In the East or right band jamb is a Niche two feet three incffes by one foot seven, with a semicircular arch of tiles alternated with wedge-shaped blocks of stone (No. 19) like tbe margins round the second reveal of the first-floor windows.

It has been mistaken for a Porter's niche, but, as it is only five feet seven inches high, and the sill one foot eight inches off the floor, it could not have been made or used for such a purpose. The suggestion of Mr. Jenkins that it held the water of lustration in Heathen times, and the consecrated water in Christian times is very probable. It accounts for the Ecclesiastical character of the Sculptures, which without doubt were tbe work of the Anglo-Saxons,* when the building was used as a Church ; and the monumental inscription seems to be a reminiscence of tbe fact. It is very improbable that Ecclesiastics, or Soldiers, or Prisoners,! carved these figures during the time the Normans used tbe building as a Castle. By the side of the niche or stoup is a flue (a No. 19) eleven inches by six, built with care to receive the bar of tbe outside gates.

Tbe flue is seven feet long and bridged over with tiles, the sides and bottom being of bard stone. The width of the gateway, the simplicity of its arrangements, and tbe broad flight of steps which led to it, show that it is not of Norman design. Archæo logists point to Hedingham Castle as an example of what an entrance to a Norman Keep is like, but it is dissimilar in all respects to that of Colchester. The entrance to Hedingham Castle is at a greater height from the ground, it was covered by an Appendage, or side tower ; tbe steps are parallel with the wall of tbe Keep, they were narrow and confined between two * On the Anglo-Saxons a3 Churchmen, see Hook's Lives of the Archbishops. Vol. i., c. v. to viii., and as Workmen, see Arch. Jour., i. 27, h., 1863, vii., 152 x. 280. t Norwich Castie was a prison in the reign of Henry 1st. (Arclwologia xii., 152.)

Norman Castles were preceded by Saxon ones and were used as prisons at a period so immediately after the Conquest as to be scarcely compatible with the erection of Norman strongholds. (Jour. Brit. Arch. Assoc, xiv., 204.) Also Note *, Page 9. walls, and tbe wall of the Keep was provided with a loophole for an obbque attack upon tbe enemy. At tbe top of tbe steps were openings which enfiladed the approach from top to bottom, and after having over come all these difficulties the besiegers bad then to turn a corner before they could attack the main door of entrance. In diametrical opposition to all these cautions at Hedingham, the Eoman entrance to Colchester Castle was by a flight of steps upwards of ten feet in width and straight out from the doorway : the steps were not covered in any way, or guarded by loopholes, but tbe approach was made easy and com modious for worshippers and not to guard against the attacks of assailants.

In 1850 Mr. Jenkins traced the foundations of tbe steps, and at tbe bottom was a Eoman gutter, fourteen inches in width, formed of Kentish rag and coated with tarras.* Upon opening the ground again in 1857 some of tbe steps were found. They are of three courses of extremely bard tiles like those in the grand staircase and in the Coliseum. The steps are fifteen inches wide and eight inches high, and the ends of the two topmost ones (a b) are imbedded in the jamb of the doorway, proving that they were built with the jamb : these and portions of other steps below them prove the correctness of view No. 14. No. 7. STEPS. ENTRANCE GATEWAY Vaults. The Vaulting of Colchester Castle is more extensive than in any Castle Keep. Basement. Tbe vaults beneath the ground floor, i.e., in tbe centre of tbe Basement (No. 11), were part of a dense mass of buffding prepared to receive * Tarras mortar was extensively used by the Romans, it was made of stone from the neighbourhood of Liege, beaten fine and mixed with quickhme and a small quantity of water. tbe superstructure, and to raise it high above all surrounding objects.