Full text of DOI:10.1007/BF03192411

Acta Theriologica 46 (1): 1–12, 2001.
PL ISSN 0001–7051
Energy balance of hibernating mouse-eared bat Myotis myotis:
a study with a TOBEC instrument
Pawe³ KOTEJA, Miros³aw JURCZYSZYN and Bronis³aw W. WO£OSZYN
Koteja P., Jurczyszyn M. and Wo³oszyn B. W. 2001. Energy balance of hibernating
mouse-eared bat Myotis myotis: a study with a TOBEC instrument. Acta Theriologica 46:
1–12.
We used a non-invasive TOBEC method (Total Body Electric Conductivity) to
estimate lean body mass and fat content in mouse-eared bats Myotis myotis (Borkhausen,
1797) hibernating in Poznañ Forts (W Poland) and in a semi-natural cave-mine
Miedzianka (SE Poland). In December, fat content averaged 5.5 g in females (body mass
= 29.4 g) and 5.3 g in males (body mass = 28.4 g). At the end of hibernation (April), fat
content averaged 2.2 g in females (body mass = 25.6 g) and 1.4 g in males (body mass =
23.7 g). Fat content did not differ between the localities either in December or in April,
but the pattern of changes of fat content was different. We calculated the rate of energy
expenditure in hibernating bats using two methods, based on independent samples (fat
content in first-time captured individuals) and based on paired observations (changes
of fat content in re-captured individuals), and discussed problems associated with the
two approaches. Both methods show that the bats need about 4.9 g of fat (191 kJ) to
sustain a 165-day hibernation. However, the rate of fat usage varied considerably
between the sites and hibernation phase. Although the average amount of fat
remaining in April would be sufficient to support at least six more weeks of
hibernation, the level of reserves was close to zero in some individuals.
Institute of Environmental Sciences, Jagiellonian University, Kraków, Poland, e-mail:
koteja@eko.uj.edu.pl (PK); Department of Systematic Zoology, A. Mickiewicz University,
Poznañ, Poland (MJ); Institute of Systematics and Evolution of Animals, Polish
Academy of Sciences, Kraków, Poland (BWW)
Key words: Chiroptera, energetics, fat, hibernation, non-invasive methods
Introduction
During hibernation, insectivorous bats inhabiting cold-temperate climate rely
on energy stored in body fat. Maintaining energy balance during the period is an
important component of their life history (Thomas 1995a, b). Consequently, it is
often assumed that behavior, physiology and even morphology of bats can be
optimized to minimize energy expenditures during hibernation (eg Nagel and Nagel
1991, Schutt 1993). However, only few studies attempted to measure total energy
expenditure during hibernation and its relation to the amount of fat reserves
(Speakman and Racey 1989, Ransome 1990, Thomas et al. 1990, Thomas 1995b).
If the entire energy used during hibernation comes from fat reserves, the rate of
energy expenditure can be estimated by monitoring changes of body mass and
[1]

2
P. Koteja et al.
composition. The task is not trivial for two reasons. First, any manipulation of
hibernating bats may stimulate arousals, which increase energy expenditure
during hibernation period (Thomas et al. 1990, Speakman et al. 1991, Thomas
1995a, b). Thus, the measurement itself may influence results of such a study.
Second, most methods of measuring fat content are destructive or highly invasive
or cannot be used in the field. Measurement of Total Body Electric Conductivity
(TOBEC) seems to be a promising solution to the second problem (Walsberg 1988,
Froncisz et al. 1994, Koteja 1996). With this non-invasive method, lean body mass is
estimated and fat content can be calculated if body mass is also known. The
measurement takes only a few minutes and can be repeated frequently on the same
individual. Although handling of the bats induce arousals, the TOBEC measure-
ments are less disturbing than any alternative technique (eg, the methods based on
labeled water require an injection and taking blood sample after some time). We
used the TOBEC method to estimate fat content and calculate energy expenditures
in hibernating mouse-eared bats Myotis myotis (Borkhausen, 1797).
Material and methods
Study site and measurement protocol
The study was performed in a semi-natural cave-mine Miedzianka (south-eastern Poland; Wo³oszyn
1962, 1964) and in the city of Poznañ forts (western Poland; Bogdanowicz and Urbañczyk 1983,
Jurczyszyn 1995, Jurczyszyn and Bajaczyk 1996). Measurements of body composition were performed
four times in Miedzianka and Poznañ during the 1995/96 winter (Table 1). Mouse-eared bats M. myotis
were taken from the walls and delivered to a place within the hibernaculum where a portable table
with instruments was located. Then, rectal body temperature (Ahlborn Therm 2244-1 with NTC-C
856-1 probe; 0.1°C) and body mass (Kern 444-33; 0.1 g) were measured and the bat was placed in a
chamber of ACAN-2 TOBEC analyzer (chamber diameter: 38 mm; Jagmar, Kraków; Piasecki et al.
1995). After taking a few ACAN readings three small pieces of paper (about 3 ´ 3 mm) with printed
numbers were glued on the individual’s back, and the bat was replaced on the wall. Usually, all the
procedure lasted a few minutes, but sometimes longer, eg when a cluster of a few individuals was taken
(the individuals were kept in a Styrofoam container before the measurement). We did not measure
body temperature immediately after taking the animals from the wall, because our primary concern
Table 1. Dates of measurements and numbers of mouse-eared bats Myotis myotis used in this study.
Location
Date
n
% females
% recaptured
Miedzianka
12 December 1995
12 January 1996
29 February 1996
18 April 1996
26
23
29
17
57.7
30.4
48.3
41.2
30.4
34.5
47.1
Poznañ
21 December 1995
16 January 1996
14 March 1996
16 April 1996
19
16
27
28
42.1
37.5
40.7
46.4
18.8
11.1
50.0

Energetics of hibernating mouse-eared bats
3
was to have the bats non-aroused during the TOBEC measurements (any movement of the animals
decreases accuracy of the measurements). Thus, the measured body temperature depended also on the
variable handling time (ranging approximately from 1 to 10 minutes). Therefore, from the values
recorded before TOBEC readings, we arbitrarily selected four lowest values for each sex in every trial.
The means calculated from those lowest readings are biased estimates of population mean values, but
they are better indicators of differences in body temperature between the sites and trials than the
averages based on all observations.
ACAN calibration
The values of Total Body Electric Conductivity (TOBEC) are affected, by principle, by the
temperature of the measured objects. In previous studies, performed with normothermic birds and
mammals, the effect of temperature was not significant (Walsberg 1988, Koteja 1996). However, the
range of body temperatures in the bats measured in this study was much larger than in normothermic
animals and the raw readings had to be corrected for the effect. Temperature correction coefficient was
obtained from repeated ACAN readings in nine arousing bats (body temperature between 3.5 and 15°C;
after measuring body temperature and taking ACAN reading, body temperature was measured again
and a next ACAN reading was taken, and so on). The coefficient was calculated as a common
regression slope from an ANCOVA model with ACAN readings as dependent variable, body tempera-
ture as independent variable, and individuals as categories (Fig. 1).
To use TOBEC values for estimating lean body mass (LBM), a number of animals must be
sacrificed for calibration of the instrument (as described in Koteja 1996). As the bats are protected in
Poland, we could use only 10 individuals for the calibration (both sexes, 5 individuals at the beginning,
and 5 at the end of hibernation) (Ministry of Environmental Protection and Natural Resources permit
#OP 4201/109/98). After taking ACAN readings the animals were killed by cervical dislocation and
stored in a freezer. Later, the bodies were dried in a vacuum drier to constant mass and fat extraction
was performed with ether in a Soxhlet apparatus. Lean body mass was calculated as a difference
between total body mass and the extracted fat mass.
Data analysis
Fat mass of individuals measured in the field was calculated by subtracting the TOBEC-estimated
lean mass from total body mass. To calculate energy balance we assumed an energy equivalent of 39
kJ/g fat. The amount of fat used during hibernation and the rate of energy expenditures was calculated
in two ways: (1) from differences between mean fat content measured in individuals which were
captured the first time during a visit (ie, individuals in which fat usage was not affected by handling on
previous visits), and (2) from fat content changes observed in individuals which were recaptured.
The effects of sex, site and trial on body mass, lean mass, fat mass and body temperature were
examined by means of ANOVA. The analysis was performed with SYSTAT 6.0 (SPSS Inc.).
Results
ACAN calibration
ACAN readings increased with body temperature of arousing bats (Fig. 1). The
common regression slope was 2.207 per 1°C (SE = 0.378). ACAN readings (scaled
for 37°C to allow comparison with normothermic mammals in future studies)
increased with lean body mass (Fig. 2a). A calibration equation was obtained by
rearranging the fitted regression equation: LM = (ACAN + 37.14)/10.12.
Based on the equation, an average (absolute) error of estimated fat content was
0.54 g, and the maximum error was 1.11 g (Fig. 2b; note, that the error includes
inaccuracy due to variation in body temperature). A slope of the relation between

4
P. Koteja et al.
200
180
160
140
120
Fig. 1. Relation between ACAN readings
and body temperature in nine mouse-eared
bats (distinguished by symbols). The solid
line represents a common regression with a
slope of 2.207/°C (from ANCOVA).
2
4
6
8
10
12
14
16
B ody temperature (oC)
230
220
210
200
190
180
a
-
ACAN = 37.14 + 10.12 LBM
R2
=
0.79
22
23
24
25
L ean body mas s (g)
b
8
6
4
2
0
Fig. 2. Calibration of ACAN-2 for mouse
eared-bats Myotis myotis: (a) relation be-
tween ACAN readings (scaled for 37°C) and
lean body mass, and (b) relation between fat
content estimated with ACAN (body mass
minus ACAN-estimated lean mass) and mea-
sured directly by ether extraction (slope with
confidence interval is given).
fitted regression:
slope =0.86 (c.i.: 0.66 1.06)
-
line of identity: slope =1.00
0
2
4
6
8
Fat mass (extracted; g)
ACAN-estimated and measured directly fat content in bats was slightly, but not
significantly lower than 1.0 (Ho: b = 1.0, p = 0.124; Fig. 2b). We did not test enough
individuals to perform a rigorous analysis of prediction errors. In mice and voles

Energetics of hibernating mouse-eared bats
5
(body mass similar to that of the bats) measured with ACAN, 95% confidence
intervals for individual prediction were
average fat mass (Koteja 1996).
1.9 g, and
0.3 g for predicting an
Field data
We performed 185 measurements on 140 individuals (Table 1). Up to 50% of
individuals were recaptured. We did the measurements four times, but our TOBEC
instrument (ACAN) was malfunctioning during the second trial in both locations. A
loose connection in electronic circuits was sporadically causing decreased readings.
The effect was too small to distinguish the affected readings, and we had to discard
all the ACAN data collected during the second trials (however, we present the data
for body temperature and body mass). We present major results in graphical format
here; full table of descriptive statistics is accessible at http://cic.isez.pan.krakow.pl/.
During the long and cold 1995/96 winter body temperatures of bats hibernating
in Poznañ were 2 to 5°C lower than in Miedzianka cave (ANOVA: F1, 124 = 58.7,
p < 0.001; Fig. 3). The variation in body temperatures was also higher in
Miedzianka. This was probably because in Poznañ most of the bats were collected
from one place, while in Miedzianka the animals were scattered over the cave. An
average handling time was somewhat longer in Miedzianka than in Poznañ (we did
20
a
Miedz ianka, females
Miedz ianka, males
Poznañ, females
P oz nañ, males
15
10
5
0
b
8
6
4
2
Fig. 3. Body temperature of the bats mea-
0
sured before ACAN readings (mean 2 SE):
(a) all individuals, (b) lowest four readings
for each sex.
December January February
March
April
T ime

6
P. Koteja et al.
not measure it precisely), which could also contribute to increased average and
variance of measured body temperature. However, the difference in average body
temperatures was also evident when the lowest four readings per sex were analyzed
separately (Fig. 3b).
Because handling could influence energy expenditure of the bats, and therefore
also their body composition, the following analysis is based only on individuals
which were captured for the first time. Average body mass was always higher in
females than in males (Fig. 4a; F1, 124 = 7.33; p = 0.008). Body mass measured in
December (early hibernation) and in April (end of hibernation) did not differ
significantly between Poznañ and Miedzianka. However, body mass measured on
February/March was significantly higher in Poznañ than in Miedzianka (F1, 39
=
6.12; p = 0.018). Thus, in Miedzianka, a large decrease of average body mass
occurred between December and March, and only a very small decrease during late
hibernation (Fig. 4a).
TOBEC-estimated lean body mass was higher in females than in males (F1, 99
=
5.17; p = 0.025), but it did not differ between the sites and remained almost
constant throughout the hibernation (Fig. 4b). The amount of body fat, calculated
as body mass minus TOBEC-estimated lean mass, also tended to be higher in
32
a
30
28
26
24
22
26
b
24
22
20
Miedzianka, females
Miedzianka, males
Poznañ, females
Poznañ, males
8
c
6
4
2
0
Fig. 4. Changes of body mass (a), lean mass
(b), and fat mass (c) in mouse-eared bats
hibernating in Miedzianka and Poznañ (mean
2 SE). The results are based only on the
individuals that were captured for the first
time.
December January February
March
April
T ime

Energetics of hibernating mouse-eared bats
7
females than in males (F1, 99 = 3.86; p = 0.052; Figs 4c and 5). As expected, average
amount of fat decreased during the hibernation (F2, 99 = 67.5; p < 0.001). The
amount of fat did not differ between Miedzianka and Poznañ at the begining
(F1, 41= 0.056; p = 0.814) or at the end of hibernation (F1, 19 = 0.380; p = 0.545).
However, at the end of February the bats in Miedzianka had significantly lower
amounts of fat than the bats in Poznañ at the beginning of March (F1, 39 = 11.6;
p = 0.002). The decrease of fat content was steady throughout the winter in
Poznañ, while in Miedzianka the entire decrease of fat mass was observed between
December and the end of February, and the values did not differ between February
and April (Fig. 4c). An analysis with lean body mass included as a covariate in the
ANOVA models gave similar results.
In one of the males from Poznañ the estimated fat content in December was only
0.9 g, and the individual appeared as severe outlier (studentized residual –3.4).
With this individual omitted, however, the analyses gave similar results.
Based on the changes of average fat content, the estimates of the rate of energy
expenditure ranged from –0.02 kJ/d to 1.84 kJ/d, depending on site and time period
(the values correspond to slopes of the lines connecting means on Fig. 4c). Total fat
usage estimated for the entire hibernation (assumed 165 days) was 5.0 g (195 kJ) in
Miedzianka and 4.8 g (188 kJ) in Poznañ.
Comparisons based on recaptured individuals revealed a pattern similar to that
based on all individuals. In Poznañ, the decrease of body mass and the rate of fat
usage was steady across the hibernation, while in Miedzianka, the rate was
significantly higher during winter months than in the spring (Fig. 6). Based on the
Females
Males
40
30
20
10
0
December
30
20
10
0
0
2
4
6
8
0
2
4
6
8
April
40
40
30
20
10
0
30
20
10
0
Fig. 5. Distribution of ACAN-estimated
fat content in hibernating mouse-eared
bats (pooled for Poznañ and Miedzian-
ka) in December (second month of hiber-
nation) and in April (end of hibernation).
0
2
4
6
8
0
2
4
6
8
T OBEC-estimated fat mass (g)

8
P. Koteja et al.
35
30
25
a
Miedzianka
Poznañ
20
26
b
24
22
20
c
+
+
1.99 0.37 kJ/d (4)
0.58 0.34 kJ/d (6)
Fig. 6. Changes of body mass (a), lean
mass (b), and fat mass (c) with its
energy equivalent) in recaptured in-
dividual mouse-eared bats hibernating
in Miedzianka and Poznañ. Bottom
panel (c) shows also estimates of en-
ergy expenditures (kJ/d). Note that
the estimates are calculated separa-
tely for individuals captured in the
first and second trial, and separately
for those captured in the second and
final trial. All results are shown as
mean 2 SE. Number of individuals
is given in brackets.
-
-
+
+
1.11 0.68 kJ/d (9)
-
1.40 0.20 kJ/d (2)
-
8
6
4
2
0
300
200
100
0
December January February
March
April
T ime
rate of fat changes in recaptured individuals, estimates of the energy cost of
hibernation ranged from 0.58 to 1.99 kJ/day, depending on site and time (Fig. 6c).
Total fat usage estimated for the entire hibernation amounted to 6.1 g in
Miedzianka and 5.6 g in Poznañ, which corresponds to 239 kJ and 217 kJ,
respectively. The values are much higher than the previous estimates based on
population average values.
Handling the bats induced arousals, which, most probably, caused increased
energy expenditures in the following period. Therefore, the above estimates of the
rate of fat usage and energy expenditures based on the recaptured individuals can
be biased upwards.
Discussion
Estimates of total energy expenditure and fat usage in hibernating bats are
scarce (Ransome 1990). Based on respirometric measurements in captive bats,
Thomas et al. (1990) estimated that a 6.5 g little brown bat Myotis lucifugus
requires about 2 g for the entire hibernation (193 days). Pistole (1988) measured fat

Energetics of hibernating mouse-eared bats
9
content in big brown bats Eptesicus fuscus in an annual cycle. The results indicate
that the bats (body mass in autumn about 20 g) used about 3 g for hibernation
(from November to April).
Our estimates of the required amount of fat in mouse-eared bats (autumn body
mass = 25 g) are much higher: about 5.8 g (228 kJ) based on changes of fat mass in
recaptured individuals, and about 4.9 g (191 kJ) based on changes of population
mean values calculated from independent samples. The difference between the
estimates obtained with the two methods may have two causes: a non-random
sampling and the effect of handling on energy expenditures of the bats.
First, the calculation based on independent samples may underestimate the real
cost of hibernation if individuals with high fat content enter the hibernaculum later
(and are not included in the autumn sampling), or if individuals with low levels of
energy reserves leave the hibernaculum earlier. Such mechanism seems to be a
plausible explanation of the paradoxical no-change of fat content between February
and April in Miedzianka (Fig. 4c). A few individuals leaving a hibernaculum in
second half of March and early April were observed in previous years (M. Jurczyszyn,
pers. obs.), but we do not know if those were indeed individuals which were forced
to interrupt hibernation because of fat depletion. Note, that the possible bias
cannot be detected or corrected by simply increasing the sample size.
Repeated measurements on the same individuals are not problem-free, either.
Although TOBEC measurements are non-invasive, handling of the bats induces
arousals. As the arousals are energetically very costly (eg Thomas et al. 1990,
Thomas 1995b), the handling associated with the measurement increases the rate
of fat usage in the following period.
To evaluate the size of the problem, let us suppose that the entire cost of
hibernation is associated with arousal cycles, and that the arousal induced by the
handling is added to the arousals that occur spontaneously (Thomas et al. 1990).
We do not have data on the cost of an arousal cycle in the mouse-eared bats, but it
can be approximated by dividing the total fat usage (estimated from independent
samples) by an estimated number of arousals per season. Jurczyszyn and Bajaczyk
60
1991/92
1992/93
1993/94
300
200
100
0
40
20
0
Fig. 7. Distribution of the length of
uninterrupted torpor in mouse-eared
bats hibernating in Poznañ forts (data
from Jurczyszyn and Bajaczyk 1996).
1
2
3
4
5
6
Duration of torpor bouts (weeks)

10
P. Koteja et al.
(1996) found that in mouse-eared bats hibernating in Poznañ Forts the average
time between arousals was about 12 days (Fig. 7; the value can be an over-
estimation, because the bats were checked only once per week and arousals more
frequent than one per week could not be detected). Thus, during the entire
hibernation (assumed 165 days) an average bat would arouse about 14 times, and a
single arousal cycle would cost 0.35 g fat (or 13.6 kJ). After subtracting the cost of
the additional arousal the estimated rate of energy expenditure between December
and February is only about 10% lower than the uncorrected, but the estimate for
the period between February and April is much more affected: 37% in Poznañ and
48% in Miedzianka. The corrected estimate of the entire cost of hibernation
amounts to 5.1 g fat (or 199 kJ) in Miedzianka and 4.3 g fat (or 169 kJ) in Poznañ,
which is similar to the estimate based on independent samples.
The rate of metabolism in hibernating bats generally decreases with body
temperature (Nagel and Nagel 1991, Thomas 1995b). Therefore, we expected that
the rate of fat usage should be higher in Miedzianka than in Poznañ, because body
temperatures of the bats hibernating there were higher than in the former site (Fig.
3). Although the estimated amount of fat required for the entire hibernation tended
to be higher in Miedzianka than in Poznañ, the differences were not significant.
Moreover, the pattern of changes in body composition across hiber- nation season
challenges the first-choice hypothesis that the difference in thermal conditions was
indeed responsible for the difference in fat usage between the localities.
At the end of hibernation season, the rate of energy expenditures in the bats
from Miedzianka decreased more than three-fold compared to the average rate
between December and February (Figs 4 and 6). On the other hand, the rate of
energy expenditure seemed steady in Poznañ. The dramatic difference between the
first and the second part of hibernation in Miedzianka cannot be explained by a
change in thermal condition (Fig. 3). In addition, it cannot be explained by a
selective migration, because the pattern was clear in within-individual comparisons
(Fig. 6). Finally, it also does not seem feasible that in Miedzianka the bats restored
fat reserves by feeding (cf Speakman and Racey 1989; Hays et al. 1992), because in
1996 the temperatures around 0°C prevailed to the beginning of April, and the bats
would hardly find a prey. Thus, it seems that the bats can to a large extent control
the rate of energy expenditures independently on ambient temperature, and
decrease it dramatically when endangered by exhaustion of fat reserves. A similar,
large variation in the rate of fat usage was also observed in greater horseshoe bats
(Ransome 1990).
The most plausible cause of the large variation in energy expenditures is a
variable frequency of arousals (eg Thomas et al. 1990). The length of torpor bouts in
the mouse-eared bats varies dramatically (Fig. 7; Jurczyszyn and Bajaczyk 1996).
Although most of the observations indicated short intervals of about one week
(60%), durations of up to 6 weeks were observed. Other authors reported even
longer torpor bouts for the species (Harmata 1987). Thus, the bats are physio-

Energetics of hibernating mouse-eared bats
11
logically able to decrease the frequency of arousals, and hence the rate of energy
expenditures, by a factor of at least five.
If the bats in Miedzianka were able to restrict the rate of energy expenditure in
the spring, why did their rate of energy expenditure was so high throughout the
winter? A plausible explanation of the high rate of energy expenditure in Miedzian-
ka may be human activity in the cave, which could cause more frequent arousals
even if the visitors did not intend to disturb the bats (Thomas 1995a). Both of the
study sites are protected biological reserves. However, whereas the entrance to
Poznañ forts is also protected by iron gate, Miedzianka cave proved to be practically
open to intruders. One of the reviewers of a previous version of the manuscript
made a comment that in such a situation our estimates of the cost of hibernation
are artifacts and have no biological meaning. Although we wished we had a better
control over our study sites, we do not quite agree with the comment. Human
activity in many, if not most of the caves in Europe is a biological fact. Thus, it
should be considered as yet another environmental factor.
In any study of hibernation energetics, a major question is whether hibernating
animals are seriously endangered by exhaustion of fat reserves (eg Speakman and
Racey 1989). At the end of the hibernation season, males remained with an average
of 1.4 g fat, and females with about 2.2 g fat (Fig. 4c). With such amounts of fat, the
bats could hibernate for two more month (ie till mid June). Mouse-eared bats
usually leave hibernacula in mid April (Jurczyszyn 1995). Because at the same time
large amount of insects becomes available, one might conclude that the bats are not
seriously endangered by exhaustion of fat reserves during hibernation.
This conclusion, however, may be a hasty oversimplification. The answer to the
question should not be based on an estimate of population average, but on an
analysis of distribution of individual values of fat content in the population. Both at
the beginning and at the end of the season, variation in the amount of fat content
was remarkably high (Fig. 5). Part of the variation is clearly due to the low accuracy
of the indirect estimates of fat content (eg, all the negative values). However, at the
end of hibernation, the modal fat content in males was between 0 and 1 g (Fig. 5).
Thus, at least in some individuals the amount of fat reserves at the end of
hibernation must have been close to zero.
Note also that a few individuals had less than two grams of estimated fat content
at the beginning of hibernation (Fig. 5a). Even if the values were highly under-
estimated (eg by 2 g), the bats would not survive the winter if they expended energy
at a population average rate. Moreover, one of the individuals used for calibration
indeed had only 1.65 g fat in December, ie, about three times less than the amount
used by an average individual during the rest of hibernation. Thus, maintaining
energy balance seems to be a challenge to at least some individuals of the mouse-
-eared bats hibernating in Poland.
Acknowledgements: The field work would not be possible without the help of several volunteer
students, especially W. Ga³osz, M. Labocha, and T. Postawa. We are also grateful to J. Weiner,
Y. Winter and two anonymous reviewers for critical comments, and to D. Stone for proofing the
language. The study was founded by KBN grant 6P04C 137 08.

12
P. Koteja et al.
References
Bogdanowicz W. and Urbañczyk Z. 1983. Some ecological aspects of bats hibernating in city of Poznañ.
Acta Theriologica 24: 371–385.
Froncisz W., Piasecki W., Koteja P., Staliñski J. and Weiner J. 1994. A new instrument for non-
-invasive measurement of total body water and fat content in small mammals. Polish Ecological
Studies 20: 323–328.
Harmata W. 1987. The frequency of winter sleep interruptions in two species of bats hibernating in
limestone tunnels. Acta Theriologica 32: 331–332.
Hays G. C., Speakman J. R. and Webb P. I. 1992. Why do brown long-eared bats (Plecopus auritis) fly
in winter? Physiological Zoology 65: 554–57.
Jurczyszyn M. 1995. [Ecology of wintering in bats from the genus Myotis Kaup (Chiroptera, Vesper-
tilionidae) within Fort I in Poznañ]. Ph D thesis, A. Mickiewicz University, Poznañ: 1–79. [In Polish]
Jurczyszyn M. and Bajaczyk R. 1996. Arousal frequency large mouse-eared bat, Myotis myotis
(Borkhousen, 1797) (Chiroptera, Mmmalia) hibernating in Fort I in Poznañ. Badania Fizjo-
graficzne nad Polsk¹ Zachodni¹, Seria C, Zoologia 43: 67–72. [In Polish with English summary]
Koteja P. 1996. The usefulness of a new TOBEC instrument (ACAN) for investigating body composition
in small mammals. Acta Theriologica 41: 107–112.
Nagel A. and Nagel R. 1991. How do bats choose optimal temperatures for hibernation. Comparative
Biochemistry and Physiology 99A: 323–326.
Piasecki W., Koteja P., Weiner J. and Froncisz W. 1995. New way of body composition analysis using
total body electrical conductivity method. Review of Scientific Instruments 66: 3037–3041.
Pistole D. H. 1988. Sexual differences in the annual lipid cycle of the big brown bat Eptesicus fuscus.
Canadian Journal of Zoology 67: 1891–1894.
Ransome R. D. 1990. The natural history of hibernating bats. Christopher Helm, London: 1–235.
Schutt W. A. Jr 1993. Digital morphology in the Chiroptera: the passive digital lock. Acta Anatomica
148: 219–227.
Speakman J. R. and Racey P. A. 1989. Hibernal ecology of the pipistrelle bat: energy expenditure,
water requirements and mass loss, implications for survival and the winter emergence flights.
Journal of Animal Ecology 58: 797–813.
Speakman J. R., Webb P. I. and Racey P. A. 1991. Effects of disturbance on the energy expenditure of
hibernating bats. Journal of Applied Ecology 28: 1087–1104.
Thomas D. W. 1995a. Hibernating bats are sensitive to nontactile human disturbance. Journal of
Mammalogy 76: 940–946.
Thomas D. W. 1995b. The physiological ecology of hibernation in vespertilionid bats. Symposia of the
Zoological Society of London 67: 219–231.
Thomas D. W., Dorais M. and Bergeron J. M. 1990. Winter energy budgets and cost of arousals for
hibernating little brown bats, Myotis lucifugus. Journal of Mammalogy 71: 475–479.
Walsberg G. E. 1988. Evaluation of a nondestructive method for determining fat scores in small birds
and mammals. Physiological Zoology 61: 153–159.
Wo³oszyn B. W. 1962. [Bats from caves of Œwiêtokrzyskie Mountains]. Przegl¹d Zoologiczny 6: 156–162.
[In Polish]
Wo³oszyn B. W. 1964. [New observations of bats in caves of Œwiêtokrzyskie Mountains]. Przegl¹d
Zoologiczny 8: 286–292. [In Polish]
Received 30 June 1999, accepted 10 June 2000.