Ethoecology of the round-tailed ground , tereticaudus

Item Type text; Dissertation-Reproduction (electronic)

Authors Drabek, Charles Martin, 1942-

Publisher The University of .

Rights Copyright © is held by the author. Digital access to this material is made possible by the University Libraries, University of Arizona. Further transmission, reproduction or presentation (such as public display or performance) of protected items is prohibited except with permission of the author.

Download date 10/10/2021 00:09:37

Link to Item http://hdl.handle.net/10150/332155 . ETHOECOLOGI OF THE ROUND-TAILED GROUND

SQUIRREL, SPERMOPHILUS TERETICAUDUS

bgr

Charles Martin Drahek

A Dissertation Submitted to the Faculty of the

DEPARTMENT OF BIOLOGICAL SCIENCES

In Partial Fulfillment of the Requirements For the Degree of

DOCTOR OF PHILOSOPHY WITH A MAJOR IN ZOOLOGY

In the Graduate College

THE UNIVERSITY OF ARIZONA

1970 THE UNIVERSITY OF ARIZONA

GRADUATE COLLEGE

I hereby recommend that this dissertation prepared under my direction by CHARLES MARTIH DMBEK______entitled Ethoecology of the Round-tailed .

Spermophilns tereticaudua,______be accepted as fulfilling the dissertation requirement of the degree of ______Doctor of Philosophy______.______

(y-MW-i iA/yv\- 1 7 d Dissertation Director Date

After inspection of the final copy of the dissertation, the following members of the Final Examination Committee concur in its approval and recommend its acceptance:*

k>v * 4 y v i {^yp ? Ay m

•y IXteu? /f 7 3

This approval and acceptance is contingent on the candidate's adequate performance and defense of this dissertation at the final oral examination. The inclusion of this sheet bound into the library copy of the dissertation is evidence of satisfactory performance at the final examination. STATEMENT BY AUTHOR ,

This dissertation has been submitted in partial fulfillment of requirements for an advanced degree at The University of Arizona and is deposited in the University Library to be made available to borrowers under rules of the Library„

Brief quotations from this dissertation are allowable without special permission, provided that accurate acknowledgment of source is made* Requests for permission for extended quotation from or repro­ duction of this manuscript in whole or in part may be granted by the head of the major department or the Dean of the Graduate College when in his judgment the proposed use of the material is in the interests of scholarship. In all other instances, however,.permission must be ob­ tained from the author.

j ACKNOWLEDGMENTS

I wish to thank Dr® E® Lendell Cockrum for his advice and contributions throughout ray graduate work. I thank Drs® Stephen M®

Russell $ Albert Re . Meads Everett H. Lindsay and Donald L. Bryan t who critically read the manuscript and offered many helpful suggestions.

Most of the field work was conducted while I resided at the

University of Arizona-Boyce Thompson Southwestern Arboretum. I greatly acknowledge the financial assistance provided through University of .

Arizona~NSF institutional grants 5701”003=»8ll~b2 and 5701'= 0Q3=-155-ay in 1967 and 1968 respectively, as well as the staff at the Arboretum. V . ■■ . ■■ - ' --'.r Finally, I wish to acknowledge my wife, Jane, who by aiding me with all phases of this study, as well as maintaining a full time teaching job, deserves more credit than the average wife.

ill TABLE OF CONTENTS

Page

LIST OF TABLES „ „ . „ . . O 0 0 vi ° • 0 ° 0 •

, „ . O o. o 0 vii LIST OF ILLUSTRATIONS ° 0 ® ® • •• «

» o o . o o o b o o • o 0 viii ABSTRACT ° « 0 ® 0 ® • 0 ° •

. , „ . . . , © o INTRODUCTION O .0 » ® •. • o • 1

DESCRIPTION OF THE STUDY AREA. o o o o o ® 0 ° ® 0 •» o • 3

. . 0. 10 METHODS AND MATERIALS. ° ° 0 o ® 0 0 ••

Study Area ..... o o 0 o o 0 e 9 o 0 0 10

Squirrel Enclosure . © 6 o 0 o O © o ■O O © 0 O o o . 10

Trapping and Marking Techniques. o O o 9 0 O 0 0 o o o o 11 Behavioral Observations. o o 0 9 o o 0 o 0 0 o o o . 15 Movement Analysis. . 0 C o O 0 0 0 O O o o o 0 o 16 '' Burrow Excavation. . o o O 0 e o 0 0 9 0 0 0 o o . 0 16 Climatological Data. ° ° ° ° ® 0 0 6 • 0 16

MAINTENANCE BEHAVIOR . . . ■0 O o o p 0 0 O 0 ® • o o o 17

. Locomotion . . „ „ . © c .. 0 0 ; 17

Grooming ...... b o OO O o 0 0 0 o 0 o O 19

Exploration. .... 0 o o O o o o 9 O o o Q 0 o 0 O 21

Ingestion...... o o O 0 O o 0 9 9 O 9 0 o 0 o o 22

Elimination. , . . . o 0 o 0 © © ' 0 .© 0 o- O. OOO 23

Alertness...... O © o o O 0 o O 0 0 0 O 23 Digging...... o O O 0 0 © o 0 0 0 0 24 Grass Gathering. . . O O 0 o 0 0 0 O. 0 0 0 O o' 25 Tail-flicking. . . . 0 o ° ° ° 0 0 0 ° ® •* * 26

ENVIRONMENTAL ADAPTATIONS. ° ° 0 °. ° 0 ® o o o 28

Burrows...... 0 0 © 0 o O 0 28 Torpor...... 0 0 c 0 0 ° 0 ° o ° © •• 32

^OOD HABITS...... 4» 0 o * o o e © 0 0 9 O 0 o 36

iv TABLE OF CONTENTS (Continued) Page

COMMUNICATION OOO0O 0OO 0 006000 OOOOOOOOO 41

V O G 8 3 . OOCOO 0 0 0006 o o 41 Olfactory . o » « . o o 45

Tactile e oo- e o o o o-O o o o 46 V ISUalooeoceoeooo 46

COMPETITIVE AND-PREDATORY FACTORS OOOOOOOOOOOOOO 48

Predators ©oo o » ® © o @ o o © o 48 Competitors © © © ® ® . © « 6 . © O o 53

BEHAVIOR;IN THE ENCLOSURE © © © © o o o o o o o.e o o o o o o : 55

ACTIVITY PATTERNS oOO.OOOOOOOO OOOCOOOOOOO 59

AtHZ112.8-3o. 0000000©C 6 0 0 © o o 59 Daily Activity© © © © © © © o 6 o o 59

POPULATION DYNAMICS e o o o g o o o o c o o o o o o o o o - o o

Population -Structure and Density. o o 64 Sex Ratios © © © © © © © © © © © © 67 Population Turnover and Dispersal 67

Home Range© © © © © © © © © © © © 6 6 69 Familiarity of Home Range 79 Intraspecific Intolerance 80

INTRASPECIFIC BEHAVIOR OOCOOOOOOOOOOOOOOOOO

Epimeletic Behavior © © © © © © © 85 Play©o©e©ooo©e©©oo© 86

Agonistic Behavior. © © © © © © © 9 6 0 O 86

SUMMARY AND CONCLUSIONS © © © © © © © © OOOCOOOOO OO 89

REFERENCES CITED 9 e 6 e Q © © 6 O G G 6 © O 6 © O 6 0 6 G O G 102 LIST OF TABLES

Table

1„ Flora of the study area « , e . .= • •• . „ * . .

2 0 Ninety of the dye marks used to identify individual ground scjumrelSo © © © © © © © © © © © © © - © © © ©

3© Mean burrow temperatures at various distances from the entrance, and depths and lengths of excavated burrow systems (N in parentheses) © © « © © © » © «

4© Food preferences noted during 166 observations of feeding round-*tail s © © © © © © © © © © © © © © © ©

5© Vertebrate fauna occurring or observed in the study area© ©©©©©©©©©©©©©©©©©

6. The population structure of the Round-tailed ground captured in the study area in June, July and August of 1967 and 1968 * © © © © © © . . . © ©

7© Population structure of resident ground squirrels ©

8= The percentage of the 1967 resident squirrels found in the study area in 1968 ©».*©© . © © © © « ©

9© The estimated home range size, standard error of the mean and range in acres for 1967 and 1968 resident squirrels © © © © © © © © © © © © © © © © © © © © ©

10. The frequency at which the home ranges of the 1968 resident squirrels overlapped with intraspecific V O li lp £> e c o o o e cneeco eeoo osooe o o

11. Comparative behavioral traits and responses to predators for various species of Soermophilus . LIST OF ILLUSTRATIONS

Figure Page

I, Location of the study area and vieinity in ' south-central Arizona * * , , * « . . , . * , , « * * a » 4

2* Mean monthly precipitation for the Station,and

Florence, Arizona « o o o o o o O o 0 O o 6 0 0 o o 8

3= Temperature ranges of the Station and Florence, AriZOha * a a a a a o a a a a a a a a a a a a a a a a 9

4a Some common postures associated with the egocentric behavior patterns of the Round-tailed ground squirrel 18

5a Mean number of round-tails observed above ground and mean ambient: temperatures during June, July and \

A U g U S t S U r V e y S a a a a a a a a a a a a a a a a a a a a a a b l

6„ Home ranges of an adult male, squirrel 1-100, during the 1968 summer months 0 . a « = = , a . a . » a 75

7 a Home ranges of an adult female, squirrel 3-60, before and after its four young left the home burrow „ . a » e a 76

80 Home ranges of a juvenile female, squirrel 30-80, before and after leaving its home burrow, , , 0 a a a 0 a 78

vii ABSTRACT

Bthoecological aspects of a natural population of Round-tailed ground squirrels, Spermonhllus tereticaudus neglectus Merriam, v?ere investigated during 1967 and 1968 in a Sonoran Desert Larrea zone,in south-central Arizona. The squirrel population in a 25 acre study area was marked for permanent and observational identification to elucidate adaptive behavior and population structure6

Maintenance behavior is described and its possible role in communication is discussed. Exploratory behavior is Shown to.be an important feature in the establishment and familiarization of a home range.

The utilization of burrows played an important role in the round-tail’s adaptation to a desert environment. Structure, mainten­ ance, temperatures and diurnal use of the subterranean systems are discussed. Although semi-colonial in behavior, the squirrels were

found to be solitary in their burrow habits.

Mesquite .trees were the primary food source, from which the round-tails obtained a succulent diet, and was supplemented by the

fruits of creosote-bushes, perennial wildflowers, insects and carrion.

The vocal repertoire included numerous sounds, which were

associated with specific behavioral patterns. Evidence Indicates that a visual communicative system has evolved among the Round-tailed ground

squirrels, and an upright posture of alert behavior functions as a

signal to warn squirrels of predators.

vili Ix

Sexually active adult males placed in a squirrel enclosure ex­ pressed sexual behavior toward adult females and intense, dominating agonistic behavior toward juvenile maleSo

The daily activity of the squirrels was bimodal, with the onset approximately an hour after sunrise, and cessation at sunset| thus, they avoided the extreme ambient temperatures 0 The squirrels were inactive above ground in October, November and December when periods of torpidity were probably frequent0

An estimated resident population of 2B1 „Squirrels per acre was . composed largely of adult and juvenile females both summers. Forty-six per cent of the 1967 residents.were present in 1968 with the adult fe­ males accountihg for the largest percentage and the juvenile males the smallest. Juvenile males left the home burrows earlier than juvenile females and seldom established residency within the study, area.

Home .range estimates of 0.74 acres for adults and 0.77 acres for juveniles were derived by the minimum area method and were found to change throughout the summer months. Influential factors were sex, age, and food preferences.. Resident squirrels expressed familiarity with

their home ranges-. and c h a s e d transients attempting to- establish resi­ dency either from the study area or to sparsely populated areas therein.

Intraspecific behavior is described, and the role agonistic tendencies and individual distances play in the social hierarchy and in the spatial distribution of areas of activity is discussed. INTRODUCTION

The limited knowledge of the North American ground squirrels is in part the result of their geographic distribution. The extant 23

North American species of Soermonhilus and five Ammospermophilus listed by Hall and Kelson (1959) are limited almost entirely to the western

United States and Mexico, an area wherein relatively few zoologists live.

The Thirteen-lined ground squirrel, S. tridecemlineatus. has the largest geographic range,, extends the farthest to the east and is readily ac­ cessible to many zoologists. Consequently, the literature is the most extensive for this species (Johnson,. 1917j Wade, 19275 Evans, 1951j

Rongstad, 1965; McCarley, 1966; Bridgwater and Penny, 1966; McCarley,

1970). In contrast, the entire distribution of the Mohave ground squirrel, S. mqhavensis, which singly shares the subgenus Xerospermo-

•philus with the round-tail, is limited to the Mohave Desert area Of

California (Hall and Kelson, 1959)» and the literature concerning this

species is nearly non-existent (Burt, 1930; Bartholomew and Hudson, I960)

Since Spermophilus is one of the few groups of that is

strictly diurnal, not particularly secretive in Its habits, relatively large in size and semi-colonial, it is surprising that more ethoecologi-

cal studies have not. been made on members of the group. The ability of

Spermophilus to undergo periods of torpor as a means of avoiding a

stressful environment has, in recent years, attracted the attention of workers investigating the phenomenon of . However, these

1 2 investigations have been conducted in the laboratory and have, not required the investigators to remain in the field to study the in its natural habitat (Bartholomew and Hudson, I960j Hudson, 1964;

Pengelley and Kelly, 1966).

The literature on S. tereticaudus indicates the scant knowledge about this ground squirrel. The distribution of the four subspecies, restricted to sparsely populated desert areas of Arizona, Nevada,

California, Mexico and Baja California, is probably the reason more, investigations have not been conducted® There has been little concern for this non-game animal with the exception of occasional pest control measures to eradicate squirrels from agricultural lands0

As recently as 1964 the reproductive cycle of S. tereticaudus

■was determined (Neal, 1965b), and its disappearance during the fall and winter months had.been the only evidence for hibernation until investi­

gations by Neal (1964), Hudson (1964) and Pengelley and Kelly (1966).

, Because the desert biome is of particular interest, due to the

stressful environmental conditions that are placed on the inhabiting

organisms, physiological adaptations of S. tereticaudus have been in­

vestigated under laboratory conditions? however, it is essential that

the adaptive behavioral characters displayed by this animal in its

natural habitat be studied to fully understand the implications of

laboratory results. The objective of this study was to-investigate

some basic aspects of the ethoeoology of S, tereticaudus for a better . 1 . ' understanding of the role behavior plays in adaptation to a desert

environment and to elucidate evolutionary implications of population

dynamics and social organization. . DESCRIPTION OP THE STUDY AREA

The study area vas situated in south-central Arizona in the northeast portion of the Sonoran Desert designated by Shreve (1951) as the Arizona Upland, The Round-tailed ground squirrels in this area are restricted to the Larrea zone of the plains and lower bajadas, being replaced in the upper bajadas by Harris' , Amraospermo- philus:harrisi (Drabek, 1967), The study site was located 12e0 miles north and 1,0 mile east of Florence5 Pinal Coe$ Arizona (Fig, l)0

The Larrea zone is part of a physiographic unit of the Arizona

Upland that Shreve (1951) refers to as the Plains, There the soil varies from a fine alluvium to a coarse gravel overlying a calcareous hardpan,

caliche. Creosote-bush, Larrea tridentata. is the dominant plant

species of the Plains, but the delimited 25 acres utilized as the study area contained a large number of honey mesquite trees, Prosoois

iuliflbra. Table 1 lists the flora found in the study area. The peren­

nial plants were collected from the area as they bloomed and were

identified at the University of Arizona herbarium.

The elevation of the area was approximately 1850 feet, and the

area was bordered on the north and south by dry washes, varying from

2 to 8 feet in depth and 4 to 20 feet in width, which drained the area.

Besides the study area in the Larrea zone north of Florence,

in which a natural population of round-tails was observed, observations were also conducted on captive squirrels introduced into an outdoor

squirrel enclosure. This was located at the University of Arizona-Boyce

3 Phoenix 4 0 mi.

DESERT BIOLOGY STATION SUPERIOR

FLORENCE JCT Picket Post Mfn

STUDY AREA

Arizona

Phoenix Globe • . Silver Bene Mine.' Area of Map

.Tucson

North Butte

South Butte

Scale in Miles FLORENCE

Fig, 1, Location of the study area and vicinity in south-central Arizona Table 1, Flora of the study area0

Common and scientific names follow Kearney and Peebles (1964)» ' ■ , , ' / ' '

- . : ■ ____ _ ; ' v ' ■. ■ ' ' ' ' '

FAMILY SPECIES COMMON NAME

' ' ' Trees and shrubs

Ulmaceae Celtis mllida . . ' Desert hackberry

Leguminosae Prosoois juliflora Honey mesquite

Leguminosae Cercidium floridum Blue palo-verde

Leguminosae Olneya tesota Ironwood

Zygophyllaceae Larrea tridentata Creosote-bush

Herbs and grasses ■

Gramineae Schismus barbatus

Gramineae Schismus arabicus

Portulacaceae Calandrinia ciliata -—

Gruciferae Thelypodium lasiophyllum ——

Cruciferae Lepidium lasiocarpum ———

Cruciferae Lepidium obloneum Table 1 Continued

FAMILY SPECIES COMMON NAME

Herbs and grasses

Cruciferae Lemdinm medium

Cruciferae Sisymbrium irio

Leguminosae Lutdnus snarsiflorus Lupine

Legumihosae Luninas c'oncinnus Lupine

Euphorbiaoeae Euphorbia sp» Spurge

Boraginaceae Pectocarya platycarpa

Boraginaceae Cryptantha sp,

Boraginaceae Plagiobothrys arizonicus

Compositae Heterotheca subaxillaris

Compositae Rayless goldenrod

Compositae Franseria spe Bur-sage.

Compositae Pectis papposa Chinchweed 7

Thompson Southwestern Arboretum, 4 o0 miles vest of Superior, Pinal Co.,

Arizona. The Arboretum is utilized by the University as a Desert

Biology Station (Fig. 1).

Climatological data for the Biology Station and Florence,

Arizona (Figs. 2 and 3) show the annual precipitatidn to be biseasonal.

The weather data recorded in the study area corresponded to that of

Florence, which received an average annual precipitation of 9.59 inches, with peaks in August and December. Each rainy season was followed by a bloom and heavy growth of the desert vegetation. The maximum temperatures occurred during, the summer months with mean- monthly-maximum temperatures remaining well oyer 100 F, The squirrels were inactive and not exposed to the lower temperatures of October,

November and December. I , 2 MA MNHY RCPTTO FR THE PRECIPITATION MONTHLY FORFIG, MEAN2. INCHES OF PRECIPITATION 2.0 2.5 0.5 3.0 1.0 1.5 STATION AND FLORENCE, ARIZONA 8 FIG. STATION RANGES OF THE TEMPERATURE AND 3. DEGREES FAHRENHEIT 40 40 IOO H 60 60 50 50 80 80 70 70 90 90 H 4 4 4 4 4 FLORENCE, ARIZONA ------FLORENCE EET ILG STATION BIOLOGY DESERT AN- HL- I M U XIM A LY-M TH N O -M N EA M ENMNHY- IIU \ MINIMUM - MEAN-MONTHLY 9 METHODS AND MATERIALS

Study Area -

Rodnd-tailed ground squirrels are abundant in the alluvial soil of the dense Larrea zone north of Florence, but the selected study

site was established in an area with a, sparse stand of creosote-bushes : and mesquite trees. This broken density allowed unobstructed viewing

of the ground surface for nearly 600 feet in all directions and offered an ideal area to conduct an observational study on ground squirrels0

In May, 1967, the study area was surveyed by tape and compass, measuring off 100 foot quadrants along ground contours with ten lines,

designated as rows one through ten, running north and south and 13 rows

(A-M) running east and wests A 12 square inch aluminum tag attached to

a 2 foot wire was placed upright in the ground where each north-south

(l-10) row crossed the 13 (A-M) east-west rows® The respective numbers

were painted on each tag (e6g6 A-l, E-6, etc.) to serve as reference

points. At the beginning of the 1968 season additional markers were

placed at 50 foot intervals throughout the area,to afford the observers

intire accurate reference points.

Squirrel Enclosure

To study more intensely the intraspecific behavior between

different age groups and sexes of S e tereticaudus. a 70 by 80 foot out­

door, enclosure was constructed at the Biology Station. Sheets

of corrugated sheet metal, 26 inches wide, were riveted together with

10 11 a 1 inch overlap and placed in a trench 5 feet in depth. This formed

the subsurface walls of the pen while one-eighth inch hardware cloth wired to 6. foot steel posts placed 10 feet apart composed the above

surface walls. A cement curb adhered the sheet metal to the posts and hardware cloth. The height of the pen was 45 inches. Sheet metal

overlapped the top 1 foot to prevent the squirrels from climbing out.

Observations on squirrels released into the pen at the be- .

ginning of each summer were made several days a week at ground level outside of the pen in 1967 and. from a tree seat built in a Chinese pistachio tree located within the pen during the 1968 season. The presence of the observer from either position appeared undisturbing to

the squirrels. Commercial chicken feed was contained in a ground level feeder for the squirrels. Other trees within the pen were a round leaf

eucalyptus, Eucalyptus nolvanthemos. two white gum eucalyptus, E.

camnaspe. and two red gum eucalyptus, E. rostrata.

Trapping and Marking Techniques

Squirrels were live trapped in non-commercial 4 by 4 by 12 inch

sheet metal traps which had one-half inch hardware cloth spot welded

on the tops and backs. The most suitable bait was found to be sun­

flower seeds. Ants made other baits of apple, melon, oats and peanut

butter mixed with oats unsatisfactory.

To mark the resident population in the study area, traps were

set at selected.sites. Trapping was done only during the peak hours

of the squirrel activity,, and the traps were overturned when the ob­

servers were not present in the area. After trapping a squirrel, the 12 trap site was located on the grid and recorded. The squirrels were then allowed to enter a 16 by 10 inch cloth bag* placed over the en­ trance of the trap, for weighing and handling. Weights were obtained to the nearest gram with a Chatillon"spring scale ranging from 0-250 grams. Body temperatures were taken with a rectal probe of a Yellow

Springs telethermometer.

Before marking, each squirrel was sexed and aged. As Neal

(1965a) pointed out, no satisfactory technique has been perfected to separate Round-tailed ground squirrels older than 8 months into age groups. He found no reproductive characters which distinguished one year olds from adults. Thus, the squirrels in this study were con­ sidered either juveniles or adults, with a juvenile based on the follow­ ing characters: small size, soft pelage, small unpigmented nipples and undeveloped gonads (Neal, 1965a).

To provide a permanent identification, each squirrel was toe clipped. Two toes are usually clipped when permanently marking and at least one-half of the digit is removed, as entire terminal phalanges are occasionally lost under natural.conditions (Layne, 1954).

The chances of this happening are evidently not great in S, tereticaudus. for of 170 squirrels captured, only one had lost a toe. In this study no more than one toe was clipped from each foot, and with the exception of three individuals, no more than two toes were clipped. Sciurids have four toes on the fore feet and five toes on the hind feet. Look­ ing at a ventral view of the animal the toes on the fore feet were designated as units one through eight, counting from left to right, while the hind toes were designated as units of ten, 10-100, counting 13 left to right (Davis, 1956)„ Thus, removing the lateral toe on the right forefoot and medial toe on the left hind foot would designate a squirrel, 1-60. This system provided 139 permanent markings. More squirrels could have been marked by clipping a single toe or three toes. The removed toes showed no sign of subsequent infection nor impairment to the digging, foraging or movement activities of the squirrels.

To mark the squirrels for observational purposes Nyanzol D dye :

(Nyanza Color and Chemical Co., Lawrence, Mass.) was used following .. the preparation described by Melchior and Iwen (1965). The only modi­ fication was in the concentration of hydrogen peroxide. It was found that 25 cc of 3% hydrogen peroxide worked as satisfactory as 3.7 cc of

30% and was much easier to use. The small size of the Round-tailed ground squirrel did not make it possible to dye corresponding toe-dip numbers on their sides, but a variety of symbols worked most satis­ factory in both application and individual recognition (Table 2). The symbols were applied with a one-quarter.inch brush, and then the squirrel was returned to the trap for several minutes to allow the dye to dry. Upon the release of each squirrel, the release direction and escape behavior were recorded including the distance and direction, moved and the hole entered.

Trapping was conducted throughout the 1967 summer and the first half of the 1968 summer. Once the resident population was marked, sub­ sequent trapping allowed re-weighing and re-marking of marked squirrels and marking of transient squirrels. Table 2« Ninety of the dye marks used to identify individual ground squirrels.

The symbols along the abscissa Were painted on the sides of the squirrels while the six body regions painted are listed on the left,. Each symbol and each body region was used alone, then combinations of symbols and regions.

' XXX • ■ 1 11 111 — = 0 V A

1 2 3

Hind legs

Fore legs

Head

Butt-tail

Neck ring

Hind-fore legs 87 88 89 90 15

Behavioral Observations

Observations began in the study area in June of 1967 and con­ tinued through July and August„ . Periodic observations were conducted during the fall and spring indaths with daily observatibnb resumed in

June of 1968 and continued into Septembere The observers sat at ground level and used 7 x 35 binoculars and a Bausch and Bomb 15X-60X zoom Balscope spotting scope, A 35mm camera with a 400mm telephoto lens was used to photograph the various squirrel postures and associated behavioral patterns. Observations were made from the same locality both seasonsj, a site which afforded two observers a full 10 acre ground view of the study area. This observation station was at a large mesquite tree which appeared to prevent the squirrels from recognizing sil­ houettes and most movements of the observers since our presence was ignored by squirrels 100 feet away, and after several weeks squirrels as close as 25 feet would forage and conduct daily activities.

Observations were conducted in the study area 6 days a week, with 4 days devoted to the peak activity of the morning hours, while afternoon and evening activity was observed.the remaining 2 days. Each day notes were taken on maintenance, predator, communicative, inter- and intraspecific behavior. Periodically, mid-afternoon-activity v/as observed to establish daily activity patterns of the squirrels. In

1967, surveys were taken every half hour during the observation periods to tally the, number of squirrels above ground. 16

Movement Analysis

Two observers spent 4 to 6 hours daily recording the squirrel activity in the area® The locality of each, squirrel seen was plotted on a grid of the study area and its movements and behavior recorded*

The total spatial observations for each squirrel were then plotted on graph paper„ Home range areas and shapes were estimated by connecting the extreme points of observations with straight lines, according to the minimum area method described by Gookrum (1962) and measured with a compensating polar planimeterc

Burrow Excavation

Burrows were excavated in both the enclosure and study area at the end of each season when the squirrels would be least disturbed. An

8 foot long commercial plumbers s snake was. used to locate the passage­ ways of the subterranean burrows before the surface was removed,. Each excavated burrow was mapped and measured for depth and length. Burrow temperatures were recorded by placing the ambient temperature probe of a telethermometer on the end of the plumber3s snake and slinking it into the burrow as far as possible:,

Climatologlcal Data

Relative humidity and ambient temperature data in the pen and study area were, obtained with a constant recording hygro-thermograph

(Fries Instrument Division), In 1968 soil surface temperatures were recorded with a telethermometer during the hours of observation. MAINTENANCE BEHAVIOR

In analyzing the social organization of any animal, it is im­ portant to know what the animal Wd s , As "Scott (1958:11) stated,

"oooothe primary phenomenon to be studied in the science of animal be­ havior is behavior itself." The following descriptions of maintenance behavior are based on patterns, observed in the enclosure and in the study area. The study of such egocentric behavior helps to understand ■ some of the patterns seen in social, behavior.

Locomotion

Ground squirrels were observed walking, running and jumping.

Squirrels walked only when they were investigating (cautious behavior) and. when moving distances of a few feet. During undisturbed foraging activity, squirrels walked in a typical quadruped fashion by advancing one forefoot followed by the opposite hind foot. The head of the squirrel was close to the ground and the.back was flexed. In a cautious- exploratory behavior the squirrel utilized a much slower walk.as it approached the object being investigated.

Running was the means of locomotion most commonly observed in found-tails. Movement to creosote-bushes, mesquite trees and burrow holes, further than a few feet away, was.a moderate running motion characterized by a series of bounds. They also moved in a faster, galloping motion, observed only during intra-specific chases and es­ cape (Fig. 4)« This running posture was an alternation of a stretched

17 Fig, 4» Some common postures associated with the egocentric behavior patterns of the Round-tailed ground squirrel

A 0 and BY are the post postures of alert behavior, C, is a running locomotion.seen in intraspecific chases and escape, D, is a foraging posture, E, is the posture of a down-alert behavior. Some common postures associated with the egocentric be­ havior patterns of the Round-tailed ground squirrel. 19 body with extended fore limbs and an arched back with, the fore limbs pushing posteriorly while the hind limbs moved anterior, laterally«

A jumping movement was used when getting into and out of trees and shrubsc Squirrels were observed jumping up to 3 feet off the ground into the low hanging limbs of mesquite trees, A jump onto the tree trunk at the termination of a run was also usede Arboreal capabilities were quite evident in the round-tail, Movement within a tree was accomplished by hanging on with the hind legs while the fore legs extended and grasped an adjacent branch. Often, squirrels were seen, swaying on distal tips of mesquite and creosote branches, eating the most recent growth. Squirrels also jumped at low flying grass­ hoppers, ■

Grooming

Grooming behavior possessed by round-tails included dusting, sunning and washing,. Each squirrel had a dusting area near its burrow which consisted of a small depression of loose dirt. The animal pawed the area with several strokes then pushed its head and body into the loose dust, sometimes in a rolling motion. Washing was performed in . quick jerky motions. After licking the forepaws, the neck and head region were rubbed, moving from the ears over the eyes and down the nose* The venter was mouthed, as well as rubbed with the forepaws, while the tail was run through biting motions of the mouth. Scratch­ ing the anterior portion of the body was accomplished with the hind feet. Squirrels emerging at the start of the day had a dark and ruffled pelage, probably due to the high humidity of the burrow 20

(Vorhiesj 1945j Studier and Baca, 1968)„ Emerging round-tails nor­ mally sat sunning at the burrow entrance for 10 to 15 minutes, then dusted and groomed= A shiny, light pelage, slick in appearance was the result of these activitiess

The dusting habits of S„ tereticaudus appestred to resemble the comfort movements of sandbathing in heteromyids as described by

Eisenberg (1963)0 Such movements are a normal part of the behavioral repertoire in most that have adapted to arid habitats

(Eisenberg, 1967)> This grooming behavior of round-tails may also serve the dual function of dressing the pelage and marking as Eisenberg has found in heteromyids. Body scent left in the dusting depressions • near the squirrel burrows may have communicatory value to other, in­ vestigating squirrelSo Periodically, squirrels rubbed their heads and necks against branches of creosote-bushes in scratching motions which could also serve to leave a scent.

Another comfort movement possessed by round-tails was that of stretching. This was most normally accomplished by depressing the back and fully extending the hind limbs posteriorly and fore limbs anteriorly.

The low-lying branches of creosote-bushes were sometimes grasped with the forepaws while the hind feet moved distally to stretch the body.

All grooming behavior was conducted within the animal's home range and most normally at or very near the home burrow entrance. Most grooming was part of the beginning activities of the day. 21

Exploration

Exploratory behavior was shown when a squirrel approached strange objects and unfamiliar holes. The posture of the body was a low crouch with a slight arch in the back and a slightly bristled tail, which often waved slowly. As the unfamiliar object was approached, the neck was fully extended while all four legs remained stationary.

This posture placed the visual, auditory and olfactory senses closest to the object. Cautious behavior was observed when a .squirrel in­ vestigated potential escape holes in the parts of its home range not regularly utilised. When foraging away from its home burrow, it in­ vestigated these proximal holes, then periodically ran and sat in them.

When startled it often used these holes rather than the home burrow.

Exploratory behavior, an important facet when gaining general

familiarity of a region, was seen in all round-tails observed. Many

arehs were repeatedly frequented. It is in these areas in which home

ranges evolve (Calhoun, 1963).

The escape behavior of captured squirrels, especially transients,

often included cautious movements. Usually the squirrel ran directly

to the nearest burrow-, flattened the body, and slowly began investigate'

ing the hole. In such instances the squirrel showed no concern for the

observers, which in some cases were only 5 feet away. Escaping

squirrels sometimes cautiously investigated several holes before one

was entered and sometimes entered one momentarily then rapidly exited

to another. . <

Traps set in the area appeared to act as novel attractants to

the squirrels, as seen in the Uinta ground squirrels, S. armatus. 22 studied by Balph (1968),. The adults gained confidence in approaching the traps much more rapidly than did the juveniles, but both initially, expressed a cautious behavior toward them0 In one instance, an adult squirrel closed a trap by kicking 2 inches of dirt into it0

Ingestion

Most round-tail feeding activity occurred the first few hours of the day. However, when patches of chinchweed. Pectls m n n o s a . were, in bloom, the last few hours before sunset were often spent foraging*

Two different feeding postures were observed. Food was usually manipu­ lated with the fore feet while the animal hunched in a sitting position

(Fig, 4)o The tail was extended on the ground and served as a sup­ portive "third leg". The squirrels walked slowly within the foraging area then sat when something was picked up and eaten, Mesquite beans were handled with the forepaws to extract the seeds,

A second feeding posture was observed when the squirrels were foraging in patches of abundant perennials such as chinchweed. The squirrels did not use their forepaws but just walked slowly and some­ times even lay prone on their venters, eating the tops of the flowers in a grazing motion. Since this type of foraging kept the squirrel *s head down, it frequently lifted its head and peered around. This was also the posture when squirrels fed on ants. When feeding on flowers and fresh shoots in trees and shrubs.they again did not handle the food but used their forepaws to sustain their precarious positions.

Other ingestive motions exhibited by the round-tails were the gnawing on bark of mesquit© trees and the shredding of meat from a carcass when carrion feeding. Occasionally a squirrel placed some- . thing iia its mouth with one forepaw while remaining in a walking position.

Elimination

The Round-tailed ground squirrels did not normally interrupt other activities to eliminatee The posture differed from that of a walking stance by a slight hunch in the back. Fecal pallets were most commonly observed around the periphery of the home burrox^ hole p and . . once a squirrel was seen scratching the soil adjacent to its burrowp defecating and covering. This indicates a similarity to the elimina­ tive habits of the California ground squirrel, S, beechevi. reported by

Linsdale (1946). The habit of elimination near the burrow entrance may serve as another communicative sign to other squirrels, aiding'them to recognise the presence of occupants during cautious-exploratory behavior, '

Alertness

Ralph and Stokes (1963) described alertness In ground squirrels, as a.single continuum of activity that runs from rest to alarm, but at rest the animal is less likely to escape than at alarm. The posture exhibited in an alert behavior, probably more than any other pattern, characterises the Round™tailed ground squirrel for everyone who has observed it in its'natural state. The assumed position is a vertical stance as exhibited, with variations, by most sciurids (King, 1955$

Ralph and Stakes, 1963$ Waring, 1966$ Wolfe, 1969)e Due to the 24

possible resemblance to fence posts* the term ^pieket-pin18 has been

used to apply to many species of ground squirrels in an alert posture

(for example* Olin* 1954).

The forepays were placed in front of the chest* sometimes in

a crossed position. The tail hung to the ground and the ^mek was often

extended* as well as the hind feet* placing the squirrel on its toes

(Pig. 4). Squirrels were observed to hold this motionless* upright

"post" position* nearly 5 minutes. The head was normally turned and

the squirrel often leaned in the direction of the stimulus* which may

have afforded binocular viewing and a directed reception for olfactory

and auditory cues.

The. round-tails reacted initially with a post position to all

potential predators* non-predators (jack rabbits and cattle)* observers

and to foreign objects (traps) placed in their home range. The

squirrels assumed a down-alert position (Balph and Stokes* 1963) when

the stimulus was identified and there was no immediate danger (Fig. 4).

Visual and auditory cues of potential predators in the immediate vicin­

ity sent squirrels directly to the entrance of their home or escape

burrows where a post position was assumed.

Digging

Digging behavior was observed in both sexes often during the

. morning hours* tot excavating one-half hour before sunset was observed

on occasions. Excavating of new burrows was seldom seen* as usually

older* unoccupied burrows were widened and utilized. The digging motions

were sporadic* with the forepaws pushing the dirt under the body between 25 the hind legs, which then kicked the dirt farther back. Only freshly dug holes had dirt at the exit and then, only a shallow ridge e One squirrel spent 10 minute^ scattering a pile of dirt which had accumu­ lated at the hole entrance, McCarley (1966) found this absence of dirt around the burrow entrances also characteristic of the Thirteen- lined ground squirrel, S„ tridecemlineatus. Dirt was not observed being moved with the nose as seen in the by King (1955) <.

The digging behavior of the ground squirrels always included frequent interruptions because they raised their heads to peer around.

This behavior afforded a periodic cue with the immediate surroundings while in a vulnerable position to a predator. Violent shaking motions of the entire body after a period of digging eliminated loose dirt from the pelage. Burrow excavations showed that much loose dirt was always left in the passageways$ probably to supply material for plugging the entrances.

Grass Gathering

Grass gathering occurred at any time during the period of the round-tail®s diurnal activity. When collecting dried grasses, presum­ ably for nest material,, the squirrels combined running movements with foraging postures, picking up dried grass stems with the forepaws and placing them into their mouths. This behavior was characterized by two actions, 1) the rapidity at which the squirrel gathered, resulting in a somewhat frantic behavior and 2) the little concern for anything else while gathering. On one occasion a juvenile female, squirrel 30-80, was observed conducting only grass gathering activities for one and 26 one-half hours. In 1 hour she made 4-0 trips to her burrow carrying grass. It took her approximately 1 minute to collect a mouthful of grass and about 30 seconds were spent in the burrow after, each trip.

Some of her trips took her within 10 feet of another squirrel? but no interaction resulted. Activity was stopped shortly, however, when a hawk passed over, .

This behavior of removing materials from an area m y represent signs,: as suggested by Calhoun (1963), by which an individual might recognise the presence of neighbors. The closer the home ranges, the greater these signs will be.

Tail-flicking

Tail-flicking was a motion seen in round-tails in which the tail was moved from side to side in an arc, Balph and Stokes (1963) '■ -v reported the tail-flicking of the Uinta ground squirrel was not asso­ ciated with any specific movement or behavioral situation and suggested it may be an intention movement of a shift in the position of the animal. However, in round-tails, tail-flicking was identified with several behavioral patterns. It was most accentuated when they were confronting ground predators or as they established claims to newly occupied burrows. The latter behavioral expression consisted of the following pattern. The squirrel sat hunched forward at the newly

■ claimed hole and emitted a single peep. This was followed by a drumming of the hind legs .on the ground while the bristled tail was slashed from side to side,: A peep was emitted about every 10 seconds, and the process was repeated. Occasionally, the hind leg drumming and tail- 27 flicking continued with peeps being emitted at a rate of one per seeondc After each drumming sequence, the squirrel turned slightly to face squirrels in other directions. This behavior sometimes con- . tinned for an hour and extended over several days.

Tail-flicking was also part, of the behavioral pattern elicited by ground predators, as observed during interactions with rattlesnakes and gopher snakes. Often .between drumming sequences the squirrel dashed back and forth in.front of the snake to prevent it from enter­ ing the squirrel’s burrow.

Tail-flicking was also associated with the cautious and explora­ tory behavioral patterns. In.these instances the tail was waved very slowly. EWIRQNMEffi’AL ADAPTATIONS

Adaptation to an arid environment requires a solution to - ' problems posed by l) lack of water? 2) widely spaced food with, concomi­ tant loss of cover and 3) the extremes of heat and cold (Eisenberg,

1967)o The physiological and morphological solutions to these problems are varied among the vertebrates, as reviewed by Sehmidt-Mielsen (1964a), but all point toward the maintenance of a constant body temperature,.at least during the. active periods, and the conservation of body fluids«,

The adaptation of the heteromyid rodents to recover metabolic water in the kidney and to live on dry food alone is a most striking example0

Burrows

Behaviorally, common adaptations in desert animals are nocturnal activity and the occupancy and maintenance of subterranean burrow sys­ tems. Since the round-tails apparently lack a kidney as efficient as the heteromylds (Schmidt-Nielsen, 1964a) and are diurnally active, the burrows represent a very important adaptation for their survivale

Yorhies (1945) showed the maximum temperature in the subterranean nest of S. •tereticaudus during the summer was 29 G, -aven when the*, ambient temperature exceeded 40 C and the soil surface temperature reached 75 C e

Burrow temperatures in the pen and the study area indicated there was more than a 10 degree drop in temperature in the first 45*0 cm.(Table 3)«

All active burrows were found to be plugged at approximately 45*0 cm.

28 ■which would help maintain constant reduced, temperatures in the sub­

terranean systems*.

Table 3„ Mean burrow, temperatures at various distances from the entrance and depths and lengths of excavated burrow systems (N in parentheses).

Entrance 30e0 cm 45 e0 cm 45.0 cm ■ (10) ^6,4 G (10): 37.1 G (10) 34»2 G plugged

I'feximujn depth Maximum length ___' : .. > ___ ■ .

(14) 37.0 cm (14) 179.0 cm (25.0— 53.0) (30.0— 419 cm)

The 15 burrows excavated had a mean depth and length of 37,0 cm

and 179 cm respectively. The declination in all burrows, was very gentle

The borrow systems usually had two entrances although as many as four

were found. The mean diameter'of 23 burrow entrances was 57 mm, and

conqpas's readings indicated entrances vjere random (M^O) in relation to

the direction of the opening. Kot all entrances■(16=25) were located at

the base of$ or even hearp creosote-bushes or other shrubs. Some open­

ings $ especially those in the abandoned mounds of the Banner-tailed

kangaroo rat. Mpodomvs s-peotabilis. dropped straight down 3 to 6

inches with three smaller burrows entering laterally. . 30

No squirrels were found in any of the burrows exeavated* but

all were known to have been used during the summer. The absenee of

an occupant during excavations Was also experienced by Rongstad .

(1965) while investigating S. trideeemlineatus burrows. No nests or

excrement caches were seen in the burrow systems of the round-tail

as described for S. trideeemlineatus by Rongstad (1965). Edwards

(1946) reported seldom finding dung in nests or passageways in the

burrow systems of S. mexicanus. which had outside defecation areas..

Defecation by round-tails was observed outside of the burrows^ but

excrement was never piled. The only excavated burrow which resembled

a nesting burrow, according to depth and length as classified by

Rongstad (1965) for the Thirteen-lined ground squirrel, was that of a

juvenile male, squirrel 5, whose burrow had a depth of 91.0 cm and a

length of 244 cm. It was superpositioned by another series of holes

and burrows. Probably the most striking thing about the burrow struc­

ture of round-tails Was the large number of branches and lateral

burrows, most ending as 1 foot cul-de-sacs. Some of these could have

been plugged, hibernating burrows as found in S. tridecemlineatus by

Johnson (191?) and corroborated by Rongstad (1965).

There was no evidence suggesting that round-tails shared burrow

systems with such sympatric, fossorial species, comparable in size

(diameter), as the kangaroo rats, D. merriami and D. soectabilis. On

one occasion a released round-tail entered a D. snectabilis burrow to - ' ' ' - : ' escape. After about 30 seconds the squirrel re-emerged from the burrow,

pursued by a Banner-tailed kangaroo rat. Both re-entered another hole

of the same mound upon seeing the observers. 31

Many paths were noted connecting.groups o£ holes and often ex- . . ' . ' . tended to mesquite trees and creosote-bushes® . Captured squirrels

released on these paths usually followed them to burrow entrances6

Such.paths may furnish the squirrels the quickest route to an escape

" . - - or home burrow when confronted with a predator while feeding® Evans

(I95I) observed paths, linking the burrow entrances of S® tridecemlineatus®

To. determine the number of refuges available to a squirrel for

escape from a predator, the active squirrel holes in 12 randomly se­ lected 100 foot square quadrants were counted® A mean of 29 squirrel

holes per acre indicated that each squirrel would have access to almost

20 holes®

During the summer months the round-tails exhibited a bimodal

daily activity pattern .and thus avoided the highest temperatures of the

day® The few squirrels seen above ground during this hot period re­

stricted their movements to shaded areas, sat at the burrow openings

with frequent entries and foraged in the shade of creosote-bushes or

mesquite trees.

Hyperthermia tolerance for short periods of time was evident in

S. tereticauduSo The maximum body temperature recorded was 43.05 G

(IO9.5 P) from a captured juvenile female, squirrel 8-20® Neal (1965c,)

reported body temperatures as high a s .41 •9 C in active adults and

40®9 0 in active juveniles in June and July respectively. Captured

animals often demonstrated excessive salivation, sometimes wetting the

entire face and neck® This behavior acts to cool the anterior parts

sufficiently to protect the most sensitive parts of the central nervous 32 system and keep these below the body temperature of the remaining body parts (Schmidt-Nielsen, 1964b).

When a squirrel became hyperthermic it had to unload the accum­ ulated heat and probably did this by entering and plugging its burrow.

The behavior at this point was probably similar to that described by

Bartholomew (1968) for Ammospermophilus leucurus. with the squirrel flattening itself .against the cooler burrow floor to lose heat to the substratum by conduction and to. the walls by radiation until the body temperature approached the usual resting level. Often squirrels active at mid-day demonstrated a flattening of the body, stretching all four limbs laterally to expose the complete venter to the soil, in a shaded area above ground. This may have been a response of a slightly over­ heated squirrel in an attempt to lose heat by conduction to the ground surface.

Thus, the utilization of a burrow by 3. tereticaudus appears to play an important role in the maintenance of body temperature and the

conservation of body fluids. By allowing some fluctuation of body temperature the round-tail can integrate the surface and subterranean temperatures to an average where active heat regulation through expendi­ ture of water is unnecessary (Schmidt-Nielsen, 1964b).

Torpor

A physiological adaptation most species of Spermophilus have made to,the environment is the ability to undergo hibernation or esti­ vation. Estivation was defined by MaeMillen (1965) as torpor occurring under conditions of moderate ambient temperatures irrespective of season 33 and duration, while hibernation.is torpor, occurring under conditions

of low ambient temperature and usually over a period of several days

to weeks. Such periods of dormancy function as a means of avoiding

the most severe aspects of the environment and vary in duration.in

different species. The period of dormancy in the Thirteen-lihed

ground squirrel,. S. tridseemlineatus, is approximately 135 days in

Texas (MeCarley, 1966) while the Mohave ground squirrel. S. mohavensis

has been reported to be active from April to August then dormant for

7 months (Ingles., 1965). The latter is of especial interest because it

is the nearest relative of S. tereticaudus. being the only other living member of the subgenus Xerosoer-moohilus (Hall and Kelson, 1959).

S. mohavensis is confined to the Mohave Desert where seasonal environ­ mental changes are not as marked and precise as in boreal zones, but

during winter food is scarce and there is an intermittent drought

(Pengelley and Kelly, 1966). Pengelley and Kelly term S. mohavensis

an bbligant hibernator, having long periods of hibernation alternating with a short period of the homothermic active state. Bartholomew and

Hudson (I960) have demonstrated under laboratory conditions that 3. mohavensis can spontaneously, become, torpid, during, the summer, even in

the presence of food and water, and consider it an estivator as well as

a hibernator.

S. tereticaudus possesses most of the typical features of a

hibernator (Hudson, 1964) and has demonstrated capabilities of torpid­

ity at different months of the year under Varying laboratory conditions

(Hudson, 1964; Hudson and Bartholomew, 1964; Neal, 1964; Pengelley and

Kelly, 1966). Hudson (1964) considered the round-*tail to be only an estivator after finding occasional incidences of torpor in late spring, summer and early fall. He felt that activity above ground observed during.. the summer months'did not lend credence to the possibility of routine and prolonged periods of dormancy in any pattern other than a nightly shallow torpore Hudson worked with 8* t e chloims and S. te teretieaudua while the subspecies of this, study was S, t. negleetus.

Pengelley and Kelly (1966) found S» teretieaudua has almost abandoned - a period of obligatory hibernation, although it hibernated for short periods under laboratory conditions during October, November and

December.,. .

Observational evidence of torpidity in a natural population is virtually impossible to obtain, especially when dealing, with a fossorial animal, but repeated daily observations would indicate any continuous absence of individuals or a group of individualse In this study> dur­ ing the summer of 1968, the home ranges of 13 squirrels, were within; unobstructed view of•the observers» The daily activity of these squirrels was tallied to determine if any were absent for prolonged periods. The observers were confident that these fpiar adult males, three adult females, three juvenile males and three juvenile females were seen when abroad. The summer observation period in 1968 extended from 11 June to 28 September and included 37 days of observation with each observation period occupying 4 to 6 hours. One adult female, ' squirrel 3-60, was seen each of the 37 days of observation. The adult- females and the juvenile males and females were observed 82,7 per cent of the days with a range of 62-100 per cent while the adult males were seen only 31®5 per.cent of the days (23-41%). Some daily absences . 35 could have been due to squirrel activity at times of the day other than when the observers were present since the absences in the adult females and juveniles ;vfere. of only 1 and 2 day periods« The: absences of the adult males* however, extended from 2 to 9 days, after which they were observed within the same home range. All the adult males in this study showed a marked absentia in the observational records compared with adult females and juveniles.

This indicates that some squirrels, especially adult males, may undergo some extended durations of torpidity in the summer months*

In the Thirt.een-lined ground squirrel, S„ trldeoemlineatus. all adult squirrels became dormant in August but sometimes reappear in September for a few days, and all adult, males are hibernating before all adult females (McCarley, 1966), Thus, the prolonged absences in the adult male round-tails may have been due to an early entry into hibernation,

A habitat selected by dominant adult males, affording favorable forag" ing areas, may allow them to build up the necessary fat deposits to enter states of torpidity sooner than other squirrels. Such entries into torpidity would be highly adaptive for summer conditions since the reduction in body temperature reduces pulmonary water loss (Hudson,

1964)$ a savings especially useful during, the summer months of dry vegetation.

Other adaptations to an arid environment possessed by the round" tails have teen suggested by Dice and Blossom. (1937) and Brown (1968),

These are the extremely light pelage which protects from solar radia™ " tion and a thick epidermis modified to reduce water loss. FOOD HABITS

With the exception of a few incidental observations reported by

Vorhies (1945)# the food habits’of S. teretie.andua were virtually un­ known,, The Round-tailed ground squirrel’s maintenance of a subterran­ ean burrow system plays an important role in the conservation of body fluids and temperature regulation* but diet# especially, a succulent one* is also of great importance in maintaining water balance. The . utilization of succulent foods and rainwater, when available are two more adaptations which enable the round-tail to exist in the arid en­ vironment of the southwesto

The following description of feeding behavior is based on 173

observations and indicates the food preferences of the round-tails varied during the summer months (Table 4), Although food preferences related closely to food availability# the honey mesquite tree# Prosopis

juliflora, served as a continuous food.source in the study area. In

68 instances (41.0%) squirrels were observed feeding on some part of mesquite trees. The parts eaten included the leaves# flowers and bark

but most frequently the beans. The legumes were eaten both while on

the tree and after they had dried and fallen to the ground® The pods were opened and. the seeds were removed and eaten. There was no evidence of food being cached. The few livestock that periodically traversed the

study area Xtiere also observed'edtihg; the mesquite pods. On several

occasions round-tails were seen breaking pieces of cow manure* sometimes dragging large chunks 10 feet to their burrow entrance* then extracting

36 - Table 4 0 Food preferences noted during 166 observations of feeding round-tails.

(N in parentheses)

Food June (N=3l) July (N=?6) August (H=59) Total B3.66

Trees and Shrubs

Celtis pallida 12.0* 1,6* 6.0*

Prosopis .luliflora 41.9% 51.3* 28.8* 41.0*

Olneya tesota 6.4* 9.3* 6.7* 7.8*

Larrea tridentata 33.8* 20.0* 15.2* 21.8% ' -s '■ ' "■ ... ■ ; ■ ’ Herbs and.Grasses

Leniditm sp. 9.7* . 1.8*

. Euphorbia sp. 2,6* . 1.6* 1,8*

Plaffibbothrys arisonicus 3,2* 1.3* . 1.2*

Aplopappus heterophrllus 1.3* . 0.6*

Franseria sp. 1,3* 0.6*

Feetis papposa 1.3% 27.1* 10.0*

Animal matter

Insects 1.3* 18,6* . 7.2* and eating the seeds which had been passed by the cows„ The squirrels sometimes ate small strips of mesquite bark that were removed from the trunk and branches. The removal resulted in some bare patches measur­ ing 4 by 10 inches.

The ubiquitous creosote-bush was the second most preferred food source in the study area. Only the fuSssy-white globular fruits were observed eaten, Vorhies (1945) felt this was the only edible part of the shrub for the squirrels. The creosote-bushes were a major source ' of food for two reasons, l) Although it bloomed most profusely during the spring, it flowered throughout the year, offering a continuous food source. Often a creosote-bush completely lacking any yellow flowers or fruits would be within 10 or 20 feet of another which had numerous fruits or was beginning a fresh bloom, 2) The dense Larrea afforded the squirrel population an abundance of creosote-bushes.

Following the onset of the summer rains in mid-July, the food preference shifted markedly. The study area then supported patches of blooming chinchweed, Pectis papposa, resulting in reduced utilization of the mesquite trees and creosote-bushes as food sources. By the middle of September all the chinchweed was dried, and during August the squirrels again fed on mesquite. The chinchweed bloom also altered the normal behavior of the squirrels by causing a disregard for the individual distance of other squirrels. As many as eight squirrels were observed feeding within several feet of one another on a patch of Pectis

Mo agonistic behavior was observed as the squirrels voraciously ate the flower heads. 39

Observations on the consumption.of animal matter* .shown as insect material in Table 4$ are attributed primarily to ants and ter­ mites. On one occasion a juvenile female* squirrel 20-70* uncovered and fed on what appeared to be a eoleopteran insect* and.in several instances squirrels were seen jumping at grasshoppers in flight. As when foraging on lush patches of chinchweed* several squirrels would feed closely together when insect larvae were found in cow manure pads6

On the morning■of August $* 1968* a group of seven squirrels consist­ ing of adult males* adult females and juvenile females were feeding within several feet of one another at the base of an ironwood tree.

An investigation of the area showed hundreds of winged insects'in the bottom sides of overturned manure pads. Fecal pellets of captured squirrels often consisted almost entirely of ant remains.

In reporting the omnivorous food habits of the White-tailed antelope squirrel* Ammospermonhllus leueurus, in southern Nevada*

Bradley (1968a) found vertebrate remains in stomach samples throughout the year. Included were.species of the following rodents: Perognathus.

Dipodomys. Peromyseus and Qnyohomys. Species representing the above genera were also sympatrie in the study area with romd-tailsy but due to their nocturnal activity no interactions were observed. The liter­ ature indicates that many ground squirrels possess carnivorous tenden­ cies (Johnson* 1922; Edwards* 1946* Packard* 1958; Bridgwater and Penny*

1966; Clark* 1968), S. tereticaudus exhibited such behavior- on three different occasions when squirrels which succumbed in the traps* were left in the study area. In one instance.a juvenile female, squirrel

8-50, pulled an entire carcass into the burrow she occupied after 40 eating on the neat of the shoulder and neck regione The carcass of a Mojave rattle snake. Crotalus scutulatus. was also eaten when left near the entrance of a burrow. Such carnivorous habits probably pro- . vided the opportunistic squirrels more moisture than vegetative matter. Lizards comprised a large part of the stomach contents of

Ammospermoohilus analyzed by Bradley* but in this study no predation was observed on the lizards (Cnemidophorus tieris, Scelooorus magister and Dipsosaurus dorsalis) that -occurred in the area. Much of the vertebrate remains found in the antelope ground squirrels may have resulted from squirrels finding torpid reptiles and carrion.

S, tereticaudus. in all probability* would be as .opportunistic in similar situations* as indicated by observations of round-tails stalk­ ing and killing English Sparrows and eating portions of them (Bradley*

1968a). A similar observation was made on the Mexican ground squirrel*

S. mexieanus by Edwards (1946)„ gommmcatxon

“Central to the notion of communication is the reception of information through a stimulus that an organism perceives from the external environment11 (Marler, 1967 $769). With communication based on the transmission and reception of a various number of stimuli, the most important stimuli to which the Round™tailed ground squirrels responded were vocal, olfactory$ tactile and visual."'

Vocal

Round™tails had a vocal repertoire that included growls, hisses, squeals, peeps and trills, each associated with a different behavioral pattern.

A low, guttural growl, much like that, of -a dog, was emitted by

some individuals when in the confinement of a trap and especially when handled. A “reptile-like” hissing sometimes accompanied the growling of trapped squirrels. A curling of the upper lip, to produce a snarl, indicated this vocalization may play a role in threat behavior in intra-

specific situations as observed in the Uinta ground squirrel by Balph and Balph (1966). However, the growl was of such low volume that it would not have been perceived by us during the observed agonistic inter­ actions which were at least 75 feet away. On. one occasion, growling was heard after a. captured adult male, squirrel 1-100, entered a burrow on

the periphery of his home range. Three minutes later another squirrel

surfaced from a partially plugged hole three and one-half feet from 41 where 1=100 had entered. This growl was probably a result of the thwarting situation* as Eisenberg (1963) sometimes witnessed in heteroymids when animals were harassed by oonspeeifies or were d@= fending a nest,

A squeal was sometimes produced by captured squirrels when first approached by the observers. Such a squealing vocalization continued and intensified as the squirrel was handled for weighing$ dyeing and toe clipping. Squeals were also heard from squirrels @n~ * gaging in intraspecific fighting* especially when a subordinate indi~ vidual was being held down or bitten on the haunches by a dominant squirrel. Often the occurrence of such encounters was located by the observers when squeals were heard from fighting squirrels over 100 feet away. The subordinate squirrel only squealed during contact and not during the escape run.

The primary motif of the vocal repertoire of the Rouzid-tailed ground squirrel was a sharp* high frequency bird-like peep. It was produced by a sudden contraction of the abdominal muscles forcing air through the larynx. Variations were heard with several different be­ havioral patterns, but a single-peep was the roost common of all the vocalisations utilised by the squirrels. It appeared to indicate the presence of a foreign object in the vicinity. For example* some squirrels sat in their burrow entrances* with only their heads out* oriented toward a recently placed trap and emitted a single peep approxi­ mately every 30 seconds. The duration of this behavior varied from . ' squirrel to squirrel but normally did not last longer than 10 minutes.

Neighboring squirrels* after responding in alert behavior* appeared to become habituated to the vocal squirrel and continued ■with their

■ - • . - . * ' ■ grooming or foraging activities.

Peeps, eight to fifteen per second, produced a sound similar to the "churr" of the Uinta ground squirrel as described by Balph and

Balph (1966) and the trill of the Thirteen-lined ground squirrel de­ scribed by McCarley (1966)» Observations of the interactions between ground predators and adult female round-tails with young did not in­ dicate that the trill was used as a distress call as McCarley (1966) showed for S« trideeemlineatus. Occasionally trills were emitted when squirrels were observed tail-flicking near newly claimed holese Thus, it appeared to be used intraspecifieally as Balph and Balph (1966) found in 3. armatus. Trills were also heard from captured animals while in the trapse

; v. • ••. When a released squirrel entered a nearby hole, often several peeps were heard. Even when the squirrel was out of sight underground these., sounds sometimes continued, faintly audible.

The literature indicates how varied the vocal responses of the ground-dwelling sciurids are to predators. Fitch (1946) found S. beecheyi to have distinctive chirps that were given in response to the sight of hawks and other large birds flying in the vicinity, including

Turkey Vultures. A different call was elicited by the sight of rattle­ snakes, a 11 snake-warning" chirp that could be heard 700 yards away 1

Waring (1966) reported that the vocal repertoire of , Marmota flaviventris. did not seem- to include an aerial predator warning. This corroborated the earlier finding by Armitage (1962) that M. flaviventris emits no alarm calls to Ospreys. The Arctic ground squirrel, S. 44 uadulatns. emits a single v/histle upon the approach of an aerial pre­ dator and repeats it at 6 to 8 second intervals if the predator alights nearby (Melchior, 1968)„ A trill is used by the Thirteen- lined ground squirrel$ S„ tridecemlineatus. during predator inter­ actions (McCarley$ 1966), and the W n t a ground squirrel S. armatUSc

■which possesses a varied repertoire of sounds (Balph and Balph, 1966), responds to airborne predators with 11 chirps11 and ground predators with

"ehurrs”. The Columbian ground squirrel, S, columbianus, voices an alarm call when potential enemies- approach, which is then passed on by other squirrels. The calls become louder and faster until the squirrel takes refuge in a burrow (Manville, 1959).

With only two exceptions (N=5l), there were no vocalizations by round-tails as a warning for ground predators, and no vocaliza­ tions were ever heard when squirrels responded to aerial predators

(N"23)« Visual communication by way of the post posture in alert be­ havior appeared to be the primary means by which squirrels were warned of aerial and ground predators. As the 11 snake-warning" chirps of the

California groimd squirrel aided Fitch .(1948) and field workers to locate rattlesnakes, the post posture of the round-tails forewarned the observers in this study; to the presence of hawks and snakes. Ap­ pearing almost as a simultaneous movement, all of the squirrels above ground would assume a post posture and peer in the direction of the predator. The same response was seen when Coyotes were heard in the distance. Squirrels appeared to differentiate between Turkey Vultures and hawks,.as the former elicited no response as they circled the study area. On.one occasion a vulture alighted on the ground, an action 45 which would have sent all the proximal squirrels underground had it been a hawk# however, they appeared undisturbed.

Since observations were conducted near ground level, it was easy to see the effective uses of the alert response as a communi­ cative signal. Squirrels in the immediate vicinity of a ground predator retained their post posture, sometimes emitting periodic peeps, whereas the more distant squirrels resumed their previous activities but frequently looked in the direction of the stimulus.

No sound communication was observed in epimeletic behavior

(parent-young) of round-tails nor was any tooth chatter heard as de­ scribed for Marmots by Waring (1966),

Olfactory

Frequent.nose-mouth and nose-rear contacts between scrotal males and adult females in the enclosure were observed. This was prob­ ably olfactory testing as seen in the courtship behavior of the

Thirteen-lined ground squirrel described by MeGarley (1966) and the

Uinta ground squirrel seen by Balph and Stokes (1963), During sexual behavior in the spring, olfactory cues probably play a very important role.

In cautious and exploratory behavior the squirrels did a great deal of sniffing, especially at the entrances of unfamiliar burrows.

Through maintenance behavior the squirrels left various signs in the form of fecal pellets, dusting areas and branches which they had rubbed, which could serve as olfactory stimuli to indicate their presence. Ol­ faction may be an important means of communication in the burrow system. 46

Tactile

As with olfaction,, the tactile stimuli to which the ground

squirrels reacted were difficult to observe® The tactile communica­

tion observed in the round—tails included the contact of wrestling

juveniles during play and intraspecific fights® The squirrels un­

doubtedly made much greater use of tactile stimuli while underground«,

Tactile vibrissae would be important to the round-tail$ which is

virtually blind in the labyrinth of burrow passageways, due to the

possession of a highly adaptive all cone eye (Walls, 1942)„

Visual

Vision appeared.to play the most important role in communica­

tion among the Round-tailed ground squirrels, as it preceded or ac­

companied all the other forms of stimulation previously mentioned® A visual perception of other ground squirrels seemed to afford the indi­ vidual some sense of security after periods of disturbance by predators®

Sciurids, except Glaucomys. are distinguished from other ro­

dents by a pure cone.retina (Walls, 1942), and some have been shown to

possess color vision .(Creseitelli and Pollack, 196$) ® Other visual

adaptations possessed by ground squirrels, as discussed by Walls

(1931, 1942), indicate the importance of vision in many behavioral

patterns, A yellow, spherically shaped lens acts as an intra-ocular

filter providing an immunity to dazzlement and less involuntary eye movement. These characters allow a squirrel to tolerate strong light

intensities without blinking and to perceive moving objects while not moving the head, when in a flattened posture or a signaling, post position® 47

It appears that the visual capabilities of round-tails serve

as the primary defense against both aerial and ground predators. Un­ like many animals that use secretive:actions to escape predators* the

ground squirrels possess behavioral patterns that afford good visual,

coverage of the environment and neighboring squirrels. This adapta­ tion in S. tereticaudus appears much like that seen in the Uinta .

ground squirrel as elaborated by Balph and Balph (1966), The open habitat of the Larrea zone promotes such behavioral patterns by pro­ viding unobstructed viewing at ground level. COMPETITIVE.AND PREDATOR! FACTORS

The reptiles, birds and mammals that occurred or were observed in the study area are those characteristic of the Lower. Sonoran Larrea

zone (Table $). The marked exception was the Porcupine, Brethizon. dorsaturn, which inhabited a desert wash on the northern boundary of the area. Occurrences of Porcupines in the desert-shrub of southern Ari­

zona are not uncommon, see Reynolds (1957)» No predation on ground

squirrels was observed during the study, but the recognition of poten­

tial predators was based on the behavioral responses of. the squirrels

and on known instances of predation on other species of ground squirrels

or small mammals as reported in the literature (Fitch, 1948j King, 1955;

Peterson, 1961; Willis, 1963; McGarley, 1966; Stebbins, 1966),

Predators

Potential predators were observed in the area on 53 occasions,

40 per cent of the observations were on Red-tailed Hawk activity. The

appearance of a hawk in the area, either circling above or passing low,

immediately elicited an alert response from every ground squirrel.

Alerted squirrels were sometimes over 500 feet apart. After assuming

a post posture, each squirrel ran immediately to its nearest burrow and

posted again. The post or down-alert posture was held by the squirrel

at the burrow entrance until the predator left, A low flying aerial

predator in the immediate vicinity, however, sent the animals'-"diving

underground without a second look. Squirrels foraging in trees and

' ■ ■■ 43 . Table 5<> Vertebrate fauna occurring or observed in the study area0

Squirrel behavior and literature accounts are the basis for the denotation (*) of poten­ tial predators. Vernacular and'scientific names for reptiles* birds* and mammals follow Stebbins (1966)* Phillips* Marshall and Monson (1964) and Gockrum (I960) respectively.

REPTILES .BIRDS '

Dlpsosaurus dorsalis. Desert Iguana Cathartes aura-, Turkey Vulture

Sceloporus magister. Desert Spiny Lizard *Buteq iamaieensis. Red-tailed Hawk

CEsmidophoroua tigris. Western Whiptail *Buteo albonotatus. Zone-tailed Hawk

*Masticophis flagellum plceus. Western Black Racer *Falco sparverius. Sparrow Hawk

*Pituophis melanolencus. Gopher snake Lpphortyx gambelli, Gambsl’s Quail

*Crotalus scutulatus. Mojave rattlesnake Zenaidura -macroura. Mourning' Dove

Zenaidura asiatica. White-winged.Dove

*Geococoyx: californicus. Roadruziner

Golaptes auratus, Gilded Flicker

Centurus uropygialis, Gila Woodpecker

' - ^Toxostoma bendirei, Bendire's Thrasher-

Toxostoma curvirostre. Curve-billed Thrasher

Lanius ludovicianus. Loggerhead Shrike . Table 5» continued

MdJ'imLS

Lenng alien!» Allen8 s antelope jack rabbit

L e m s californicus„ Black-tailed, jack rabbit

Sylvilaeus audobonl, fesert cottontail

SneriGoohilus tereticaudus. Round-tailed ground squirrel

Perognathus' ampins. Arizona pocket niouse

Perognathus baileyif Bailey8s pocket mouse

Perognathus penicillatus. Desert pocket mouse

Dipodomys speotabilis. Banner-tailed kangaroo rat

Dipodomys merriami. Jferriam's kangaroo rat

Onychomys torrldua. Southern grasshopper mouse

Peromyscus eremicus. Cactus mouse

Erethizon dorsatum. Porcupine

*Cania latrans. Coyote

*Tazldea taxus. Badger ! K

. 51 shrubs also ran to their burrows. This response to aerial predators was always the same and invariably, no vocalisations were heard.

Usually the hawks only circled above, but on occasion they passed through the area 1 to 2 feet above the ground, alighted in a tree momentarily, then glided low for several hundred feet in another dir­ ection, Squirrels that had to run greater distances to an escape burrow sometimes flattened their bodies at the burrow entrances rather than posted. The light pelage of the round-tails .blended with the background soil, and the posture was apparently an attempt to escape detection. As MeOarley (1966) observed in S. tridecemlineatus. the flattened posture makes a motionless squirrel almost undetectable.

All ground predators elicited the same alert reaction of stand­ ing upright, then running to the burrow entrance. When a snake was in the immediate vicinity, squirrels closest to the snake violently flicked their tails from side to side and drummed their hind legs at their burrow entrances while facing the snake. If the snake continued to move in their directionthe squirrels dashed back and forth several feet in front of it, as if to prevent it from entering a hole. They ran to within a foot of its tail but never touched it. This behavior, similar to that described for S„ beecheyj by Pitch (1948)$ continued until the snake entered a burrow. Often, several neighboring squirrels were seen around a snake, indicating that the immediate danger repre­ sented by a predator pre-empted concern for individual distances. One female, squirrel 20-70, rapidly left her burrow within 20 feet of the observers followed immediately by a Black Racer, MagtieotMs flagellum. 52

She proceeded to harass the snake -without any apparent concern for our presence*

Potential reptilian predators were observed in the study area on 14 occasions, 50 per cent of the observations were on Mojave rattle™ snake(a). Grotalus scutulatus. Snakes were seen coiled near burrow entrances early in the morning and a strike was elicited from a

Gopher snake when a caged squirrel was placed near it* Sudden close™• quarter, encounters with snakes in burrows must have often occurred, putting the ground squirrel at a tremendous disadvantage*

A Bendire6s Thrasher was observed fighting with an adult squirrel as it jumped in the air with wings spread and feet turned toward the squirrel* On another occasion an adult squirrel ran be­ tween a juvenile and a pursuing Thrasher*

The round-tails always demonstrated an alert behavior toward . passing Roadrunners, and one chased a squirrel around the base of a desert hackberry several times while making stabbing motions with its bill. MeCarley (1966) observed roadrunner predation on the Thirteen- lined ground squirrel.

Coyotes and Badgers are also considered in predator roles,, as their extensive diggings, seen periodically in the area,, indicated that ground squirrels could have been caught in blind-ending burrows*

The squirrels did not appear to object to the presence of the jack rabbits, cottontails and cattle which on occasion passed through the area. The nearest squirrels posted momentarily then resumed ac­ tivities once the object was recognized as harmless. 53

The round-tails were subject to predation at all times, and it appeared that when above ground they guarded against it very effec­ tively, Their population density and visual communication provided a great number of eyes to perceive predators and alert neighboring squirrels„ During above ground activity high numbers were probably a basic element of survival* This effective guard against aerial and ground predators indicated that predation is probably greatest by reptiles in the burrow systems.

Although, the amount of predation in the study area was not known, observational evidence of very low reduction in the population indicated that it must have been small while the study was underway.

Due to the numerous observations of predators in the area, especially the hawk activity, it is felt that the.observer's presence did not keep predators away.

Competitors

Due to the nocturnal activity of the syapatrie rodents in the area, interspecific encounters were not observed. No evidence was found to indicate active round-tail burrow systems-were shared with other rodents| however, all three, species of lizards occurring in the area were observed utilizing active squirrel burrows. Squirrels did use burrows in the abandoned mounds of the Banner-tailsd kangaroo rat,

Dipodomys spectabilis. but they only retreated to active burrows of this kangaroo rat out of the necessity to escape. Holdenried (195?) found the Spotted ground squirrel, S, spilosoma. to inhabit active as well as old burrows of D. spectabilia. 54

Some food eompetit-ion would be expected^ but the speraopbilio pocket mice and'kangaroo rats probably fed very little on the succulent plants so important to the round-tail. The food habits of D, spectabilis and D, merrismi Indicate both kangaroo rats gather primarily the seeds from grasses (Vorhies and Taylor, 192% Reynolds, 1958)6

These studies also found that mesquite beans were highly preferred food, but the lack of arboreal capacities prevented the rats from harvesting the succulent beans, Reynolds (1958) felt that the round-tail did not seriously compete with D , merrxami for food. BEHAVIOR IN THE ENCLOSURE

To observe more intensely the behavioral patterns associated with the encounters of adult and juvenile male and female Roundstailed, ground squirrelss a squirrel enclosure constructed at the University8s

Desert Biology Station was utilized. Squirrels were placed in the

70 by 80 foot enclosure each summer, in June, and observations 3 days a week, during the hours of peak activity, continued into August. The following behavioral patterns are based on observations of nine adult males, seven adult females and three juvenile mles.

The preparation of the genital, tract for breeding in round-tails apparently occurs during hibernation, as Neal (1965b) found motile sperma­ tozoa in the oauda epididymides of scrotal males in early January. He found live spermatozoa in squirrels until 8 April then the testes began - - R-i-' regressing about mid-April. However, captivity delayed the regression of the male reproductive organs and captive males remained sexually active until the end of August. It was not reported if females were examined to see if this was also true.

Some of the squirrels that were to be placed in the enclosure were collected in the spring, kept in captivity and released in the enclosure in June, At this time- some of the adult males had retained a scrotal condition, with the testes in the scrota. This suggested that the level of gonadotrophins was relatively high. They exhibited sexual behavior toward adult females and an emphatic agonistic behavior toward other squirrels, especially juvenile males.

55 56

The sexually active males consistently ran toward the femalea$

stopped short of them, then extended their necks and began sniffing, the mouth region of the females« Is a male approached# its tail yas usually stiffened in a vertical position while the female slowly flicked her tail in a vertical motion* The male sniffed and licked her mouth and at this point the female usually retreated# running to the edge of the enclosure or into a burrow. When the male approached more slowly# a prolonged oral contact resulted before the male sniffed and licked, the rear, of the female. The male then grabbed the haunches of the female with, his fore legs and attempted eopulatory movements# but the females invariably broke away and ran.

This suggested that the round-tail copulation occurs above ground as in the Thirteen-lineti ground squirrel (MoGarley, 1966) and not below ground as is suspected for the Uinta ground squirrel by

Balph and Stokes (1963). They also found that in the Uinta ground

squirrel a slow approach by the male and his associated posture appar­ ently has an "appeasing" function# with the female permitting contact.

It was not known if the unreceptive behavior of the females was due to a low gonadotrophin level or the confining conditions of the enclosure. However# the labia of the genital opening were not enlarged as Neal (1964.) observed during the breeding season.

The ground squirrels placed in the enclosure dug their burrows# not differing in structure from those in the study area, and used sep­ arate holes, although some holes entered a common burrow system. The general pattern of the agonistic behavior was similar to that observed in the study area but much more intense. The adult males in a scrotal 57 condition -were dominant over the adult females and juveniles6 One scrotal male, squirrel 1-40, dominated all the other squirrels sharing . the enclosure, including another scrotal adult male. This male harassed juvenile squirrels to the extent that it dug at the entrance of a hole into which a juvenile had been chased, eliciting squeals from the subordinate animal, When the juvenile was caught away from the hole, 1-40 prevented it from re-entering.

Encounters between 1-40 and.a juvenile male, squirrel 3-90, ' were observed on June 25, 1968, Intense agonistic behavior was.ex­ pressed by the dominant 1-40 throughout the day. At 0950 hours 3-90 was observed being harassed by 1-40 which chased and bit it whenever it attempted to enter a hole. Each encounter brought loud squeals from

3-90 which then retreated to a shaded area in the pen or attempted to climb the enclosing fence. Periodic fighting continued throughout the day, resulting whenever 3-90 attempted to enter a hole. At 1927 hours,

3-90 was observed lying on its side near a wall of the enclosure, its right hind leg quivering and its eyes half closed. When I approached, it crawled 4 feet, dragging its hind legs, then rolled on its back and turned all four feet upward in a’defensive posture. It-allowed.itself to be handled and appeared to be in a state of complete exhaustion demonstrating a very rapid, deep respiration. This condition apparently resulted from the constant dominating harassment from 1-40 and an ex­ posure to the high ambient temperatures which had reached a maximum of .

41.8 C. This squirrel was removed from the pen, caged and fed lettuce - leaves and sunflower seeds for several days until it recovered. 58

On June 28, 3-90 was again placed in the enclosure,, This time it was successful' in excavating and occupying an abandoned squirrel hole^ as the adult male, 1-40i> was not seen until the following day0 .

At this time 1-40 was no longer in a scrotal condition, and no agonistic behavior toward 3-90 was expressed® Even when 3-90 remained abroad foraging for 30 minutes it was not chased by 1-40,

The above observations are probably indicative of the extreme intolerance adult males have toward other squirrels, especially sub­ adult males, during the height of sexual activity. Such behavior may affect the breeding population by limiting the breeding role of in­ experienced males (Brown, 1966)* ACTIVITY PATTERNS

Amual

Round"*tails were active above ground through. September, but periodic observations in the study area during October, November and

December indicated the;squirrels were inactive these months. Pengelley and Kelly (1966) found that under laboratory conditions the. round-tails went into a state of torpor (hibernation) in October, November and

December, but due to frequent interruptions of the hibernating state they termed S. tereticaudus a poor hibernator. They indicated that the round™tail has almost abandoned a period of obligatory hibernation and retains only traces of an endogenous circannian rhythm. Because this frequency of arousal is a function of the ambient temperature (Pengelley and Kelly, 1966), some squirrels might be observed above ground during unseasonably warm periods in October, November and December, This is corroborated by observations of S. tereticaudus active above ground in his study area at the end of November and early January (Neal, 1964)«

Neal (1964) discussed the reproductive cycle of the round-tail and found males were reproductively active in early January but breeding did not begin until mid-February.

Mlz AsMsiii

Daily activity patterns of the round-tail were studied during the 1967 season. A surveillance of 10 acres in the study area, un­ obstructed from the observer’s view, was made each one-half hour to

59 60 tally the number of squirrels above ground. These surveys were con­ ducted on various days to include every daylight hour, and at the time of each survey general weather conditions were noted. These were later combined with recorded relative humidity and ambient temperature readings.

S. tereticaudus is a diurnal animal and may be above ground during any daylight hour, but the number of active squirrels and the amount of activity was influenced by the ambient temperature. During the summer months, there was a bimodal pattern of activity (Fig. 5), whereas observations in the fall and spring indicate the activity peak was at mid-day. Figure 5 represents the general activity pattern of the round-tails based on the mean number of squirrels active at the time the surveys were taken on 23 random days from June 28 to August 5,

1967. The ambient temperatures are the mean temperatures for the hours the survey was taken.

During the summer the squirrels were not active at sunrise, the coolest time of day, but rather activity began approximately 1 hour after and peaked 3 to 4 hours later. Mot all squirrels emerged from their burrows at the same time, and even sibling juveniles sometimes emerged in intervals separated by as much as 30 minutes. Some individ­ uals consistently emerged earlier than others. Upon emergence, most squirrels spent 10 to 15 minutes at the entrance of their burrows, either sitting outside or with just their heads out. Once active, for­ aging was the main activity while grooming and burrow excavation were secondary. Squirrels often moved throughout their entire home ranges during the morning activity, and they sometimes investigated the main Lumber of Squirrels 40 Fig. 5. Mean number of round-tails observed above ground and mean ambient temperatures duringand ambientground temperatures mean above observed of round-tails number Mean 5.Fig. sunrise See textdiscussion.for See surveys.and August July June, min temperature ambient squirrel number

Hoursof Day TSSO sunset

40

Temperature in Decrees Centigrade Decrees in Temperature & 62 burrow entrance of inactive neighboring squirrels without consequence,

However, when neighboring squirrels were also above ground, such tres­ passing resulted in chases*

Squirrel activity began dropping off 3 to 4 hours after sunrise and continued to decrease until late afternoon* On most of the cloud­ less summer days the ambient shade temperature reached 38 G by 1200 hours while temperatures in the sun exceeded 42 0, and the relative humidity dropped below 30%c Surface temperatures of the soil recorded as high as 62 G during the peak of the afternoon heat. Through the afternoon hours above ground activity all but completely ceased.

Squirrels seldom left the immediate vicinity of the main burrow entrance during the afternoon activity, but an occasional animal was seen on a short run from one shade source to another. The squirrels that were out during the afternoon hours consistently returned to their burrows with­ in 5 minutes and re-emerged after 10 to 15 minutes^ It is during these hours that the squirrels would have to be most concerned with maintain­ ing an optimum body temperature as described in the section on burrows.

The highest temperatures of the day peaked in the summer months at 1500 to 1600 hours. An increment in squirrel activity became notice­ able 2 hours later and continued until dark. Most squirrels had termi­ nated surface activity one half hour before sunset, but occasionally some were active until dark, although they remained near their burrow entrances. This late cessation contrasted with the delayed onset of morning activity. The evening activity of a squirrel primarily in­ volved movement throughout the home range and foraging but seldom grooming and- digging. 63

In late July and throughout August cloudy afternoons became numerous and the squirrel activity increased. However? this increment in the morning, afternoon and evening activity correlated with the ambient temperature and not the cloud cover. On days following a heavy rain, the activity was noticeably increased, even under clear skies, when the temperatures were unusually low,

A strong wind delayed the onset of activity at the beginning of the day, but a wind arising later did not appear to send the squirrels underground. Rainfall sent squirrels below ground, and the onset of activity was delayed following a night of heavy rains and low tem­ peratures. POPULATION DYNAMICS

Population Structure and tensi