llllllllllllllllllllllllllll|l||llllllllllllllllllllllll

31293 01787 981

LIBRARY
Michigan State
University

 

 

 

 

This is to certify that the

thesis entitled
Nocturnal and Diurnal Rhythms in the
Unstriped Nile Grass Rat, Arvicanthis niloticus

presented by
Julie Anne Blanchong

 

has been accepted towards fulfillment
of the requirements for

 

Masters degree in locum— J

' " 7
1%" {A 1 0/211”

Major professor

M I
Date arch 4 1999

 

0-7639 MS U is an Affirmative Action/Equal Opportunity Institution

 

PLACE IN RETURN BOX to remove this checkout from your record.
TO AVOID FINE return on or before date due.
MAY BE RECALLED with earlier due date if requested.

 

DATE DUE

DATE DUE

DATE DUE

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

1“ WWW“

NOCTURNAL AND DIURNAL RHYTHMS IN THE UNSTRIPED
NILE GRASS RAT, Arvicanthis niloticus

By

Julie Anne Blanchong

A THESIS

Submitted to
Michigan State University
in partial fulfillment of the requirements
for the degree of

MASTER OF SCIENCES
Department of Zoology

1999

ABSTRACT

NOCTURNAL AND DIURNAL RHYTHMS IN THE UNSTRIPED
NILE GRASS RAT, Arvicanthis niloticus

By

Julie Anne Blanchong

In a laboratory population of unstriped Nile grass rats, Arvicanthis niloticus,
individuals with two distinctly different patterns of wheel-running exist, one that is
diurnal and the other that is relatively nocturnal. In the first experiment, I found that these
patterns are strongly influenced by parentage and by sex. Specifically, offspring of two
nocturnal parents were significantly more likely to express a nocturnal pattern of wheel-
running than were offspring of diurnal parents, and more females than males were
nocturnal. In the second experiment, I found that diurnal and nocturnal wheel-runners
were indistinguishable with respect to the timing of post-partum mating, which always
occurred in the hours before lights-on. Furthermore, pre-weaning young and their parents
all exhibited diurnal patterns of general activity even if they were nocturnal when housed
with a wheel. In the third experiment, I discovered that adult female A. niloticus with
nocturnal patterns of wheel-running were nocturnal with respect to general activity and
core body temperature when a running wheel was available, but diurnal when the running
wheel was removed. Finally, a field study revealed that A. m'loticus are almost
exclusively diurnal in their natural habitat. Together these results suggest that individuals
of this species are fundamentally diurnal, but that access to a running wheel shifts some

individuals to a more nocturnal pattern.

ACKNOWLEDGMENTS

This research was supported by NIMH grant #MHS3433 and a Michigan State
University AURIG to Laura Smale. I thank my advisor Laura Smale for support
throughout my projects. I also thank the other members of my committee, Kay Holekamp
and Antonio Nunez, for guidance in designing experiments and reporting results. I am
also grateful to Terri McElhinny and Megan Mahoney who were my co—investigators in
many of these experiments. I thank Colleen Novak for helpful comments and David
McFarlane for technical support. I also thank personnel in the Office of the President of
Kenya for granting me permission to conduct the field research and the Senior Warden of
the Masai Mara National Reserve for his cooperation. I also thank Kay Holekamp and the

National Science Foundation for the use of facilities in the Masa Mara.

iii

TABLE OF CONTENTS

LIST OF TABLES ......................................................................................................... vi
LIST OF FIGURES ....................................................................................................... vii
CHAPTER 1

INTRODUCTION TO CIRCADIAN RHYTHMS AND ARVICANT HIS NILOTIC US ....1

CHAPTER 2
FACTORS INFLUENCING WHEEL-RUNNING PATTERNS IN A. NILOTICUS
Introduction ..................................................................................................................... 5
Methods
General Methods .................................................................................................. 6
Experiment 1 ........................................................................................................ 6
Experiment 2 ........................................................................................................ 8
Results
Experiment 1 ........................................................................................................ 8
Experiment 2 ...................................................................................................... 15
Discussion ..................................................................................................................... 20
CHAPTER 3

TIMING OF MATING IN DIURNAL AND NOCTURNAL MATING COUPLES AND
DEVELOPMENT OF RHYTHMS IN JUVENILES

Introduction ................................................................................................................... 24
Methods ......................................................................................................................... 25
Results ........................................................................................................................... 26
Discussion ..................................................................................................................... 28
CHAPTER 4

GROSS MOTOR ACTIVITY AND BODY TEMPERATURE RHYTHMS IN
DIURNAL AND NOCTURNAL INDIVIDUALS

Introduction ................................................................................................................... 29
Methods ......................................................................................................................... 30
Results ........................................................................................................................... 3 1
Discussion ..................................................................................................................... 34
CHAPTER 5

iv

ACTIVITY PATTERNS OF A. NILOTI C US IN THEIR NATURAL HABITAT

Introduction ................................................................................................................... 40
Methods ......................................................................................................................... 41
Results ........................................................................................................................... 42
Discussion ..................................................................................................................... 5 1
CONCLUSIONS ........................................................................................................... 55
REFERENCES .............................................................................................................. 59

LIST OF TABLES

Table 1. Analysis of Differences Between Groups With Respect to the Hour After Lights-
Out That Wheel-Running Stopped .......................................................... 14

Table 2. Average Number of Animals Captured Per Site ............................................... 49

Table 3. Number of Animals Captured in Period 1 and Period 2 at Isolated Sites .......... 50

vi

LIST OF FIGURES

Figure 1. Actograms of typical (A) diurnal and (B) nocturnal wheel-running patterns of
A. niloticus ..................................................................................................................... 10

Figure 2. Histogram representing percentages of nocturnal male and female offspring
from each category of mating couple ............................................................................. 11

Figure 3. Histogram representing numbers of offspring of each category of mating couple
with activity offsets occurring in each of 8 one-hour intervals following lights-
out ................................................................................................................................. 12

Figure 4. Hour of activity offset and SEM (standard error of the mean) for animals
classified as nocturnal or diurnal at the end of their first and third months in running
wheels ............................................................................................................................ 16

Figure 5. Average number of wheel revolutions (SEM) per day for animals that ceased
wheel-running in each of 8 one-hour intervals following lights-out ................................ 17

Figure 6. Average number of wheel revolutions per hour of animals that stopped running
1 hour (open circles) or 8 hours (dark circles) after lights-out ........................................ 18

Figure 7. Actogram depicting a nocturnal animal maintained in constant darkness ....... 19

Figure 8. Average activity patterns (# of 10 minute intervals off the nest per hour) of
offspring from (A) diurnal and (B) nocturnal parents over time. (C) Average activity of
parents at day 20 ............................................................................................................ 27

Figure 9. Average daily patterns of general activity and SEM for diurnal (open circles),
intermediate (triangles) and nocturnal (dark circles) animals without a running wheel
(phase 1) ........................................................................................................................ 32

Figure 10. Average daily patterns of general activity and SEM for diurnal (open circles),
intermediate (triangles) and nocturnal (dark circles) animals with a running wheel (phase
2) ................................................................................................................................... 33

vii

Figure 11. Representative actogram of general activity during phasesl,2 and 3 for a
diurnal animal (lights-on 0700, lights-out 1900) ............................................................ 35

Figure 12. Representative actogram of general activity during phasesl,2 and 3 for a
nocturnal animal (lights-on 0700, lights-out 1900) ......................................................... 36

Figure 13. Average daily patterns of core body temperature and SEM for diurnal (open
circles), intermediate (triangles) and nocturnal (dark circles) animals without a running
wheel (phase 1) .............................................................................................................. 37

Figure 14. Average daily patterns of core body temperature and SEM for diurnal (open
circles), intermediate (triangles) and nocturnal (dark circles) animals with a running
wheel (phase 2) .............................................................................................................. 38

Figure 15. Total number of A. niloticus captured in timer and non-timer traps in each of
six 4-hour intervals throughout the 24-hour day (sun rises 0630, sets 1915) ................... 44

Figure 16. Percent of male (black bars) and female (gray bars) A. niloticus caught in
timer and non-timer traps in each of six 4-hour intervals throughout the 24-hour day (sun
rises 0630, sets 1915) ..................................................................................................... 46

Figure 17. Percent of adult (black bars) and juvenile (gray bars) A. niloticus captured in
timer and non-timer traps in each of six 4-hour intervals throughout the 24-hour day (sun
rises 0630, sets 1915) ..................................................................................................... 47

Figure 18. Number of A. niloticus (gray bars) and other rodent species (black bars)
captured in timer-traps in each hour of the 24-hour day (sun rises 0630, sets 1915) ....... 48

viii

CHAPTER I

INTRODUCTION TO CIRCADIAN RHYTHMS AND Arvicanthis niloticus

An animal’s environment fluctuates in a rhythmic and relatively predictable
manner over the course of the 24-hour day. As a result, animals face temporally varying
challenges related to finding food, reproducing and avoiding predators. In order to
enhance their chances for survival animals must ‘do the right thing at the right time’
(Moore-Ede et al., 1982). To accomplish this, animals possess circadian systems
responsible for measuring time and synchronizing internal processes with daily cycles in
the environment (Moore-Ede et al., 1982). These systems contain physiological
pacemakers synchronized with changes in the environment (Moore-Ede et al., 1982).
These internal ‘clocks’ enable animals to anticipate changes in the environment, prepare
in advance for periods of activity and promote internal coordination of physiological and
behavioral processes (Moore-Ede et al., 1982).

Most animals are specialized to be active at a particular phase of the 24-hour day.
An animal’s activity is generally organized in one of three broad patterns: diurnal, in
which activity is concentrated in the daytime when it is light; nocturnal, in which activity
is concentrated during the night when it is dark, or crepuscular where activity is clustered
at dawn and dusk (Moore-Ede et al., 1982). Concentration of activity at a particular time
of day leads to specializations of other behaviors as well. For example, diurnal animals
have sharp vision and tend to rely on social anti-predator strategies while nocturnal
animals rely less on their visual systems and often use cryptic strategies for predator

avbidance (Daan, 1981). Because so many differences exist between diurnal and

nocturnal lifestyles, evolutionary transitions from one of these behavior patterns to the
other have been relatively rare.

The first mammals were nocturnal, and this remains the most common pattern of
activity in modern mammals (DeCoursey, 1990). In the order Rodentia most families
have remained nocturnal. Notable exceptions include most members of the family
Sciuridae, some genera of murids and many South American rodents in the families
Caviidae, Dasyproctidae, Chinchillidae, Capromyidae and Octodontidae (Nowak and
Paradiso, 1983). While most species of rodents can be categorized as diurnal, nocturnal or
crepuscular there is some evidence of species in which diurnal and nocturnal individuals
coexist. Among cotton rats, Sigmodon hispidus, some individuals are diurnal while others
are nocturnal in the laboratory and the field (Johnston and Zucker, 1983). In their lab
study, Johnston and Zucker (1983), found that half of the juvenile cotton rats showed
diurnal patterns of wheel-running while half were nocturnal. As adults, most, but not all,
animals showed predominantly nocturnal patterns. However, nocturnal adults sometimes
switched to become diurnal either temporarily or permanently. Octodon degus are diurnal
South American rodents in which morning and evening ‘chronotypes’ coexist in both the
laboratory and the field (Labyak et al., 1997). These different chronotypes exhibit distinct
differences in the patterns of their activity and temperature rhythms (Labyak et al.,
1997).

A variety of species of voles, mice and lemmings show seasonal variations in
their timing of activity in order to avoid the harsh seasonal microclimates (Rowsemitt,
1986; Madison, 1985). Specifically, during the summer when days are longer and

temperatures are higher, these animals tend to be more crepuscular, and more active

 

during the cooler parts of the day, whereas in the colder winter when days are shorter they
are more diurnal, and express relatively high levels of activity during the middle of the
day. For example, montane voles, Microtus montanus, from northern latitudes showed
crepuscular patterns of activity under long-day conditions in the laboratory and in
summer field conditions, while they were more diurnal under short-day lab conditions
and winter field conditions (Claypool, 1984). Some species may also switch from one
pattern of activity to another in response to the presence of other species. The golden
spiny mouse, Acomys russatus, native to hot, arid regions, is normally nocturnal, but
when it is sympatric with A. cahrinus it switches and becomes diurnal (Haim and
Rozenfeld, 1993). A. russatus also change their activity patterns in response to seasonal
changes in photoperiod and ambient temperature, in a manner similar to that described
above in some other rodent species (Friedman, 1997).

The unstriped Nile grass rat, Arvicanthis niloticus, appears to be another species
of rodent in which individuals vary with respect to the distribution of their activity
relative to the light-dark cycle. These tropical rodents are widespread in equatorial Africa
and are usually the dominant rodent in their ecosystem (Kingdom, 1974). They are stout-
bodied rodents five to seven inches in length not including the tail. The coat is sandy
colored and peppered with black. Their heads are blunt with a short muzzle and circular
ears (Kingdom, 1974). A. niloticus are highly opportunistic and almost entirely
herbivorous, though analysis of stomach contents sometimes reveals traces of insects
(Senzota, 1982). They favor drier grasslands as the major habitat, live in burrows
constructed beneath bushes that provide cover, and travel in runways that radiate outward

from burrow entrances (Packer, 1983).

J

Although considerable research has been done on A. niloticus, there is a lack of
consensus with regard to the pattern of its activity rhythms. Systematic field data have not
been reported on this issue, however, while investigating other aspects of this species’
ecology, some researchers have concluded that A. niloticus are primarily diurnal (Quilici
et al., 1969; Delany and Kansiimeruhanga, 1970; Kingdon, 1974; Packer, 1983; Rabiu
and Fisher, 1989). However, Senzota (1990) describes them as diurnal with crepuscular
tendencies, while others have reported them to be partially nocturnal (Ansell, 1960;
Vesey-Fitzgerald, 1966; Rosevear, 1969; Harrison, 1972). Still others assert they are
chiefly nocturnal (Schmutterer, 1969; Ghobrail and Hodeib, 1982).

This thesis describes my efforts to systematically characterize the daily rhythm
patterns of A. niloticus in both a laboratory and a field setting. Chapter 2 describes my
efforts to determine if an individual’s pattern of activity is influenced by its parentage or
by its sex. Chapter 3 describes the timing of mating in diurnal and nocturnal A. niloticus
as well as the ontogeny of rhythrnicity in juveniles of this species. In Chapter 4, patterns
of gross motor activity and core body temperature are compared in diurnal and nocturnal
individuals. Finally, Chapter 5 describes my efforts to characterize this species’ pattern of

activity in its natural habitat in Kenya, Africa.

CHAPTER 2
FACTORS INFLUENCING WHEEL-RUNNING

PATTERNS IN A. NILOTIC US

INTRODUCTION

Almost all animals can be categorized as diurnal, nocturnal or crepuscular.
However, there are some rodent species within which both diurnal and nocturnal patterns
are exhibited (e.g. cotton rats, Sigmodon hispidus, Johnston and Zucker, 1983; degus,
Octodon degus, Laybak et al., 1997). The unstriped grass rat, Arvicanthis niloticus,
appears to be another such species.

Within a population brought to the laboratory from the Masai Mara National
Reserve (MMNR) in Kenya, most individuals tested in running wheels were diurnal, with
elevated levels of activity during the daylight that decreased dramatically shortly after
dark. One female, however, showed a distinctly different and nocturnal pattern of wheel-
running with high levels of activity several hours after the lights went out. She was mated
with two males and 65% (15/23) of her offspring also expressed this nocturnal pattern of
wheel-running. Only a small proportion of the offspring of other breeding pairs expressed
nocturnal patterns of running (7%, 3/43) .

Though the coexistence of diurnal and nocturnal individuals has been reported in
a number of rodent populations, nothing is currently known about the causes of this
intraspecific variability. In this experiment, the range of activity patterns seen in a

laboratory population of A. niloticus is described as well as efforts to determine if a grass

hi

rat’s wheel-running pattern can be predicted by the activity patterns of its parents, by its

sex or by its age.

METHODS
General Methods

In August 1993, 29 A. niloticus were captured in Masai Mara National Reserve,
Kenya and imported to the USA. These animals were housed at Michigan State
University and breeding couples were formed to establish a laboratory colony of animals.
All animals were maintained in a 12: 12 light-dark cycle and maintained at a constant
temperature of 22 degrees Celcius. Animals were housed in 34 x 30 x 16 cm cages
containing aspen chip bedding. All animals were provided with water, Harlan Teklad
rodent diet 8640, carrots and oats. Mating couples were supplemented with Harlan Teklad
high-fiber rabbit diet 7015 once a week and were provided with plastic tube or metal
shelters. Juveniles were housed with their parents until 21 days of age at which time they
were weaned and placed as a group into a separate cage. At 50 days of age male and
female siblings were separated.

To monitor wheel-running, animals were housed individually and provided with a
seven cm diameter running wheel to which a magnet was attached. With each revolution
of the wheel, the magnet closed a switch attached to the cage. The number of running
wheel revolutions was recorded in five minute bins using the Dataquest IH Minimitter

system (SunRiver, OR).

EXperiment 1

I established mating couples of A. niloticus with either (1) two diurnal (DD; 11: 4),
(2) two nocturnal (NN; n: 4) or (3) one diurnal and one nocturnal (DN, n: 6; 3N females,
3D females) animals. Prior to pairing, the rhythms of these breeders were determined by
testing them individually in waning-wheels for two weeks. I housed offspring of these
animals with both parents until weaning (21 days of age), at which time they were placed
into individual cages with running wheels (11: 38, 52, and 38 from DD, DN and NN
mating couples respectively). Wheel-running was recorded continuously for one month to
establish whether patterns were diurnal or nocturnal and to assess the hour after lights-out
in which animals became inactive.

These analyses were based on data averaged over the last five days of the first
month of testing in running wheels. I defined diurnal animals as those whose activity by
the fifth hour after lights-out was less than ten percent of the 24 hour peak. Nocturnal
animals were those whose activity during and/or following the fifth hour after lights-out
was greater than twenty percent of the daily peak. I defined the hour after lights-out in
which an animal ceased to be active as the hour during which activity dropped to less
than ten percent of the 24 hour peak.

I continued to monitor a random subset of offspring from each category of mating
couple for an additional two months to test for the effects of age on the pattern of activity
(n=12, 17 and 19 offspring of DD, DN and NN mating couples, respectively). I classified
these animals as diurnal or nocturnal at the end of the third month, again by averaging the
hourly activity over the last five days of each testing period. These data were also
analyzed to determine the hour after lights-out in which animals stopped wheel-running.

In order to be certain that the differences between diurnal and nocturnal animals

were not the result of masking, three nocturnal offspring were placed into constant
darkness after three months in a 12:12 light-dark cycle.

Data were analyzed using chi-square tests with Yate’s corrections for continuity,
analyses of variance (AN OVA) with repeated measures, Pearson correlations and t-tests.
Results were considered significant when p<0.05.

Experiment 2

I tested whether the influence of fathers on their offsprings’ rhythm patterns was
mediated by the experience of growing up in the presence of the father. I tested this by
comparing the activity patterns of animals that were raised by their diurnal mother with or
without their nocturnal father. The parents’ wheel-running patterns were determined by
testing in running wheels for one month prior to pairing. Four mating couples consisting
of a diurnal female and a nocturnal male were established. When a litter was born the
father was removed, and the female was left to raise the pups on her own in two of the
mating couples. For the other two mating couples, both parents remained with pups until
they were weaned. All pups were weaned at twenty-one days of age and immediately
placed in individual running wheels where they were kept for one month in order to
assess their pattern of activity. After one month these wheel-running patterns were
assessed to determine if they were nocturnal or diurnal as in the first experiment. Patterns
that developed in animals that were raised with diurnal mothers with or without the

nocturnal fathers were compared.

RESULTS

E’xmriment 1

Almost all animals exhibited clear patterns of rhythmic wheel-running that fit
characteristic diurnal (Figure 1a) or nocturnal (Figure lb) patterns. Of 128 animals
tested, six could not be categorized as clearly diurnal or nocturnal and were classified as
intermediate. Among the remaining 122 animals, nocturnal patterns of activity were
expressed by 14% (5/37) of offspring from DD parents, 53% (25/47) from DN parents
and 68% (26/38) from NN parents. The ratio of diurnal to nocturnal offspring differed
significantly among the three categories of mating couples (x22=26.9, p<0.001).
Specifically, there were more nocturnal offspring from NN mating couples than from DD
mating couples (x21: 20.0, p<0.001) and from DN parents than from DD parents (x2,=
15.8, p<0.001). When male and female offspring were analyzed separately 1 found a
difference among mating couple types with respect to the proportion of both their male
offspring (x22: 9.3, p<0.01) and their female offspring (x22: 13.6, p<0.005) that were
nocturnal. In addition, I found that when animals from all mating couples were pooled,
females were more likely to express a nocturnal pattern of activity than were males (x21:
12.5, p<0.001; Figure 2). Specifically, among offspring of NN parents, more females
(89%) than males (50%) were nocturnal (x21: 5.9, p<0.01; Figure 2). The sex difference
among offspring of DN (x21: 2.5, NS; Figure 2) and DD parents was not significant (x21:
0.3, NS; Figure 2).

In order to examine the influence of parentage and sex more precisely, I compared
offspring of different mating couple types with respect to the hour after lights-out in
which they ceased to be active (Figure 3). This value differed as a function of the parents’
rhythm patterns. Offspring of NN parents continued their activity after lights-out longer

than did offspring of DN parents (x28: 17.2, p<0.05) or DD parents (x23: 26.3,

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Figure ' l. Actograms of typical (A) diurnal and (B) nocturnal wheel-running patterns of
. A. niloticus.

10

100

 

 

E males
females
80 —
C)
.E
h
8
a so —
E
c
I—
.2
8 4o _
c
o..—
O
o\°
20 —
0

 

 

 

 

 

 

   

Mating couple categories

Figure 2. Histogram representing percentages of nocturnal male and female offspring
from each category of mating couple.

 

Number of animals

 

 

 

.0 1 2 3 4 5 6 7 8 9
Last hour of activity following lights-out

Figure 3. Histogram representing numbers of offspring of each category of mating couple
With activity ofl'sets occurring in each of 8 one-hour intervals following lights-out.

p<0.001). In addition, offspring of DN parents continued to be active longer after lights-
out than offspring of DD parents (x28: 27.6, p<0.001). Overall, the distribution of animals
that stopped wheel-running at various times after lights-out was distinctly bimodal, with
many animals stopping at one hour or six hours or more, while relatively few stopped at
intermediate times (Figure 3). When I analyzed the sexes separately, I found that female
offspring of the different types of mating couple ceased wheel-running at different times
after lights-out (Table 1). Specifically, females of NN parents continued to be active
longer than did females from DD parents. I found no significant differences when
comparing female offspring of DN parents to those of DD parents or NN parents with
respect to the time after lights-out when wheel-running stopped (Table 1). This analysis,
in contrast to my more powerful categorical analyses, also failed to detect significant
differences between male offspring of the three different categories of mating couples.
Overall, males and females differed with respect to the hour after lights-out in which they
became inactive (Table 1). Specifically, females continued to be active longer after lights-
out than did males.

To determine if activity patterns changed as animals aged, I conducted a repeated
measures AN OVA with which I compared the hour at which wheel-running stopped after
one and three months in wheels, in animals originally classified as diurnal or nocturnal.
The pattern of activity expressed by an individual in month one changed slightly, but
significantly by month three (F2: 3.132, p=0.049). This difference is due to 10 out of 38
animals that switched their patterns completely. When the data was analyzed without
these animals I found that there was no effect of age on the pattern of activity (F2: 2.362,

p=0.105). I also found an interaction between the initial pattern of activity (diurnal or

13

Table 1. Analysis of Differences Between Groups With Respect to the Hour After
Lights-Out That Wheel-Running Stopped.

 

 

 

 

Animals Comparison x3 df p value
Females DD:DN:NN 34.0 16 <0.0l
DDZNN 18.0 8 <0.025
DDIDN 3.8 8 NS
DNINN 13.2 8 NS
Males DDiDNINN 22.3 16 NS
All ‘ MZF 22.4 3 <0.005
DD M:F 7.7 8 NS
DN MP 4.1 8 NS
NN Mil: 1 1.7 8 NS

 

 

 

14

nocturnal) and the direction of change in the pattern over time (F2: 19.647, p<0.001).
Specifically, animals classified as diurnal in month one remained active longer after
lights-out at month three than at month one, while animals initially classified as nocturnal
became inactive sooner after lights-out at month three than at month one (Figure 4).
Despite these changes, at the end of three months, animals initially classified as diurnal
remained significantly different from those initially classified as nocturnal (t44= 4.898,
p=0.001, Figure 4).

A one-way AN OVA comparing the total number of daily wheel revolutions by
animals that stopped running in each of the first eight hours after lights—out showed that
there was a difference in the total amount of wheel-running among animals that stopped
activity at different times after the lights went out (F7: 8.04, p: 0.001). Animals that
continued to run for eight hours after lights-out ran approximately 49% more than did
animals that ceased running in the first hour after lights-out (Figure 5; t57=5.78, p=0.001).
However, the heightened activity among nocturnal individuals was disproportionately
concentrated in the dark period of the 12:12 light-dark cycle. This can be seen in the
comparison of hourly wheel-running rates of individuals that stopped running one or
eight hours after lights out (Figure 6). Furthermore, the number of hours that animals
were active after lights—out was actually negatively correlated with the number of hours
they were active during the light period (r: 0.594, p: 0.001, Pearson Correlation). Thus,
nocturnal animals were not simply more active than diurnal ones, but rather, the
fundamental waveform of their rhythm differed. This basic nocturnal waveform was

maintained in the three animals that were released into constant darkness (Figure 7).

Experiment 2

15

 

 

 

 

 

 

 

 

A + Nocturnal
2 8 - r —o— Diurnal
LU
(D
r' 7-
c
g 6"
g _
35
e 5~
0
a? 4 _
...>.
E
u— 3 ~
0
b
8
I 2 ‘
1 , I I
1 3
Month

Figure 4. Hour of activity offset and SEM (standard error of the mean) for animals
classified as noctumal or diurnal at the end of their first and third months in running

wheels.

16

 

2500

2000-

1500 -

1000 —

 

 

Average Daily Revolutions

 

 

 

o 1 2 3 4 5 6 7 a 9
Hour After Lights-Out

Figure 5. Average number of wheel revolutions (SEM) per day for animals that ceased
wheel-running in each of 8 one-hour intervals following lights-out.

\
I ¢Ou~ putt < \a-ilu of Ti

 

 

 

 

 

250
200 _
150 4 9
b j. '.
E -' o
100 - ‘
E 5. . .030” .-
0 0.0. .
I '0- -
50 1 O 0.0.
'O O
.6 .
0., 0.0.... O"O"O'O-O
2 4 6 8 10 12 14 16 18 20 22 24
-

 

 

Time

Figure 6. Average number of wheel revolutions per hour of animals that stopped running
1 hour (open circles) or 8 hours (dark circles) after lights-out.

l8

11/247?@:@@:@@.@@ RCTOGRRM Page 1

 

Figure 7. Actogram depicting a nocturnal animal maintained in constant darkness

Three litters (n: 11 offspring) were raised with the father present and three litters (n=10

offspring) were raised without the father present. The proportion of animals in these two
groups that were nocturnal was almost identical. Eight out of eleven pups raised with the
father present expressed a nocturnal pattern of wheel-running while eight out of ten pups

raised without the father were nocturnal.

DISCUSSION

These results suggest that the group-living murid rodent species, A. niloticus,
contains individuals with two relatively distinct patterns of daily wheel-running activity.
The most dramatic difference between these nocturnal and diurnal individuals occurred
during the hours after lights-out. Most animals increased wheel-running immediately after
lights-out (Figure 6), but then some stopped running within one or two hours while others
continued for approximately six to eight hours, and relatively few animals stopped at
intermediate times (Figure 3). In contrast to these evening bouts of activity, which were
bimodally distributed, activity levels at other times of day were more normally
distributed. Animals that ran for a long time at night ran more overall (Figure 5), but the
elevated running was disproportionately concentrated in the evening (Figure 6). In fact,
there was an inverse correlation between the number of hours that animals ran during the
light and dark periods of the day.

One possible interpretation of the differences between nocturnal and diurnal
wheel-running patterns involves masking. For example, nocturnal individuals might find
the light aversive or frightening, and may therefore wait until dark before becoming

active. This interpretation was tested by transferring nocturnal individuals to constant

20

darkness, which did not change the fundamental waveform (Figure 7). That is, the
evening bout of wheel-running remained lengthy, and daytime wheel—running remained
minimal. Earlier work found that constant darkness similarly did not substantially alter
the waveform of the wheel-running rhythm of diurnal A. niloticus (Katona and Smale,
1997). Differential masking effects of light are therefore unlikely to account for
differences between nocturnal and diurnal wheel-running patterns.

Individual differences in A. niloticus wheel-running patterns were strongly
influenced by parentage and by sex. Specifically, animals with two nocturnal parents were
significantly more likely to be nocturnal than were animals with two diurnal parents, and
females were more likely to be nocturnal than were males. Although the data provided by
this study cannot definitively rule out possible mechanisms by which mothers and fathers
influence the development of rhythm pattern, these data can help to clarify some
possibilities. If the pattern of wheel-running was influenced solely by rearing
environment it would be expected that the majority of offspring raised only in the
presence of one diurnal parent would have shown a diurnal pattern of wheel-running. The
pattern of results obtained in experiment two was inconsistent with this prediction. The
results of this experiment suggest that the influence of the father on the rhythms of his
offspring is not mediated by his presence, but, most likely, by his genes.

Several genetic mechanisms could account for the effective of selective breeding
on rhythm patterns in A. niloticus. First, a single autosomal gene could account for the
influence of parentage on whether an animal is nocturnal or diurnal. If this is the case,
however, then that gene does not have complete penetrance. If it did, then either NN or

DD mating couples would have two recessive alleles, and 100% of their offspring would

21

exhibit the same pattern as the parents, which was not what I found. However, it remains
possible that a single gene responsible for the activity pattern may have incomplete
penetrance such that a small subset of the population that is homozygous for this gene
would express the alternate phenotype. For example, some NN individuals may be
diurnal, or some DD individuals could be nocturnal because of a non-genetic source of
variability. Furthermore, the sex difference suggests that if a single allele is involved,
then its penetrance would have to be different in males and females. A second possibility
is that a single gene with a sex-linked pattern of dominance determines whether an animal
is nocturnal or diurnal. Specifically, a nocturnal allele could be dominant in females and
recessive in males. This pattern has been documented for the inheritance of some forms
of baldness in humans and for the spotting pattern of Ayrshire cows (Dodson, 1956;
Griffiths et al., 1993). According to this hypothesis, males and females heterozygous for
this gene would exhibit different phenotypes. This hypothesis is consistent with this data.
A third possibility consistent with the results is that multiple genes are involved in
determining whether an animal exhibits a nocturnal or diurnal pattern of wheel-running.
Populations within which individuals with dramatically different patterns of
rhythrnicity coexist have been described in a variety of other rodent species, including
degus (Octodon degus; Labyak et al., 1997), cotton rats (Sigmondon hispidus; Johnston
and Zucker, 1983), a variety of vole species (Madison, 1985), inbred strains of rats
(W ollnik, 1991) and selected lines of mice (Bult et al., 1993). Rhythm “chronotypes”
have also been described in degus, a diurnal species in which a relatively rare evening
chronotype exhibits delayed nighttime drops in body temperature and wheel—runnin g

aetivity (Labyak et al., 1997). This species is different from A. niloticus in that the

22

evening chronotypes are at one end of a relatively normal distribution, rather than one end

of a bimodal distribution.

23

CHAPTER 3
TIMING OF MATING IN DIURN AL AND N OCTURNAL INDIVIDUALS AND

DEVELOPMENT OF RHYTHMS IN JUVENILES

INTRODUCTION

The previous experiments have demonstrated that the timing of wheel-running
differs dramatically between individual A. niloticus and that this difference is influenced
by their parentage and by their sex. The coexistence of individuals with distinctly
different patterns of wheel-running raises the question of whether other aspects of
temporal organization are different in nocturnal compared to diurnal wheel-runners.
Nocturnal and diurnal species typically differ with respect to patterns of a host of
variables, including hormone secretion, sleep, body temperature, parturition, mating and
general activity (Moore-Ede et al., 1982; McElhinny, 1996). In this and the following
chapter, I compared nocturnal and diurnal wheel-runners with respect to the timing of
reproductive parameters as well as rhythms of body temperature and general activity.

The time of mating coincides with the time of peak activity in most animals. For
example, female rats are most sexually receptive in the first six to nine hours after dusk
(Keuhn and Beach, 1963; Hardy, 1972), when they are most active. Previous
investigation of the time of mating of diurnal A. niloticus indicates that, unlike nocturnal
rodents, they mate in the early morning prior to lights-on (McElhinny et al., 1996). In this
study, the time of mating was compared between A. niloticus that expressed diurnal and

nectumal wheel-running patterns in order to determine if diurnal and nocturnal

24

individuals’ mate at different times of day. I also monitored the development of rhythm
patterns in pups before they were weaned, and compared these to their wheel-running

patterns expressed after they were weaned.

METHODS

Diurnal (n=2) and nocturnal (n=4) mating couples whose activity rhythms had
been determined by testing in running wheels for two weeks were paired and placed in
aquaria with aspen chip bedding and a plastic tube for shelter. All animals were kept in a
12:12 light-dark cycle with a dim red light to allow for videotaping when the lights were
off. Around-the-clock videotaping of mating couples began eighteen days after animals
were paired in order to capture the time of parturition (21 days after mating) and post-
partum mating. The time of parturition was recorded as the time at which the first pup
could be seen. The time of mating was recorded as the first incidence of lordosis by the
female in response to male mounting. Five days after parturition, litter size was culled to
two and one of the pup’s fur was marked with Nyanzol-D hair dye to allow for individual
identification of pups on videos. Videotaping continued until litters were weaned 21 days
after they were born. Videotapes from days 10, 15 and 20 after parturition were analyzed
in ten minute intervals for the entire 24-hour day in order to determine each individual’s
pattern of activity (adults and juveniles). An animal was considered active if it was off the
nest area for at least five minutes of the ten minute interval. At day 21, pups were placed
in individual running wheels for one month to determine their pattern of wheel-running.
Data were analyzed using t-tests and chi-square tests, and were considered significant

When p<0.05.

25

RESULTS
Parturition occurred at many different times throughout the 24-hour day during

lights-on and lights-out for both diurnal (4 litters) and nocturnal (8 litters) mating couples.
There was no difference between diurnal and nocturnal mating couples in the time of day
in which parturition occurred (t5= 0.68, NS). All post-partum mating occurred in the early
hours of the morning. This was true for animals that had exhibited nocturnal as well as
diurnal patterns of wheel-running. In contrast to other nocturnal rodents, which mate in
the first hours after lights-out, nocturnal A. niloticus mated an average of one hour and 36
minutes (range= 4 hrs prior- 1 hr after) before lights-on. Diurnal mating couples mated an
average of 3 hours and 5 minutes (range= 5- 1 hr) prior to lights-on. Diurnal and
nocturnal animals thus, both mated in the early morning.

When videotapes were analyzed to evaluate juvenile activity patterns at 10, 15 and
20 days of age it was apparent that all animals, including those that were subsequently
nocturnal in running wheels, exhibited diurnal activity (Figure 8a,b). Clear rhythms did
not exist at day 10 when juveniles rarely left the nest, but were apparent in all animals by
day 15. Surprisingly, all parents were diurnal with respect to their patterns of general
activity (Figure So), even those that had been nocturnal in running wheels. The temporal
distribution of activity of all juveniles paralleled that of their parents throughout
development (Figure 8). Juveniles exhibited little activity at day 10, and activity increased
progressively until weaning (Figure 8a,b). When juveniles were placed in individual
running wheels at weaning, 12/13 and 2/6 with nocturnal and diurnal parents,
respectively, exhibited nocturnal patterns of wheel-running. More offspring from

necturnal parents exhibited nocturnal patterns that did those with diurnal parents

26

 

 

 

 

 

 

 

+Day10 /T (A)
s. -o--Dav15 Y ‘
+Day20 Y I \
. ’l l l
/
>~ / \ lo \
E J \ r I Ol
'5 3 r Fw /\I l
i I \ / V l
t O f l
a 2 I ‘v
E O O \ / \/ l
1.. I, 0 000 k /0 \\
\ 9 0.0 t \ \l o\
o.

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

6
—O—- Day 10 (B)
5 1 0 Day 15
+ Day 20 /Y
1
2.‘
E
‘o‘
<
3‘
5
o
I
0 2 4 6 8 10 12 14 16 18 20 22 24
7
+ Diurnal Males i
6 ‘ o Nocturnal Males l (C)
+ Diurnal Females i
—q Nocturnal Femaies .
5 4 J
2:
E
8
<
2‘
5
o
I

 

 

 

 

Figure 8. Average activity patterns (# of 10 minute intervals off the nest per hour) of
offspring fi'om (A) diurnal and (B) nocturnal parents over time. (C) Average activity of
parents at day 20.

27

(18,: 12.8, p<0.005).

DISCUSSION

All A. niloticus engaged in post-partum copulation in the early morning regardless
of their wheel-running pattern whereas parturition occurred throughout the 24-hour day in
both groups. In addition, juveniles in this species were active almost exclusively when
their parents were active until the time of weaning. When placed in running wheels at
weaning, offspring expressed their own independent patterns of wheel-running that
usually, but not always, corresponded to their parents’ wheel-running patterns. Very low
levels of general activity were apparent in offspring at 10 days of age. Rhythms appeared
at day 15 and increased in amplitude at day 20. All animals, adults and their offspring,
expressed diurnal activity in this experiment when they did not have a running wheel,
even those with nocturnal wheel-running patterns. These data suggest that these nocturnal
patterns may not be expressed until: 1. animals are separated from conspecifics, 2.
animals are provided with a running wheel or 3. until they are 21 days of age. Alternative
number three is contradicted by the fact that nocturnal parents were diurnal when paired
without a running wheel present. However, none of these alternatives could be
definitively ruled out by this experiment, which was not designed to address this issue.

The next experiments examined these issues more explicitly.

28

CHAPTER 4
GROSS MOTOR ACTIVITY AND BODY TEMPERATURE RHYTHMS IN

DIURNAL AND NOCTURNAL INDIVIDUALS

INTRODUCTION

In the previous experiments individual A. niloticus showed distinctly different
patterns of wheel-running that were strongly influenced by parentage and the sex of the
animal. However, all animals mated at the same time of day regardless of the activity
pattern they expressed in running wheels. Therefore, while wheel-running can be useful
in the study of circadian rhythms, its relationship to other activities is uncertain, and it
does not always parallel general activity. A number of researchers have found that
patterns of wheel-running do not always resemble the pattern of activity without a wheel.
For example, Roper (1976) found that Mongolian gerbils exhibit a nocturnal wheel-
running pattern while Lewill (1974) described gerbils as diurnal on the basis of direct
observation of their behavior in the absence of a running wheel. In the wild, rabbits are
consistently nocturnal, but in the laboratory they have been found to show a wide range
of wheel-running patterns (Kennedy et al., 1994). In addition, Wistar-Imamichi rats
showed different activity patterns when ambulatory and wheel-running activites were
compared (Shinoda et al., 1988; Shinoda and Miura, 1994). In the next experiment, gross
motor activity and body temperature rhythms of A. niloticus were investigated in order to
determine if animals exhibiting diurnal and nocturnal wheel-running patterns differ when

a running wheel is no longer present.

29

METHODS

Adult female A. niloticus used in this study were first tested in running wheels for
at least one month to determine their pattern of wheel-running. Animals were classified as
diurnal (n= 4), nocturnal (n= 5) or intermediate (n= 4). All animals were anesthetized
with methoxyfurane (Metofane) and implanted intraperitoneally with paraffin-coated
Minimitter transmitters. Both muscle wall and the skin were sutured with dissolvable gut
sutures and reinforced with autoclips. Incisions were treated with Nolvasan topical
antibiotic and animals were given subcutaneous injections of 1.0 cc Lactated Ringer’s
solution and 0.2 cc of the analgesic, buprenorphine hydrochloride (Buprenex).
Transmitters emitted signals correlated with gross motor activity and body temperature.
These signals were picked up by receivers placed beneath the animal’s cage and
transmitted to a computer equipped with the Dataquest III Minimitter System, (Sun
River, OR).

All animals in the study went through the same three phases of treatment. First,
immediately following surgery, animals were placed in cages without a running wheel
and gross motor activity and body temperature were recorded for three weeks. Second, at
the beginning of the fourth week animals received running wheels, and gross motor
activity, body temperature and wheel-running were recorded for two weeks. Third, for the
final week of the study, the running wheel was removed and gross motor activity and
body temperature were recorded as in phase one.

Hourly levels of gross motor activity were averaged over a five day period for
each phase of the experiment, and the percentage of activity that occurred in each hour of

the 24-hour day was determined for each animal. A repeated measures ANOVA and F-

30

tests were performed to compare the amount of activity in the first five hours after lights-
out in the three groups and across the three phases of treatment. Average hourly body
temperatures were calculated for each animal from the same five day period of each phase
of the experiment.
RESULTS

During phase 3, transmitters in two of the nocturnal animals stopped functioning,
and data from these two animals were, therefore, excluded from statistical analysis.
However, their patterns of general activity and body temperature during phase 1 and
phase32 resembled those of the other nocturnal animals. In the absence of a rtmning wheel
(phases 1,3) all animals exhibited diurnal patterns of gross motor activity, including those
whose original wheel-running patterns had been nocturnal (Figure 9). When animals were
given access to running wheels (phase 2) all animals originally classified as nocturnal
showed nocturnal patterns of general activity while animals originally classified as
diurnal remained diurnal (Figure 10). The percent of general activity that occurred in the
first five hours after lights out was influenced by the phase of treatment (F 2= 14.360, p=
0.001), and by an interaction between the phase of treatment and the initial classification
of animals as diurnal, nocturnal or intermediate (F 4= 17.018, p= 0.001). Pairwise
comparisons were calculated using F-tests. For animals originally classified as nocturnal,
the percentage of total activity that occurred in the first five hours after lights-out changed
from phase 1 (wheel absent) to phase 2 (wheel present) (F12: 126.41, p<0.001), and from
phase 2 to phase 3 (wheel absent) (F L2: 145.08, p<0.001). Specifically, during phase 2,
when wheels were present, all animals that were originally classified as nocturnal shifted

their patterns of activity so that they became more nocturnal (Figure l 1). Animals

31

 

12

“)4

°/o of Total Activity

 

 

 

 

 

 

0 2 4 6 8 10 12 14 16 18 20 22 24
_ —

 

Time

Figure 9. Average daily patterns of general activity and SEM for diurnal (open circles),
intermediate (triangles) and nocturnal (dark circles) animals without a running wheel

(phase 1).

32

 

 

 

 

 

 

 

20
15-
3:
IE
2
_ 10~
£9.
0
j.
H-
O
o\°
5-
0-
0 2 4 6 8 10 12141618 202224
_ -
Time

Figure 10. Average daily patterns of general activity and SEM for diurnal (open circles),
intermediate (triangles) and nocturnal (dark circles) animals with a running wheel (phase

2).

33

classified as diurnal or intermediate did not change across phases (Figure 12). In addition,
nocturnal animals differed from both diurnal and intermediate animals during phase 2
(wheel present) with respect to the percentage of activity that occurred in the first five
hours after lights-out (F2: 571.88, p<0.005; F2: 411.28, p<0.005 respectively). There
were no other differences between groups within any phases of the experiment.

The body temperature rhythms of all animals were diurnal without running wheels
(Figure 13). When a running wheel was made available (phase 2), the body temperatures
of animals categorized as nocturnal remained elevated for several hours after lights-out,
as did their activity (Figure 14). The temporal pattern of diurnal animals’ body
temperature peaks did not change across the phases of the experiment. The body
temperature rhythms paralleled patterns of gross motor activity in all animals at all phases
of treatment. Body temperatures were elevated for all animals when a running wheel was
present. This elevation occurred during the day in animals in the diurnal category, and at
night in animals in the nocturnal category (Figure 14). Without a running wheel, animals
in the diurnal category appeared to have a higher overall body temperature than did

animals classified as nocturnal (Figure 13).

DISCUSSION
As described above, wheel-running patterns do not always parallel patterns of
general activity (Kennedy et al., 1994; Lewill, 1974; Mather, 1981; Roper, 1976;
Sherwin, 1998; Shinoda etal., 1988; Shinoda and Miura, 1994). This also appears to be
true for A. niloticus. Regardless of initial wheel-running patterns, all animals implanted

with abdominal transmitters showed diurnal patterns of gross motor activity and body

34

97/03/25 00:00:00.00 ACTOGRAH Page 1
Channel 120

i—i—i—i—i—

 

6.00

 

, .00
lme (Hours)

Figure 11. Representative actogram of general activity during phases 1,2 and 3 for a
. diurnal animal (lights-on 0700, lights-out 1900).

35

98/01/17 00:00:00.00 RCTOGRHH Page 1
Channel 106

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

"4.666 8.000_ 12.00 18.00 26.66
Time (Hours)

Figure 12. Representative actogram of general activity during phases 1,2 and 3 for a
nocturnal animal (lights-on 0700, lights-out 1900).

36

39

 

 

 

Average Temperature

 

 

 

 

 

 

38-

O

374

36-

35 I I I I T I T I I I I I
0 2 4 6 8 10 12 14 16 18 20 22 24
_ -

Time

Figure 13. Average daily patterns of core body temperature and SEM for diurnal (open
circles), intermediate (triangles) and nocturnal (dark circles) animals without a running
wheel (phase 1).

37

40

 

38 -

 

Average Temperature

 

 

 

 

 

10 12 14 16 18 20 22 24

 

 

 

Time

Figure 14. Average daily patterns of core body temperature and SEM for diumal'(open
circles), intermediate (triangles) and nocturnal (dark circles) animals with a running
wheel (phase 2).

38

temperature when running wheels were removed. Specifically, animals with diurnal
patterns of wheel-running exhibited diurnal patterns of gross motor activity throughout
the three phases of the study, while animals with nocturnal wheel-running patterns were
diurnal without a running wheel (phase 1,3), but returned to a nocturnal pattern of general
activity when a wheel was present (phase 2).

In all species examined, body temperature rhythms are generated by an
endogenous circadian mechanism (Moore-Ede et al., 1982). My experiment took place in
a light-dark cycle so I cannot say definitively if the body temperature rhythms of A.
niloticus are endogenous or not. However, previous studies on other species have
consistently demonstrated that the circadian rhythm in body temperature and locomotor
activity are closely related (Renfinetti and Menaker, 1992). Peaks in body temperature
always coincided with peaks in activity in all animals in this review. Body temperature
rhythms were assessed in A. niloticus with and without running wheels, and compared
between diurnal and nocturnal animals. In this study, I found that all A. niloticus express
diurnal patterns of gross motor activity and body temperature in the absence of a running
wheel. I also found that when housed without running wheels, animals originally
classified as diurnal had relatively higher average body temperatures than did those
classified as nocturnal (Figure 13). This is the only difference that has been found

between these two groups of individuals in the absence of a running wheel.

39

CHAPTER 5

ACTIVITY PATTERNS OF A. niloticus IN THEIR NATURAL HABITAT

INTRODUCTION

Arvicanthis niloticus is an herbivorous murid rodent inhabiting dry savanna,
woodland and grassland habitats in tropical Africa (Rosevear, 1969; Kingdon, 1974).
Males and females have been observed to live in groups in equal numbers (Delany and
Munro, 1985; Senzota, 1983) and associations have been described as random with
respect to age and sex (Senzota, 1990). A. niloticus reside in underground burrows they
construct at the bases of bushes, trees, rocks, banks, trash piles and termitaria (Delany and
Neal, 1966; Packer, 1983; Senzota, 1983) and maintain runways that radiate outward
from burrow entrances (V esey-Fitzgerald, 1966; Senzota, 1990).

As described in the previous experiments, I have studied patterns of wheel-
running, gross motor activity, and body temperature in members of a captive colony of
A. niloticus descended from animals trapped in East Africa in 1993. In these laboratory
animals, individuals express distinctly different diurnal or nocturnal patterns of wheel-
running. The pattern expressed by an individual depends on the pattern expressed by its
parents and on its sex. However, gross motor activity and the peak in body temperature
are diurnal in all animals housed without running wheels, including those that are
nocturnal with a wheel. This raises the question of whether any A. niloticus are nocturnal
in nature, where running wheels are not available. To determine if diurnal and nocturnal
A. niloticus actually coexist in a natural setting, and to systematically characterize their

rhythms in the wild, I conducted a field study in the region from which the founders of

40

this colony were collected, the Masai Mara National Reserve (MMNR), Kenya.
METHODS

From June 15- August 22, 1998, 20 colonies of A. niloticus were systematically
sampled at MMNR for two periods of eight consecutive days approximately one month
apart. During each of these two sampling periods, traps were set between 0700h-1900h or
l900h-0700h. Sherman traps baited with a teaspoon of powdered cats were placed at the
entrances to animal burrows and in runways the animals had constructed. Two kinds of
traps were used. The first was a standard 9 x 3 x 3.5 inch Sherman trap. The second was a
Sherman trap with a timer mounted on the side of the trap that was attached to a button
on the floor of the trap that was triggered by the opening and closing of the trap door
(Barry et al., 1989). When the trap was set the time of day was recorded, the timer was set
at 00:00 and the trap door was opened which depressed the timer button and started the
timer’s counter. When an animal entered the trap the door closed which released the
button and stopped the timer. The timer, therefore, recorded the interval between when
the trap was set and when the animal was captured. The time of day that the animal was
trapped was determined to the nearest minute by adding the elapsed time to the time at
which the trap was set.

Day and night trapping methods were different in two respects. First, traps were
checked repeatedly during day trapping periods and only at the end of the period for night
trapping. Traps were not checked at night because MMNR regulations precluded walking
after dark when many large and dangerous predators (e.g. lions) were most active and
difficult to see. During the day, however, traps were checked every hour between 1100b

and 1600b in order to avoid animal mortality due to extreme heat. When animals were

41

found between these hours they were identified and released, and the trap was removed.
The second difference between day and night sampling protocols involved the
deployment of non-timer traps during the day, but not at night. During the day, timer-
traps (x= 8 .5/ site) were set closest to burrow openings and a small number of non-timer
traps (x= 3.6/ site) were placed around the periphery. These additional traps were needed
during the day to ensure that traps would not be saturated before the sampling period
ended. This was not an issue at night because so few animals entered the traps (see
results). Furthermore, at night, when traps could not be checked, only timer-traps could
provide information on the time the animal entered the trap. To account for this
difference in trap deployment, analyses were done on timer-trap data exclusively as well
as on the full data set. The distributions of activity that resulted from these two analyses
were then compared. 3

Upon capture, the time the animal was captured was recorded. The first time an
animal was caught it was anesthetized with the inhalant Metofane in order to keep it
from struggling and inflicting harm to itself during marking. The animal was sexed,
weighed, its physical condition noted (size of testes for males; pregnancy and lactation
for females) and its ear marked with an ear-punch. Animals were then released at the site
of capture. Animals were considered to be adults if they weighed >40 grams.

Data were analyzed using chi-square tests, and were considered significant when
p<0.05.

RESULTS
One hundred seven A. niloticus were trapped at 20 different sites a total of 372

times. This number includes pilot data obtained while locating animal burrows. Seventy-

42

seven individuals were systematically captured 287 times using both’timer and non-timer
traps following the protocols described in the methods. The time of all of these trappings
could be assigned to one of six four-hour intervals (2300h-0300h, 0300h-0700h, 0700b-
1100h, 1100h-1500h, 1500h-1900h, 1900h-2300h). These data were used to determine
the time of day at which animals were most likely to be captured overall, as well as to
compare males with females, and adults with juveniles with respect to the time of day at
which they were most likely to be captured. One hundred fifty-one captures of 56
individuals involved traps with timers in which the exact time of capture was recorded
(mean number of timer-trap captures per animal: 2.66, SEM: 0.28). These 151 captures
were used to analyze the distribution of activity at a finer scale as well as to compare
timer-trap data to data that included non-timer trappings during the daytime.

During the field study it became light at approximately 0630h and dark at
approximately 1915h. Almost all 287 A. niloticus trappings occurred during these
daylight hours (Figure 15). All individuals trapped appeared to be diurnal. Only five out
of 77 animals (two adult females, one adult male and two juvenile males) were captured
after dark or before sunrise. Each of these five individuals were only caught once at these
times, and all were also captured during the daytime (34 times for all five animals). Using
data from both timer and non—timer traps, a comparison of the proportion of traps
containing animals during the light versus the dark showed that animals were
significantly more likely to be captured during the day than at night (x22=150.6, p<0.001).
When the 24-hour day was divided into four-hour intervals there was a significant
difference between intervals with respect to the probability a trap would contain a grass

rat (x25=195.6, p<0.001) (Figure 15). Males and females did not differ with respect to the

43

140

 

 

120 ~
100 —

80-

 

 

60~

# Captured

404

20—

 

 

 

 

 

 

 

 

23000300 03000700 0700-1100 1100-1500 1500-1900 1900-2300
— _

 

Time intervals

Figure 15. Total number of A. niloticus captured in timer and non-timer traps in each of
six 4-hour intervals throughout the 24-hour day (sun rises 0630, sets 1915).

44

four-hour interval in which they were most often captured (x25=0.67, NS) (Figure 16).
There was also no difference between adults and juveniles in the four-hour interval in
which they were most likely to be trapped (x25=2.43, NS) (Figure 17). The fundamental
pattern observed when only timer-trap data was exanrined (Figure 18) was the same as the
pattern obtained from the combination of timer and non-timer trap data (Figure 15).

Occasionally, other species including shrews and various species of mice were
captured at sites in which A. niloticus were also captured. However, A. niloticus were
captured more frequently (287 trappings) than were members of all other species (49
trappings). In addition, all trappings of other species in A. niloticus study sites occurred
during the night (Figure 18).

Although this study was not designed to elucidate social structure or population
dynamics, some observations were consistent with the notion that A. niloticus live in
social groups containing at least one adult male and female and their offspring. These
animals were found living in underground burrows constructed under thorny bushes.
Group size and makeup appeared to depend on the proximity of habitable bushes. At six
sites, bushes were distinctly isolated from other bushes harboring A. niloticus. Animals
trapped at these sites generally included a single adult male and female pair and juveniles
presumed to be their offspring (Table 2). When these sites were sampled again
approximately one month later the original animals were usually still present and often a
second group of offspring was present as well, suggesting that the population was
expanding at some sites. This was most clearly seen at bushes that were isolated from
other bushes, and a clear assessment could be made as to which animals were living there

(Table 3). At six trapping sites where bushes were in close proximity more animals were

45

50

 

 

40-

E 30—

3

o.

to

O

°\° 204
10—
o-

 

 

2300-0300 0300-0700 0700-1 100 1 100-1500 1500-1900 1900—2300
_ _

 

Time Intervals

Figure 16. Percent of male (black bars) and female (gray bars) A. niloticus caught in
timer and non-timer traps in each of six 4-hour intervals throughout the 24-hour day (sun
rises 0630, sets 1915).

46

 

50

404

30-

20-

% Captured

 

 

 

o _ . . - .
2300-0300 0300-0700 0700-1 100 1 100-1500 1500-1900 1900-2300

— —
Time Intervals

 

 

Figure 17. Percent of adult (black bars) and juvenile (gray bars) A. m'loticus captured in
timer and non-timer traps in each of six 4—hour intervals throughout the 24-hour day (sun

rises 0630, sets 1915).

47

 

20

18-

16-

14-

‘12 ~

 

# Animals Captured

     

 

 

L8
(3:
7f

1.3

 

r if: .:
LT. ‘ 'I ' I: ’:
, .A 1 f“ a“.
- - J' . 12.--
:1 I? 1 -1 I '3 :i- 5‘55
o :.; ;,:- 8;”, ." 1:11

A

012 3 4 5 6 7 8 9101112131415161718192021222324

Time

Figure 18. Number of A. niloticus (gray bars) and other rodent species (black bars)
captured in timer-traps in each hour of the 24-hour day (sun rises 0630, sets 1915).

48

Table 2. Average Number of Animals Captured Per Site.

 

 

 

 

 

 

 

 

Trapping Site: Adult Males Adult Females Juvenile Males Juvenile Females

(SEM) (SEM) (SEM) (SEM)
Isolated (n=6) 1.2 (0.2) 0.8 (0.2) 1.0 (0.4) 0.7 (0.3)
Interconnected 4.2 (0.7) 2.0 (0.8) 1.3 (0.5) 0.8 (0.3)
(n=6)

 

49

 

Table 3. Number of Animals Captured in Period 1 and Period 2 at Isolated Sites.

 

Adult Male Adult Female Juvenile Male Juvenile Female

 

Period 1 6 4 4 l

 

Period 2:

# still present 5 4 3 l

 

Period 2:
# new animals

 

 

 

 

 

 

 

50

captured (Table 2). At these sites, animals were observed moving from bush to bush, and
it was not possible to determine under which specific bush each animal was living. At the
remaining eight sites too few animals were captured to conclude anything about their
group makeup. Two animals were captured in the same trap on six occasions. The pairs
were as follows: adult male and adult female (twice), adult female and unidentified adult,
adult male and juvenile male, two juvenile males and two juvenile females. Based on
these findings animals appeared to associate in a variety of age and sex combinations.
DISCUSSION

A. niloticus are clearly diurnal in their natural habitat in the Masai Mara National
Reserve in Kenya. These animals were captured throughout the daylight hours and very
rarely during the night (Figures 15 and 18). Five A. niloticus were captured, each on one
occasion, in the late evening or early morning hours, and each of these animals was
repeatedly captured during the day. One potential problem interpreting these results is
that sampling methods for day versus night trapping were somewhat different. This was
necessary because at night, a variety of dangerous animals (e. g. lions) are active, making
it dangerous to check traps in the dark. During the day, traps had to be checked at hourly
intervals because the extreme heat would have killed animals had they been left in traps.
In addition, more traps were set during the day than at night to avoid trap saturation
which would have biased against sampling times late in the light period. Trap saturation
was avoided by the use of a small number of non-timer traps around the periphery of the
burrow systems during the day. Timer-traps in daytime and nighttime sampling periods
were set in the same places at the entrances to burrows. One way to try to account for

these differences in sampling methods was to compare the distribution of activity

51

including and excluding non-timer trap data. There was no difference in the resulting
distributions of activity when the data were compared (Figures 15 and 18).

This discrepancy with regard to the activity pattern of this species may be due to
the taxonomic confusion for the genus Arvicanthis. Different authorities have described
from one to five different species throughout the range of this genus (Misonne, 1974;
Corbet and Hill, 1980; Honacki et al., 1982). Because of this, studies of these animals in
other regions of Africa may not be describing the same species characterized here.
Specifically, the papers that describe A. niloticus as primarily nocturnal (Schumutterer,
1969; Ghobrail and Hodeib, 1982) both took place in Sudan and might actually be a
different species from the animals in southern Kenya that I have found to be diurnal. In
order to be confident the lab and the field animals were from the same species, my field
study was done at the same sites from which our lab colony was collected. This was the
first study done with the specific purpose of describing the activity pattern of this species.

The diurnal pattern of activity in A. niloticus is unusual and quite different from
the nocturnal pattern expressed by most murid rodents. One possible explanation for A.
niloticus’ diurnal pattern of activity may be predator avoidance. A. niloticus are preyed
upon by numerous species of birds, reptiles and mammals. The following have all been
observed to be their predators: spitting cobra (Naja nigricolis), black-backed jackal
(Cam's mesomelas), long-crested hawk eagle (Elanus caeruleus), black headed heron
(Arc/ea melanocephala) and dwarf mongooses (Helogale parvula) (Packer, 1983). These
predators have different patterns of activity which may make A. niloticus vulnerable to
predation throughout most of the day and night. A. niloticus may feed during the middle

of the day because this is a time of rest for numerous birds of prey and small carnivores

52

(Kingdon, 1974).

Group living is also uncommon in murid rodents, yet A. niloticus living in
isolated bushes generally consisted of ‘family groups’ of an adult male and female pair
and juveniles presumed to be their offspring (Table 2). In addition, these pairs were often
found together with a second group of offspring when sites were re-sampled
approximately one month later (Table 3). This raises the possibility that this species may
form pair bonds for at least short periods of time. Many studies of the social organization
of various species of voles have led to the hypothesis that the nature of the habitat and
population density affect an animal’s mating system (Getz, 1978). Prairie voles, Microtus
ochrogaster, which live in large, continuous habitats in which they compete for resources
with other larger herbivores, have been described as monogamous at low population
densities and in the laboratory (Thomas and Birney, 1979.) Getz (197 8) however, stated
that at high p0pulation densities this probably would not hold true. M. califomicus,
another species of vole that lives in habitats similar to those of prairie voles also tends
toward monogamy (Lidicker, 1979). M. pennsylvanicus, however, which live in smaller,
more isolated, ephemeral areas are described as polygynous (Tamarin, 1984). A. niloticus
live in a habitat in which they, like M. ochrogaster, are in constant competition with other
large herbivores for resources and may also tend toward monogamous behavior, at least
when population density is low (personal observations).

This field data provides no evidence of the nocturnal pattern evident in the wheel-
running behavior of some A. niloticus in captivity (Katona and Smale, 1997; Blanchong
et al., unpublished). This apparent discrepancy could be explained in a number of ways.

Perhaps the nocturnal wheel-running pattern, which was rare in wild—caught animals

53

(Katona and Smale, 1997), reflects a temporal variant which is rare in wild-populations.
If I had sampled a larger population, or if I had sampled at different seasons or at different
phases of a population cycle, I might have detected nocturnal individuals. Alternatively,
the nocturnal patterns of wheel-running seen in some captive A. niloticus may not reflect
patterns of general activity that these animals would exhibit in free-living conditions.
Wheel-running has often been used as an easy method of assessing patterns of activity in
rodents and it generally provides accurate information as to whether rodents are diurnal or
nocturnal (e.g. DeCoursey, 1973; Johnston and Zucker, 1983; Kavanau, 1971; Kenagy,
1978; O’Reily et al., 1986; Rowsemitt,1986). However this may not always be the case
(Kennedy et al., 1994; Lewell 1974; Mather, 1981; Roper, 1976; Sherwin, 1998; Shindoa
et al., 1987; Shindoa et al., 1988; Shindoa and Miura,1994). The current field data
reinforce the view that one should not assume that wheel-running provides accurate

information as to whether animals are diurnal or nocturnal in nature.

54

CONCLUSIONS

In summary, 1 have demonstrated that individual A. niloticus differed with respect
to their patterns of wheel-running, and that these differences were influenced by
’ parentage and by sex. However, when a running wheel was not present the pattern of
activity for all animals was almost exclusively diurnal. A running wheel caused a shift in
the timing of activity of only some individuals, specifically those that expressed nocturnal
wheel-running patterns, but diurnal gross motor activity patterns. This may be related to
the higher average body temperature found for animals classified as diurnal versus those
that were nocturnal in the absence of a running wheel. In addition, I found that all animals
observed in a field setting showed diurnal activity. Surprisingly, the pattern of diurnal
activity exhibited by A. niloticus in the field differed dramatically from the diurnal
pattern of activity seen in the laboratory. In the field, animals’ activity peaked in the
middle of the day while in the lab diurnal animals showed a more crepuscular pattern in
which there were two peaks of activity, one shortly after the lights came on and the
second shortly before the lights went out.

This combination of results provides a new way of looking at possible
mechanisms underlying individual differences in wheel-running patterns in A. niloticus. It
raises the possibility that some individuals shift to the nocturnal pattern because they run
in their wheels more than other individuals. That is, high levels of running may feed back
to change the basic waveform of the rhythm. This hypothesis is consistent with the
positive correlation between the number of hours an animal runs after lights-out and the
overall level of wheel-running (e.g. Figure 5). This correlation, however, is far fi'om

perfect, raising the possibility that some individuals may be more sensitive than others to

55

feedback effects of wheel-running on the pattern of rhythrnicity. The mechanisms
underlying the influence of parentage and sex on rhythm patterns (Chapter 2) could
involve effects of these variables on either the levels of wheel-running, or sensitivity to
feedback effects of wheel-running.

Feedback effects of wheel-running on circadian rhythms have been documented in
a number of nocturnal rodents (Pratt and Goldman, 1986; Mistleberger, 1991; Yamada et
al., 1988; Reebs and Mrosovsky, 1989). Most of this work has focused on properties of
rhythms under free-running conditions. For example, in hamsters, wheel-running in
response to a novel wheel phase shifts the endogenous rhythm in a precise phase
dependent manner (Reebs and Mrosovsky, 1989). These shifts only occur in animals that
run above threshold levels when presented with the novel wheel (Mrosovsky and Biello,
1994). Such effects were not found in A. niloticus when animals kept in constant darkness
were enclosed in novel wheels for three hour periods at a variety of circadian times
(Smale, unpublished observations). A dramatic effect of wheel-running on the pattern of
entrainment was recently reported in Octodon degus (Kas and Edgar, 1998), a normally
diurnal species (Fulk, 1976). In this species, animals that were diurnal without a wheel
became more active at night than during the day when given a running wheel (Kas and
Edgar, 1998). The effect of running wheels on the pattern of activity thus appears to be
somewhat similar in A. niloticus and 0. degus, two rather distantly related rodents that
independently evolved a diurnal lifestyle.

From the point of view of trying to understand animals in their natural habitat, the
effect that wheel-running has on the fundamental pattern of activity in nocturnal A.

niloticus is a puzzle. The switching system that responds to wheel-running in the

56

laboratory could theoretically have evolved to respond to some other stimuli in nature. It
could reflect some ancestral nocturnal mechanism that is no longer active in modern A.
niloticus in their natural habitat, or it could be a byproduct of selection for some other
characteristic(s). My field study provides some evidence against the first of these
hypotheses. I found no evidence that any modern A. niloticus are nocturnal. In fact, in the
field, these animals appear to be even more diurnal and less crepuscular than they are in
the lab. These animals were rarely trapped before sunrise, their activity progressively
increased until it peaked in the middle of the day, and was essentially over by sunset.
Although the details of the activity pattern deduced from trapping are unlikely to perfectly
reflect natural activity patterns (Hicks et al., 1998), the huge difference between day and
night trapping results provides clear evidence that these animals are diurnal (Figure 15).
Of special note is the dramatic difference between the number of animals trapped in the 5
hours after traps were set in the morning compared to evening. However, it remains
possible that I would have detected nocturnal patterns of activity in some animals had I
sampled a larger population, or if I had examined animals at a different season, or when
the population density was dramatically higher or lower.

In summary, a nocturnal pattern of wheel-running in captive A. niloticus was
prevalent in offspring of nocturnal parents, and was more common in females than in
males. However, a nocturnal pattern was only expressed in these animals when a running
wheel was present. Nocturnal wheel-runners were indistinguishable from diurnal ones
when other behaviors and body temperatures were monitored in the absence of a running
wheel. In captivity, it therefore appears that running in a wheel feeds back to change the

mechanisms responsible for the temporal organization of wheel-running, as well as other

57

behaviors. The role that this system might play in the regulation of rhythms in their
natural habitat is unclear. The activity pattern deduced from two months of trapping a

free-living population of A. niloticus suggests that all individuals are diurnal.

58

REFERENCES

59

REFERENCES

Ansell, W. F. H. 1960. Mammals of Northern Rhodesia, The Government Printer Lusaka,
Rhodesia.

Barry, R. E., JR., Fressola, A. A., and Bruseo, J. A. 1989. Determining the time of
capture for small mammals. Journal of Mammalogy, 70: 660-661.

Bult, A. L., Hiestand, VanderZee, E. A. and Lynch, C. B. 1993. Circadian rhythms differ
between selected mouse lines: a model to study the role of vasopressin neurons in
the suprachiasmaticc nuclei. Brain Research Bulletin, 32: 623-627.

Claypool, L. E. 1984. The environmental and physiological deterrrrinants of activity
patterns in Microtus montanus the montane vole. Ph.D. Thesis. University of
Utah, Salt Lake City, UT.

Corbet, G. B. and Hill, I. E. 1980. A World List of Mammalian Species, Publs. the British
Museum (Nat. Hist..)No 813, London.

Daan, S. 1981. Adaptive daily strategies in behavior. Pp. 275-298, in A handbook of
behavioral neurobiology, Volume 4: Biological rhythms (J. Aschoff, ed.). Plenum
Press, New York.

DeCoursey, P. J. 1973. Free-running rhythms and patterns of circadian entrainment in
three species of diurnal rodents. J. Interdiscipl. Cycle Res., 4: 67-77.

DeCoursey, P. J. 1990. Circadian photoentrainment in nocturnal mammals: ecological
overtones. Biol. Behav., 15: 213-238.

Delany, M. J. and Kansiimeruhanga, W. D. K. 1970. Observations of the ecology of
rodents from a small arable plot near Kampala, Uganda. Rev Zool Bot Afr., 81:
417-425.

Delany, M. J. and Monro, R. H. 1985. Movement and spatial distribution of the Nile rat
(Arvicanthis niloticus) in Kenya. Journal of Tropical Ecology, 1: 111-130.

60

Delany, M. J. and Neal, B. R. 1966. A Review of the Muridae (Order Rodentia) of
Uganda. Brit. Mus. (Nat. Hist.) Zool. Bull, 13: 295-355.

Dodson, E. O. 1956. Genetics, the Modern Science of Heredity, W.B. Saunders Co.,
Philadelphia & London.

Friedman, D., Hairn, A. and Zisapel, N. 1997. Temporal segregation in coexisting spiny

mice (genus Acomys): role of photoperiod and heterospecific odor. Physiology &
Behavior, 62: 407-411.

Fulk, G. W. 1976. Notes on the activity, reproduction, and social behavior of Octodon
degus. Journal of Mammalogy, 57: 495-505.

Getz, L. 1978. Speculation on social structure and p0pulation cycles of microtine rodents.
Biologist, 60: 134-147.

Ghobrail, L. I. and Hodieb, A. S. K. 1982. Seasonal variation in the breeding of the Nile
rat (Arvicanthis niloticus). Mammalia, 49: 537-542.

Griffiths, A. J. F., Miller, J. H., Suzuki, D. T., Lewontin, R. C. and Gelbart, W. M. 1993.
An Introduction to Genetic Analysis, W.H. Freeman and Co., New York.

Hairn, A. and Rozenfeld, F. M. 1993. Temporal segregation in coexisting Acomys
species: the role of odor. Physiology & Behavior, 54: 1159-1161.

Hardy, D. F. 1972. Sexual behavior in continuously cycling rats. Behaviour, 41: 228—297.

Harrison, D. L. 1972. The Mammals of Arabia. VIII, Ernest Benn Ltd, London.

Hicks, N. G., Menzel, M. A. and Laerm, J. 1998. Bias in the deterrrrination of temporal
activity patterns of syntopic Peromyscus in the southern Appalachians. Journal of
Mammalogy, 79: 1016-1020.

Honacki, J. H., Kinman, K. E. and Koeppl, J. W. 1982. Mammal Species of the World,
Allen Press, Lawrence, Kansas.

61

Johnston, P. G. and Zucker, I. 1983. Lability and diversity of circadian rhythms of the
cotton rat Sigmodon hispidus. American Journal of Physiology, 244: 338-346.

Kas, M. J. H. and Edgar, D. M. 1999. A nonphotic stimulus inverts the diurnal-nocturnal
phase preference in Octodon degus. The Journal of Neuroscience, 19: 328-333.

Katona, C. and Smale, L. 1997. Wheel-running in Arvicanthis niloticus. Physiology &
Behavior, 61(3): 365-372.

Kavanau, J. L. 1971. Locomotion and activity phasing of some medium—sized mammals.
J. Mamm, 52: 386-403.

Kenagy, G. J. 1978. Seasonality of endogenous circadian rhythms in a diurnal rodent

Ammospermophilus leucurus and a nocturnal rodent Dipodomys merriami. J.
Comp. Physiol, 128: 21-36.

Kennedy, G. A., Hudson, R., and Armstrong, S. M. 1994. Circadian wheel running
activity in two strains of domestic rabbit. Physiology & Behavior, 55: 385-389.

Keuhn, R. E. and Beach, F. A. 1963. Quantitative measurement of sexual receptivity in
female rats. Behaviour, 21: 282-294.

Kingdon, J. 1974. East African Mammals: an Atlas of Evolution in Africa, Volume IIB,
The University of Chicago Press, Chicago.

Labyak, S. E., Lee, T. M., and Goel, N. 1997. Rhythm chronotypes in a diurnal rodent,
Octodon degus. American Journal of Physiology, 273: R1058-R1066.

Lewill, C. J. 1974. Activity rhythms of golden hamsters (Mesocricetus auratus) and
Mongolian gerbils (Meriones unguiculatus) by direct observations. Journal of
Zoology. 174: 520-523.

Lidicker, W., JR. 1979. Analysis of two freely-growing enclosed populations of the
California vole. Journal of Mammalogy, 60: 447-466.

62

Madison, D. M. 1985. Activity rhythms and spacing. Pp. 373-419, in Biology of New
World Microtus (R. H. Tamarin, ed). Special Publication No. 8 of the American
Society of Mammalogists.

Mather, J. G. 1981. Wheel-running activity: a new interpretation. Mammal. Rev., 11: 41-
51.

McElhinny, T. L., Smale, L., and Holekamp, K. E. 1996. Patterns of body temperature,
activity, and reproductive behavior in a tropical murid rodent, Arvicanthis
niloticus. Physiology & Behvaior, 62: 91—96.

Misone, X. 1974. Part 6, Order Rodentia. In The Mammals of Africa: an Identification

Manual, J. Meester and H.W. Setzer, eds, Smithsonian Institution, Washington.

Moore-Ede, M. C., Sulzman, F. M., and Fuller, C. A. 1982. The Clocks That Time Us,
Harvard University Press, Cambridge, MA and London, England.

Mistlberger, R. E. 1991. Effects of daily schedules of forced activity on free-running
rhythms in the rat. Journal of Biological Rhythms, 6: 71-80.

Mrosovsky, N. and Biello, S. M. 1994. Nonphotic phase shifting in the old and the cold.
Chronobiology International, 1 1: 232-252.

Nowak, R. M. and Paradiso, J. L. 1983. Walker’s Mammals of the World, Johns Hopkins
University Press, Baltimore and London.

O’Reilly, H. O. , Armstrong, S. M., and Coleman, G. J. 1986. Restricted feeding and
circadian activity rhyhtms of a predatory marsupial, Dasyuroides byrnei.
Physiology & Behavior, 38: 471-476.

Packer, C. 1983. Demographic changes in a colony of Nile grass rats (Arvicanthis
niloticus) in Tanzania. Journal of Mammalogy, 64: 159—161.

Pratt, B. L. and Goldman, B. D. 1986. Activity rhythms and photoperiodism of Syrian
hamsters in a simulated burrow system. Physiology & Behavior, 36: 83-89.

63

Quilici M, Ranque Ph and Camerlynck P (1969) Elevage au laboratiore d’Arvicanthis
niloticus. Mammalia 33:245-247.

Rabiu, S. and Fisher, M. 1989. The breeding season and diet of Arvicanthis in northern
Nigeria. Journal of Tropical Ecology, 5: 375-386.

Reebs, S. G. and Mrosovsky, N. 1989. Effects of induced wheel running on the circadian
activity rhythms of Syrian hamsters: entrainment and phase response curve.
Journal of Biological Rhythms, 4: 39-48.

Renfinetti, R. and Menaker, M. 1992. The circadian rhythm of body temperature.
Physiology & Behavior, 51: 613-637.

Roper, T. J. 1976. Sex differences in circadian wheel running rhythms in the Mongolian
gerbil. Physiology & Behavior, 17: 549-551.

Rosevear, D. R. 1969. The Rodents of West Africa, Trustees of the British Museum
(Natural History), London.

Rowsemitt, C. N. 1986. Seasonal variations in activity rhythms of male voles: mediation
by gonadal hormones. Physiology & Behavior, 37: 797-803.

Schmutterer, H. 1969. Pests of Crops in Northeast and Central Africa, Gustav Fischer-
Verlag, Portland.

Senzota, R. B. M. 1982. The habitat and food habits of the grass rats (Arvicanthis
niloticus) in the Serengeti National Park, Tanzania. African Journal of Ecology,
20: 241-252.

Senzota, R. B. M. 1983. A case of rodent-ungulate resource partitioning. Journal of
Mammalogy, 64: 326-329.

Senzota, R. B. M. 1990. Activity patterns and social behavior of the grass rats
[Arvicanthis niloticus (Desmarest)] in the Serengeti National Park, Tanzania.
Tropical Ecology, 31: 35-40.

Sherwin, C . M. 1998. Voluntary wheel running: a review and novel interpretation.
Animal Behaviour 56: 11-27.

Shinoda, M., Miura, T., and Tadokoro, S. 1988. Differences of behavioral rhythms
observed by flat cage and running-wheel cage in female rat. Experimental
Animals, 37: 463-468.

Shinoda, M. and Miura, T. 1994. A comparison between wheel-running and ambulatory
activities in young, adult and old rats based on long-term observation.
Experimental Animals, 43: 79-84.

Tamarin, R. H. 1984. Animal Population Regulation Through Behavioral Interactions.
Pp. 698-720. in Special Publications No. 7 of the American Society of
Mammalogists.

Thomas, J. and Bimey, E. 1979. Parental care and mating system of the prairie vole,
Microtus ochrogaster. Behav. Ecol. Sociobiol., 5: 171-186.

Vesey-Fitzgerald, D. F. 1966. The habits and habitats of small rodents in the Congo river
catchment region of Zambia and Tanzania. Zoo. Afr., 2: 111-122.

Wollnik, F. 1991. Strain differences in the pattern and intensity of wheel running activity
in laboratory rats. Experientia, 47: 593-598.

Yamada, N., Kazutaka, 8., Chi, K., Takahashi, S. and Takahashi, K. 1988. Free-access to
a running wheel shortens the period of free-running rhythm in blinded rats.
Physiology & Behavior, 42: 87-91.

65

IE5

"‘igiillllllllllll