THE PARASOL TAIL AND THERMOREGULATORY BEHAVIOR OF THE CAPE GROUND XERUS INAUIUS'

ALBERT F. BENNETT, RAYMOND B. HUEY, HENRY JOHN-ALDER, AND KENNETH A. NAGY School of Biological Sciences, University of California, Irvine, California 92717; Department of Zoology, University of Washington, Seattle, Washington 98195; Department of Developmental and Cell Biology, University of California, Irvine, California 92717; Department of Biology, University of California, Los Angeles, California 90024 (Accepted 6/2/83)

The Cape , Xerus inauris, a diurnally active inhabitant of hot arid regions of , reduces environmental heat load by using its bushy tail as a portable parasol. During late spring in the southern Kalahari, operative envi- ronmental temperature (T,) was measured on thermal mannequins of differing posture and tail orientation. Simultaneously, the thermoregulatory behavior of a free-living colony of was observed. The tail is raised to shade the body and the squirrels turn their backs to the sun whenever T, exceeds approximately 40 C. The shade provided by the tail reduces T, by over 5 C and allows squirrels to extend greatly their periods of continuous surface foraging (up to 7 h instead of 3 h). At midday, squirrels must frequently retreat to their burrows as T, of tail-shaded then also exceeds 40 C.

INTRODUCTION during the hottest period of the day, thereby Most small living in deserts greatly increasing its foraging time. are nocturnal and thereby avoid exposure Cape ground squirrels spend their day to heat during the hottest times of the day. foraging for small plants and seeds around Nevertheless, a few species, such as ground their burrow entrances (Zumpt 1968; Her- squirrels, are diurnal. Living in open areas zig-Straschil 1978). They dig large burrow on the desert floor, they may be exposed systems with multiple entries and tunnels to intense environmental radiation, par- that usually are 60-70 cm below ground ticularly solar radiation, and high soil tem- surface (Zumpt 1970; Herzig-Straschil peratures. Here we report observations on 1978). These ground squirrels are very so- the thermoregulatory behavior of the Cape cial, with generally six to 10 individuals ground squirrel (Xerus inauris), a species inhabiting one colony (Nel 1975; Herzig- widespread throughout the arid regions of Straschil 1978). In the Kalahari semides- southern Africa (Smithers 197 1; Herzig- ert, the site of our study, these colonies are Straschil 1979; de Graaff 1981). This usually located in open pans, river beds, squirrel reduces its environmental heat load or river banks (Smithers 1971; de Graaff by using its bushy tail as a portable parasol 1981). Because these areas are generally without shade, foraging animals are often We are indebted to G. de Graaff and Warden E. exposed to very high ambient temperatures. LeRiche for their support, assistance, and permission to conduct these studies. We thank W. D. Haacke, The most conspicuous anatomical fea- M. Haacke, C. K. Brain, and M. G. L. Mills for ture of this is its large, bushy tail, assistance; M. Chappell and T. Bradley for comments which is dorsoventrally flattened and nearly on the manuscript; and R. D. Stevenson for discus- as long as its body (Herzig-Straschil 1978; sion. This research was funded by National Science de Graaff 1981). The tail gives rise to the Foundation grants DEB 81-09667 and PCM 81-02331, and National Institutes of Health Grant KO4 animal's common name in Afrikaans: AMOO351. The Datsun-Nissan Co. (T. Dreyer, pro- "waaierstert " or "fan-tailed meer- motions manager) generously provided a four-wheel kat" (Synman 1940; de Graaff 1981). Its drive vehicle for our field research. use for creating shade for the squirrel has Physiol. 2001.57(1):57-62. 1984. been noted previously (Powell 1925; O 1984 by The University of Chicago. All Smithers 1971; Marsh, Louw, and Berry rights reserved. 0031-935X/84/5701-8341$02.00 1978; Herzig-Straschil 1978, 1979). A sit- 58 A. BENNETT,R. HUEY, H. JOHN-ALDER,AND K. NAGY ting animal holds its tail erect behind its skins of two adult female squirrels on back; when horizontal, animals hold their frames fashioned from baling wire. The tails over their bodies (fig. 1). The squirrels mannequins were the same size and shape often orient their backs to the sun, thus as live squirrels and were modeled into a completely shading their bodies from di- sitting position, the most common posture rect solar radiation. Changes in tail posi- of foraging squirrels. Thermocouples were tion and orientation appropriate to ther- mounted in the thoracic region of the hol- moregulation at different times of day have low internal cavity, and temperatures were also been noted (Marsh et al. 1978; Herzig- monitored with a Wescor thermocou~le Straschil 1979). We have made the first thermometer. The mannequins were thor- measurements of the actual effectiveness oughly dried and placed in full sun in an of the tail as a heat shield in reducing en- area directly adjacent to the observed col- vironmental heat load on these animals and ony. During the middle of the day, tail po- have correlated changes in patterns of be- sitions of the models were exchanged: tem- havior with specific levels of operative en- peratures of the mannequins rapidly vironmental temperature. reversed and attained the same level as the other mannequin prior to the change. When MATERIAL AND METHODS both manneauins were in the same tail A colony of at least 11 Xerus inauris was posture and position, their T,'s were less observed during late November (spring) than 1 C different. 1981 in the dry bed of the Nossob River The tendency of squirrels to orient their in Kalahari Gemsbok National Park on the position (front, back, right side, or left side) border of the Republic of and with respect to the sun was computed us- the Republic of (12 km north- ing two-tailed x2 tests (adjusted for con- east of Twee Rivieren, R.S.A.). The num- tinuity), assuming an expected ratio of 1:3 ber of squirrels active and their posture for squirrels orienting randomly to the sun. and tail position and orientation with re- Probability of nonrandom orientation is in- spect to the sun were surveyed with bin- dicated in the text. oculars every 15 min from sunrise to sunset on a 400 x 110-m quadrat. Animals in RESULTS transit are excluded from reported census Observations were made on a warm, dry, data. The area had no trees or other ob- and cloudless day, typical for late spring jects to provide shade and was exposed to in the Kalahari. Air temperature in the full sun throughout the day. Soil temper- shade at 1 m above ground level rose from atures at the ground surface and 10 and 15 C at 0600 hours to a maximum of 33 C 50 cm below ground were measured every at 1530 hours (local time); intermittent 30 min with buried thermocouples and a breezes blew from the south. Soil surface Wescor thermocouple thermometer. temperatures exceeded 50 C from 1100 to Operative environmental temperature 1600 hours, peaking at 61.7 C at 1230 (T,) (Bakken 1976, 1980), an integrated hours, but temperatures at burrow depth measure of radiant and convective thermal remained at 2 7 C for the entire day. input to an organism, was measured with The first squirrel emerged at 0748 hours pelt-covered taxidermic mannequins local time, well after sunrise (0549 hours). (Bakken et al. 1981; Chappell and Bar- The number of active animals, their tail tholomew 1981~)during the squirrels' .ac- position, and temperatures of the man- tive period. The T, is the equilibrium tem- nequins at each census period are shown perature a squirrel would attain if it lacked in figure 2. The daily behavior pattern of metabolic heat production and evapora- the squirrels can be separated into a tem- tive water loss. Thus T, serves as a mea- poral series of five relatively discrete pos- sure of environmental heat loads. Two tures: tail down on soil surface, tail up on mannequins were constructed, one in the surface, tail up and burrow shuttling, tail tail-up (shaded) position and the other in up on surface, and tail down on surface. the tail-down (unshaded) position. Man- From the 0800 to the 0915 hours census, nequins were prepared by mounting the squirrels foraging on the soil surface had FIG. 1.-Use of the tail as a sun shade by the Cape ground squirrel (Xerus imuris). Top, The erected tail covers the entire dorsal surface of a sitting animal. Bottom, The tail is held erect over the back of a horizontal squirrel. Note the shading of the body and head. tail down, tail up, tail up, tail down, tail up, 4 *4 On *4 entering burrows $4 on *4 on * A. 1 2, surface surface surface surface

Time of Day

FIG. 2.-Thermoregulatory behavior and operative environmental temperature in the Cape ground squirrel. A, The number of individuals active above ground at each 15-min census interval. Open areas indicate animals with tails down; shaded areas, individuals with tails up over their backs. B, Operative environmental tem- peratures of two thermal mannequins of ground squirrels. 0 = the model with the tail down, 0 = the model with the tail up. The shaded band indicates the approximate range of ambient temperatures that begins to elicit a rise in body temperature and salivation in this species. their tails down, and their orientation with cause temperature in these burrows was regard to the sun was random (14 of 34 only 27 C, environmentally acquired heat observations with back to sun, P > .05). loads could thus be unloaded in the cool Between 0915 and 0930 hours, the T, of burrow tunnels, a situation very similar to the tail-down (unshaded) mannequin ex- that observed in the antelope ground ceeded 40 C. Temperatures in this range squirrel, Ammospermophilus leucurus in still air impose severe heat loads on this (Bartholomew 1964; Chappell and Bar- species: exposure of these squirrels to 40- tholomew 19816). Foraging bouts between 41 C in a dry, controlled temperature cab- 1130 and 1530 hours averaged only 9.6 min inet causes a rise in body temperature from in duration (n = 24; SE = 0.92). During a normal value of 37.9 C and profuse sal- surface activity at midday, tails were usu- ivation (Bolwig 1958). Field behavior ally held erect. changed dramatically at this time: squir- When T, of the tail-up mannequin de- rels erected their tails over their backs and clined below approximately 40 C during sat with their backs toward the sun (46 of midafternoon, the squirrels stopped shut- 58 observations between 0930 and 1130 tling in and out of their burrows and re- hours, P < .001). This posture and tail mained on the surface. They again ori- orientation reduced T, by 6-8 C (fig. 2). ented with backs toward the sun (46 of 50 By the 1130 hours census the T, of the observations, 1530-1730 hours, P < .001) tail-up (shaded) mannequin also exceeded and continued to shade their backs with 40 C, and even squirrels with tails up thus erected tails. At 1730 hours, T, of the tail- encountered high heat loads. During this down mannequin also declined below 40 period, squirrels began to disappear from C. The squirrels soon began to lower their the surface (fig. 2) and to shuttle to and tails, and their orientation with backs to from their cool burrows, emerging for only the sun was much less striking (10 of 20 brief foraging bouts on the surface. Be- observations, 1800-1900 hours, P > .05). BEHAVIORAL THERMOREGULATION IN A GROUND SQUIRREL 6 1 The sun set at 1904 hours, and the last ally in Cape ground squirrels. They forage squirrels entered burrows at 1911 hours. actively throughout the day during the winter (Herzig-Straschil 1979), engage in DISCUSSION burrow-shuttling behavior at midday dur- The tail of the Cape ground squirrel is a ing the spring, and remain underground highly effective parasol, significantly re- during the hottest parts of the day in the ducing environmental heat loads. Tail el- summer (Bolwig 1959). The suite of ther- evation decreased T, by an average of 5.6 moregulator~adjustments outlined here C between 0930 and 1730 hours (maximum (parasol tail, sun orientation, and burrow decrement = 8.3 C) (fig. 2). shuttling) greatly increases time for for- The activity and behavior of the squir- aging and social behavior for this species. rels are highly correlated with environmen- But environmental heat loads during the tal heat load. Squirrels show no orientation hottest parts of the year are apparently too to solar position and do not erect their tails great for these behaviors to be completely early in the morning and late in the after- effective. These behavioral and morpho- noon when T, is low. When T, exceeds logical adjustments can lessen but not about 40 C, they erect their tails and turn eliminate thermoregulatory difficulties in their backs to the sun, reducing heat input the intensely hot deserts that these squir- considerably. This behavioral use of the rels inhabit. They reinforce the notion that tail as a heat shield more than doubles the the behavior and morphology of organisms time available to squirrels for continuous (Bartholomew 1964; Porter and Tracy surface foraging, adding 4 h to the 3 h 1974), not just their physical environment, sustainable with tail-down behavior alone must be considered in analysis of adap- (fig: 2). The use of the parasol tail, sun- tations of organisms to climate. orientation behavior, and burrows as ther- This is the first study to report simul- mal retreats should also considerably re- taneous measurements of operative envi- duce water loss by these animals, which ronmental temperatures and continuous rarely have access to drinking water. Shad- observations of thermoregulatory field be- ing by the tail and burrow shuttling are havior. The correlations between T, and also used by the antelope ground squirrel, behavior in Xemcs inauris are very striking: Ammospermophilus leucumcs, during hot sun orientation, tail carriage, and timing weather (Hudson 1962; Bartholomew 1964; of surface activity appear to be highly pre- Chappell and Bartholomew 1981a, 1981 b). dictable with reference to this single mea- The tail of Ammospemzophilus is less bushy surement. T, is an integrated measure of and is much shorter, extending less than heat load that can easily be estimated in half the body length and reducing T, by a remote field situations with minimal maximum decrement of only 3 C. These equipment (Bakken 1980; Bakken et al. observations and those of Golightly and 198 1; Chappell and Bartholomew 1981). Ohmart (1978) provide empirical support We anticipate that the use of thermal man- for the suggestion (Muchlinski and Shump nequins will provide many opportunities 1979) that the sciurid tail may be an im- in natural contexts for the interpretation portant thermoregulatory organ, not only of thermoregulatory behavior as well as for in cold but also in hot environments. the documentation of environmental re- Diurnal surface activity varies season- striction of activity.

LITERATURE CITED

BAKKEN,G. S. 1976. A heat-transfer analysis of an- erative temperature affecting small animals. imals: unifying concepts and the application of Ecology 62:s 1 1-3 18. metabolism chamber data to held ecology. J. The- BARTHOLOMEW,G. A. 1964. The roles of physiology oret. Biol. 60:337-384. and behavior in the maintenance of homeostasis -. 1980. The use of standard operative tem- in the desert environment. Symp. Soc. Exp. Biol. perature in the study of the thermal energetics of 18:7-29. birds. Physiol. 2001. 53:108-119. BOLWIG,N. 1958. Aspects of animal ecology in the BAKKEN,G. S., W. A. BUTTEMER,W. R. DAWSON, Kalahari. Koedoe 1: 1 15-135. and D. M. GATES. 1981. Heated taxidermic . 1959. Further observations on the physio- mounts: a means of measuring the standard op- logical and behavioral characteristics of small an- 62 A. BENNETT, R. HUEY, H. JOHN-ALDER, AND K. NAGY

imals in the southern Kalahari. Koedoe 270-76. The sciurid tail: a possible thermoregulatory CHAPPELL,M. A,, and G. A. BARTHOLOMEW.1981a. mechanism. J. . 60:652-654. Standard operative temperatures and thermal en- NEL, J. A. J. 1975. Aspects of the social ethology of ergetics of the antelope ground squirrel Ammo- some Kalahari Z. Tierpsychol. 37:322- spemaophilus leucurus. Physiol. Zool. 54:81-93. 331. . 19816. Activity and thermoregulation of the PORTER,W. P, and C. R. TRACY. 1974. Modeling antelope ground squirrel Ammospemzophilus leu- the effects of temperature changes on the ecology curus in winter and summer. Physiol. Zool. of the garter snake and leopard frog. Pages 594- 54:2 15-223. 609 in J. W. GIBBONSand R. R. SCHARITZ,eds. DE GRAAFF,G. 1981. The rodents of Southern Af- Thermal ecology. Atomic Energy Commission, rica. Buttenvorths, Durban. Washington, D.C. GOLIGHTLY,R. T., JR., and R. D. OHMART.1978. POWELL,W. 1925. Rodents: description, habits, and Heterothermy in free-ranging Abert's squirrels methods of destruction. Dep. Public Health, Pre- (Sciurus aberti). Ecology 59:897-909. toria, Union South Afr., Pub. 32 1. HERZIG-STRASCHIL,B. 1978. On the biology of Xe- SMITHERS,R. H. N. 1971. The mammals of Bo- rus inauris (Zimmermann 1780) (Rodentia, Sciu- tswana. Trustees Nat. Hist. Mus. Rhodesia, ridae). Z. Saugetierkunde 43262-278. Salisbury, Mus. Mem. 4. -. 1979. Xerus inauris (Rodentia, Sciuridae): SYNMAN,P. S. 1940. The study and control of the an inhabitant of the arid regions of southern Af- vectors of rabies in South Africa. Onderstepoort rica. Folia Zool. 28: 119-124. J. Veterinary Sci. Anim. Ind. 15:9-140. HUDSON,J. W. 1962. The role of water in the biology ZUMPT,I. E 1968. The feeding habits of the yellow of the antelope ground squirrel Citellus leacunrs. mongoose, Cynictis penicillata, the suricate, Su- Univ. California Pub. Zool. 64:l-56. ricata suricatta, and the Cape ground squirrel, MARSH,A. C., G. LOUW,and H. H. BERRY.1978. Xenrs inauris. J. South Afr. Veterinary Med. Ass. Aspects of renal physiology, nutrition and ther- 39:89-91. moregulation in the ground squirrel Xerus in- . 1970. The ground squirrel. Afr. Wild Life auris. Madoqua 2:129-135. 24:115-121. MUCHLINSKI,A. E., and K. A. SHUMP,JR. 1979.