Arnold1988_Article_SocialThermoregulationDuringHi

Prévia do material em texto

J Comp Physiol B (1988) 158:151-156 
Journal of 
C o m p a r a t i v e 8~ho~,~,, Systemic~ 
and Environ- 
Phys io logy B mona, Physiology 
�9 Springer-Verlag 1988 
Social thermoregulation during hibernation 
in alpine marmots (Marmota marmota) 
Walter Arnold 
Max-Planck-Institut f/ir Verhaltensphysiologie, Abteilung Wickler, D-8130 Seewiesen, Federal Republic of Germany 
Accepted January 13, 1988 
Summary. Body temperature (Tb) of socially hiber- 
nating alpine marmots, a pair and two family 
groups, was monitored continuously from October 
to March with implanted temperature-sensitive ra- 
diotransmitters. At the same time, the animals' 
behaviour was observed. The recurrent entrances 
into and arousals from hibernation were highly 
synchronised within groups. Group members al- 
ways lay huddled together when euthermic and 
also when torpid with a few exceptions at higher 
ambient temperatures (Ta). Body contact with eu- 
thermic nestmates warmed torpid marmots passi- 
vely. The Tb of animals reentering hibernation did 
not fall to values close to Ta as long as euthermic 
group members were present. Although animals 
presumably save energy through social thermore- 
gulation, especially when euthermic, these benefits 
are not necessarily mutual among group members. 
Differences in thermoregulatory behaviour of indi- 
viduals described in this study could be responsible 
for differential weight losses during winter as 
found in the natural habitat (Arnold 1986). 
Introduction 
In hibernating mammals, short phases of normal 
body temperature (Tb) alternate regularly with lon- 
ger torpid phases during which Tb is slightly above 
ambient temperature (Ta) (Morrison and Galster 
1975). Most energy is consumed in warming the 
body to a state of euthermia and maintaining this 
(Benedict and Lee 1938; Kayser 1953; Bailey and 
Davis 1965; Tucker 1965). Individual energy con- 
sumption could be reduced by social hibernation 
if group members become euthermic at the same 
time and warm each other. Energy expenditure 
during torpor is minimal only if Tb is close to T~t, 
the hypothalamic temperature threshold for meta- 
bolic heat production. Although some studies 
failed to demonstrate the existence of a Tse t in the 
early stages of a period of hibernation (South et al. 
1975) or throughout winter (Pivorun 1986), others 
suggest the continuous operation of the mamma- 
lian thermoregulatory system during hibernation 
at all Tb'S experienced (Florant and Heller 1977; 
Florant et al. 1978). For instance, yellow-bellied 
marmots (M.flaviventris) increase their metabolic 
heat production in proportion to the degree of hy- 
pothalamic cooling, and lowering T, well below 
Tso t elicits complete arousal (Heller and Colliver 
1974; Florant and Heller 1977; South et al. 1975). 
Alpine marmots respond similarly to low T, with 
augmented oxygen consumption (Arnold et al., un- 
published). Thus huddling and mutual warming 
should be advantageous not only during euthermia 
but also during torpor when Ta < Tset. 
With the exception of the woodchuck 
(M. monax), all species of the genus Marmota are 
social and hibernate in groups (Bibikow 1968; 
Rausch und Rausch 1971; Barash 1973; Anderson 
etal. 1976; Johns and Armitage 1979; Holmes 
1984; Arnold 1986). It has been shown for alpine 
marmots that hibernating in large groups reduces 
winter mortality and enhances female fertility in 
the following spring, most likely due to reduced 
weight loss during social hibernation (Arnold 
1986). On the other hand, the presence of first year 
juveniles in a hibernating group increases the 
weight losses of all older group members (Arnold 
1986). The present study investigates the thermo- 
regulatory behaviour of socially hibernating alpine 
marmots in the laboratory in order to understand 
how the costs and benefits of joint hibernation, 
as found in the natural habitat, could arise. 
Materials and methods 
Animal maintenance. Temperature measurements. In autumn 
1983 three groups of alpine marmots, an adult pair (group 1), 
parents and three first-year juveniles (group 2) and parents, 
a male 2-year-old and two first-year offspring (group 3) were 
152 W. Arnold: Social thermoregulation during hibernation in alpine marmots 
caught in the National Park of Berchtesgaden, West Germany. 
The adult female in group 3 escaped before onset of hiberna- 
tion. All other animals were released again at the capture sites 
in spring 1984. Each group was housed in a wooden box; one 
juvenile from group 2 was kept singly to keep the number of 
juveniles in groups 2 and 3 equal. Boards with an opening di- 
vided each box into two equal chambers, one provided with 
hay as nesting material. This nest chamber was appro~mately 
the size of a natural den (60x 60 x75 cm; cf. Bibikow 1968; 
Kapitonov 1978). Thermocouples mounted in the floor of the 
boxes measured T~. 
Groups 1 and 2 and the single juvenile were kept in a tem- 
perature controlled coldroom from 12 Oct 1983 to 28 March 
1984, group 3 from 12 Oct 1983 to 5 March 1984 under dim 
red light (0.3 lux) and with a relative humidity of 80%. 7", 
was occasionally varied for experimental purposes (Y= 6 ~ 
S D = I ~ range=4-10 ~ Throughout winter the animals 
were without food or water. They could at all times be observed 
with video cameras through the plexiglas box tops. All animals 
were weighed to the nearest 50 g at the beginning and end 
of the maintenance period in the coldroom and on 9 Feb 1984. 
Transmitters (weight 19 g) with a temperature-dependent 
pulse rate were implanted under anesthesia into the abdominal 
cavities of the 10 group animals. Transmitters were encased 
in synthetic resin and coated with a l : l mixture of beeswax 
and paraffin. The signals were received via antennas on the 
lids of all boxes. For each animal, body core temperature data 
were punched on tape every 100 s. Average body temperatures 
were obtained graphically from plots. The reentry phase of 
a short hibernation cycle (Morrison and Galster 1975) was as- 
sumed to end at the time when a constant Tb is reached. The 
subsequent part of a torpor episode until arousal, or preceding 
passive warming by euthermic nestmates, is called "deep hiber- 
nation". 
Disturbances and statistical evaluation of data. On 29 Nov 1983 
the temperature-control unit of the coldroom failed. All boxes 
were transferred to a second coldroom. Group 2 animals were 
euthermic at this time, all others arose prematurely. On 1 Dec 
1983 the boxes were placed back into the original coldroom. 
This interrupted the reentry into hibernation of some animals 
in group I and group 2 briefly, but in group 3 for a longer 
period of time (Fig. 1). The weighing on 9 Feb 1984 caused 
group 1 to end the torpid phase on the next day (Fig. 1). Also 
on 9 Feb 1984 the lone juvenile, at this time euthermic, was 
returned to group 2, and the critically thin adult male was re- 
moved from group 3, maintained singly and fed. These manipu- 
lations may have influenced further hibernation in groups 2 
and 3. 
Only undisturbed short hibernation cycles were used in 
the statistical analysis. In calculating average duration of torpor 
periods, shortened autumn (Oct) and spring cycles (March) and 
those which were apparently shortened due to falling T, (cycles 
3 and 4 in group 3, 6 in group 2, Fig. 1), were also excluded. 
Every cycle of an animal counted as an independent event. 
When not otherwise stated, two-tailed Mann-Whitney U-tests 
(Siegel 1956) were used throughout to find significant differ- 
ences. For all group comparisons, T, was not detectably differ- 
ent. 
Results 
Course of hibernation in social groups 
Hibernation began after a few days in the cold- 
room at 7-8 ~ Circadian periodicity of Tb disap- 
? 
o 
E 
0J 
group I 
40 1 2 3 4 5 6 7 8 9 10 11 
0 
2040IQ d 
0 
group 2 
1 2 3 4 5 6 7 8 9 10 11 12 40 
I od o ~ 
2O 
0 
40 J 
I ad 
2O 
0 
4o2o Jju~ ? 
0 
4020 l luvo~ 
0 
group 3 I 2 3 Z, 5 6 7 89 
2 O 
0 
2 O 
0 
2 O 
0 
2 O 
0 _ _ . 
i , , i i 
Oct Nov Dec Jon Feb Mar 
Fig. 1. Tu of animals in group 1, 2 and 3 throughout the study. 
+ = failure of coldroom. * = weighing day; addition of a juve- 
nile female without transmitter to group 2; removal of adult 
male from group 3. Lines above each group = running number 
of analysed short hibernation cycles. 2y = 2-year-old 
peared with the first entrance into torpor (Pohl 
1964). Body temperature decreased almost to Ta 
and remained near ira during deep hibernation. 
The animals moved only in the euthermic phases. 
First and last cycles were shorter, typically for spr- 
ing and autumn (Fig. 1) (Morrison and Galster 
1975; Pivorun 1976; Torke and Twente 1977; 
French 1982a). As in the field situation, in spring 
the adult males in all groups ended dormancy first 
and juveniles last (Arnold 1986, Fig. 1). 
Within groups, all animals usually changed al- 
most simultaneously from euthermia to torpor and 
back (Fig. 1). Immediately after arousal, the mar- 
mots most frequently showed grooming and nest- 
building behaviour, the latter being particularly in- 
tensive when T, was low. Tb was highest at the 
beginning of euthermia but then decreased slightly 
and irregularly. Reentry into hibernation lasted on 
average 121 h (SD=30) and arousal on average 
40 6.9 h (SD = 1.8). During undisturbed midwinter cy- 
cles, animals in group 1 and group 2 were torpid 
on average for 344 h (SD = 43) and euthermic for 
24.8 h (SD = 10.5). The adult male was mainly re- 
sponsible for shorter torpor phases (2=225 h, 
SD= 32, P<0.001) and the tendency to longer eu- 
thermic periods (2 = 30.8 h, SD = 26.4) in group 3. 
Its weight loss during winter (46.5%) was consider- 
ably higher than in all other animals (2--25.2%, 
SD = 1.8). 
Huddling and its influence on Tb 
When euthermic, all animals in a nest always main- 
tained close body contact. They rarely left the nest 
and then only briefly, mainly to urinate (cf. Torke 
and Twente 1977). The first animal to become eu- 
thermic lay beside or on still torpid ones, groomed 
them and covered them with hay. The huddling 
behaviour during torpor was similar to that during 
euthermia but seemed to be influenced by To. After 
providing a Ta of 10 ~ both group I animals arose 
prematurely (cycle 8) and discontinued body con- 
tact and the typical curled posture in the next tor- 
por period, similar to hypothalamically heated 
marmots (Mills and South 1972). With relatively 
high T,'s (7-10 ~ at the beginning of the reentry 
phases, some animals from group 2 avoided body 
contact and lay prone in the outer chamber (adult 
female, cycles 6, 8, 9; adult male, cycles 8, 9; juve- 
nile female, cycles 8, 9). Subsequently lowering T, 
to 5-7 ~ apparently led to arousal of the group 2 
parents, which lay alone (adult female, cycles 8, 
9; adult male, cycle 9). When they were able to 
move about again, they lay down against the juve- 
niles and covered the whole group with hay. 
Marmots lying without body contact with 
others cooled faster (P= 0.005), and their TD dur- 
ing deep hibernation was closer to Ta than in 
huddled animals (P= 0.01). Consistently, an indi- 
vidual took longer to reenter deep hibernation the 
longer its nestmates remained euthermic (Spear- 
man rank correlation coefficient (r~)= 0.602, N = 
55, P<0.001). With one or more animals in body 
contact being euthermic, an irregular course of Tb, 
similar to reentry into hibernation, was detectable 
at the end of a torpor phase (Fig. 2). This rise 
of Tb caused by passive warming correlated with 
the duration of this warming (rs=0.732, N=70, 
P>0.001). As separated animals and those arous- 
ing first were not warmed passively, they had to 
overcome greater temperature ranges to attain 
euthermia than huddled animals arousing subse- 
quently (P = 0.004). 
35 
10 
3O 
E 
25 
C~ 
E 2O 
0 
r 
r 
W. Arnold: Social thermoregulation during hibernation in alpine marmots 153 
i i i I I ~ I I r I 
days 
Fig. 2. Detailed illustration of synchronised euthermic periods 
in a huddling group. Note passive warming through body con- 
tact with euthermic nestmates of torpid animals prior to own 
arousal and those reentering hibernation 
Individual differences 
In group 1 the male was always warmer than the 
female during deep hibernation (P< 0.05) and eu- 
thermic periods (P<0.001). He also arose before 
the female in all cycles (Binomial test, P<0.001). 
The parents in group 2 usually reentered hiberna- 
tion after the juveniles (Friedman test, P<0.05) 
and maintained higher Tb'S during deep hiberna- 
tion (Friedman test, P < 0.05). Group 2 individuals 
seemed to arouse in random order, yet in undis- 
turbed midwinter cycles the juveniles arose before 
the adults in most cases (Friedman test, P<0.01). 
In group 3 usually the adult or the 2-year-old male 
reentered hibernation last (Binomial test, P = 0.02), 
and subsequently was the warmest in the nest (Bi- 
nomial test, P=0.06). In every cycle, the adult 
male arose long before the others (Binomial test, 
P = 0.002). Tb'S were clearly higher in group 3 than 
in groups 1 and 2 during euthermia (P<0.05) as 
well as during deep hibernation (P<0.001). Juve- 
niles had shorter euthermic periods than adults 
(P < 0.001). The extremely longer euthermic phases 
of the group 3 adult male were excluded for this 
comparison, to be conservative. 
Discussion 
The observed synchrony of short hibernation cy- 
cles within groups may have resulted from the ani- 
154 w. Arnold: Social thermoregulation during hibernation in alpine marmots 
mals' increasing sensitivity to external stimuli to- 
wards the end of deep hibernation (Twente and 
Twente 1965; Morrison and Galster 1975). Percep- 
tion of peripheral thermal stimuli during torpor 
is known in marmots and was hypothesized to be 
important because thermoregulatory responses eli- 
cited by central receptors would be too retarded, 
due to the great heat capacity of their large bodies 
(Luecke and South 1972; Mills and South 1972). 
Alternatively, peripheral sensitivity could be an ad- 
aptation to social thermoregulation, as this ability 
is not found in the solitary hibernators Spermophi- 
lus tridecemlineatus and S. lateralis (Lyman and 
O'Brien 1972). 
In many homeotherms individual heat loss is 
reduced by huddling in groups, in fact all the more 
effectively the colder the environment and the 
more animals involved (for review see Madison 
1984). The same benefit should apply to huddling 
alpine marmots and be most pronounced during 
simultaneous euthermia and arousal, the periods 
of highest energy consumption in hibernators. Ju- 
veniles particularly should profit from huddling, 
as they have the lowest fat reserves (Bibikow 1968; 
Armitage et al. 1976), and their smaller size in- 
volves greater heat loss (Mills and South 1972; 
Th/iti 1978). Furthermore, passive warming by eu- 
thermic animals served to reduce the thermal gap 
over which an individual had to actively warm it- 
self when arousing. The distinct, irregular rise of 
Tb due to passive warming clearly contrasted with 
the smooth, steep increase during active arousal 
(Fig. 2) and is very unlikely to stem from an indi- 
vidual's own metabolic heat production. Such a 
time course of Tb was always found only in animals 
with body contact with euthermic nestmates. The 
slight rises of Tb during deep hibernation in the 
group 2 animals during cycle 8 (Fig. 1) occurred 
because the juvenile without a transmitter, pre- 
viously kept alone, was returned to its family while 
it was euthermic. 
It may appear that an animal could derive the 
greatest advantage from passive warming the later 
it arouses after the other group members. This 
would lead to desynchronised cycles. However, the 
resulting energy costs to the individual are very 
likely to be higher than the potential benefit. 
Firstly, euthermic animals would have to thermo- 
regulate alone against the cold environment forlonger periods and secondly, in the presence of 
euthermic nestmates, the Tb of those reentering 
hibernation could rarely fall sufficiently to allow 
minimum metabolism (cf. Fig. 2). For the same 
reason, animals reentering hibernation should 
avoid body contact with euthermic group members 
in order to cool as fast as possible. However, this 
could be disadvantageous in the following situa- 
tion: If Ta and hence Tb drops below about 8 ~ 
torpid marmots increase their energy consumption 
again (Florant and Heller 1977; Arnold et al., un- 
published). The animals would then profit from 
social thermoregulation but movement is possible 
only at the cost of at least partial arousal. Al- 
though some study animals seemed to adjust their 
spacing in relation to T,, lying scattered at high 
Ta and closing up again with decreasing Ta, hud- 
dling behaviour was not predictable. A T, of 6 
or 7 ~ did not necessarily elicit close body contact 
of all group members at the reentry into hiberna- 
tion. This might be due to the recognised individ- 
ual and temporal variability of Tse t (Heller and 
Colliver 1974; Florant and Heller 1977). However, 
further investigations are necessary to determine 
whether alpine marmots actually respond to small 
deviations of T b from Tse t with huddling behav- 
iour, as suggested by this study. Lying singly dur- 
ing deep hibernation will rarely be beneficial in 
natural hibernacula because Ta in such a situation 
will decrease continuously throughout winter to 
almost 0 ~ in spring and be below 8 ~ for about 
two thirds of the hibernation season (Arnold et al., 
unpublished). Under such harsh conditions, close 
body contact could even be rewarding during reen- 
try into hibernation, because thermoregulatory re- 
sponses are to be expected, if Tb falls faster than 
the gradually declining Ts~ t (Heller et al. 1977; 
Florant et al. 1978). 
The costs and benefits of social hibernation dis- 
cussed so far would apply equally to all individuals 
if the sequence of reentry into and arousal from 
hibernation and the level of Tb were random in 
relation to nestmates. However, this was not the 
case in any of the groups in this study, indicating 
how differential weight losses, found in the colder 
natural environment (Arnold 1986), could come 
about. Throughout all hibernation cycles, the 
group 1 female profited thermally from the male, 
in contrast to group 2. Similar differences between 
the pair-males' thermoregulatory support are ap- 
parent in the field, where a female will lose less 
weight the more her male loses (Arnold 1986). 
Adults remain euthermic longer on account of 
their greater body mass (French 1982a; 1982b; 
1985), hence juveniles hardly ever thermoregulate 
against the cold for lack of a warming adult. The 
group 2 parents apparently warmed the juveniles 
during deep hibernation and followed them closely 
when these arose first. If parents were lying alone 
when Ta was decreasing, they would arise to huddle 
again with the juveniles. Similar behaviour could 
w. Arnold: Social thermoregulation during hibernation in alpine marmots 155 
be responsible for the higher weight losses found 
in the na tu ra l habi ta t , o f m a r m o t s associa ted with 
juveniles. Energy costs would be incurred if pa ren t s 
shor tened their per iods of t o r p o r or ma in t a ined 
higher Tb dur ing deep h ibe rna t ion because of the 
presence o f juveniles. H ighe r Tb means shor te r tor- 
pid phases (Twente and Twente 1965) and fre- 
quency of a rousa l rises when Tb is act ively regu- 
lated above Ta (Pengelley and Kel ley 1966; La- 
chiver and B o u l o u a r d 1967; Soivio et al. 1968). 
G r o u p s 1 and 2 h iberna ted as long as free-liv- 
ing an imals and had similar dura t ions o f to rp id 
and eu thermic phases as an alpine m a r m o t kep t 
in a c o l d r o o m dur ing winter 1982/83. In contras t , 
h ibe rna t ion in g r o u p 3 was obvious ly dis turbed. 
The adul t ma le ' s except ional ly shor t per iods o f tor- 
p o r and long eu thermic periods, leading to its ex- 
t r ao rd ina ry high weight Joss, obvious ly inf luenced 
the o ther g roup m e m b e r ' s h ibe rna t ion (Fig. 1). 
Thus differences between adul t and juvenile ther- 
m o r e g u l a t o r y b e h a v i o u r in this g roup should be 
in te rpre ted with care. However , they are similar 
to those in g roup 2 and suggest tha t the male 
2-year-o ld behaved like a parent . 
An individual ' s to ta l energy c o n s u m p t i o n pre- 
s umab ly decreases with the size of a h iberna t ing 
group. However , wi th increasing g roup size, d imin- 
ishing re turns are to be expected, due not only 
to the physical laws govern ing energy exchanges 
but also because the shor t h ibe rna t ion cycles neces- 
sarily b e c o m e less exact ly synchronised. T o predic t 
the op t ima l g roup size under field condi t ions, we 
need to de te rmine how the me tabo l i c rate o f indi- 
viduals depends on their t h e r m o r e g u l a t o r y behav- 
iour relative to their g roup members . Only such 
m e a s u r e m e n t s will pe rmi t exact quant i f ica t ion o f 
the costs and benefits o f social h ibe rna t ion in m a r - 
mots . 
Acknowledgements. I thank the Nationalparkamt Berchtesga- 
den and the Oberforstdirektion Mfinchen for permitting the 
capture of the study animals. I am grateful to H. Wiesner, 
who implanted the radiotransmitters, and to P. Heinecke for 
technical assistance. B. Knauer expertly prepared the figures. 
I am indebted to F. Trillmich and three anonymous reviewers 
for critical comments on earlier drafts of the manuscript. This 
study was supported and financed by the Max-Planck-Gesell- 
schaft. 
References 
Anderson DC, Armitage KB, Hoffman RS (1976) Socioecology 
of marmots: female reproductive strategies. Ecology 
57: 552-560 
Armitage KB, Downhower JF, Svendsen GE (1976) Seasonal 
changes in weights of marmots. Am Midl Nat 96 : 36-51 
Arnold W (1986) Sozio6kologie des Alpenmurmeltieres. PhD 
thesis, Faculty of Biology, Ludwig-Maximilian-Universit/it 
Miinchen 
Bailey ED, Davis DE (1965) The utilization of body fat during 
hibernation in woodchucks. Can J Zool 43 : 701 707 
Barash DP (1973) The social biology of the Olympic marmot. 
Anita Behav Monogr 6:171 249 
Benedict FG, Lee RC (1938) Hibernation and marmot physiol- 
ogy. Carnegie Inst Washington Publ No 497, Washington 
Bibikow DI (1968) Die Murmeltiere. Die neue Brehm Bficherei. 
Ziemsen, Stuttgart 
Florant GL, Heller HC (1977) CNS regulation of body l:emper- 
ature in euthermic and hibernating marmots (Marmotafla- 
viventris). Am J Physiol 232:203-208 
Florant GL, Turner BM, Heller HC (1978) Temperature regula- 
tion during wakefulness, sleep, and hibernation in marmots. 
Am J Physiol 235:R82 R88 
French AR (1982a) Intraspecific differences in the pattern of 
hibernation in the ground-squirrel Spermophilus betdingi. J 
Comp Physiol 148:83-91 
French AR (1982b) Effects of temperature on the duration 
of arousal episodes during hibernation. J Appl Physiol (Re- 
spirat Environ Exercise Physiol) 52:216-220 
French AR (1985) Allometries of the duration of torpid and 
euthermic intervals during mammalian hibernation: a test 
of the theory of metabolic control of the timing of changes 
in body temperature. J Comp Physiol B 156 : 13-19 
Heller HC, Colliver GW (1974) CNS regulation of body tem- 
perature during hibernation. Am J Physiol 227:583-589 
Heller HC, Colliver GW, Beard J (1977) Thermoregulation dur- 
ing entrance into hibernation. Pflfigers Arch 369: 55-59 
Holmes WG (1984) The ecological basis of monogamy in Alas- 
kan hoary marmots. In: Murie JO, Michener GR (eds) The 
biology of ground-dwelling squirrels. University of Ne- 
braska Press, Lincoln London, pp 250-274 
Johns DW, Armitage KB (1979) Behavioral ecology of alpineyellow-bellied marmots. Behav Ecol Sociobiol 5:133--I 57 
Kapitonov VI (1978) Winter burrow digging of the Kamchatka 
marmot in northwestern Verkhoyansk Russian-SFSR 
USSR. Byull Mosk O-Va Ispyt Prir Odt Biol 83:43-51 
Kayser CH (1953) L'hibernation des mammifSres. Annie Biol 
29:109-150 
Lachiver F, Boulouard R (1967) Evolution de la periodicit6 
des phases de sommeil et de reveil chez de Lerot (Eliomys 
quercinus) au cours du sommeil hivernal et de la l+thargie 
induite en +t6. J Physiol (Paris) 59:25~251 
Luecke RH, South FE (1972) A possible model for thermoregu- 
lation during deep hibernation. In: South FE, Hannen JP, 
Willis JR, Pengelley ET, Alpert NR (eds) Hibernation and 
hypothermia, perspectives and challenges. Elsevier, Amster- 
dam, pp 577-604 
Lyman CP, O'Brien RC (1972) Sensitivity to low temperature 
in hibernating rodents. Am J Physiol 222:864-869 
Madison DM (i984) Group nesting and its ecological and evo- 
lutionary significance in overwintering microtine rodents. 
In : Merrit JE (ed) Winter ecology of small mammals. Carn 
Mus Nat Hist Spec Publ/0:267-274 
Mills SH, South FE (1972) Central regulation of temperature 
in hibernation and normothermia. Cryobiology 9:393-403 
Morrison P, Galster W (1975) Patterns of hibernation in the 
arctic ground squirrel. Can J Zool 53:1345-1355 
Pengelley ET, Kelley KH (1966) A 'circadian' rhythm in hiber- 
nating species of the genus Citellus with observations on 
their physiological evolution. Comp Biochem Physiol 
19:603-617 
Pivorun EB (1976) A biotelemetry study of the thermoregulato- 
ry patterns of Tamias striatus and Eutamias minimus during 
hibernation. Comp Biochem Physiol 53A:265-271 
156 W. Arnold: Social thermoregulation during hibernation in alpine marmots 
Pivorun EB (1986) Hypothalamic thermosensitivity in hibernat- 
ing chipmunks, Tamias striatus. Physiol Zool 59:194-200 
Pohl H (1964) Diurnal rhythms and hibernation. Ann Acad 
Sci Fenn Ser AIV Biologica 71 : 361-373 
Rausch RL, Rausch VR (1971) The somatic chromosomes of 
some North American marmots (Sciuridae), with remarks 
on the relationship of Marrnota broweri Hall and Gilmore. 
Mammalia 35 : 85-101 
Siegel S (1956) Nonparametric statistics for the behavioral 
sciences. McGraw Hill, New York 
Soivio A, Th/iti H, Kristoffersson R (1968) Studies on the peri- 
odicity of hibernation in the hedgehog (Erinaceus europaeus 
L). III. Hibernation in a constant ambient temperature of 
- 5 ~ Ann Zool Fenn 5 : 224-226 
South FE, Hartner WC, Luecke RH (1975) Responses to pre- 
optic temperature manipulation in the awake and hibernat- 
ing marmot. Am J Physiol 229:150-160 
Th/iti H (1978) Seasonal differences in Oz consumption and 
respiratory quotient in a hibernator (Erinaceus europaeus 
L). Ann Zool Fenn 15:69-75 
Torke KG, Twente JW (1977) Behavior of Spermophilus latera- 
lis between periods of hibernation. J Mammal 58: 385-390 
Tucker VA (1965) The relation between the torpor cycle and 
heat exchange in the California pocket mouse Perognathus 
californicus. J Cell Comp Physiol 65:405-414 
Twente JW, Twente JA (1965) Effects of core temperature upon 
duration of hibernation of Citellus lateralis. J Appl Physiol 
20:411---416