Natural History of the Ground Squirrels of California

Front matter

REPRINTED FROM Volume VII, Numbers 11–12, Pages 597–708, OF THE Monthly Bulletin of the State Commission of Horticulture | NATURAL HISTORY | OF THE | GROUND SQUIRRELS | OF CALIFORNIA | By JOSEPH GRINNELL and JOSEPH DIXON | Contribution from the University of California Museum of Vertebrate Zoology

CONTENTS.

Page

Illustrations

ILLUSTRATIONS.

Plates.

Page

Figures in text.

Page

Note.—Figures indicating page numbers refer to folios at foot of pages

Introduction

NATURAL HISTORY OF THE GROUND SQUIRRELS OF CALIFORNIA.

By JOSEPH GRINNELL and JOSEPH DIXON.

Contribution from the Museum of Vertebrate Zoology, of the University of California.

INTRODUCTION.

Human occupancy of a new country always tends to upset the primitive balance of things. Man either purposefully or incidentally begins at once to modify the original complement of animal and plant life both through destruction of native species and by bringing in with him alien kinds. Some native species become more and more restricted in range, even disappearing altogether; others tend to increase and spread, finding conditions for their existence to be improved through man's activities.

In the case of the ground squirrels of California, we have a group of mammals which seems to have in many places benefited by human invasion. This is probably due to the destruction by man of the many predatory animals, such as hawks, eagles, coyotes and badgers, which under original conditions kept the small herbivorous mammals in check, and in part to the improved food supply made available to the ground squirrels through his cultivation of crops. Because of the destructiveness of these rodents to the planted crops and native forage upon which man is dependent to a large extent directly or indirectly for his own food supply, the problem of ground squirrel control has become one of very immediate agricultural and pastoral importance.

It would seem that knowledge, as full as possible, of the ground squirrels of California is necessary to determining the most successful means of controlling them and to applying these means properly to the varying conditions throughout our state. This knowledge should include the main distinctions by which each may be known from its relatives, the distribution of each of the species, the extent of the burrows, the breeding rate, the food habits, and, indeed, every other class of facts obtainable relative to their natural history. It is not often apparent, in advance, which facts will and which will not prove of critical importance in economic work.

To illustrate the value of a thorough knowledge of the food habits of the animal in question, when the most efficient method of controlling destructive rodents is sought, we need only to point to the present method used in poisoning the California Ground Squirrel by the use of barley coated with strychnine, rather than barley soaked in a strychnine solution. By applying a knowledge of the food habits of this animal it was possible greatly to increase the effectiveness of poisoned grain because of the discovery by Stanley E. Piper, of the United States Biological Survey, that this squirrel is more readily poisoned through the membranous walls of its cheek-pouches when merely carrying the poisoned grain than through the stomach after the poisoned grain has been eaten. Strychnine-coated barley has not, however, been found so successful when applied to the Oregon Ground Squirrel, which animal apparently does not habitually gather and store seeds and grain to a large extent, but is active chiefly during the spring and early summer when green vegetation, upon which it depends for food, is to be had. Some sort of green baits might be expected to be more effective with this species.

It is very desirable at this time that we know more about the extent of æstivation and hibernation among our ground squirrels. We know little or nothing of the effect of gases upon squirrels in this condition; and it is obvious that such part of a squirrel population as is dormant at the time of a poison campaign will escape destruction and become a nucleus of reinfestation.

During the past spring and summer our work of excavating squirrel burrows after the occupants had been gassed, has shown that one frequent cause of failure in the ordinary waste-ball method of applying carbon bisulphid is due to the fact that sudden elevation in the course of the burrow, of as much as two feet in some cases, prevents the heavy gas from reaching the animal. This condition was found to occur much more frequently than is generally supposed.

The instances just cited all suggest that the present knowledge of our ground squirrels is far from complete, and they serve to emphasize the above contention that a more thorough knowledge of this subject is essential to securing the most intelligent and efficient methods of control of these our chief rodent pests. It is the purpose of the present paper to supply the information available from all sources in as much detail as it has proven feasible for the authors to secure it at this time. The facts and inferences are given just as they came, irrespective of whether or not they show obvious and immediate economic bearing. They are here available to everyone who is concerned with methods of control.

Acknowledgments

ACKNOWLEDGMENTS.

The present paper has been prepared at the suggestion of Mr. G. H. Hecke, California State Commissioner of Horticulture, and Mr. W. C. Jacobsen, Superintendent of Rodent Control under this commission. Both these men have rendered many valuable helps during the progress of our work and have co-operated to facilitate its final publication. Their emphasis all along has been upon the need of a summary of the facts relative to rodent natural history at this particular time, when efforts from every direction are being concentrated upon the problem of increased crop production.

Upon the facilities and auspices of the Museum of Vertebrate Zoology of the University of California the writers have been dependent for the opportunity of carrying through their undertaking; and behind this has been the continual financial and moral support of the founder and constant patron of the Museum, Miss Annie M. Alexander. The collections of specimens and the field records contained in this Museum, gathered during the past ten years, have been absolutely indispensable to the present accomplisliment. A total of 1263 study skins of ground squirrels taken within this state have been examined in the course of our work.

In the spring and summer of 1911, two field assistants from the United States Public Health and Marine-Hospital Service were assigned by Dr. Rupert Blue, then Surgeon in Command, San Francisco, to accompany the Museum party working in south-central California, for the purpose of increasing the common knowledge of the distribution and manner of occurrence of the rodents of the region. Acknowledgment is here made for the use of certain facts from the manuscript report of that year's work made to Dr. Blue by the Director of the Museum.

During the past year, Dr. W. C. Billings, Surgeon, in temporary charge, United States Public Health Service, San Francisco, has very greatly aided us in having his field men send us specimens of ground squirrels, both alive and dead, for experimentation and study.

To Professor Harvey M. Hall, of the University of California, we are indebted for identifying numerous seeds obtained from the cheek-pouches of ground squirrels.

Through special provision of the State Commissioner of Horticulture we have been fortunate in being able to have the accompanying five colored plates of ground squirrels drawn by the eminent animal artist, Mr. Louis Agassiz Fuertes. We thank Mr. Fuertes for the special pains he has taken in executing these drawings.

Color terms used in our descriptions are taken from Ridgway's Color Standards and Color Nomenclature (1912).

Information used by us from published sources is credited through the system of author, year and page references to the list of "Literature Cited" which appears at the end of this paper. The authority for important information obtained from field notebooks or from letters is given in parenthesis together with the abbreviation for the word manuscript—"MS."

September 13, 1918.

Key to the Ground Squirrels of California

KEY TO THE GROUND SQUIRRELS OF CALIFORNIA.

1. Size large: body alone more than 9 inches (228 mm.) long; tail more than 6 inches (152 mm.), not counting hairs; ears tall and rather pointed; tail bushy.

2'. No blackish patch on back.

3'. Tail shorter: without hairs, 7½ inches (190 mm.) or less; a more or less sharply defined triangular whitish patch on each shoulder.

4'. Size smaller: body alone less than 10¾ inches (273 mm.); general tone of coloration lighter.

1'. Size medium or small: body alone less than 9 inches (228 mm.); tail less than 6 inches (152 mm.), not bushy; ears low and rounded, or else reduced to mere rims.

6. Body with conspicuous lengthwise stripes; coloration varied.

7. One white stripe on each side of body; under side of tail conspicuously white; head not more or less yellowish or coppery.

8'. Larger: body alone 6¼ inches (158 mm.) or more; general coloration light clay color.

7′. Three stripes on each side of body, one white and two black; head more or less yellowish or coppery.

10. Tail longer: without hairs, about 3½ inches (89 mm.).

6'. Body without any stripes: general coloration nearly or quite uniform.

12. Size larger: body alone more than 7 inches (178 mm.).

12'. Size smaller: body alone less than 7 inches (178 mm.).

14. Tail round and relatively long: over 3 inches (76 mm.).

14'. Tail flat-haired and relatively short: under 3 inches (76 mm.).

Note.—Figures indicating page numbers refer to folios at foot of pages.

California Ground Squirrel

CALIFORNIA (OR BEECHEY) GROUND SQUIRRELS.

CALIFORNIA GROUND SQUIRREL.

Citellus beecheyi beecheyi (Richardson).

PLATE I.

Other names.—Digger Squirrel, part; Beechey Ground Squirrel; Beechey's Marmot; Beechey Spermophile; Spermophilus beecheyi, part; Arctomys beecheyi; Spermophilus grammurus beecheyi, part; Citellus variegatus beecheyi; Citellus grammurus beecheyi; Otospermophilus beecheyi.

Field characters.—A large ground-dwelling squirrel, with long bushy tail, good-sized ears, and general brownish coloration; dull whitish area on side of neck and shoulder, and fine dappled pattern of coloration on back and sides, to be seen in close view. Length of body alone about 10½ inches, with tail (without hairs) about 6½ inches more.

Description.—Adults in summer pelage: Top of head, stripe down middle of hind neck, whole back, sides, and rump, of a general wood brown tone of coloration, but variegated in fine pattern on back, rump and sides by mottlings of snuff brown and buffy white; these mottlings usually line up in transverse rows, the rows being most distinct across the rump; a large area centering on side of neck and involving

Fig1 Fig. 1. Ears of ground squirrels to show characters of size and shape in different species. a, California Ground Squirrel; b, Fisher Ground Squirrel; c, Oregon Ground Squirrel; d, Stephens Soft-haired Ground Squirrel; e, Mohave Ground Squirrel; f, Yuma Round-tailed Ground Squirrel; g. Sierra Golden-mantled Ground Squirrel; h. Desert Antelope Ground Squirrel. All natural size and drawn direct from specimens. Note: a and b are extreme examples; the average difference existing between the California and the Fisher ground squirrels is much less; individuals of each race can be found which will overlap some individuals of the other in size and shape.

shoulder, and a faint stripe backward a short distance from upper margin of this area, dull white. Cheeks dirty white, changing to wood brown color between eye and ear; eyelids white; whiskers black. Ears tall and conspicuous, finely haired; color of ear inside, pale pinkish buff; back of ear, front half, black, becoming dull cinnamon buff at base and on hinder margin; fine black hairs extending above rim at tip of ear sometimes so numerous and long as to form a small tuft; whole lower surface of body, inner sides of fore and hind legs, and upper sides of feet, pinkish buff; hairs of breast and belly gray at base, this resulting in a darker tone on this area. Palms of fore feet naked; soles of hind feet thinly haired behind tubercles, or else wholly naked, due apparently to wearing away of the hairs altogether; claws brownish black, horn-color toward tips. Tail bushy, though not nearly so much so as in the tree squirrels, flat haired, parallel-sided, and square or round ended; hairs along sides of tail about 41 mm. (1⅝ inches) long, at end of tail the same; general color of tail both above and below buffy grizzled gray, in other words mixed black and buffy white in fine pattern; the buif tone is deeper below than above; close inspection shows the hairing of the tail to be concentrically banded, three black bands and four light ones, the outermost black band being broadest, and the outermost light one constituting a peripheral whitish fringe.

Color variations.—As far as we can see, the two sexes are identical in coloration, save as caused by the greater rate of wear to which the pelage of the female is subject during the season when the young are being reared. Wear progresses in some cases until most of the colored ends of the hairs are gone, and a dingy light brown color is acquired, including also the tail. Molting begins anteriorly and progresses backward.

The material we have studied seems to show but one decided molt in adults each year, and this takes place during July and August. Young, however, seem to undergo two molts in the first six months of their lives. When one-third grown their pelage is characterized by a fluffy texture and a yellowish tone of color, but the general pattern is closely similar to that of adults; when nearly full grown the young are smooth-coated and show rather brighter tones of brown and clearer white shoulder patches that even fresh-pelaged adults.

There are not infrequent special, or "sport" variations, in the Beechey Ground Squirrel, such as albinos, either complete or partial, which have been reported from time to time. We have been told of "black" ground squirrels; and there is in the Museum of Vertebrate Zoology an adult male, from Stanislaus County, taken by W. C. Jacobsen, January 30, 1918, which is of a curious light pinkish-buff tone of coloration, save for the outermost concentric black band on the tail and for the whitish shoulder patches and a suggestion of dappling on the back.

The color description given above was taken from specimens from the vicinity of San Francisco Bay. Specimens from other parts of the general range of the Beechey Ground Squirrel depart from this slightly in different respects. Two specimens at hand from Marysville Buttes, Sutter County, are of paler, grayer tone of general color. A series of skins from the western slope of the central Sierra Nevada averages whiter underneath and darker brown on middle of back. Examples from the southern San Joaquin Valley are paler in tone of coloration and exhibit whiter shoulder patches, thus evidently constituting intergrades towards the Fisher Ground Squirrel. A series from the coast district of southern California, from Santa Barbara to San Diego, shows darker brown back, but whiter under surface of body, and the white shoulder patches are more conspicuously contrasted. Some San Diego County examples in rather worn pelage show a curious reddish tinge on the rump.

Measurements.—Average and extreme measurements, in millimeters, of twenty full-grown specimens from west-central California are as follows: Ten males: total length, 435 (405–475); tail vertebræ, 164 (150–175); hind foot, 57 (52–60); ear from crown, 20.5 (17–24); greatest length of skull, 59.1 (56.9–61.1); zygomatic breadth, 36.9 (35.0–39.2); interorbital width, 14.1 (13.3–15.0). Ten females: total length, 423 (400–460); tail vertebræ, 162 (150–175); hind foot, 57 (55–58); ear from crown, 18.4 (16–20); greatest length of skull, 56.2 (53.8–59.5); zygomatic breadth, 35.8 (34.3–37.6); interorbital width, 13.9 (13.0–14.7).

It will be seen from the above figures that females are decidedly smaller bodied than males though in tail length they are about the same. The skulls of the oldest individuals, particularly males, show greatest general size, greatest zygomatic breadth (as compared with total length of skull), broadest jugals, stoutest postorbital processes, highest developed sagittal crest, most nearly approaching parietal ridges (these meeting also farthest forward), and broadest frontal region (which also shows a concave or "dished" upper surface). In other words, old animals have skulls which are more massive and angular than those of young ones. We find that relative age of an individual can be recognized approximately by the relative degree of development of the above characters. Amount of wear on the crowns of the molariform teeth and of advance in coalescence of the contiguous bones along certain sutures also give criteria for determination of age.

Weights.—Average and extreme weights, in grams, of twenty full-grown specimens from west-central California are as follows: Ten males, 696 (600–923); ten females, 592 (491–774). Averages in ounces: males, about 24 (1½ pounds); females, about 20 (1¼ pounds).

It appears that males are 17 per cent heavier than females. The heaviest specimen out of a total of 36 weighed, was an old male tipping the scales at 923 grams, or 32½ ounces, or a trifle over two pounds. This animal was shot June 12, 1918, in a slaughter yard at Mendota, and was exceedingly fat.

Type locality.—"Neighborhood of San Francisco and Monterey" (Richardson, 1829, p. 170).

Distribution area.—The greater part of central and southern California west of the desert divides. Altitudinally, ranges from sea-level up regularly to about 6,000 feet, and locally and sparsely to as high as 8,200 feet (in Yosemite National Park). As regards life-zone, the California Ground Squirrel is most abundant in the Upper Sonoran zone, less so in the Lower Sonoran and Transition, and but relatively rare and local in Canadian (see fig. 23).

More in detail, this squirrel is limited to the northward in the coast belt abruptly at the south sides of the Golden Gate, San Francisco Bay, Carquinez Strait, and Suisun Bay. Its range extends northward over the eastern half of the Sacramento Valley to the Marysville Buttes and its limits thence swing northeastwardly through the Feather River country to the southern border of Lassen County. From this last point south it covers both slopes of the Sierras nearly to the Yosemite region, but thence south to Tulare County, only the western slope. In the vicinity of Lake Tahoe it gets a little way into the state of Nevada. To the southward it covers most of the San Joaquin Valley, and the coast belt south throughout the San Diegan district to the Mexican border, and beyond this even to the San Pedro Martir mountains.

Along the eastern border of its range, from Tulare County south to Riverside County, the race beecheyi grades into the race fisheri. The dotted line on the map (fig. 17) marks approximately the center of the area of intergradation between the two races. As will be seen, the limits of beecheyi swing west across the southern San Joaquin Valley and thence around south so as to exclude the Bakersfield region and the Tehachapi, Tejon, San Bernardino and San Jacinto mountains.

Specimens examined.—A total of 149 specimens from the following localities in California: San Francisco County: Ingleside Race-track, 1. Alameda County: vicinity of Berkeley, 9. Contra Costa County: Walnut Creek, 12; west side Mount Diablo, 2. San Mateo County: Sierra Morena, 1; Pescadero Creek, 1. Santa Clara County: Palo Alto, 1. San Benito County: Cook, 2. Monterey County: Monterey, 6. Sutter County: Marysville Buttes, 2. Stanislaus County: Claribell Station, 1. San Joaquin County: eight miles southwest of Tracy, 1. Sierra County: near Sierraville, 1. Placer County: Dutch Flat, 1; Blue Canyon, 1; Cisco, 2. El Dorado County: Fallen Leaf Lake, 1; Kyburz Station, 2. Tuolumne County: Aspen Valley, 6,400 ft., 1. Mariposa County: Merced Grove, 1; Crane Flat, 6,300 ft., 1; Indian Creek, 6,100 ft., 1; Yosemite Valley, 3; Merced Lake, 7,500 ft., 2; Mono Meadow, 7,300 ft., 1; Chinquapin, 6,200 ft., 1; El Portal, 2,000 ft., 2; Coulterville, 2; Pleasant Valley, 1. Merced County: Snelling, 1; Los Baños, 1. Madera County: Raymond, 2. Fresno County: Mendota, 4; Panoche Creek, at 502 ft., 1; Friant, 1; Kings River, 5,000 ft., 2. Ventura County: Matilija, 4; Ventura, 3. Los Angeles County: vicinity of Pasadena, 7; near Azusa, 2. San Bernardino County: near Colton, 4. Riverside County: Thomas Mountain, 6,800 ft., 1. San Diego County: Warner Pass, 3; Grapevine Spring, 1; Witch Creek, 8; Julian, 8; Cuyamaca Mountains, 4; San Diego, 2; Point Loma, 10; Chula Vista, 3; near mouth Tiajuana River, 1; Dulzura, 7; Campo, 1; Jacumba, 1; Mountain Spring, 4.

The California Ground Squirrel is probably known by sight to more people than any other one of our four hundred kinds of native mammals. It inhabits open ground in well-settled territory and it forages abroad during the daylight hours when its movements are most likely to attract attention. Numbers are to be seen from the windows of passing trains, and the traveller by automobile is often thrilled by the narrow escapes of those heedless individuals which dash across the road immediately in advance of him, not infrequently to their own undoing. Then, too, this squirrel has, perhaps, been more widely advertised than any of our other mammals. A few years ago it came into prominence as a proven disseminator of the dreaded bubonic plague, and it has become notorious for its exceeding destructiveness to cultivated crops.

The term "Digger Squirrel" is often applied to this species, more especially in the foothill and mountain regions, in recognition of its burrowing habits, to distinguish it from the tree-inhabiting gray and red squirrels. The book name, Beechey Ground Squirrel, much used in the literature relating to it, is derived from the accepted scientific name Citellus beecheyi. This name, beecheyi, was bestowed upon the animal by its original describer (Richardson, 1829, p. 170) "in honour of the able and scientific Commander of the Blossom," Captain F. W. Beechey. The British ship "Blossom" cruised the Pacific Ocean northward even to Bering Strait during the years 1825 to 1828. Collections of specimens were brought back from many localities visited, including San Francisco and Monterey; among these specimens was one or more of the squirrels in question. These were evidently preserved for the most part by Mr. Collie, surgeon of the ship, who is quoted by Richardson as stating that "this kind of Spermophile 'burrows in great numbers in the sandy declivities and dry plains in the neighbourhood of San Francisco and Monterey, in California, close to the houses. They frequently stand up on their hind legs when looking round about them. In running, they carry the tail generally straight out, but when passing over any little inequality, it is raised, as if to prevent it being soiled. In rainy weather, and when the fields are wet and dirty, they come out but little above ground.'" And further information is given, according, for the most part, with what anyone can see for himself today in the same general region. This attests to the acuteness of observation of Mr. Collie, and also shows how the squirrels had already, some ninety years ago, begun to impress people with their numbers and boldness.

The California Ground Squirrel may be distinguished from other members of the squirrel family by the combination in it of the fallowing characters: essentially ground-dwelling habits, relatively large size, long bushy tail, tall pointed ears, and generally grayish coloration with a three-cornered silvery white patch on each shoulder. Close inspection discloses a finely dappled pattern of coloration (see Fig. 2) such as is not shown in any tree squirrel or in any of our other ground squirrels except its near relatives, the Douglas, Fisher, Rock and Catalina Island squirrels. The detailed descriptions, measurements, etc., as given in the accompanying small-type paragraphs, should be studied for further particulars in this connection.

Fig2 Fig. 2. A typical example of the California Ground Squirrel, the rodent of chief economic importance in the state. Note the tall, rather pointed ears, the whitish eyelids, the grizzled-white shoulder patch, the mottled back and rump, and the bushy tail. Photographed from life by J. Dixon.

The species now under discussion is restricted in its distribution mainly to the state of California. It extends a little ways south into Lower California; and to the eastward it barely crosses the Nevada line in the vicinity of Lake Tahoe. To the northward in the coast belt it is cut off sharply by the Golden Gate and San Francisco Bay, but in the interior it extends to the headwaters of the Feather River. Southeastwardly toward the deserts the race beecheyi blends into the race fisheri, which in turn is wholly cut off by the hottest deserts beyond. (See map, fig. 17.) In the coast belt north of San Francisco Bay and from the upper Sacramento Valley northward the Beechey Squirrel is replaced by the Douglas Ground Squirrel.

Altitudinally the California Ground Squirrel ranges from sea level, as on the shores of Monterey Bay, up to an altitude of at least 8200 feet, as in the Yosemite National Park. It is most abundant on the plains of the San Joaquin and in the Coast Ranges and Sierra foothills. As regards life-zone, the metropolis of the species lies in the Lower and Upper Sonoran (see fig. 23). It is less numerous in the yellow pine belt (Transition zone), and is but rarely or sparingly represented in the Canadian zone, still higher on the mountains. Its preferences as to local conditions are not closely limited, except that it avoids dense chaparral and thick woods. It frequents pasture lands, grain fields, orchards, sparsely tree-covered slopes, small mountain meadows, rock outcrops on the tops of ridges, and even granite talus slopes. It is always most abundant, however, in the open situations, and its decided preferences are such that it thickly populates much of the best farming and grazing lands in the state, to the great reduction of their producing value from the human standpoint.

This squirrel secures shelter for itself and young, and safety from its enemies, by burrowing in the ground. Where possible it chooses to excavate its retreats in hillsides or in low earth banks. Here some at least of the necessary digging can be done in a horizontal direction. But, of course, those members of the species which live on the plains or on small flats or meadows in the foothills or mountains must dig down vertically for considerable distances to gain the requisite protection. Many of the squirrels which live in the granite country make their homes under large boulders or in rock taluses where a minimum of burrowing is necessary to insure safe retreats. On wooded hillsides special safety from enemies that dig is secured by location of the burrows under tree-roots or old stumps.

In the foothill region at the southern end of the San Joaquin Valley there seems to be a decided tendency on the part of the ground squirrel to select alluvial fans for home sites. This preference may be accounted for by the fact that the plants upon which the squirrel feeds make a better growth in the deep alluvial soil there than they do on the adjoining hillsides which are often steep and with but shallow cover of soil. For the same reason the alluvial fans afford easier digging to considerable depth and hence better protection. In seven burrows in different localities, in which the squirrels were gassed and then dug out (J. Dixon, MS), the extent, diameter and depth were found to vary and to depend largely upon the nature of the soil. In shallow adobe or clayey soil, underlaid by broken rock, the burrows were found to be short, of

Fig3 Fig. 3. Four characteristic poses of a California Ground Squirrel. Photographed by J. Dixon at Berkeley in August, 1918.

small diameter, and not reaching to any considerable depth. Those in alluvial or sandy soil were found to be of large diameter, of greater extent, and to reach to much greater depths.

The most conspicuous signs of activity on the part of ground squirrels in any locality are the large mounds of earth that have accumulated in the course of excavating the burrows. This earth is commonly thrown out in a fan-shaped pile directly in front of, and to the sides of, the main entrance to the burrow (see fig. 4). These mounds of earth are often three or four feet in diameter and from six to ten inches above the general level. They vary greatly in size, but average larger in sandy soil than in clayey or rocky ground. The size of the mound is, however, no reliable index to the length or size of the burrow except in those cases where the burrow is of a straight or simple pattern. In

Fig4 Fig. 4. Mound and burrow entrance of a "digger" squirrel, in sandy ground. Mounds of earth such as these are often three or four feet in diameter and rise from six to ten inches above the general level. The route taken to the feeding grounds being used at this particular time is indicated by the numerous tracks at the left-hand side of the entrance. Photographed at Tipton, Tulare County, May 23, 1918.

colonial or intercommunicating burrows the dirt is not always thrown out at those entrances which allow of the shortest possible "haul."

Most of the work of tunnel excavation is carried on during the spring months, as is shown by the mounds of fresh, soft earth accumulated at the mouths of the burrows in that season. In the lowlands, where there is a large crop of wild oats in the springtime, this newly excavated earth supports a ranker growth than the surrounding parts of the field, so that, as one of our party wrote in his field notes, "the plain looks like a cemetery overgrown with grass," with these taller stands of oats about the squirrel holes suggesting grave mounds.

To some extent the ground squirrels, like the pocket gophers, thus act on wild land as natural cultivators of the soil, and may thus serve a useful purpose. On the other hand, their burrows are frequently the cause of much destructive erosion on hillsides during heavy rainstorms. Numerous small landslides have been noted on steep hillsides on the campus at Berkeley, that were plainly caused by the presence of squirrel burrows which had concentrated and conducted the water in narrow channels instead of permitting it to spread out and soak in or run off in the natural way. The presence of squirrels along irrigation canals results in the embankments becoming undermined by their burrows, with ensuing disastrous breaks in the canals, especially at times of high water.