ANALYSIS OF SLEEP CYCLES
IN THE RODENT

By
HENRY B. VAN TWYVER

A DISSERTATION PRESENTED TO THE GRADUATE COUNCIL OF

THE UNIVERSITY OF FLORIDA
m PABTTAL FULFILLMENT OF THE REQUIREMENTS FOR THE
DEGREE OF DOCTOR OF PHILOSOPHY

UNIVERSITY OF FLORIDA

December, 1967

UNIVERSITY OF FLORIDA

3 1262 08552 2240

To Kathy and Bob,
who someday will do much better,

ACKNOWLEDGMENTS

The author wishes to express his gratitude to Dr. W.
B. Webb who supervised the research described in this dis-
sertation and to the members or the supervisory committee.
These were: Drs . Donald C. Goodman, Bradford N. Bunnell,
Robert L. King, James A. Horel, and Madelaine E. Carey.

Special thanks go to Dr. W. R. Knight of Hiram College
who provided the thirteen lined ground squirrels. The help
of Robert Conant, Jimmie Vandegrift, Mario Perez, Walter Young,
and Patricia Van Twyver in various phases of data collecting
and scoring is also gratefully acknowledged.

Appreciation is also expressed to Mrs. Dorothy Vandegrift
for her patience in typing this manuscript.

During the course of this investigation, the author was
supported by a Center for Neurobiologicai Sciences traineeship.
This study was aided by a NASA institutional grant to the
University of Florida, project number 2972 A36 to W. B. Webb
and H. B. Van Twyver. The contribution of the University of
Florida Computing Center is acknowledged.

xn

TABLE OF CONTENTS

ACKNOWLEDGMENTS iii

LIST OF TABLES v

LIST OF FIGURES vi

INTRODUCTION <, 1

METHOD 29

RESULTS 39

HISTOLOGY 83

DISCUSSION 85

SUMMARY 95

APPENDICES 97

REFERENCES 142

BIOGRAPHICAL SKETCH „ . » 147

IV

LIST OF TABLES

Table Papie

1 Summary of Values for the Mouse 50

2 Summary of Values for the Hamster 51

3 Summary of Values for the Rat 52

4 Summary of Values for the Squirrel 53

5 Summary of Values for the Chinchilla 54

6 Analysis of Variance of Minutes of Total Sleep Time... 56

7 Analysis of Variance of Sleep Cycle Lengths 58

8 Cycle Times, Heart Rates, and Respiration Rates for
Five Rodent Species, Ordered on the Basis of Cycle

Time 60

9 Analysis of Variance of Day-Two Total Sleep Time 61

10 Analysis of Variance of Length of Total Sleep Periods. 66

11 Analysis of Variance of Length of Slow-Wave Sleep
Periods 67

12 Analysis of Variance of Percentage of Paradoxical Sleep 70

13 Analysis of Variance of Length of Paradoxical Sleep
Periods 73

LISX 0? FIGURES

figure

1 Relationship betv;een speed of sleep cycle and two

other vital processes in several mammalian species 16

2 Percentage of paradoxical sleep for several mamma-
lian species 24

3 Examples of EEG records during waking and sleep in
Citellus tridecemlineatus 40

4 Hourly distributions of mean minutes of waking and
sleeping patterns in Citellus tridecemlineatus 44

5 Hourly distributions of mean minutes of sleep patterns

in Rattus norveKJcus on days 1 and 2 45

6 Hourly distributions of mean minutes of sleep patterns

in Mus musculus 46

7 Hourly distributions of mean minutes of sleep patterns

in Mesocricetus auratus . 47

8 Hourly distributions of mean minutes of sleep patterns

in Chinchilla lanic;er 48

9 Comparison of mean percentages of total sleep time

for five rodent species 62

10 Comparison of mean total sleep period lengths for five
rodent species 68

11 Comparison of mean slow-wave sleep period lengths for
five rodent species 69

12 Comparison of mean paradoxical sleep percentages for

five rodent species 72

13 CciTiparison of mean paradoxical sleep period lengths

for five rodent species 74

VI

INTRODUCTION

The electroencephalograph (EEC) has made objective studies
of sleep possible. High, but not perfect, correlations have
been shown between EEG patterns and behavioral indices of waking
and sleep in normal animals. With this technique, relatively
unambiguous studies of an essentially subjectively determined
state of the organism are possible. In addition, studies of
changes of electrical activity of the brain during sleep are now
possible. In recent years, the investigation of sleeping and
waking patterns has been given new impetus by the discovery of
a state of sleep which has distinct electroencephalographic and
physiological characteristics from the classically defined
sleep patterns. This state is referred to as "rapid eye move-
ment sleep" (RE-iS) (Aserinsky & Kleitman, 1955), "activated"
sleep (Dement, 1953), and "paradoxical" or "rhombencephalic"
sleep (Jouvet, 1962). The terminology used to describe this
neuroencephalographic and behavioral pattern of sleeping is
confusing and often contradictory. At this point, agreement on
a single phrase to describe it has not been reached. Hartmann
(1965), and more recently, Jouvet (1967), have discussed and
enumerated the terms now in use. In the present paper, the
terminology will follow that of Jouvet. The term slow-wave
sleep (SWS) will be used to describe the classical sleep pattern

marked by the appearance of slow-wave activity in the EEG
(1-2 cps), while paradoxical sleep (?S) will describe the phase
of sleep which is characterized by an activated EEG and which
has been shown to be associated with dreaming in human subjects.

Priority for discovery of the association between the
electroencephalographic pattern of PS and dreaming is not com-
pletely clear at this time. In 1937, Loomis, ej^ al^. » described
a stage of sleep (their stage B) having EEG characteristics
similar to the waking EEG and associated with dream recall.
They referred to this state as "dreaming sleep." According to
Jouvet (1967b), a German investigator, R. Klaue, made a similar
discovery at about this time. Both of these reports were over-
looked and it was not until 1955, when Aserinsky and Kleitman
reported that RS4s were associated with an "activated" EEG
pattern, that interest became intense in this area. Since then,
EEG studies have been completed on a variety of animals, including
many primates and extending as far down on the phylogenetic
scale as the frog.

It appears reasonable that understanding of the phenomenon
of PS may be increased by study of PS characteristics in a wide
variety of animals. Results of such studies will be reviewed
in this chapter. Unfortunately, the purpose of these studies
has varied. Kany of the studies were done for the sole purpose
of determining the existence or absence of the PS state in each
particular animal. Several were involved with the developmental

aspects of sleep, while some relate changes In amounts of PS
or SWS to other developmental changes of the organism. For
this reason, cross comparative data for many of the animals
which have been studied are not available. For example, not
all investigators have reported percentages, frequencies, and
stage characteristics^ which would be of value in making com-
parisons among animals. Variability in experimental controls
used, age and strain of animal, and conditions of recording
make direct comparisons difficult and risky in some cases.
In addition, many studies involve only short-term recording
periods which make hazardous the derivation of normative data.
Such limited recording periods can result in bias relative to
when the recording was made and the periodicity of the animal.
Data collected from nocturnal animals, for example, cannot be
directly compared with data from diurnal animals if the short-
terra recording period includes only daytime hours (cf . Roldan
6. Weiss, 1962; Roldan, et ai . , 1963; Weiss & Roldan, 1964).
For valid comparisons to be made, 24-hour recordings are
necessary.

Tha EEG Sta<^es of Sleep

There are three ways that patterns can be extracted from
EEG records. ESG data may be evaluated by extracting patterns
without regard to behavior on the basis of groups of similar

wave-foras, by searching for patterns which correspond to par-
ticular behavioral states of the organism, or by separate
analysis of the two (Sterman, et_ al_. , 1965) with a subsequent
attempt to synthesize them. It is common to find a certain
amount of discrepancy between EEG patterns and behavioral
patterns. For example, sleep EEG patterns may be seen in an
animal which by behavioral criteria is awake. Many animals
show such patterns while standing with the eyelids open.
This is particularly true in the ruminants, but occurs in
the rodents as well. Alternatively, the neonatal kitten has
an essentially flat EEG, but may be behaviorally asleep so
far as outward signs are concerned. In addition, the use of
certain drugs can dissociate behavior and EEG indices of sleep.
Additional confusion may be added by the effects of brain lesions;
and finally, the EEG during PS resembles wakefulness, so that
an additional index is needed.

As discussed above, EEG wave- forms may be grouped into
patterns which then can be used to describe the electrographic
characteristics of sleep. Dement and Kleitman (1937a) described
four such patterns for human sleep. This system is most widely
used today. Stage 1 was characterized by a rapid cortical
EEG tracing with no spindling. Stage 2, a transitional stage,
consisted of a rapid cortical tracing upon which spindles were
superimposed. Stage 3 presented a pattern of slow-wave ac-
tivity with a moderate amount of spindling, and stage 4

showed EEC waves of large amplitude and frequencies predominating
in the 1-2 cps range with no spindles. Stage 1 sleep with
rapid eye movements was associated with dreaming and corresponds
to PS in lower mammals. The four stages of sleep usually
occurred in progression and it was assumed that stage 4 was
the deepest stage of sleep. A similar system with five stages
was used by Loomis, et al. (1937).

Weitzman (1961), in a short-term study of Macaca mulatta.
partitioned this animal's EEG sleep patterns into four stages.
These were: a moderately fast cortical tracing with bursts
of slow waves similar to the spindling seen in human sleep,
followed by the disappearance of spindle bursts and the appearance
of 5-7 cps higher amplitude tracings. This in turn was followed
by periods of high synchronization with frequencies in the
1-3 cps range. During the initial stages of sleep, spindle
bursts were recorded at more than four times the amplitude
of human spindle bursts reported by Dement and Kleitman (1957a).
Presumably, this was due to potential differences caused by
electrode placements and not to comparable differences in
electrical brain potentials. It is suggested that spindling

Although the question of dreaming is not testable in animals,
many people have observed that the twitching often seen in sleep-
ing animals, and later shown to occur during PS, suggests that they
dream. Objective criteria, such as the waking EEG pattern asso-
ciated with RE"!, and the well documented rise in arousal threshold
which are similar in man and aninial, suggest that the two states
represent common processes. That the PS state occurs prior to
any possible visual imagery (Valatx, et al . , 1964) attests to the
involvement of more than simple visuaT~<irearaing per se.

represents a conunon process in both animals. During PS, eye
zrovements, twitching, and facial grimacing were associated
with a rapid cortical tracing and constituted the fourth stage
of sleep. Reite, ejt al . (1965), in a more extensive study,
also defined four stages of sleep in Macaca nemestrina.
These did not differ substantially from those described above.
Adey, ^ ^. (1963), in a preliminary report, pointed out the
similarity of the chimpanzee sleep pattern to that of man but
defined only three stages of sleep in this animal. A stage
of sleep corresponding to stage 1 in man was not reported but
the other stages were not significantly different. They also
suggested that PS could be partitioned into two stages on the
basis of long bursts of rhythmic activity recorded from the
amygdala. It seems unlikely on an evolutionary basis, that
the chimpanzee sleep pattern should be less complex than
that of the monkey. However, these authors do report a tran-
sitional period of "alpha fragmentation" which signalled the
onset of behavioral sleep, and may correspond to stage 1 sleep
in man.

Bert, et al^. (1966) have suggested that this complex
pattern of wave-form changes which occurs as the monkey progresses
from behavioral arousal to deep sleep, and which, as noted above,
has been divided into four or five distinct stages, is common
to all primates. They found, in a primitive primate, Galago
senepalensis (bush baby) , an EEG pattern very similar to that

described above for man, chimpanzee, and monkey. Differences
In amounts of a given pattern were noted, there being less
stage I sleep than in the higher primates, and fewer, more
variable spindles were seen in stage 2.

From the evidence cited above, it can be concluded
that the pattern of EEG changes in the primate consists of
four stages. Differences in EEG sleep patterns among primates
consist only of differences in lengths of the different
stages.

Fewer sleep stages have been identified for the phylogenet*
ically less complex animals. However, it is likely that the
number of sleep stages identified depends upon number and
location of recording electrodes. In the smaller animals,
physical location of the electrodes as well as real differences
in brain characteristics may have an effect on the type of re-
cordings obtained, and hence on the number of sleep stages
defined. A systematic study comparing EEG patterns of animals
having electrodes placed over functionally similar brain areas
has not been done. These limitations must be considered in a
comparative analysis of sleep stages.

Dement (1938) defined two phases of sleep in the cat,
one with slow waves which resembled the deeper stages of
sleep in man, and a stage of PS with a rapid cortical tracing.
Sterman, jet al. (1965) also reported only two stages of sleep

in the cat in addition to a drowsy pattern with slow waves.
They also pointed out the possibility of partitioning PS into
two stages on the basis of rhythmical cortical bursting activity
much like that recorded from the amygdala by Adey, et al. in the
chimpanzee. Similarly, two EEG stages (SWS and PS) have been
reported for the dog by Bonamini , £t a^,- (1962) and for the
goat (Ruckebusch, 1962a>, but only after a one-month period
of adaptation to the experimental chamber. In a previous paper,
Ruckebusch and Bost (1962) described only a SWS stage in the
goat, but noted an imperfect correlation between slow-wave
EEG activity and sleep, and between rapid EEG tracings and
waking. Discrepancies between behavioral and EEG manifestations
of sleep were presumably due to periods of rumination and PS.
Two stages of sleep were also defined for the lamb by Ruckebusch
(1962b, 1963) and D. Jouvet and Vaiatx (1962). Both of these
authors reported that "going to sleep" in the lamb was marked
by periods of 8-15 cps spindling in cortical leads thus sug-
gesting that a third stage could have been differentiated.
Ruckebusch (1963b) also investigated the sleep EEG characteristics
of the donkey and was able to define both a SWS stage, and a
stage of PS of brief duration which occurred only while the
animals were lying on their sides. For the rabbit (Faure, 1962;
Faure, e_t aj_. , 1962a and b ; Roidan & Weiss, 1963; Khazan &
Sawyer, 1963), many data exist to describe a SWS stage with
spindles and a PS stage. Several studies have been done on

the rat and it is agreed that two patterns (PS and SWS) suffice -
to describe the EEG sleep of this anirnsl also (Swisher, 1962;
Roldan &VJeiss, 1962; Soulairac, 21. 2l.- ' ^^^3; Soulairac, e^ al. ,
1965; Van Twyver, et £l. , 1966; Duncan, et aJL. , 1967), This
is also true for the mouse (Weiss & Roldan, 1964) and guinea
pig (Jouvet, e^ aJi- , 1966).

Essentially similar EEG patterns characterize the sleep
of the opossum, a primitive mammal (Affanni & Vaccarezza, 1964;
Snyder, 1967). Following an extended period of "going to
sleep" marked by intermittent slow waves and rapid cortical
activity, slow-wave activity was recorded in a pattern which
followed the characteristic mammalian pattern. This was often
followed by the PS phase. It is likely that the extended length
of the "going to sleep" phase had to do with the fact that
these animals were trapped wild, unlike most of the domesticated
animals described in other studies.

Taken together, these studies suggest that EEG pattern
characteristics differ in complexity among the mammals. Most
notable is the increase in number of stages needed to describe
the sleep-EEG of primates as compared with the carnivores,
ruminants, and rodents. Neglecting the absence of a clearly
defined stage of spindle sleep having long duration in the
rodents, the major difference in EEG characteristics between pri-
mates and lower mammals seems to be in the initial stage of "going
to sleep," The primates differ in that there is a tendency

10

for desynchronization of the EEC during these early minutes
of sleep.

In birds, Klein, £t £l . (1964) described two EEG patterns
of sleep which corresponded to the STviS and PS stages found in
mammals. No spindles were seen during sleep however, and this
may mark a departure from the classical sleep pattern common to
all mammals studied so far. Hermann, Jouvet, and Klein (1964),
found evidence only for slow-wave sleep in the tortoise, there
being no indication of the existence of a PS stage. Recently,
Tauber, e_t £l, (1966, 1967) have described two stages of sleep
in the chameleon (Chameleo jacksoni and C_^ melleri) and iguana
(Ctenosauro pectinata) as evidenced by REM and changes in arousal
thresholds and cardiac irregularity. Howover, no corresponding
EEG changes or EMG hypotonicity were noted during these periods.
They speculated that the absence of highly developed foveate vision
may account for the absence of a stage of PS in the tortoise as
shown by Jouvet. In these studies, REM activity was used as a
criterion for presence of the PS state. Presumably, REMs occur
only in animals with highly developed visual systems.

Finally, Hobson (1967), in a careful study, found no
evidence for either SWS or PS in the bullfrog (Rana catesbiana).
The EEG pattern of this animal, recorded from the optic tectum
and olfactory bulbs consisted of desynchronized low voltage
activity during periods of quiescence and higher voltage

11

slow waves during alert behavior. Long- term studies of arousal
thresholds did not reveal any decreased sensitivity to stimu-
lation which would suggest that the frog was sleeping in a manner
similar to that of evolutionarily higher animals.

Several summary statements can be made about the EEC pat-
terns of animals in terms of evolutionary position. In the
first place, a great deal of similarity exists in the kind
of brain wave patterns among mammals. Primates, especially
man, display the most complex patterns and four stages of sleep
are defined in all of them. In the carnivores, ruminants,
rabbits, and rodents, there are basically two patterns of sleep.
Early EEG desynchronization common to primates is absent in
these groups while a tendency for less predominant spindling
is noted in the rodents. In birds or reptiles, spindling does
not occur during SWS. The existence or absence of PS in
reptiles requires further study. In the frog, it appears
that sleep does not exist, in the sense that the term is used
to describe man-.malian behavior.

In addition to differences in EEG patterns of animals
of different complexity, differences in the distribution
and amounts of these patterns have been noted. The next sec-
tion is concerned with this problem.

12

Polygraohlc and B';h3;vioral
Characteristics of PS

The differences in EEG characteristics ranging from
wakefulness to stage 4 sleep in primates represent a continuous
process rather than discrete steps. This is emphasized by
the difficulty of judging this data (Dement & Kleitman, 1957a)
and a certain amount of indecision must exist at points of
transition between stages. However, it is generally agreed
that the onset of PS which always follows a certain period
of SWS in all adult mammals recorded to date, is not gradual
but abrupt. The sudden onset and cessation of PS suggests
that it is not simply a continuation of SWS but is truly a
dichotomous state. In addition, studies of evoked potentials
and arousal thresholds support this contention although agreement
among experiments is not unanimous. Many other characteristics
peculiar to this state have been described. JouveC (1967),
in a recent review, has enumerated these phenomena, many of
which can serve to identify the PS state in chronic recordings.
These are: atonia of the neck muscles, decrease in spinal re-
flexes, clonic movements, decrease in blood pressure, heart
rate changes, increase in cortical blood flow, respiratory
arrhythmia, and changes in the galvanic skin response. Arousal
thresholds are greatly increased. In the lower mammals, a
synchronous rhythm can be recorded during PS from the hippo-
campus as well as from other deep nuclei. Changes also occur

13

in the olfactory bulb of the cat and in the amygdala of the
chimpanzee (Adey, e_t ^. , 1963).

Phasic characteristics such as REM, sudden pupillary
mydriasis, and spiking activity throughout the visual system
also can be used to identify the PS state. However, identi-
fication of PS state by means of phasic criteria such as REM
alone may be misleading. It may be possible for long lapses
in phasic movements to occur, thus resulting in an underestimate
of the length of the period. Reite, et £l. (1955) used REM
as an index in conjunction with other measurements for the mon'cey
but imposed the requirement that the PS period be accompanied
by REM to be scored as such. This may have resulted in
underestimating the amounts of PS (11.1 per cent as compared
with the value of 16 per cent found by Weitzman).

Distribution of
the Stages of Sleep

Diurnal rhythms of activity in living organisms have been
well documented. A review of this is contained in a monograph
by Clouds ley- Thompson (1961, pp. 38-80). Day-night differences
in patterns of sleeping occur in the laboratory animal also.
For example, under continuous light, Sterman, jet _al_. (1965)
found a reliable tendency for waking EHG patterns to predominate
during the night and a reciprocal tendency for sleeping patterns

14

to occur during the day in the cat. Similarly, a day-night sleep
ratio of 2.22 for young adult rats was reported by Van Twyver
and Webb (1967).

A sub-circcdian cycle of EEG stages of sleep has
also been defined in the human (Dement & Kleitmcn, 1957a),
cat (Dement, 1958; Sternian, £t £l. , 1965), rat (Roldan &
Weiss, 1962; Roldan, Weiss & Fiflcova, 1963; Weiss & Roldan,
1964), and monkey (Reite, £t al . , 1965). Estimates of
this "sleep-EEG" cycle have also been made for the rabbit
(Roldan & Weiss, 1963) and mouse (Weiss &. Fifkova, 1964).
The "sleep-EEG" cycle reflects the tendency for sleep stages
to recur periodically. In adult mammals, as noted earlier,
PS always follows a certain minimal period of SWS , although
this is not true of neonates. Cyclic variations in EEG pat-
terns occur mainly during the day in the cat (Sterman, jet al_. ,
1965) when sleep is most predominant, Batini, ejt £l. (1967)
found a wide variability of sleep cycle time in the monkey at
night and never saw a complete cycle during the day. I'Thether
this is true for other animals as well is not known since
the cycle times for other animals are based on short-term
recordings. Cycle time is usually expressed as the mean number
of minutes between the end of one PS episode and the end of
the next period. It varies from about 90 minutes in man and
monkey to about 8.5 minutes in the mouse. Roldan and Weiss
(1954) have suggested that the speed of the cycle is metabolic mIIv

15

determined since it parallels heart rate and respiration values
in different animals. Figure 1 presents this relationship for
the different animals which have been studied to date.

Hartmann (1967) has inferred a cycle time of 124 minutes
for the adult elephant. This estimate V7as based on visual
observation of several animals for three nights in a zoo, and
not on ESG studies of this animal.

Very little information is available on the lengths
of the SWS and PS episodes in man or the animals or even of
the total amounts of time that they sleep each day. Assuming
that no sleep is gained outside the laboratory, then the
"golden mean" for sleep in man is 7.5 hours (Agnew & Webb,
1967), or about 31 per cent of the day. The laboratory cat
sleeps about 58 per cenc of the time (Sterman, £jt a^^. , 1965),
or 55.1 per cent (Ursin, 1967). Sleep length values for the
rabbit (Roldan & Weiss, 1963) and the rodent (Weiss & Roldan,
1964) have also been estimated but these are based only on
short-term recording. No estimate of the total amounts of
time slept per day is possible from these studies. According
to D. Jouvet, et al. (1966), the well-adapted guinea pig sleeps
approximately 52.5 per cent of the day. An estimate of total
sleep time for the rat was made by Van Twyver, £t £l. (1966)
and Van Twyver and Webb (1967). Total sleep time amounted
to 44 per cent, 51 par cent, or 41 per cent during baseline
recordings depending on the age of the animal.

16

(T X

O

q:

— 3snoy\!

— IVd

1-

CJ

cr

cr

\

X

cr

III

UJ

^

1-

<

h-

q:

UJ

III

2

_1

h-

o

( )

<

f-
<

i-

ir

Q_

cr

a.

CJ

<

CO

LU

UJ

UJ

— iiaavd

— ivo

(5
!-•

>

CD
to

03

tn
o

CI

o

a

*j

u

0}

o
o

c

u

'J
CI

1

o

o

o

o

o

o

o

o

^

o

CD

CVJ

— A3>{N05AI

NV^M

c
o

a.

•r-f

JU

re

!0

CJ

•1-1
o

NV/\I dOd 3nnVA 1N3D d3cl

to

17

Significant reductions in total sleep time (TST) have
been shown in the lamb with the onset of rumination as will be
discussed later in this paper. The donkey also sleeps a very
short amount of time each day, only 13.11 per cent, according
to Ruckebusch (1963b). Further discussion of this point will
be reserved for the final section of this paper.

Ontoganetic Changes in PS

Comparisons of the amount of PS in neonatal and adult
animals have been made in the kitten by Jouvet, e^ al. (1961)
and Valatx, £t al . (1964). In neonatal kittens, recorded from
the ages of 4 hours to 6 weeks, progressive changes in E2G
characteristics were shown. At 4 hours after birth, the
ESG consisted mostly of high frequency low amplitude waves
and did not vary much during sleep or waking. Often, however,
during behavioral sleep, the electromyographic (EMG) tracing
of the neck muscles became flat, indicating atonia. Muscle
tv/itchcs and HEZ-ls also occurred, signalling the stage of PS.
This occupied about 90 per cent of total sleep time in the
neonatal cat, more than three times the percentage of PS obtained
in the adult animal. In addition, episodes of PS were found
to occur spontaneously after a period of waking and showed a
lack of dependence upon periods of quiet sleep. This is in
contrast to studies of adult animals which have shown that PS

18

is always preceded by a period of SWS .

Meier and Berger (1965) and Berger and Meier (1966) studied
the development of sleep patterns in the neonatal rhesus monkey
continuously for the first week of life and from that time until
1 month of age on alternate days. They found that the new-
born monkey demonstrated EEG patterns typical of the adult
rhesus, but they identified only two sleep stages. They
classified the sleep-ESG patterns into a stage of SWS marked
by high voltage slow waves in the 1-3 cps range and one of PS
which did not differ from the waking pattern except for minor
amplitude changes. The failure to find as many E2G pattern
differences as reflected by the fewer number of stages in the
developing monkey as are found in the adult, points to the
difficulty of comparing very young animals with mature ones.
They found an increase in mean per cent of PS from 31 per cent
of TST on day 1 to a value of 43 per cent on the second day of
life. This was relatively constant until day 7 when the
percentage began to fail to the value of 35 per cent on day
23. The amount for juvenile monkeys was 27 per cent. In ad-
dition, gradual increases in the lengths of SWS, PS, and
awake periods were shown. The tendency for PS to occur
independently of SWS in the newborn monkey changed to one of
dependence with age. On day 1, only 8 per cent of PS periods
followed SWS, but by day 21, 36 per cent did, a value which
was also obtained for the juvenile monkey. In the lamb (Ruckebusch,

19

1963a), a similar decrease in PS (15.8 to 3.17 per cent) was
shown with ontogenetic development, but the decrease was de-
pendent upon the occurrence of rumination. He also noted
an increased tendency for PS episodes to follow SWS episodes
with development, as well as an increase in length and decrease
in number of PS episodes . The decrease in PS occurred over a
15-day period. During this time, SWS decreased from 57-40
per cent of the day as well. Since the EEG pattern during
rumination consisted mainly of slow-wave activity similar to
sleep, during periods of behavioral motionless, it could be
speculated that this state may be considered a stage of sleep.
Only minor body movements occurred during this state. This
would partially explain the sudden reduction in SWS as noted.
The reason for the sudden decrease in PS however, is not clear.
Jouvet and Valatx (1962) found a similar decrease in PS per
cent in the lamb. PS amounted to 10 per cent of TST in the
neonatal lamb and presents a pattern of low amounts of PS that
is still further reduced by weaning. These data serve to define
a limitation of any comparative study of PS. Animals must
be studied in comparable periods of maturational development
if valid comparisons are to be made.

Theories of P-radoxlcal Sleep

Several theories of the function of PS have been developed.

20

Oswald (1962) has speculated that the cycle of deep to light
sleep, or "SEG-sleep cycle" in man, which has a period of ap-
proximately 90 minutes, is a consolidation of the innate cycle
of sleep and wakefulness found in the human neonate. Aserinsky
and Kleitman (1955) found this cycle to have a period of about
1 hour. Due to social pressures, the portions of each hour
devoted to sleep in the neonate become consolidated within a
single 8 hour period. However, the tendency for an excitability
cycle of about an hour continues. This is manifested by the
rhythmical change in EEG excitability during sleep which
makes up the adult sleep cycle. The PS portion of it cor-
responds to waking. This theory is supported by ontogenetic
studies of sleep in the kitten and monkey which have shown
developmental increases in length and decreases in number of
sleep periods. Developmental studies of periodicity per se,
have as yet not been done in animals. Added support to this
theory has also been provided by Sterman (1967) who showed that
fetal activity periods in utero occur during periods of PS
in the mother. >

Another suggestion about the origin of the sleep cycle
involves the idea of environmental influences upon the sleep
period itself (Redding, et al . , 1964; Snyder, 1967). Since
the sleeping anim^^l is relatively susceptible to predators,
changing levels of vigilance might have biological utility.
In this way, shifts in arousal could allow a security check

21

at a time when the animal is most vulnerable. This speculation
hov;ever, almost demands that the distribution of excitability
stages be random within the sleep cycle. Such is not the case
V7ith the human (Dement & Kleitman, 1957a), cat (Stertnan, et al. ,
1965), macaque (Reite, jet aj^. , 1965), or rodent (Weiss & Roldan,
1984). Distribution of excitability within the sleep cycle
is not random as reflected by the tendency for episodes of PS
to occur periodically. Decreases in excitability of the reticular
formation have been shown during PS in the rat (Roldan, _et a_l. ,
1963) and rabbit (Roldan & Weiss, 1963). This also would
indicate that PS is not simply a period of increased arousal.
In addition, arousal threshold to a shock is greater during
PS than during SWS in the rat (Dillon & Webb, 1965) although
threshold to environmentally meaningful stimuli may not
necessarily be higher. If paradoxical sleep reflects a de-
crease in depth of sleep and serves the function of environmental
monitoring, lower thresholds would be expected.

Ephron and Carrington (1966, 1967) have postulated a
homcostatic theory of PS. They suggested that an optimal
level of "cortical tonus," a term which has similar meaning
to "excitation," is maintained by PS during sleep. During
waking, "overaf ferentation" (sensory excitation)results in
high levels of corticnl tonus. Slow-wave sleep then serves
to reduce this by returning the nervjus system to a srite of
quiescence. This in turn causes excitation of certain neurons

22

causing a gradual return to higher levels of tonus. This builds
up gradually until activity triggers an episode of PS. PS
then serves the function of increasing cortical tonus by
providing "endogenous af ferentation." These changes in level
of tonus, they suggest, are accompanied by corresponding weaken-
ing and strengthening of conscious processes. These are reflected
by the gradual buildup of dreaming during SWS which reaches a
maximum during PS . This part of the theory is supported by
sleep-EEG cycling and by the increased probability of a dream
report as the onset of PS approaches. Actually, the electro-
graphic aspects of it are no different from the consolidation
hypothesis described above. However, Ephron and Carrington
also suggest that changes in EEG stages have survival value
because they keep the organism ready for action in the face
of danger by allowing periodical security checks . In this
sense, the notion is similar to the survival theory and suffers
from the same limitations. A study investigating thresholds
to danger signals is needed to verify that PS is indeed a
deeper stage of sleep. The most damaging criticisms of the
"cortical tonus" theory are the lack of sequential dependence
of PS and SWS in neonates and the greatly increased arousal
thresholds during PS. An adequate criticism of the Ephron-
Carrington theory follows the 1967 article (op. cit. . pp. 94-
47 by E. S. Tauber) .

23

Percentage of PS in Different Species

Jouvet (1967b) has advanced a phylogenetic statement of
paradoxical sleep which suggests that PS is decreased among
the hunted animals and increased in the hunters. This is
based mostly on his own studies of sleep patterns in several
species. He also notes a tendency for the percentage of PS
to increase with phylogenetic position from hen to cat. The
percentage of PS, however, decreases markedly in man (about
20 per cent, his figure) as compared with cat (27 per cent).
This difference, which is the reverse of that predicted, is
notable since one would expect the greatest increase in per-
centage to occur here if per cent of PS increases in phylo-
genetic development. In Figure 2, an attempt has been made
to show the obtained percentages of PS in the animals which
have been adequately studied to date. Since certain studies
suffer from lack of control, or for other of the limitations
noted earlier in this paper, preference has been given to the
work which appeared most reliable to the author. Even with
this limitation, it has been necessary to include some work
which can be questioned due to shortcomings in technique. As
shown in Figure 2, there is no simple relationship between
phylogenetic position and the percentage of total sleep tiir.
spent in PS. If one considers all of the studies available
to date, and considers the range of individual values as well.

24

lAinssodo

r
o
ro

-r

3SnOIAI

IV d

iiaavd

Old VBNino

d33HS

IVO

A3><N0iM

NVIAI

« «

^ «

U

« a

X c

o <d

« i

a B

r-l

« (d

60 Vi

Q 4)

U >

C 41

i) n
u

41 O

41
U

s

60

ISl/Sd

25

the picture becomes even more confusing. In the most care-
fully controlled studies, variability in percentage of PS
is great. For example, Williams, £t £l. (1954) found that
young adults varied from as low as 14,42 per cent to as high
as 29.90 per cent, while Sterman, e_t al_. found variability
amounting to about 8.4 per cent in the cat in a well controlled
study. For the adult monkey, obtained mean values varied
from a low of 11 per cent (Reite, et al., 1965) to 16 per cent
O^eitzman, 1961), Reite included drowsy states in the
category of sleep. Sterman has shown that drowsy HEG activity
is reciprocally related to sleep patterns in the cat, thus
demonstrating that this pattern does not reflect a state of
sleep. Including it as sleep time would ^end to reduce the
percentage of PS. The value of 27 per cent was obtained for
the juvenile rhesus monkey by Meier and Berger (1965) and Berger
and Meier (1966), This value is more in line with an evolu-
tionary theory of PS .

Bert, et al. (1966) present no values for the Galago, nor do
Adey, ejt a_l, (1963) for the chimpanzee. For the lamb (Ruckebusch,
1962b, 1963; Jouvet & Valatx, 1962), there is general agreement
that the percentage of PS is moderately high in the neonatal
stages but shows a sudden reduction with the development of
the pre-stomachs and the onset of rumination. At this time.

This figure was computed from their data by the author.

26

the percentage o£ PS falls £rom 15.8 per cent to 3.7 per cent
of behavioral sleep or from 10 per cent to 2.5 per cent ac-
cording to the two investigators. Per cent of PS for another
ruminant, the goat, appears correspondingly low although it
is not clearly stated (Ruckebusch, 1962a). Numerous studies
of PS have been completed on the rat (Roldan & Weiss, 1962;
Roldan, Weiss & Fifkova, 1963; Soulairac, et^ al . , 1963; Weiss
& Roldan, 1964; Soulairac, et al . , 1965) but curiously, only
one estimate of PS per cent in addition to the value shown
in Figure 1 has been given. Roldan and Weiss (1964)
estimated this to be 16 per cent. According to these authors,
the percentage of PS for the mouse, based on short-term
recording r., was approximately 17.5 per cent.

Without doubt, the variability in percentage of PS
found by different investigators for the same species of animal
is due, at least in part, to the degree of adaptation of the
animal to the recording session. Also, variance could be
caused by restricted recording periods although this is not
clear. Percentages of PS may also vary during different parts
of the day but to date no systematic study of this has been
undertaken. Age of the animal may also have affected the
percentages of PS obtained. A discussion of this was presented
in the section on ontogenetic changes in sleep.

In summary, a decrease in percentage of PS can be noted

27

in the less complex animals (birds, possibly reptiles). With
the exception of the reduction in ruminants, no convincing
decline in PS has been demonstrated in mammals of different
levels of development. Data from ruminants suggest the in-
volvement of maturational factors associated with alimentary
conditions. The guinea pig also falls out of line with the
other rodents in per cent of PS although unlike the ruminants,
TST is not depressed. Little information is available on the
length of sleep periods in different animals and the distribu-
tion of these periods across the day, so an analysis of this
question cannot be made at this time. The decline in "sleep
cycle" length in the less complex animals would appear to
correspond to metabolic factors and thus supports the suggestion
of Roldan and Weiss. Sleep cycles however, may occur only
during the part of the day when animals sleep most.

The experiments which are reported in the research section
of this paper were motivated by the desire to obtain sleep-EEG
recordings from several animal species under controlled condi-
tions. The basic intention was simply to describe the sleeping
and waking patterns in animals which were anatomically similar
but had different environmental backgrounds. The animals (all
rodents) were chosen on the basis of their size and availability
as well as behavioral description as stated in Walker's Mammals
of the World (1964).

28

A variety of rodents was studied. They differed on the
basis of behavioral characteristics (burrowing rodents, sur-
face-dwelling rodents) and speed of bodily processes such as
heart rate and respiration rate.

METHOD
Subjects

Five species of rodent were used in these studies.

1. Hamster: Mesocricetus auratus. The six male hamsters
used in this study were obtained from Manor Farms (Staatsburg,
N.Y.) as young adults. Their mean pre-operative weight

was 116.30 grams (range 100-130 grams). Birth dates were not
known, but weight and physical characteristics indicated that
they were all post-puberal adults.

2. Rat: Rattus norvegicus. Six young-adult male rats
of approximately 120 days of age at the time of electrode
implantation were used. They were of the Long-Evans strain
and were obtained from Rockland Farms (Gilbertsville, Pa.).
Body weight averaged 232 grams.

3. Mouse: Mus musculus . Six young-adult mice of the
CRI strain obtained from the University of Florida animal farm
were used. They were all between 100 and 120 days of age at
time of surgery and averaged 32 grams in body weight.

4. Squirrel: Citellus tridecemlineatus. The squirrels
used in this study were obtained from W. R. Knight of Hiram
College, Hiram, Ohio. They were trapped wild, but had been
housed in our laboratories for approximately one year prior
to use in sleep studies. They had not participated In other

29

30

experiments and had been undisturbed except for routine care.
All squirrels were males with mean body weight of 186.17 grains.
An exact estimate of age In these animals would be difficult
but all were at least 18 months old. Body weight varied from
161 to 200 grams at time of Implantation.

5. Chinchilla: Chinchilla lanlger. These animals
were purchased from a commercial chinchilla farm (Dan Harris,
Fairmont, Minn.). Pre-operative body weights averaged 410
grams. Ages ranged from 8->14 months at time of surgery. All
were Judged to be young adults on this basis. Spector (1956)
reports the age of puberty in this species to be 4 months,
and Walker (1964) reports a life expectancy of 10 to 40 years
in domestic breeds. These commercially raised chinchillas
had all been rejected as prime fur animals because of minor
blemishes in their fur, but were healthy and showed no other
signs of abnormality.

Electrode Implantation Procedure

All animals were operated on under Nembutal (sodium
pentobarbital) anesthesia, administered intraperltoneally
(IP) . Atropine sulfate was also administered IP to prevent
respiratory problems. Minor variations in technique of im-
plantation were necessitated by the size of the different animals
and relative location of reference points for stereotaxic

31

implantation. Basically, the procedure involved placing two
cortical screw electrodes and one pair of stereotaxically located
depth electrodes for hippocarapal recordings. In the chinchilla,
muscle electrodes were also implanted bilaterally in the neck
for EMG recordings. Clean, but not sterile, surgical techniques
were used for all animals. Each animal was anesthetized according
to the following doses: hamster-90 mg./kgm. body weight, rat-
60 mg./kgm., raice-50 mg./kgm., ground 8quirrel-50 mg./kgm.,
and chinchilla-40 mg./kgm. These amounts were not absolute
because occasionally supplementary injections were required.
In several chinchillas, local injections of Xylocaine (lido-
caine hydrochloride) were used as supplementary anesthesia.

Electrodes were implanted using either a Baltimore
(Instrument Company) or Kopf (Instrument Company) stereotaxic
instrument. Cortical recording screws were always located
unilaterally over frontal and parietal areas. Distances between
these screws were consistent within species but varied slightly
between species. Clear differences in EEC patterns were noted
between sleep and waking during subsequent recording sessions
in all animals. Bipolar steel recording electrodes, insulated
except for the tips, were also directed toward the hippocampus
of all animals. These were made from 0.36 mm. diameter
nichrome wire crimped onto female Amphenol mini-connector
pins which were then embedded in a plastic block containing the
cortical recording leads. The whole assembly was chronically

32

fixed to the skull with stainless steel screws and acrylic cement.
Recording lead wires could be attached and removed as necessary.
This procedure was varied somewhat for the mouse. Due to the
small size of this animal, smaller electrode assemblies were
necessary. Commercially prepared "tripolar" stainless steel
electrode units of very small diameter and overall dimensions
were used (Plastic Products Company, Roanoke, Va.). One
.265 mm. diameter wire was directed to the hippocampus and the
other two were attached to 1.50 mm. diameter stainless steel
screws inserted into the skull. These were placed as described
for the other animals . Monopolar hippocampal potentials were
then referred to the frontal cortical lead during recording
sessions.

In chinchillas, neck muscle electrodes were constructed
of 32 gauge multistranded, stainless steel wire insulated with
Teflon. Two small loops of wire were inserted bilaterally into
the neck muscles positioned far enough to the side to allow
simultaneous recording of both EMG and heart beats.

Stereotaxic coordinates for the rat were obtained from Konig
and Klippel (1963). These were: 2.58 mm. anterior to the
ear bars, 2.5 mm. lateral, and 2.5 mm. below the surface of the
cortex. Hamster coordinates were estimated from histological
material obtained in other studies. (Dr. B. N. Bunnell provided
these.) These coordinates were: 2 mm. anterior from bregma,
2.75 mm. lateral, and 2.5 mm. below the surface of the cortex.

33

For ground squirrels, the values were: 5.7 nan. anterior to ear
bars, 3.7 mm. lateral, and 2.7 mm. below the surface of the
cortex (estimated by Dr. W. R. Knight on the basis of other
stereotaxic operations on this animal) .

No infoxtnation was available to define the stereotaxic
coordinates for hippocampal placements in the mouse or chinchilla.
These were determined during pilot operations in the following
manner. After fixing the head in the stereotaxic headholder,
the skull was removed unilaterally from an anesthetized mouse
and the cortex and corpus callosum removed by aspiration. The
hippocampus was then visualized and located in reference to
skull landmarks. A length of steel wire fixed to the electrode
carrier was used as a reference marker to determine the anterior,
lateral, and dorso- ventral distances from bregma, the midline
and surface of the cortex. Following this, the coordinates
were verified in another mouse. This was done by electrolytically
depositing iron from the tip of a wire introduced into the
hippocampus according to the obtained coordinates . The mouse
was then perfused with a solution of 1 per cent formalin and
ferrocyanate and the brain examined grossly under a dissecting
microscope. In two such tests, the deposit was found to be
within the dorsal hippocampus.

In the chinchilla, the following method was used to determine
the hippocampal coordinates. A chinchilla head with brain intact
was fixed in formalin and set up in the headholder. The

34

head was positioned with the incisor bar 8 mm. below ear bar
zero to level the skull. Steel pins were inserted through holes
cut in the skull at specific distances from skull landmarks.
These were placed 2 mm. posterior to bregma and 2 mm. lateral
to the midline, and also at 2 mm. posterior, 4 mm. lateral.
After removing the brain from the skull, it was examined grossly
by dissection under a dissecting microscope. The hippocampal
coordinates were then determined by measurement with a millimeter
scale. These were: 4 mm. posterior to bregma, 5 mm. lateral
to the midline, and 4 mm. below the surface of the cortex.

Apparatus

EEGs were recorded either with a Grass model III D or
model IV BEG unit. Recordings were made on as many as six animals
simultaneously. The animals were housed in individual cages
within an electrically shielded, sound-attenuating recording
booth that had a transparent face for visual observation.
A one-way vision glass was also installed. During recording
sessions, the room was kept completely dark except for a light
inside the recording booth and a small dim light over the EEG
machine. During the night, the observation window was covered
and the light inside the booth extinguished.

Cage size was determined by the size of the animal. For
most animals, small plastic freezer cannisters with air vents

35

cut in the sides or top were used. These were 11 cm. in
diameter and 16 cm. high for mice, 16 cm. x 18 cm. for ground
squirrels, or 18 cm. x 18 cm. for rats. Hamsters were recorded
in 18 cm. x 23 cm. square steel cages. Special recording
chambers were constructed for the chinchillas in order to
further reduce the possibility of environmental disturbances
in sleep patterns. These were adapted from picnic ice chests
measuring 30 x 50 x 30 cm. high. This was necessary because
these animals appeared very excitable in the laboratory. They
were not observed to sleep at any time during one month of
occasional checks, suggesting that the presence of the ex-
perimenter (E) was disturbing to them. Windows and one-way
vision mirrors were installed in each box for visual observation
while recording. Each box was illuminated from within by a
7\ watt lamp, and was ventilated by means of a hose connected
to a centrally located electric fan. Temperature control tests
were run in the boxes to insure that temperature variation was
not excessive. It was found that by controlling room temperature
and varying the draft of the fan drawing air through the boxes,
temperature could be controlled at 72 +4 F. During other
runs, room temperature was held to 70 + 2 F. A weak masking
noise was also used to reduce the possibility of disturbance
in sleep pattern due to occasional room noise. Ho masking noise
was used for the other animals studied.

36

Procedure

All animals were given at least one week of recovery time
from surgery. They were then habituatea lo the recording
situation. Hamsters, rats, and mice were adapted for two days,
ground squirrels and chinchillas for one week. On the day that
recording was to begin, electrode leads were attached at least
three hours before the start of data collection. Data were
collected continuously for 48 hours starting at 10 A.M.
A 12 hour day-night light cycle was maintained. Lights were
on at 9 A.M. and were turned off at 9 P.M. each day. The
light cycle was controlled during all days that the animals
were in the laboratory.

Ground squirrels were recorded for a second 48-hour period.
This was done for two reasons. In the first place, it was
decided that these animals required more than one week of adap-
tation, and second, because of difficulty in scoring the records
due to poor "theta" rhythms in several animals. The second
session followed one month of adaptation with dummy recording
wires attached to the animals' heads during the final two
adaptation days. Michelle clips were attached bilaterally to
the skin of the neck for El-IG recordings.

Approximately two months after completion of recording
session one for the rats, a second 48-hour period of EEC
recordings was made. Since two rats had lost their skull caps

37

by this time, .a.y the remaining four animals were run.

During daytime hours of all recording sessions, visual
observation of the animals was carried out continuously. The
animals were judged as awake, asleep, or in paradoxical sleep
every minute from 9 A.M. to 9 P.M. Judgments were recorded
on the ESG record as it came off the machine. Comments on the
behavior of the animals such as the waking activity engaged
in were also noted. The visual observation data were used later
as aids in scoring EEG records.

All EEG records were divided into one-minute sections and
scored visually according to whether the animal was judged
to be awake, in slow-wave sleep, or paradoxical sleep. Dif-
ferences in EEG patterns of less than 30 seconds were over-
looked by this method. In cases where a minute of data was
approximately evenly divided into two stages, an attempt was
made to preserve the continuity of the ongoing pattern. Thus,
a short (30 seconds or less) period of slow-wave activity
occurring in a long waking pattern was not scored as such.
These periods tended to occur mostly during the time just prior
to sleep onset. In a similar fashion, rapid activity was some-
times seen during periods of slow-wave sleep. An estimate
of reliability in scoring was obtained by comparing the inde-
pendent judgments of two scorers on data of each species.
Approximately 5 per cent of the data were double scored.
Agreement ranged from 89 per cent to 97 per cent on a minute
by minute comparison.

38

Visual observations of behavior were used as an aid in
scoring the EEG records. Agreement between the two was not
perfect, however. Disagreements usually occurred during
periods prior to and just following sleep onset. During these
periods of "going to sleep," slow-wave activity often occurred
with the eyes open. Chinchillas were regularly seen to sleep
in this condition. In cases where the EEG pattern was at
variance with the behavioral judgment, preference was always
given to the EEG index. Clear EEG pattern differences, cor-
responding to the awake, SWS, and PS phases were found for all
subjects.

Electrode placements were evaluated histologically for
each animal several months after data collection. The
animals were given lethal doses of Nembutal and perfused
intracardially with isotonic saline followed by 10 per cent
formalin. The tissue was embedded in celloidon, sectioned
at 30 micra and frontal sections through the electrode tracks
were stained with cresyl violet.

One hamster died before perfusion and the brain was lost
for histology. Since the chinchillas were to be used in other
studies, only two representative brains were examined.

RESULTS
Body Position During Sleep

With the exception of chinchillas, the rodents studied generally
slept in a relaxed position. In rats, mice, squirrels, and hamsters,
a fairly common position was seen. Most often the head would be
tucked under the body with the animal in a sterno-abdominal posi-
tion. Squirrels and hamsters slept predominantly in this posture,
often with the head completely invisible under the body. Occasionally,
SWS was recorded in rats and mice in more erect positions, often
with the eyes open. Sometimes, rats, mice, and hamsters were
observed to sleep on their sides as well, but this was less fre-
quent. The utility of the head tucked position in burrowing ro-
dents is not clear, but may serve to shield the eyes during sleep.

Natural sleeping positions, and hence the sleep pattern, could
have been affected by the electrode assembly attached to the
animal's head. However, since squirrels slept predominantly in
this position, and all animals were allowed at least two weeks
recovery time from surgery, the effect should have been minimal.
Most frequently the chinchilla was observed to sleep upright
with head lowered and ears flat against the side of the head.
Often the eyelids were open. After a period of SWS in this position,
a PS episode of short duration would sometimes occur. As EKG

39

40

A. Awake

DEEP •^^*-V^^'V^rt*/vv^^v^^4.^\^^A,v^,^^
EMG,EKG-n»"iV'{ i;iilili|i|iiniiiim|n.|(i|i |M.i1'''i<Mi||ii||iiii.J<nMiiiii,|(n<.i |(<|.||«>ih<i..i.«.'

^■^i^O.S*->f W»^IW

B. Awake. Immobile

ff^W^t'^!"*l*^>ff*"i'li""'"|''"*l"'HfMii| |l<W^!Wi^Mi><r'i|ififf>!Wf-*itiNiMfwwifiyM.Hif)ti(Hf>|fM>i|fMi(W| tmm,

'""''' " ■ ■ I .. ■^lit|;iljr.u;i,i|i,i,ii...iii'.|iiii"-

c, sws

T!1l|l'M'1ii*ttfit<i!Mj|i!ii|iii|i|iMiif|ii;t)niitiniti',iifvMiv;"|itiitvt!iii|"Wliill'n'i'tiim

D. PS, Slow Wave Bursts

• • • • •

I I 10 Sec. I

Figure 3. Examples of EEG records during waking and
sleep in Citellus tridecemlineatus .

41

activity decreased, the animal usually awakened spontaneously.
This may have been caused by loss of balance. On the other hand,
awakening occurred only after 20 or 30 seconds of H4G flattening
in some instances. At certain times, usually during the afternoon
hours, chinchillas were observed to sleep while lying on their
sides. PS episodes of longer duration, marked by complete neck
muscle atonia were usually seen at these times.

E5G Pattern Characteristics

In the ground squirrel, a distinctive EEG pattern of alternating
slow-wave and rapid activity was manifested. These periods con-
sisted of intermittent bursts of sleep and wakefulness of very
short duration. A clear behavior pattern was also evident at these
times. The squirrel usually stood immobile in an erect manner in
the "freezing" position. The eyes were almost always open but
occasional blinking was seen. This pattern was particularly evident
during the first day of recording and decreased in amount after
a longer adaptation period. Figure 3 shows samples of a ground
squirrel EEG record and illustrates this phenomenon. Redding,
et al. (1964) observed a similar EEG pattern in the dog and referred
to it as "somnus alterans." This pattern is different in character
from the drowsy state reported for the cat (Sterman, et^ al^' , 1965).
Since the hourly distribution of this stage paralleled the distribu-
tion of awake periods, it was classified as a waking state. In

42

contrast to those of the ground squirrel, EEG patterns of rats,
mice, hamsters, and chinchillas could be described by the three
classical pattern designations of awake, SUS , and PS.

In all animals a high frequency, low amplitude cortical tracing
characterized the waking record. Rapid activity was also recorded
from the hippocampus during these periods, but was often interrupted
by periods of synchronization or "theta" activity, usually associated
with sniffing. Hippocampal theta rhythm was not as clear in four
ground squirrels and absent in two of them. This was later found
to be caused by the electrode location. (See Histology section.)

With the onset of sleep, the characteristic pattern of slow-wave
activity was noted in all animals. Muscle activity in squirrels
and chinchillas usually decreased in amplitude with the onset of
sleep but this depended on the animal's posture.

Paradoxical sleep could be clearly identified in all animals
studied. This was evidenced by a rapid cortical EEG with a
synchronous hippocampal rhythm. In the squirrels, as was noted
previously, the hippocampal theta rhythm was ambiguous. In this
species, E-IG activity of the neck was used in conjunction with
cortical tracings to identify PS episodes. Visual observations
provided an additional source of information in defining this
stage. Neck muscle EMGs in squirrels always flattened during PS.
In chinchillas, the reduction was not clear, but depended upon the
sleeping position of the animal.

43

An additional pattern during PS, especially noted in the ground
squirrel, is illustrated in Figure 3, line D. Dots mark periods
of bursting activity recorded from surface and deep recording
electrodes. The bursts were of approximately 1-second duration
with frequency in the 7-9 cps range. Similar activity was observed
in other animals but was seen most clearly in the squirrel.

Sleep Patterns of Five Rodent Species

Descriptive measures of sleep were summarized for all animals.
The following measures were used to define and compare sleep
patterns.

1. Percentage of total time asleep during 24 hours.

2. Percentage of PS. This was expressed as percentage of
total sleep time for each animal.

3. Length of the sleep episode. Defined as the number of
minutes which elapsed between the onset of sleep to reawakening.

4. Length of SWS period.

5. Length of PS period.

6. Number of sleeping and waking periods.

7. Ratio of day sleep to night sleep.

Hourly distributions of mean minutes of sleep time for the
two recording days are given in Figures 4 to 8 for all species.
These figures show that sleep patterns were variable from hour
to hour on day 1, with a tendency for the hourly distributions of

o o

3>IV/VW NIW

o o

C\J —

d331S dO NIW

45

o^.

m

0

c

0

10

in

03

u

3

3

u

C

•<-•

•H

tn

g

%

C

u

«s

0

0)

c

e

w

<w

3

o

4J

(0

(«

C

o:

0

•H

c

w

<IF«

3

X>

<o

..^

C •

u

W CM

u

(U

w

*J 73

•rl

4J C

"O

TO «

0.

>^

f-l

t-l

a

U

di ce

3

fli >,

O

!-( 0}

X

(«

"0

3
60

d331S dO NI^M

o

«

«

9
C

6«

C r^
«9 9

g|

(01
01 9

o
•^ c

9

.fi a

•f^ C

u u

*■> V
CO u

a
-< a

U V
9 «

5 '^
X «a

9
60

d33ns dO NIW

47

«*.< .

O w

3

CO 4i>

« «

W M

9 9

I

(0

s of mean m
esocricetus

<i

V)

S S

^

q:

0

3

•H c

O

4J .<»4

X

9

z

Ui

•w w

5

tJ «o

<

(—

a

(0

Hourly
sleep

9
60

<»<

-^

u>

48

(0 iri

« 4t

W U

S v4

e c

^4 CD

g -4

■?

a.

c a

(0

ns of mea
Chlnchill

(rt

o

^

a:

3

•H e

4J •rl

o

O

3

z

I

XI «
•rl C

z

V trf

4J «
« 4J

UJ

t4 4J

5

tJ «

1-

>s*^

s

•-' o.

h «

9 41

ID

O r-l

00

4)

9
M

d33nS dO NIW

49

sleep to become less variable on day 2. As shown in these figures,
a high degree of dependence between minutes of SWS and PS existed
in all recordings. Hourly distributions of PS closely approximated
SWS distributions in form for all species studied.

All animals except hamsters showed at least two distinctive
waking periods. These were most noticeable on day 2. Hamsters
were predominantly awake in a single period between 8 P.M. and
midnight, and were highly stable in distribution of sleep time
from day 1 to day 2 .

Means of obtained values for the different measurements are
given in Tables 1 to 5 for all species. Individual values which
contributed to these means are in Appendices A to E.

Figure 4 shows the sleeping and waking pattern distributions
for the ground squirrel. As shown by this figure, the drowsy
pattern was distributed reciprocally with respect to sleep patterns
and followed the waking pattern distribution, thus suggesting
that it be considered a waking state.

Statistical Treatment of ESG Pattern Data

All E£G data were punched onto IBM cards and summarized by
computer. Amounts and frequencies of occurrence of each stage
were analyzed for each animal, and summary statistics obtained for
individuals and groups . These values were used to compare the
sleep characteristics of the five species studied. In squirrels,

50

Table 1
Summary of Values for the Mouse

Day One Day Two

Mean Per Cent of Total Sleep Time

56.85 51.32

SD «• 5.51 SD - 6.91

Mean Length of Total Sleep Period

8.50 7.03

SD - 1.81 SD =- 0.98

Mean Length of SWS Period

6.37 5.22

SD = 0.879 SD - 0.277

Mean Length of PS Period

2.35 2.28

SD - 0.320 SD - 0.141

Mean Day-Night Sleep Ratio

1.45 1.42

SD - 0.710 SD - 0.101
Mean Per Cent of PS

9.002 10.21

SD =» 3.11 SD - 2.10

Mean Sleep Cycle Time

12.825 11.190

51

Table 2
Suianiary of Values for the Hamster

Day One Day Two

Mean Per Cent of Total Sleep Time

59.25 60.97

SD - 5.32 SD - 1.72
Mean Length of Total Sleep Period

12.31 10.52

SD - 1.71 SD - 3.21

Mean Length of SWS Period

6.10 5.05

SD - 0.854 SD - 0.190

Mean Length of PS Period

4.02 3.68

SD » 0.27 SD » 0.42
Mean Day-Night Sleep Ratio

1.876 1.558

SD - 0.353 SD » 0.056

Mean Per Cent of PS

23.37 23.47

SD - 5.12 SD - 3.75

Mean Sleep Cycle Time

10.562 10.336

52

Table 3
Suimnary of Values for the Rat

Day One Day Two

Day Three

Day Four

46.67
SD - 6.75

51.91
SD =- 3.47

Mean Per Cent of Total Sleep Time
51.94 58.46

SD = 8.39 SD - 3.34
Mean Length of Total Sleep Period

5.73 7.28 5.59 5.58

SD - 1.14 SD - 0.63 SD » 0.89 SD " 0.81

Mean Length of SWS

3.867

4.62

3.40

3.68

SD = 0.60

SD "

0.360

SD - 0.510

SD =

0.529

Mean Length of PS

2.33

2.48

2.65

2.48

SD - 0.24

SD -

0.24

SD - 0.14

SD -

0.10

Mean Day-Night Sleep Ratio

1.49 1.46 1.20 1.46

SD » 0.244 SD =• 0.178 SD = 0.210 SD - 0.316
Mean Per Cent of PS

18.46 20.65 29.39 25.45

SD - 2.86 SD - 3.68 SD - 4.04 SD " 3.69
Mean Sleep Cycle Time

8.071 9.150 7.013 7.524

53

Table 4
Summary of Values for the Squirrel

Day One Day Two

Mean Per Cent of Total Sleep Time

53.64 60.46

SD = 8.92 SD =• 10.24

Mean Length of Total Sleep Period

11.48 12.98

SD - 2.144 SD - 3.46

Kean Length of SWS Period

5.40 5.98

SD = 0.716 SD » 1.27

Mean Length of PS Period

3.65 3.65

SD " 0.35 SD = 0.24
Mean Day-Night Sleep Ratio

0.799 1.015

SD = 0.259 SD - 0.105

Mean Per Cent of PS

25.66 23.71

SD =» 2.89 SD - 3.88

Mean Sleep Cycle Time

12.720 11.128

54

Table 5
Summary of Values for the Chinchilla

Day One Day Two

Mean Per Cent of Total Sleep Time

42.80 52.23

SD = 2.77 SD - 2.99

Mean Length of Total Sleep Period

5.28 6.23

SD = 0.84 SD " 0.80

Mean Length of SWS Period

4.48 4.53

SD - 0.673 SD - 0.412

Mean Length of PS Period

2.22 2.15

SD - 0.30 SD - 0.14

Mean Day-Night Sleep Ratio

1.21 1.26

SD - 0.285 SD - 0.28
Mean Per Cent of PS

10.21 14.57

SD - 3.74 SD " 2.54

Mean Sleep Cycle Time

6.714 5.935

55

only the second 48-hour recording period was considered. Minutes
of sleeping and waking stages for different species were compared
on the basis of periodicity, total amounts, and average lengths
of the different stages.

Periodicity of Sleep

Diurnal periodicity

As shown in Figures 4 to 8, sleep time was distributed with
relationship to the day-night light cycle. This periodicity may
be expressed as the ratio of day to night sleep. Ratios greater
than 1.0 indicate nocturnal sleep-wake patterns, while diurnal
animals (day active) would have ratios less than 1.0. These values
are given in Tables 1 to 5 for each species studied.

The effect of daily periodicity was statistically analyzed
for all groups of animals simultaneously (Table 6). Day-night
differences in mean number of minutes asleep were statistically
significant (F = 106.14; df = 1 , 40; EMS = 3,213.70) as was
the interaction between species and diurnal period (F " 16.25;
df «- 1, 40; U-iS = 3,213.7). The interaction reflects differences
in day-night distributions of sleep time among groups, and hence
demonstrates group differences in day-night periodicity of sleep.
As shown in the tables of summary values, all groups except squirrels
showed day-night sleep ratios greater than 1.0. Sleep time in
squirrels was equally distributed among dark and light periods,
although predominant periods of waking occurred. These were from 1 P.M. to

\0
1-1

at
H

01

B

«
«

S

n

0)

3
C

«
M

o

Id

56

CO

t

1

S.

St

a

ss

o

r-«

•o

t-4

CM

CM

•

•

•

•

•

CO

«

»<

1

sf

1

I

l>.

<n

i

vO
O

f-l

r-4

1

z

CM

1

o

U-l

o

o

<JN

o

f.

o

o

r^

NO

1-4

m

o

CO

ON

o

00

r^

1-t

o>

o»

•

•

•

•

•

«

•

•

•

•

•

04

CO

vO

CM

o

o

l-H

M

in

CM

r^

(/)

o\

>o

«M

<t

00

CM

r^

r-<

«n

r-t

1^

s

)

cs

vO

•

CM

NO

ON

1-1

CM

CM

o

<n

r^

vO

t^

o

CM

T-*

CM

cn

m

1-4

N

es*

fH

m

<M

ON

St

m

o

•-I

-*

Irt

•-4

St

ir»

f-i

St

m

•o

r>i

<N

o\

CM

CM

CM

r-l

o

O

o

o

o

o

O

8

o

O

o

o

o

o

o

n

(^

o

o

<n

1^

m

m

1-1

ON

1-1

o

tM

00

n

vO

v£>

ON

St

d

p«.

CM

m

1-t

<^

M

00

vO

t-i

vf>

CM

vO

<M

CM

St

•*

en

irt

CM

V

CO

00

f-4

r*>

vO

CM

ir>

m

1-1

00

n

o

ON

m

ON

>o

fO

1-1

CM

St

1-1

CO

o

CM

>*

a.

r*.

o

vO

1-1

CM

sf

r*

<*

o

CO

1-4

CO

CM

w*

!-•

00

<n

CM

1

<-i

o

10

•

o.

§

V

M

a

O

K

t

•o

"O

O

CI)

CO

o

1-t

0)

>N

►n

•rt

t-i

O.

&

JS

U

41

3

(0

a

0)

P4

O

CO

CO

CO

K

PU

K

X

ITI

o

c

M

1-1

«

•

V

u

«

e

M

«

0>

K

4)

u

CI

•rH

u

1-1

O.

u

c

a

M

•v4

>4

V

u

»

O

o

x

oa

3

o

M

3

o

m

U

o

3

4J

u

M

«i

>>

O

M

3

O

u

>N

«

^

o.

o

V

o.

h

T^

CO

U

Vi

•H

ki

u

«

a

k<

M

PQ

CO

H

»

O

o

H

Q

o

H

Q

CO

H

4C

57

4 P.M. and 7 P.M. to 1 A.M. as shown in Figure 4 for day 2.

Sleep cycles

Sleep EB6 cycles were computed by counting the number o£ minutes
from the end of one PS episode to the end of the next one. Only
periods uninterrupted by waking were counted. Mean cycle times
for groups are also given in the summary tables. Complete cycles
tended to occur only during that part of the day when sleep pre-
dominated for each species. For squirrels, complete cycles were
more uniformly distributed across the day. Mean cycle times
were computed by averaging the total number of cycles across all
animals within each group. This was done to minimize the effect
of several animals in which complete cycles were few and variable.

An analysis of variance (Table 7) revealed significant dif-
ferences in sleep cycle times among groups (F = 12.44; df = 4, 25;
EMS ■ 3.23). A posteriori comparisons of group means were also
computed; however, differences in means did not suggest a pattern
that was related to other measured parameters of sleep . (See
Appendix G.)

The expected relationship between sleep cycle time, heart
rate, and respiration rate was not found. Mice, for example, with
the most rapid heart and respiration rates, had one of the longest
cycle times. Certain pairs of species, markedly different in
respiration or heart rate, were found to be not different in
cycle time. The failure of these data to support the hypothesized

58

I

«

JO

m

ji
u

60

e

V

u
u

l-l

0)
U

c

1-4

>
O

(0

l-l

«

o

en

CM

C«J

«

•

o

CO

•*

csjI

C>J

«

CO

U
Wi

3
O
M

f^

o

r^

m

00

r»

■

•

•

r^

o

o

•*

vO

00

CM

f-l

m

V

l-l

1-1

u

<a

o

0

4J

4)

u

5

a.

u

CO

H

V

59

relationship between these variables may be seen by Inspection of
Table 8, which Is a suninary o£ species comparisons of these
values. Statistical tests on these variables In general bore
out this lack of predicted relationship. Statistical Independence
was indicated by the low correlation, not significantly different
from zero, found between respiration rate and sleep cycle time.
This correlation was 0.06 (Pearson product -moment r) . The
correlation between the group mean heart rate and mean cycle time
was 0.70, which is in the opposite direction to that predicted
by the hypothesis of Weiss and Roldan (1964).

Total Amounts of Sleep

An analysis of variance indicated significant group differences
in total amounts of sleep (Table 6). This analysis was performed
on combined data for both days. However, total sleep time increased
from day 1 to day 2 for one species. The chinchillas, which were
found to be significantly lower than hamsters on individual com-
parisons of total sleep time, increased on day 2. For this reason,
the day-2 minutes of sleep time were subjected to an Individual
analysis (Table 9). A barely significant effect due to total sleep
time was found among groups (F « 3.003; df - 4, 25; EMS = 8,664).
Day-2 mean percentages of sleep suggested no obvious group differences
which could have contributed to this (Figure 9, right hand bars
for all species). Mean percentage of day-2 sleep was 52.23 per cent

60

m

«

-r*

U

«

0.

M

4J

e

«

•0

0

ptf

%

•H

Pit «

. B

U -H

0 H

U-l

«

n 1-4

«l u

*J >.

« u

»i

MH

c 0

00

0

•H 0)

V

■U 1^

r-t

Qt U

.0

Vi n>

«

•^ PQ

H

a

w «

« X

(Hi **

•0 e

e 0

««

•0

- «

m u

« «>

«J •o

5<S

w

h

Id

«

s

«

CO

«i

B

a

»-t

f-i

v-i

00

j=

<*

u

•

c

VO

en

•r4

in

x:

tH

u

4J

0

CD

in

tti

«

*

00

00

U

01

CM

4J

vO

<|c

10

•
f-l

&

C0

t-*

m

X

«*

0

u

m

u

•

0

1>t

CSJ

CM

r*

3

i-»

00

C

04

CO

CM

01

n

w

•

4c

3

CM

0

en

0

r-4

0

NO

X

yo

•-•

■ic

41

0)

B

4J

•H

fd .^

H

<N

^^.

41

•

c -^

r-l

<-s

4)

c

0 S

U

•

4J

T*

tH ^

^

C

s

^

4J 0)

e

(0

M -<

a

«^

*J

4J

•H 0

41

M

<

0. >»

«

«

4)

u u

r-t

«

e

« >•

M

s

OS

CO

a

c

4)

«

0)
>

0)

0)
3
r-l
««
>

4)

a.

o
u
u

4)

0.

CO

e
o

V4

c

U3
O

4)
>-<
V
?

(0

•r4
U
4)

4J
(0
«

c

c

"O 4>

41 a
M <

CO

4) "O

3 V

CO O

61

«

CO

^

CO

u

M
3
O
CO

■X

(O

o

o

1

1

•

CO

CO

^

f-<

vO

o

\0

1

vO

CO

CM

«*

in

o\

N

CV|

CM

T-t

r»

r*.

o

vO

o

vO

O

•*

vO

CM

o

i-(

CO

f-«

CM

(0

0)

CO

•H

01

U

•H

0)

O

a<

«

CO

CO u

c

o

0)

^K

•-•

^ «

«a

4J

*J N^

u

<S

^

5

f*T=y^^»KMyj>'t£ii:sp!^VS''9

p n-I^i-

, ■ I BUI Jl III J 11 ^>CHy»fri.'it»UajA..JJI^gg?M^

t ) • • • ■ ■« _<t ■ • ___• •••••••••••■ »

j

■ .-kJ-

- " . ~ • • • • • • • • • • •_•■ ^^^

................. ■•.•■l; (^?-

/(e-a yo 0

0

63

for the chinchilla. Other day-2 values ranged from a low of 51.32
per cent for the mouse to a high of 60.97 per cent for the hamster.
This suggested either that overall differences in the analysis
of both days together were due to chance, or else a complex change
in TST occurred from day 1 to day 2. In other words, the day 1
differences probably had different causes than did the day-2
differences. Since the day-2 analysis revealed significant
differences, individual comparisons among group means of TST were
made. No simple comparisons of means were significant. According
to Winer (1962, p. 208), an overall F ratio that is significant
indicates that at least one possible comparison of means or groups
of means is significant. In view of other results of this study,
the animals were grouped according to whether they were hibernators
or non-hibernators, and compared. This test revealed a significant
difference between these groups (p < .05), showing that ground
squirrels and hamsters combined, slept significantly more on day 2
than did the other groups combined. A similar test was also made
of the overall amounts of sleep for both days. In this test also,
squirrels and hamsters were found to differ from rats, mice, and
chinchillas on total minutes of sleep. These results are sug-
gestive and not definitive, especially since differences in group
means were very small. The interpretation of these analyses is
that hibernating rodents may sleep more than non-hibernators.
This effect was masked during the overall analysis, in part, by
an effect of adaptation occurring in chinchillas. (See section on

64

Generality.) In general, the amounts of time slept per group were
very similar, with a suggestion of higher total amounts for squirrels
and hamsters. In addition, variability of total amount of sleep
per day was small. Means and standard deviations of total minutes
of sleep for individual animals are shown in Appendix F.

Stability of total minutes of sleep for day and night periods
can be estimated by comparing day and night variability for each
species of rodent. The means and standard deviations for total
minutes of sleep were calculated for each group studied. These
are given in Appendix J. These standard deviations did not suggest
any marked differences in stability of total sleep time between
diurnal periods.

Length of Sleep Periods

The number of sleep periods also varied among the different
species studied. The mean number of periods per day, for days 1
and 2 combined were: mice, 102.60; rat, 123.42; hamster, 80.58;
squirrel, 64.56; chinchilla, 121.25. These values varied in the
way that would be predicted on the basis of other sleep characteris-
tics defined in this study. For example, animals with short sleep
episodes had higher numbers of episodes to attain the values of
total sleep time which were approximately equal for all species.
There was no significant difference in number of periods between
the first and second days of recording. Individual frequencies

65

of sleep periods are given in Appendix K.

Analyses of variance were also computed on the length of
the sleep episode and length of SWS periods. The results of
these analyses are shown in Tables 10 and 11. Figures 10 and
11 present the means and standard deviations graphically.
Significant differences between species were found for all of
these measures. The pattern of differences which is Illustrated
by the graphs were, in general, supported by individual comparison
tests. Squirrels and hamsters had longer periods of SWS and PS
and longer total sleep episodes. (A sleep episode is defined
as the period extending from onset of sleep to reawakening and
thus always includes SWS periods; it may or may not include
periods of PS.) The pattern was somewhat disturbed because
of the extended length of SWS periods in mice. Mean lengths of
SWS in this species did not differ from hamsters and squirrels
as would be suggested by the other comparative data. The most
stable differences among species were found in length of PS periods,

Description of Paradoxical Sleep

Paradoxical sleep, expressed as percentage of TST also varied
among groups. This was highly significant (Table 12) (F - 30.60;
df - 4, 25; EMS ■» 17.50). Squirrels and hamsters had the highest
percentages followed by rats and then chinchillas and mice.
This pattern of differences also suggests that hibernating animals

66

CO

ON

OS

o

M

«

V
«

S

o «

r-l ^

0) 00

^ c

H

g

01
■H

o

«s

l-l

O

C«l

o

O

NO

N

r-*

CO

o

1^

CO

•

•

•

•

•

S

1

f-l

NO

t

o

00

CM

•

M-l

ON

1 "

U^

o

1 -

-*

lO

•o

CN|

CM

<n

CM

o

o

o

en

o

o

•

•

•

M

«^J

o

CO

eg

o

CM

ON

CO

\o

o

in

o

n

vO

lO

>*

i-H

■-I

CO

>»

(0

Q

0)

(A

CO

X

C

«

CO

V

V

a

e

4)

u

a*

T*

VI

•H

1-1

Vi

u

2

u

o

J5

«

U

o

3

«

VI

4J

>»

4)

Vi

o

0)

a

Vi

s*

<

a

Vi

(O

pq

CO

H

»

Q

CO

M

o
o

V

in
o

V

*

67

t*4

J3
H

0)
•O

O
•H

M

0)

P4

I-l

CO

§!

cd

I
o

1-4
CO

00

c

01

u

c

(0
5

o

s

CO

X

w

CO

u
o

CO

$

1

SO

Sf

o

CO

vO

1

•

•

1

z

•

1

r*

o

r>.

>*

vO

o

St-

o

I-l

CO

(y>

*n

>*

CO

1

•

•

1

•

•

•

vO

o

o

OJ

o

ON

<>4

sT

in

SI

>-*

>*

m

c^

CM

r«>

vO

r-l

ro

ON

o

-*

o>

>*

m

00

CO

so

oo

*

•

•

•

•

•

•

r^

ir»

f>j

r*

o

o\

r*.

St

CM

CM

f-t

(0

>%

s

(0

CO

m

CO

X

c

03

m

01

0>

c

V

01

•H

Vi

1-4

•<-i

u

3

U

o

X

M

u

o

4J

V

u

4-1

>>

0)

u

0)

Ou

u

1-t

S

a

M

M

CO

H

»

CO

H

V

^

O

hH

Lj» •• • • • • • • • • *»*^

>#.-i,A-^».., A«>»M»«*-«rt^»v%

!*• • " • • • • •! >W. /

a . ' . ' . ft •••.*•...•• • • • • • - -■ -"^

n. •• .* • • • ■ • • *.•.•.•.•.".

-T"

Sdinuii/y

r\

CO Lo ^~

Sdjnu/i/y

70

C«J

4)

1-1
jQ
l«
H

p.
«
Oi

i-l
CO

u

X

•8
««

M

cd

a

c
u

«
u

c

ct

•rl
U

o

•-4

-t

CO

CO

■ t

z

Z

' E

o

vj-

o

\o

Ot

m

Ik.

1

1

1

•

*

1

vO

o

0^

<n

<n

li^

lr»

•^

sj-

ox

W

•

•

•

•

•

X

1

m

r>.

1

I-)

so

o

en

r-)

CJ

>^

r-t

m

«M

O*

1 ■*

in

o

1 -

sf

•n

"O

CM

<N

<n

cs

CO

so

r»

vO

OS

<*>

<f

f*.

CV4

>*

toj

T-t

r^

<n

m

*

•

•

•

•

•

•

w

<T>

CVJ

»>>

O

l-i

m

<n

lo-

•*

«n

\£>

C<l

>o

r^

in

r^

-»

«n

CM

N

CM

u

>^

.s

M

CO

M
CO

K

c

«a

m

<u

<u

u

c

V

u

&!

•r4

u

•H

T*

U

M

»

U

o

X

»

u

O

3

u

4)

u

4J

>%

0)

>4

O

&

o.

u

•H

Cd

a

M

M

(0

H

»

Q

CO

H

71

slept differently than non-hibernators . However, the pattern was
not completely clear. A posteriori comparisons of individual
means revealed that the following pairs of means differed sig-
nificantly at p <.05: Ha -Mo, Ha-Ch, GS-Mo, GS-Ch, and Ra-Mo.
In general, the differences suggested by Figure 12 were supported
by statistical analysis. The failure to find differences between
rat-hamster and rat-squirrel however, casts doubt upon the idea
that hibernators are distinctly different from non-hibernators
in percentage of PS. Further confusion is added by the PS per-
centages of days 3 and 4 for the rat. These were 25.30 per cent
and 25.45 per cent, well within the range of hibernating rodents
found here.

An analysis of variance was also computed on the lengths of
PS periods. This is given in Table 13. Figure 13 shows the
mean length of PS period for groups. Significant differences
due to species were demonstrated. Individual comparisons tests
showed that hamsters and squirrels had significantly longer PS
periods than did the other rodents studied (Appendix L) .

Sequential Pattern of Stages

Without exception, periods of PS followed SWS periods. Short
periods of arousal occurred during episodes of SWS and PS . These
were mostly associated with, but sometimes occurred without,
noticeable body movements. The sequential cycle of waking, SWS,

ddsis leioi 1 0 0/^

73

b4

en

(0
H

CO

O
•H
U
0)

a

i-i

u
o

Cd

u

0)

u

c

(0

I-I

o

09

fl
u
>»

X

M
CO

u
u

3

o

to

(O

j«

Z

z

!{|

00

1^

ir»

00

m

o

ON

ON

1

•

00

•

•

«

o
m

•
o

1-1

1

o

CN|

I

§

o

§

•

•

«

•

•

r^

o

o

o

o

ON

•*

in

o

I-I

«*

»n

C>4

CM

CN4

ON

<n

vO

<V4

00

f-l

CO

-*

-*

O

-it

o

CO

o

•

•

•

•

•

•

•

m

o

in

CM

o

o

<>J

«*i

<n

a

>.

(Q

O

(0

OT

aa

CO

»<

c

CO

ca

0)

<u

c

V

a>

•H

u

t4

•r4

u

»

U

o

x:

m

u

o

4J

<u

u

4J

>>

u

u

V

a

u

•H

S

a

u

m

CO

H

»

CO

H

o
o

V

a

|i-. , y,fi .. I..I -I

.EJ:-:"

■^-■■■*,'pgj^-)^<i-^--w -"-^ 9 ■.■^t.Tr:g''ijsraga

T fb

-.•a-.-."-'-*.' .•-'.'-:-;*I-.- .•.•.•.•.•.•.-.•.• .•.•-".•J 0*-' ;:

fj • • • • • • • • " • • • • • •.•_•..-•• • ■ • • • •* ^^

t«> , . 1 • • • • • • • • • • • • ■ •»".■.•.•..... • • • ;■ ^^ 1 ■- -^

i^ • • I • • • • • • • • • • • • " •*•.".•.-••• « • • * h *>m^ ^•

rill iiTiiiriiiiiiTwuTi I it-f»-*^*«^-=^'«^-^*^**^-^

^sx^Bt^iopps^Beamm

CD :

• • • • * • "*•--*••••- • • • * • • " •_•».. ••3 V

Ct3 -n

"Ko

I Ha •■ •■••■••••

r !■■!■< ii7i ainrii 111 ■'■imrMftiii Tn *fwi II

'■iiiiiffii^wifirri-.'^Vr-

^

00

-r

sdinui'sAl

75

and then PS, as reported for the rat, rabbit, and mouse (Roldan &
Weiss, 1962, 1963; Weiss £> Roldan, 1964), was not always seen.
They reported relatively fixed sequences of waking, SWS , PS,
and then waking occurring in cycles in these animals. In the
present study, long periods of wakefulness often followed periods
of SWS without interposed periods of PS. In addition, periods
of SWS often followed PS periods without interruption by an arousal
period, in contrast to the findings of these authors. It is true,
however, that PS episodes often ended with body movements and
a short arousal period as* reported by these authors.

The data of the present study do not support the notion that
sleep is "programmed" in fixed cycles.

Regularity of pattern sequences was dependent upon the time
of day. During hours of the day when sleep predominated, se-
quential patterning was highest.

Relationship Between Length
of Waking and Sleep Periods

The total day was divided into minutes of waking and sleep
for all animals studied. Because of the sequential nature of
these mutually exclusive categories, the number of sleep and
wake periods was necessarily equal. Hence, if the total number
of sleep episodes is known, the number of awake periods is known
also. These are equal in theory but can differ by one, in

76

finite measurements. Other statistical analyses of this study
demonstrated only very slight differences in total minutes of
sleep. This adds an additional restriction to the data with
respect to the relationship between length of sleep and waking
periods. That is, if total minutes and frequencies of sleep
periods are fixed, the mean of the waking periods is automatically
defined also. Thus, comparisons of mean lengths of waking and
sleep periods are not possible between species.

Individual distributions of sleep and wake episodes do not
suffer from these restrictions. Although the total number of
waking and sleep periods which can occur in succession are
equal by definition, there are no restrictions on the lengths
of individual periods. A detailed analysis was made to determine
the relationship between waking period length and the length of
the sleep period which followed it. An attempt was made to answer
the question: In normal sleep patterns, can the length of the
sleep period be predicted by the length of the preceding waking
period? Table 14 is a frequency distribution of sleep episode
lengths which followed waking periods of given lengths. In general,
the results of this analysis suggested a positive relationship
between waking period length and the length of the succeeding
sleep period. A more detailed discussion of this is presented
in Appendix I .

77

Relationship Between Length of
Sleep Periods and Percentage of Paradoxical Sleep

Examination of the summary data for percentage of paradoxical
sleep and sleep period duration indicated that these two variables
tended to co-vary. The chinchilla, for example, with a low
percentage of PS also tended to have shorter sleep period lengths,
while the ground squirrels and hamsters with longer sleep periods
had corresponding higher percentages of PS. The correlation
between these measures was found to be 0.72 (Pearson product-
moment r) indicating a fairly high statistical dependence between
them. The correlation between length of the SWS period alone and
percentage of PS was 0.66. These analyses would indicate that
a significant portion of the variance in PS percentage is asso-
ciated with the length of the sleep episode. In order to more
closely define this relationship, a finer analysis was performed.
Table 15 shows the results of this analysis. The lengths of all
SWS periods for all animals were measured. Frequencies of periods
of different lengths were determined and are given in these
tables. In addition, the frequency and lengths of PS periods
wbich followed SWS periods of different lengths were determined.
The results of this analysis are presented In Appendix M. There
was a general tendency for PS periods to follow SWS periods of
increasing length. A more complete discussion of this relation-
ship is presented in Appendix M.

78

Summary of Obtained Results

Descriptive values of sleep in the five rodent species
under consideration have been presented. These results can
be summarized as follows .

1. It was found that mice, rats, hamsters, and chinchillas
were nocturnal while squirrels slept approximately equal amounts
both day and night.

2. Total sleep time increased significantly from day 1 to
day 2 in the chinchilla. This probably indicates an effect of
adaptation.

3. Rodents which are hibernators (hamster, squirrel), slept
significantly more than non-hibernating rodents (rat, mouse,
chinchilla) on day 2 and days 1 and 2 combined.

4. Percentage of PS was higher for hibernating rodents
than for mice and chinchillas. Rats were intermediate in value,
but further study is needed to check this.

5. Significant differences in length of PS, SWS , and TST
periods were found. In general, hibernating rodents had longer
periods than non-hibernators .

6. Sleep cycle times varied significantly but no meaningful
relationship between cycle time and respiration or heart rate
was evident.

7. Differences in the mean number of sleep periods for different

79

species were found. In general, hamsters and squirrels had fewer
periods than did the other rodents studied.

Generality of Obtained Data

In this study, the characteristics of rodent sleeping and
waking EEG patterns have been analyzed for purposes of comparison.
In addition, normative values were presented for descriptive
purposes. The value of obtained norms is dependent upon their
generality, which in turn depends upon the effects of the recording
situation on the sleep pattern. This can be estimated in part
by comparing the first and second day sleep patterns. If an
effect of adaptation was present, it should, for example, have
been reflected by increased amounts of sleep during the second
day, or by similar changes in other measured values. Analyses
of variance were computed on all data for different summary statistics.
These are shown in Tables 6 and 10 to 13. No significant effects
were found for percentage of PS, mean length of PS period, mean
length of total sleep episode, or mean length of SWS periods.
Minutes of total time asleep differed significantly from day 1
to day 2 for all groups analyzed together (Table 6) . Examination
of the day-l-day-2 minutes of sleep were made to assess possible
changes of sleep time for each species individually. This was
done because examination of the distributions of means and
variances of TST for different species suggested that not all

80

increased on day 2. The greatest change in TST occurred in the
hamster (1.72 per cent). Mice, in contrast, had a mean reduction
of 5.53 per cent from day 1 to day 2. Individual tests of day-1-
day-2 differences in TST were made (t test for correlated means).
It was found that chinchillas increased significantly from day 1
to day 2 (t - 4.52, df - 5, EMS = 30.04). The increase approached
significance for the rat (t = 1.99, df - 5, QiS =• 47.04). For
hamsters, mice, and squirrels, the changes due to days were not
statistically significant. (The t values for these tests are
as follows: hamster, t ^ 0.62, df " 5, EMS = 40.10; mouse,
t - 1.62, df = 5, EMS = 49.10; squirrels, t " 1.74, df » 5,
QIS = 52.44.) It was concluded therefore that day-l-day-2
differences in total minutes of sleep existed only in chinchillas.
Other groups may have changed slightly, but the changes were
insignificant.

Group means of all descriptive measurements were presented
graphically in Figures 9 to 13. Statistical tests of the effect
of days upon the different measures of sleep have been presented,
and except for the increase in TST in chinchillas as noted above,
no change in descriptive measurements of sleep across days was
found. Another way to assess the day to day effect of the
laboratory situation on sleep patterns is to compare the mean
distributions of minutes of sleep across each day. Figures 4
through 8 allow this comparison for the different species. As

81

shown by these figures, hamsters and chinchillas were more stable
in pattern from day to day than other animals.

The statistical analysis of minutes of sleep for groups, days,
and diurnal periods provides an estimate of this stability. In
this analysis, each subject was represented by four scores. Two
of these scores are measures of sleep time within different
periods of day 1, and two represent measures of sleep time for
the same periods of day 2. If a lack of stability from day to
day existed in these data, a significant days x periods interaction
would be predicted. This was not found, thus demonstrating that
diurnal distributions of sleep were stable from day 1 to day 2.

It can be concluded that one measure of sleep (total sleep
time) changed from day 1 to day 2 (chinchilla, possibly rat),
but that other measures did not. For between-species comparisons,
day 2 values of TST would be most accurate and distribution of
patterns over time during day 2 would be more representative of
normal sleep patterns than day 1 values.

All rodents studied spent significant portions of the two
days asleep. A high degree of stability was shown for some
aspects of sleep, while others changed across days. Because
of the length of adaptation periods and lack of day to day changes
in measured parameters of sleep, it is reasonable to conclude
that stable levels of sleep were reached on day 2. It can be
concluded therefore, that descriptive values of sleep pattern
have general species validity for laboratory housed animals.

82

Unfortunately, no reasonable generalization can be made to rodents
in the natural environment. At present, limitations in technique
do not allow such determinations.

HISTOLOGY

Histological analysis of electrode positions revealed that
for most animals the electrodes were placed within the dorsal
hippocampus. Two representative chinchilla brains were examined.
Both of these had electrodes in the most dorsal aspects of the
hippocampus. One of these was approximately in the center of its
lateral extent while the other was positioned more laterally.

In all squirrels except one, the electrodes were positioned
much too far anteriorly and were uniformly located in the corpus
callosum. In one squirrel, the electrode track was not discernable,
probably because this animal had lost its electrode prior to perfusion.
In another squirrel, the electrode was located just above the
hippocampus at its lateral end. In spite of the poorly located
electrodes, theta activity was observed in several squirrels
during EEG recordings, although of an ambiguous nature. Syn-
chronized theta activity can be recorded from neocortex in the
rat with certain electrode placements (Swisher, 1962) so that
this result is not particularly surprising.

For mice, electrodes were generally more laterally placed
but with one exception were situated well within the hippocampus.
In one mouse, the hippocampus was not penetrated by the electrode
which rested just above it. Rats and hamsters had electrodes
uniformly well placed in the dorsal hippocampus in its most dorsal

83

84

aspect. These electrode placements varied only slightly in the
medio-lateral dimension.

Electrode assemblies were fixed to the skull with machine
screws. Due to the very thin skull in mice, the screws had
penetrated the skull and depressed the brain in these animals.
Small lesion-like depressions, approximately the size of the
screws, were seen in histological analysis. Whether the effect
of this could have disturbed the sleep patterns during later
recording is very unlikely, but must be considered.

In one squirrel, a large cyst in the left temporal area was
seen. This animal had died two months after the final recording
in which it participated and almost four months after surgery.
The sleep pattern for this animal was not unusual, thus suggesting
that the infection which was responsible for the damage was not
well advanced at the time of recording.

DISCUSSION

The results of this study suggest that ground squirrels
and hamsters, both hibernating rodents, have similar sleep patterns
differing substantially from the non-hibernating rodents studied.
This general finding was true for length and percentage values
of almost all comparative analyses. These animals appeared to
sleep more deeply than non-hibernating rodents. During the time
that these animals were in the laboratory, distinct differences
in arousal thresholds were obvious. Chinchillas were never
observed sleeping during one month of routine care and handling
prior to their use in sleep studies. Squirrels and hamsters on
the other hand, once asleep were observed to remain asleep during
the most disturbing environmental conditions. Sleeping squirrels
and hamsters could sometimes be picked up bodily without awakening,
whereas the mere presence of E in the room ensured the wakefulness
of chinchillas. Differences in depth of sleep may reflect metabolic
differences between these two types of rodent. Mice and rats
were intermediate in depth of sleep as informally observed.

The incidental observations of depth of sleep among rodents
were very consistent. They agree with the observations on posture
during sleep which suggested that hibernators sleep more deeply,
and with the statistical analyses of sleep periods and percentages.
In order to more fully test this proposition, arousal thresholds

85

86

during sleep must be determined experimentally under controlled
conditions for hibernating and non-hibernating rodents. Large
differences in threshold values between species would be predicted.

Of the five species studied, data from two can be compared
with other experiments. Only rats and mice have been studied
previously. Agreement between data of the present study and
other studies is marginal. Paradoxical sleep percentages of
rats and mice differed markedly from the values reported by
Weiss and Roldan (1964). They reported values of 16 per cent and
17.5 per cent for rats and mice observed in short-term studies.
Their recording period covered only the daytime hours. In order
to determine whether the reduced percentage of PS observed for
the mouse in the present study was due to differences in recording
times, a further analysis of PS percentage was made. Percentages
of PS for daytime hours alone were computed on the present data,
but did not differ from the value computed for the entire day
(approximately 10 per cent). This ruled out the time of recording
as a possible source of error.

Weiss and Roldan reported very short sleep cycles in the
mouse and PS periods of very brief duration. If a significant
number of PS periods were of less than 30-second duration in the
present study, they would have been overlooked by our scoring
system and might have resulted in lowered PS percentages. For
this reason, day-2 EEC records of two mice were rescored in segments

87

of 10 seconds Instead of the original l-minute segments. This
analysis resulted in percentage values within 1.0 per cent of
the initially determined values, confirming that an artifact of
scoring did not reduce the PS percentages in the present study.
It can be concluded that discrepancies in PS for the mouse were
not due to time of recording nor to scoring systems used. Another
possibility that cannot be evaluated at this time is that strain
differences were responsible. Alternatively, differences in
adaptation of the mice to the laboratory, or in control of environ-
mental disturbances between the two studies could have been
responsible. It is not possible to assess these possibilities
at this time.

Equally unclear are the causes of differences in PS percentages
determined for the rat in the present study and by Weiss and Roldan
(16 per cent). The results of the present study are more in
agreement with the results of Van Twyver, et al^. (1966), who
reported 23 per cent for this animal. Further, different mean
lengths of PS periods for rats and mice were found in this study,
as compared with Weiss and Roldan. The above authors reported
values of 2.1 minutes and 1.5 minutes as compared with the stable
values of 2.43 minutes and 2.32 minutes reported for rats and mice
in the present study.

Paradoxical sleep percentages in this study ranged from
about 10 per cent in the mouse to 25 per cent in squirrels. This
range is almost as great as that observed for the entire range

88

of mammalian species studied to date, and argues against any
simple evolutionary progression of PS percentage as suggested by
Jouvet (1967b) . PS percentages In other animals have been shown
to vary In a fashion which precludes a simple trend of Increasing
percentages for animals of higher evolutionary position. This
range Includes the lowest value found (guinea pig, 2.5 per cent;
Jouvet, £t al^. , 1966), to the highest (opossum, 30 per cent; Snyder,
1967). Primate percentages reported have been intermediate.
Thus, the evidence discounts any simple trend of Increasing
amounts of PS in higher animals.

Alternatively, PS amounts have been shown to be more associated
with alimentary factors such as rumination and weaning, than with
taxonomlc classification. High correlations between PS percentage
and lengths of sleep periods were demonstrated in the present
study for animals of similar ontogenetic and phylogenetlc standing.
This finding is best explained by the Ephron-Carrington (1966,
1967) hypothesis that PS represents a process of activation which
counteracts the effects of Inactivatlon during SWS. Longer mean
sleep periods were associated with higher percentages of PS in
this study, as would be predicted by this hypothesized relationship.
This study does not bear upon the notion expressed by them that
activation builds up gradually during SWS. Percentage of PS was
closely related with other characteristics of sleep in the present
study. A similar relationship was found in the opossum (Snyder,
1967) , which had a high percentage of PS associated with longer

89

sleep times.

An hypothesis associating PS percentages and other sleep
characteristics to the general level of "excitability" of the
species is suggested by these data. Deep sleepers, it is hypo-
thesized, sleep in longer periods and get higher percentages of PS.
This notion is supported by data of the present study which found
higher PS percentages in animals thought to sleep more deeply
than others. During PS, according to many authors, arousal
thresholds increase markedly compared with SWS arousal thresholds.
Animals which live relatively unconcealed from predators must rely
on quick responses to danger signals. Shortened sleep periods
may serve to prevent the occurrence of PS episodes with accompanying
loss of responsiveness to environmental suimuli. Sheep, guinea
pigs, and chinchillas fall in this category and have low per-
centages of PS. Mice do not fit this general picture, and further
studies are necessary to resolve this discrepancy. Although
well-controlled animal studies are few, and at this point, data
are not conclusive, general support exists for Jouvet's impression
(1967b) that predators enjoy deeper sleep and higher levels of FS
than do hunted animals. However, PS percentages may be more
related to length and depth of sleep period and only secondarily
related to the animals' food-getting status.

All of the rodents studied slept approximately equal amounts
of time each day. According to Agnew and Webb (1967), the average
amount of total time slept by young adult humans is 7V to 8 hours.

90

or about 31 to 33 per cent of the day. Very little is knox^rn about
the total time spent in sleep for different animals. The data
which are available with a few puzzling exceptions, suggest that
they all sleep approximately 50 per cent of the day under laboratory
conditions. Other evidence (Kleitman, 1963) suggests that humans
under comparable conditions sleep about 50 per cent of the day,
also. Determinations of sleep time require care to ensure that
the animal studied is not being deprived of sleep by the recording
situation nor recovering from sleep deprivation induced by pre-
recording handling. This may have contributed a great deal more
than is commonly realized to many studies of sleep.

In cats, 55 to 58 per cent of the day is spent in sleep
(Sterman, e£ al . , 1965; Ursin, 1967). Baseline studies of primate
sleep patterns have generally been done with the assumption that
they do not sleep during the day. This assumption may be invalid
for humans as well as for monkeys. Juvenile rhesus monkeys however,
sleep from 45 to 55 per cent of the day as reported by Meier and
Berger (1965). Values for rabbits, rats, and guinea pigs also
fall within this general range (Van Twyver, et aj^. , 1966; Duncan,
^al., 1967; Khazan & Sawyer, 1963; Jouvet, et a_l . , 1966).

Ruminants may sleep less than the approximate value of 50
per cent common to other mammals. However, a confounding factor
argues against this. Ruckebusch (1962), for example, reported
very low values in one lamb (32 minutes total sleep per day) which
was 30 days of age. In a second study, Ruckebusch (1953a),

91

reported that total sleep percentages never exceeded 40 per cent
in lambs. However, the sheep spends a great deal of time in
rumination, during which slow-wave activity, indistinguishable
from SWS is recorded. This suggests that rumination may be
a state of sleep and that the values reported are unrealistically
low. Total sleep percentages reported for the donkey by Rucke-
busch (1963b) are also low (13 per cent). This may have resulted
from faulty assumptions about the behavioral position and slow-wave
sleep. It Is not clear from his description, but it appears
that in this study, the decision of whether the donkey was awake
or asleep was based only upon behavioral observations, although
EEGs were reported for different stages of sleep. These were
probably acute recordings but this is not clear from his report.
SWS may have been obtained by the donkey while in the standing
position. A more controlled test is necessary to determine this.

The available data on mammals suggest that total amounts of
sleep time are approximately the same for all species. Small
differences in total sleep time obtained in the present study may
have been due to depth of sleep. Squirrels and hamsters may
simply have been more impervious to minor changes in the environment
ahd hence had more endogenously determined sleep patterns. The
reduced amounts of sleep in man, it is suggested, reflect environmental
factors. It is hypothesized that a minimal amount of sleep is a
basic requirement for all mammalian species. Animals under ad libitum
conditions, with no restrictions upon the amount of free time to

92

sleep, will sleep more than this. The laboratory environment
provides this kind of situation. Under such conditions, approximately
50 per cent of the day is spent sleeping. In man, however, and
in rats trained to reduce their sleep time (Van Twyver & Webb,
1967), the requirements of the environment reduce this amount
of sleep. Studies investigating the limits of sleep time
reduction in the rat are currently being run to test this hypo-
thesis.

The alternating sequence of slow waves and then fast waves
observed in squirrels and classified as a waking state was related
to a behavior pattern which occurs in this rodent in its natural
environment. The "freezing" pattern is often seen during orientation
to danger signals. Quick alterations between sleep and waking
of the sort suggested by the EEG data, may have functional
utility in allowing the squirrel to maintain very long periods
of this protective posture while at the same time maintaining
maximum surveillance of the environment.

In the present study, sleep-EEG cycles were determined by
measuring the elapsed time between the end of one PS episode and
the end of the next one. Careful examination of procedures used
in other studies reveal a general lack of correspondence between
experiments in the way that this periodicity has been determined.
The concept of rhythmicity in appearance of sleep stages was initially
suggested by Dement and Kleitman (1957a). They determined this
periodicity in man and cat by measuring cycle time between PS

93

periods as was done here (e.g., frcn the end of one cycle to
the end of the next). However, other studies have not defined
this in a similar fashion. Weiss and Roldan measured the time
between the end of one PS period and the beginning of the next
one to define sleep cycle time. Waking periods could also occur
in this interval, adding even more confusion. Reite, ejt al . (1965)
reported the cycle time incidentally for the monkey but did not
measure it, while the method used by Sterman, e_t al . (1965) for
the cat is unclear. In the present study, no meaningful relation-
ship between sleep cycle time and the speed of other bodily pro-
cesses was found for different animals. If a functional
relationship exists between cycle time and heart rate or respiration
rate, it may hold only for broad comparisons. Cycle times in the
present study of the rodent varied only from mean values of
6 to 13 minutes. This range is relatively small considering
the greater differences in cycle time which have been reported
between primates and rodents. The general relationship in sleep
cycle time and other metabolic indices illustrated in Figure 1
and hypothesized for rodents by Weiss and Roldsn is difficult to
interpret because the values reflected do not have common deter-
minations. One wonders what meaning can be attached to the
relationship between them. At the present time, it must be
concluded, the concept of sleep-EEG cycle times as reflecting
metabolic processes common to heart rate and respiration is open
to question.

94

The present study also disagrees with the notion expressed
by Weiss and Roldan (1964) that sleep stages follow a fixed
pattern. Sequential patterning is more common during hours of
high sleep time but as demonstrated by this study, patterns do
not occur in a fixed sequence during most of the day. In
general, the data of the present study were notable for the high
degree of stability between days.

SUMMARY

An EEG study investigated the sleeping and waking patterns
of five species of rodent under controlled conditions. Electrodes for
electroencephalography were implanted surgically into brain structures of
six each of the following species: Mus Musculus, Rattus norvegicus .
Mesocricetus auratus , Citellus tridecemlineatus , and Chinchilla laniger.
The animals were recorded either for two or four days under condi-
tions of day-night simulation. During the light part of the cycle
they were observed continuously and judgments of sleep, waking,
and paradoxical sleep were recorded every minute. EEG patterns
supplemented by observational data were divided into 1-minute
segments, scored and analyzed statistically. Normative values for
sleep characteristics in the five species were determined and
compared. It was concluded on the basis of stability of pattern
percentages that these data were representative for each species
under laboratory conditions.

An additional EEG waking pattern, distinct from the classical
one, was reported for Citellus. This pattern consisted of rapid
alternations in sleep and waking, and occurred v?hen this animal
was alert but immobile for extended periods.

Summary statistics of total percentage of sleep, percentage
of paradoxical sleep, and sleep period lengths for the five
species, suggested that they could be grouped on the basis of

95

96

hibernators and non-hibernators. Hibernators generally had signifi-
cantly longer sleep periods, higher percentages of paradoxical
sleep and slightly higher percentages of total sleep time. It
was suggested that these animals slept more deeply than non-
hibernators.

Day-night periodicity of sleep was observed in all animals
studied except Citellus. This rodent's sleep was more evenly
distributed in day and night periods.

No meaningful relationship between another type of periodicity,
the "sleep-EEG cycle" and respiration and heart rate, as suggested
by Weiss and Roldan was found for these species. It was concluded
that if these variables are related at all, the relationship
holds only for broad comparisons and not for comparisons of animals
of similar size and phylogenetic position.

It was also concluded, on the basis of the present study
and other studies of sleep, that no progressive increase of PS,
as suggested by Jouvet, occurs with increases of phylogenetic
complexity.

APPENDIX A
INDIVIDUAL VALUES FOR THE MOUSE

Sublect

Day 1

Day 2

Percent (

of Total

Sleep Time

1

57.92

38.12

2

45.90

47.92

3

57.99

52.92

k

61 .25

52.36

5

55.21

58.54

6

62.85

58.06

Mean Lenqth

of Total

Sli

eep Period

1

6.5

5.5

2

5.9

6.7

3

8.5

7.5

4

9.6

6.6

5

9.5

8.6

6

11 .2

7.3

Mean L<

snqth of

SWS

Period

1

5.8

4.8

2

4.6

5.0

3

6.9

5.5

4

6.9

4.9

5

6.8

5.7

6

7.2

5.4

97

98

APPENDIX A (Continued)

Subject Day 1 Day 2
Mean Length of PS Period

1
2

3
k

5
6

I

1
2

3

k

5
6

1
2

3

k

5

6

2.1

2.7
2.1

1.9
2.7
2.6

2.3
2.5
2.2
2.2
2.2
2.3

Day-Night Sleep Ratios

1.46 1.28
1.82 1.52
1.21 1.45
1.34 1.56
1.67 1.35
1.23 1.35

Percentage of PS

3.84 6.19

11.04 11.59

6.47 10.37
8.28 12.86

12.45 11.03

11.93 9.21

APPENDIX B
INDIVIDUAL VALUES FOR THE HAMPSTER

Sub iect Day ] Day 2

Percent

of Total

Sleep Time

1

66. 74

62.08

2

66.60

57.71

3

53.96

60.69

k

56.88

60.90

5

56.04

61.39

6

55.28

63.06

Mean Lenqi

th of Tota

1 Sl^

eep Period

1

13.4

7.8

2

12.0

6.9

3

12.5

16.8

k

8.8

10.0

5

13.2

9.8

6

14.0

11.8

Mean

Length of

SWS

Period

1

7.0

4.4

2

6.6

4.7

3

7.1

5.8

4

4.9

5.1

5

5.2

5.2

6

5.8

5.1

99

100

APPENDIX B (Continued)

Subiect

Day 1

Day 2

Mean

Length of PS

Period

1

4.0

2.6

2

4.6

4.7

3

3.8

3.9

k

3.2

3.4

5

3.7

3.7

6

4.8

3.8

Day

-Niqht Sleep

Ratios

1

1.52

1.49

2

1.69

1.60

3

2.44

1.63

k

1.68

1.51

5

1.65

1.50

6

2.29

1.62

Percentage c

if PS

1

19.35

18.12

2

22.11

20.82

3

16.60

28.26

4

21.98

22.12

5

29.12

23.08

6

31.03

28.41

APPENDIX C
INDIVIDUAL VALUES FOR THE RAT
Subject Day 1 Day 2 Day 3 Day k

Percentage

of Total

Sleep Time

1

63.68

63.82

- •

-

2

5^.72

57.43

56.07

56.53

3

53.61

53.12

-

-

h

35.61

60.00

44.10

47.57

5

50.00

59.93

38.06

49.79

6

54.03

56.46

47.85

53.75

Mean Length

of Total

Sleep Period

1

7.7

7.7

-

-

2

5.0

7.1

-

-

3

6.0

8.0

-

-

k

k.o

6.3

-

-

5

5.7

7.7

-

-

6

6.2

7.0

-

-

Mean Length of SWS Period

1

5.0

5.3

-

-

2

3.9

4.8

4.1

4.3

3

3.8

4.8

-

-

4

2.9

4.4

3.2

2.8

5

3.6

4.1

2.7

3.6

6

4.0

4.3

3.6

4.0

101

102

APPENDIX C (Cont

inued)

Subiect

Day 1

Day 2

Day 2

Day 4

Mean Lenqi

th of PS

Period

1

2.0

2.2

-

-

2

2.k

2.6

-

-

3

2.4

2.6

-

-

k

2.4

2.3

-

-

5

2.4

2.8

-

-

6

2.7

2.4

-

-

Day

-Ni

qht Slee

p Ratios

1

1.41

1.70

-

-

2

1.08

1.20

1.15

1.83

3

1.56

1.64

-

-

4

1.88

1.48

0.89

1.08

5

1.39

1 .28

1.32

1.26

6

1.60

1.46

1.45

1.67

Percentage

of PS

!

15.59

20.67

-

-

2

14.09

26.72

21.81

24.57

3

21.76

21.18

-

-

4

17.93

22.69

27.40

31.53

5

20.83

15.18

31.02

24.13

6

20.57

17.47

21.34

21 .58

APPENDIX D
INDIVIDUAL VALUES FOR THE SQUIRREL

Subject Day 1 Day 2

Percent

Ol

' Total

Slee

p Time

1

59.03

65.76

2

67.15

75.42

3

38.19

56.46

k

48.89

46.18

5

55.00

67.64

6

53.61

50.62

Mean Lenq'

th

of Tota

1 SI

eep Period

1

11.8

16.9

2

12.7

16.2

3

6.8

8.3

k

13.1

15.1

5

\2.k

12.8

6

12.1

8.6

Mean

Le

nqth of

SWS

Period

1

4.9

5.6

2

6.2

8.2

3

4.1

4.6

k

5.9

7.0

5

5.9

5.7

6

5.4

4.8

103

104

APPENDIX D (Continued)

Subject Day 1 Day 2
Mean Length of PS Period

1 3.9 3.6

2 3.5 3.9

3 3.1 3.4

4 4.2 4.0

5 3.6 3.6

6 3.6 3.4
Day-Night Sleep Ratios

1 1.14 1.22

2 1.01 0.92

3 1.21 1.07

4 0.54 0.92

5 0.89 0.99

6 0.58 0.98
Percentage of PS

1 30.24 29.14

2 25.75 17.68

3 21.09 23.00

4 26.22 25.56

5 25.76 26.59

6 24.87 20.44

APPENDIX E
INDIVIDUAL VALUES FOR THE CHINCHILLA

Sub ject Day 1 Day 2

Percent

of Total

S leep Time

1

40.14

55.49

2

43.19

48.26

3

44.93

52.85

k

47.29

49.72

5

38.33

50.49

6

42.85

56.53

Mean Length

of Tota'

1 Sleep Period

1

5.7

6.8

2

6.4

6.6

3

5.7

7.4

k

5.2

6.1

5

3.7

4.9

6

5.0

5.6

Mean Len

qth of SWS Period

1

5.2

5.1

2

5.0

4.8

3

4.9

4.6

4

4.5

4.4

5

3.3

3.9

6

4.0

4.4

105

106

APPENDIX E (Continued)

Subject

Day 1

Day 2

Mean

Lenqth of PS

Period

1

1.8

2.2

2

2.7

2.2

3

1.9

1.9

k

2.4

2.2

5

2.5

2.3

6

2.0

2.1

Day

-Niqht Sleep

Rat

;ios

1

1.58

1.78

2

1.38

1.07

3

1.50

1.17

4

0.96

1.44

5

0.99

1.16

6

0.87

0.95

Percentage

of

PS.

1

4.67

12.39

2

10.13

15.68

3

16.09

19.7]

4

8.61

13.27

5

8.51

13.48

6

12.64

12.89

APPENDIX F

Basic Data for Analysis of Total Sleep Time;
Minutes of Sleep per Period

MO

HA

RA

Day

■ 1

Day

_2

Day

Niqht

Day

Niqht

1

495

339

308

241

2

427

235

4)6

274

3

458

377

451

311

k

505

377

460

294

5

497

298

484

359

6

499

406

481

355

1

579

382

535

359

2

603

356

512

319

3

55 i

226

542

332

k

513

306

527

350

5

502

305

530

354

6

554

242

561

347

1

536

381

512

301

2

409

379

471

392

3

470

302

537

327

4

335

178

457

308

5

419

301

465

362

6

479

299

545

374

107

108
APPENDIX F (Continued)

Day

J_

Day

_2

Day

Niqht

Day

Niqht

SCI

1

452

398

521

426

2

485

482

519

567

3

301

249

42 1

392

k

259

477

318

347

5

373

419

484

490

6

283

489

360

369

CH

1

370

208

489

310

2

322

300

403

292

3

349

298

457

304

4

402

279

350

366

5

297

255

361

366

6

301

316

378

436

APPENDIX G
Basic Data for the Analysis of Sleep Cycle Times

MO

HA

RA

Day 1

Day 2

1

13.7

16.0

2

12.2

11.6

3

12.1

13.1

k

10.5

10.4

5

13.8

10.1

6

]k.2

10.1

1

12.8

7.6

2

14.8

14.8

3

7.8

11.3

4

8.2

9.5

5

9.6

10.0

6

11.0

11. 1

1

8.2

10.5

2

8.8

8.6

3

8.3

10.3

4

6.0

8.7

5

6.9

7.6

6

8.9

9.2

109

110

APPENDIX G (Continued)

Day 1

Day 2

SQ

1

12.7

10.4

2

12.4

12.2

3

12.1

12.4

k

]k.]

13.7

5

]2A

10.2

6

12.5

9.9

CH

1

12.7

10.4

2

12.4

12.2

3

12. 1

12.4

k

14.1

13.7

5

12.4

10.2

6

12.5

9.9

APPENDIX H
PR0CEDUR2 FOR DETERMINATION OF HEART RATE
AND RESPIRATION RATE

As stated in the body of the text, respiration and heart
rate values for certain rodents were obtained from Spector (1955).
Determinations of heart rates for squirrels and chinchillas were
made by counting 60 one-minute segments of heart rate recordings
for four individual animals. All segments were selected from
periods of sleeping EEG records (SWS), Respiration values were
determined by attaching a mercury filled Silastic strain gauge
around the body of different animals and recording changes in re-
sistence induced in a bridge circuit. This circuit was connected
directly to the EEG machine and written onto the EEG record.
Respiration frequencies for four animals recorded in this fashion
were then counted for 50 one-minute segments of SWS .

Mean values of respiration and heart rate indicated in Table
8 of the text were then computed from these values.

Ill

APPENDIX I
RELATIONSHIP BETWEEN LENGTH OF AWAICE
AND SUCCEEDING SLEEP PERIODS

The tables in this appendix are plots of sleep period
frequencies which followed waking periods of different lengths,
Waking period length is given in the first column (min.).
In column 2 (hist.)» the frequency of sleep episodes is shown.
The distributions of lengths of these episodes are given for
each l-ininute interval of wake periods in columns 1 to 11.
Column 11 includes periods which were equal to or greater than
11 minutes. Column 0 may be disregarded.

Mean lengths of sleep periods following each of five dif-
ferent waking period lengths were calculated. These means
increased slightly with longer waking period lengths but the
increase was only suggestive. The maximum increase amounted
to only 1.67 minutes (mice) from 1 to 25 minutes of waking
period length. Other animals showed decreases in sleep period
length. It can be concluded that if a relationship exists
between sleep and awake period lengths, it is very slight.

112

APPENDIX I (Continued)

MIC.

ii:

K I :\ HIST

11

0

3

0

2

0

0

n

0

0

0

0

0

i.

4bl

2

S8

56

41

32

32

30

24

16

16

9

105

2

172

0

35

24

18

19

14

2

12

7

0

4

37

3

9 0

c

24

14

7

8

7

4

5

1

2

1

17

4

66

1

14

9

0

6

6

3

4

0

1

U

22

5

44

12

9

5

4

2

4

0

0

2

1

K

-*

6

45

0

11

8

3

0

2

3

2

1

Q

4

11

7

37

0

6

8

5

3

3

3

2

2

2

0

3

0

27

1

5

2

7

3

1

2

3

0

1

0

2

9

3 7

S

4

5

3

2

2

2

1

2

1

7

10

18

C

5

1

0

2

3

1

1

1

0

i

3

11

19

0

7

1

2

3

0

2

0

1

c

0

^

-/

1?

21

;";

6

1

4

2

0

2

0

0

1

5

13

12

c

3

2

1

2

1

n

0

1

0

0

2

14

16

c

5

3

3

0

1

2

1

0

0

1

0

15

13

0

2

1
1

G

0

1

1

1

T^

1

1

4

16

14

0

3

3

2

0

1

1

0

0

0

2

17

10

0

1

1

3

1

1

0

1

0

0

1

1

18

10

2

1

2

2

1

0

0

0

0

2

IsJ

9

0

0

0

2

0

2

0

0

1

*-*

0

4

20

10

0

1

2

2

1

0

0

1

2

0

1

21

B

r

w

3

0

0

1

Q

u

0

0

0

0

4

22

3

/^

0

2

0

0

c

0

1

0

0

0

0

2-5

12

0

4

1

1

0

1

0

1

0

0

o
u

4

24

7

c

2

1

1

0

1

0

0

n

■^

0

0

2

25

6

0

1

0

0

0

0

0

0

i

0

4

2 6

6

0

2

1

1

0

1

0

Q

1

0

0

0

27

3

0

1

1

0

0

0

0

0

0

0

0

1

2S

7

c

2

1

0

Q

2

■J

0

Pi

0

0

2

29

5

c

0

C

2

1

0

1

1

0

0

0

0

30

5

c

0

2

0

0

2

0

1

0

G

0

p,

31

39

0

10

11

7

3

2

3

0

0

1

0

2

CRGSS HIST

■64 171 124 96 89 66 69 36 29 25 258

APPENDIX I (Continued)

114

RATS CAYS 1 AND 2

MN

HIST

0

5

1

662

2

243

3

105

4

63

3

45

6

50

7

32

8

22

9

27

10

21

11

19

12

2 3

13

14

14

9

lb

17

16

14

17

9

13

11

19

9

20

6

21

10

22

7

23

5

24

2

2 5

4

26

4

27

2

28

2

29

4

30

5

31

32

CRCSS HIST

0

1

2

3

4

5

6

7

8

9

10

11

c

I

C

"t
^

Q

1

0

0

0

1

0

c

0

129

37

56

50

41

50

31

31

23

30

134

n

49

36

19

14

23

18

19

6

9

5

45

n
O

25

10

11

7

10

10

5

4

5

1

17

1

15

IG

7

4

6

2

1

5

2

2

8

0

13

9

2

3

0

T

J

4

4

3

I

3

la

6

2

4

3

0

2

5

I

4

0

c

6

5

3

1

0

2

1

2

2

4

6

0

3

6

2

2

1

0

1

2

1^

1

3

0

7

5

3

3

2

1

0

0

1

2

3

c

5

2

3

3

0

1

1

1

G

0

5

0

1

2

2

2

1

3

3

1

1

0

3

n.
1^

4

3

1

3

3

1

0

1

Q

2

5

0

3

1

2

1

1

0

1

0

0

2

3

c

1

1

1

1

0

1

0

1

r

0

O

^

c

6

2

1

i

1

2

1

0

,'■;

0

3

0

5

0

2

0

2

1

0

1

i.

0

1

i.

2

Q

1

2

1

1

2

0

0

0

1

0

I

1

1

2

2

2

0

1

3

1

1

c

4

G

G

1

1

1

1

0

0

0

\_

0

1

I

2

0

0

1

0

0

0

1

c

5

1

1

Q

n

0

1

0

0

0

2

2

1

C

0

1

0

G

0

0

0

3

v-*

0

L

1

0

1

0

0

0

0

0

2

p,

0

1

0

0

0

0

0

0

r*
V

1

0

1

i

0

1

0

0

0

1

G

0

0

0

1

3

0

0

0

0

0

0

0

0

0

1

C

1

0

0

0

0

0

0

0

c

0

1

A
^

0

1

0

G

0

0

0

0

0

1

c

c

0

1

0

1

0

0

0

1^

0

1

G

1

0

1

0

1

0

1

0

0

c

3

7

3

3

2

1

2

3

-.

^

0

1

1

318

20 5

131

105

107

105

78

69

5 3

56

262

APPENDIX I (Continued)

.15

RATS CAYS 3 A;'siD 4

I\ til ST

13 II

0

0

rs
O

0

C

0

c

3

0

0

0

0

i

3 87

1

74

64

43

37

3 3

35

25

10

15

9

41

2

193

I

34

17

27

20

19

16

15

9

7

4

29

3

1 C 3

1

20

12

13

12

7

6

3

4

2

3

12

4

48

o

6

6

4

5

5

0

1

3

1

12

5

44

0

7

0

6

7

4

2

2

0

1

1

6

6

18

0

6

3

1

I

2

1

0

2

0

0

2

7

29

c

5

7

0

5

4

I

I

2

I

1

2

3

16

r\

3

2

0

2

3

3

I

0

I

0

I

9

15

0

4

4

2

I

0

r_

c

0

G

0

4

10

9

n

3

1

i

0

1

1

c

2

G

G

11

12

C

2

3

I

I

4

^

u

0

9

I

0

0

12

iO

Q

1

1

2

0

2

1

I

0

p

0

2

13

10

0

1

C

1

I

3

0

2

0

I

0

1

14

8

0

1

3

I

0

Q

0

2

0

I

0

15

6

0

0

I

2

1

G

1

I

vJ

0

G

16

8

0

1

1

I

1

2

I

1

i.

0

c

0

G

17

10

0

1

2

0

2

1

X

1

1

0

G

0

2

IS

6

0

2

I

2

0

1

u

0

0

G

0

0

19

9

0

4

0

0

0

0

2

0

0

I

1

1

20

9

c

I

I

1

0

0

0

1

1

0

0

4

21

5

0

1.

G

I

I

0

1

c

0

0

0

1

22

5

c

0

C

1

2

I

0

0

0

Q

1

0

23

6

0

0

2

1

I

I

0

0

0

I

0

24

9

c

1

2

0

2

1

i

Q

1

1

0

0

2 5

4

0

1

1

0

^
u

0

1

i-

0

1

0

0

2 6

3

0

1

0

0

1

0

0

0

0

1

27

2

0

0

I

0

0

0

0

0

0

i

0

Q

23

3

0

1

0

Q

0

0

0

1

0

0

0

1
4.

2 9

4

0

0

1

I

1

I

0

0

0

0

0

30

1

0

Q

0

0

I

0

0

0

0

0

0

0

31

30

c

12

4

4

2

2

1

2

2

0

0

I

CROSS HIST

3 192 147 116 106 99

85 67 33 3 5 23 125

116

APPENDIX I (Continued)

HAMSTERS

F I M HIST

9 10 11

2

C

0

C

0

0

0

0

0

0

c

1

1

1

3 46

1

5 2

33

24

2 3

15

25

16

12

12

a

120

2

132

c

16

14

11

14

3

4

7

3

4

7

44

73

1

15

10

3

4

7

6

2

2

1

3

19

4

64

r.

14

9

5

4

4

4

2

3

1

0

18

5

38

n

4

3

3

6

4

2

2

0

6

3

5

6

34

0

5

3

3

3

2

4

1

2

1

2

8

7

27

0

5

0

3

3

1

1

0

0

3

1

10

8

22

0

4

X

4

0

n
U

2

1

0

0

10

9

18

c

1

4

1

1

3

0

1

1

1

0

Zi

10

10

0

4

1

1

2

2

1

0

1

1

0

6

11

17

0

3

2

1

1

0

3

0

0

2

0

0

6

12

18

c

3

2

1

4

1

0

0

0

0

1

6

13

10

n

1

1

->

0

1

0

0

0

0

0

5

14

11

2

0

r\

0

1

1

n

0

0

7

15

11

0

1

c

0

1

1

1

2

0

0

2

3

15

9

0

1

3

0

0

2

1

0

1

0

1

17

10

c

2

1

0

0

0

0

0

1

0

0

6

18

10

0

4

0

0

0

0

1

c

0

r

0

5

19

9

0

1

C

2

3

1

0

0

0

0

1

1

20

8

n

1

0

1

0

1

0

0

1

1

3

21

4

0

1

0

1

0

0

0

1

0

'.J

0

1

22

5

0

1

c

0

0

1

0

1

1

1

0

0

23

3

c

0

0

0

0

0

0

0

1

1

0

0

24

7

c

1

1

0

0

0

1

0

1

0

0

3

25

5

c

1

0

0

0

0

1

0

0

0

3

2 6

2

0

1

A.

c

0

0

0

c

0

0

\

2 7

4

0

0

0

0

1

0

0

0

0

o

"J

3

26

4

c

0

0

0

]_

0

2

0

0

u

<J

1

29

3

0

0

0

0

2

0

0

0

0

0

0

^
i

30

2

0

Q

0

1

0

0

0

0

Q

0

1

31

42

0

8

6

1

1

3

1

1

4

1

2

14

CRUSS HIST

2 151 97 66

HO

54 58 39 38 35 32 317

117

APPENDIX I (Continued)

GRQUMD SQUIRRELS
MNhlST C 1 2 3 4 5 6 7 8 91011

0

4

1

291

2

120

3

73

4

43

'i

3 8

6

24

7

32

8

17

9

18

10

19

11

10

12

11

13

IS

14

7

15

12

16

6

17

10

13

5

19

5

20

9

21

6

22

4

23

6

24

6

25

6

26

4

27

6

23

6

29

2

30

2

31

48

CRUSS hi I ST

0

0

0

c

G

1

c

0

1

0

0

2

1

4 0

2 5

25

14

14

10

14

p

9

10

121

c

21

11

12

5

6

5

4

2

2

0

52

1

11

11

5

3

4

^

3

1

3

2

26

0

12

4

4

3

2

0

1

4

2

1

15

c

p

5

2

3

3

3

0

C

G

2

12

0

5

2

2

1

1

1

0

1

0

11

v>

5

3

5

3

2

0

1
i.

2

2

0

9

0

2

4

->

0

1

2

0

2

'j

1

3

c

2

2

3

0

0

0

1

1

0

1

8

0

7

2

1

2

0

0

1

0

i.

0

5

0

1

1

i

1

1

0

1

1

0

0

3

0

1

3

0

1

0

0

2

0

c

0

4

c

5

2

1

0

3

2

0

c

0

0

5

0

0

1

0

1

1

0

0

0

0

0

4

0

1

4

1

0

1

0

0

3

1

0

1

0

2

4

0

0

0

G

0

0

c

0

0

G

3

C

0

2

1

0

1

1

0

0

2

c

2

Q

u

1

0

1

0

c

0

0

1

0

0

G

c

0

0

1

0

1

c

3

n

^

3

1

1

0

0

1

o

0

0

1

c

1

0

2

0

2

0

0

0

0

0

1

0

1

0

c

0

]_

0

0

0

c

0

2

c

I

r
o

1

1

■j

1

0

0

0

G

2

0

1

1

0

1

0

0

0

1

0

0

2

c

0

1

0

0

1

2

1

0

c

0

1

0

0

0

0

2

0

0

0

0

0

0

2

0

1

I

0

1

0

I

c

0

'J

0

2

0

2

2

0

0

0

0

0

0

1

0

1

0

1

0

1

0

0

0

0

0

■J

0

0

1

0

0

0

0

0

0

0

0

0

1

0

13

8

7

4

2

3

3

0

0

1

7

2

152

IOC

76

50

47

35

34

29

21

18

309

118
APPENDIX I (Continued)

CHIiNiCHIi-LAS
N^ Is' HIST 0 1 2 3 4 5 6 7 8 910 11

0

1

0

0

0

1

0

0

0

0

0

0

0

0

1

4 92

n
^^^

9B

85

54

51

32

32

19

27

17

15

62

2

212

1

50

44

23

14

12

12

6

9

8

a

25

3

120

0

2 9

14

12

14

7

8

11

4

2

7

12

4

no

0

17

9

11

4

7

6

2

5

1

6

12

5

bO

i.

2 3

12

12

4

4

4

0

1

3

2

14

6

61

1

10

5

7

4

7

6

3

7

6

0

5

7

48

Q

6

4

2

6

3

4

2

4

4

3

10

3

39

c

5

3

4

3

3

1

1

2

4

2

11

9

41

0

5

6

4

5

2

3

5

0

2

9

10

25

0

4

3

3

1

1

2

3

2

Q

2

4

11

23

c

3

2

3

Q

0

2

1

2

1

X

2

2

12

38

0

10

4

5

4

0

3

0

3

0

4

5

13

17

0

3

2

1

1

0

0

1

1

1

i

1

6

14

19

0

6

5

2

0

0

0

1

0

0

0

5

15

11

G

2

3

0

1

2

1

0

0

1

0

1

16

11

0

2

0

2

1

0

2

1

1

G

0

2

17

13

0

i.

2

1

2

1

1

1

1

1

1

1

18

7

0

0

P.

2

2

]_

0

0

1

0

0

1

19

10

c

2

C

2

1

0

0

3

0

2

0

2 0

8

c

0

1

4

0

0

1

0

0

0

1

1

21

7

0

3

0

1

1

0

0

0

0

1

0

1

2 2

8

0

0

2

i

1

0

2

1

0

1

0

23

7

Q

2

C

2

1

1

1

0

0

0

0

24

3

C

0

c

0

1

0

C

0

0

0

0

2

2 '3

7

0

»

L

1

C

2

0

1

0

0

0

0

2

26

4

0

0

0

1

0

1

1

0

0

c

0

1

27

9

Q

0

2

0

0

2

0

0

1

1

0

3

28

2

c

0

0

2

0

0

0

0

0

0

0

29

5

0

G

2

0

0

0

0

0

0

1

1

1

2

0

0

0

1

0

0

0

0

0

0

0

1

31

43

0

7

6

2

4

3

0

1

2

6

0

12

CRCSS HIST 3 294 217 163 129 90 93 59 76 59 5^ 211

APPENDIX J

Means and Standard Deviations of Day and Night
Minutes of Sleep for Five Rodent Species

Day

Mouse

457
(SD = 36.7)

Hamster

544
(SD = 23.7)

Rat

470
(SD = 43.05)

Squi rre 1

398
(SD = 81.3)

Chinchi 1 la

373
(SD = 36.47

Night

323
(SD = 34.8)

323
(SD = 33.5)

326
(SD = 38.86)

412
(SD = 52. 15)

303
(SD = if6.26)

119

APPENDIX K

Frequency of Sleep Periods Per Day

MO

HA

RA

Day 1

Day 2

1

129

100

2

1 12

103

3

98

101

k

92

114

5

85

98

6

81

115

1

72

114

2

80

121

3

62

52

k

93

88

5

61

90

6

57

77

1

118

106

2

156

122

3

129

108

4

129

122

5

126

107

6

126

132

120

121

APPENDIX K (Continued

sa

CH

Day 1

Day 2

1

72

55

2

76

67

3

81

98

k

56

44

5

64

74

6

Gk

84

1

101

118

2

97

105

3

]li+

103

k

131

117

5

150

149

6

124

146

APPENDIX L

List of Individual Comparison Tests for Analyses of
Different Sleep Parameters --(P <.05)

1. Comparison of Total Sleep Time (Both Days Simultaneously)

Compar i son
MO - HA
MO - RA
MO - SQ.
MO - CH
HA - RA

HA - sa

HA - CH

RA - sa

RA - CH

Sa - CH

(HA ^ SQ.) - (RA + MO + CH)
2 3

Difference Between Means

- 86.8 NS

- 16.1 NS
+44.6 NS
+94.7 NS
+70.7 NS
+42.2 NS
-181.5 "
+28.5 NS
+110.8 NS
+139.3 NS
+ 91.91 "

Comparison of Total Sleep Time for Day 2 Alone

MO - HA -139 NS

MO - RA -103 NS

MO - SQ -160 NS

MO - CH - 13 NS

HA - RA + 64 NS

HA - SQ + 9 NS

122

123

APPENDIX L (Continued)

Compa r i son Difference Between Means

HA - CH +126 NS

RA - SQ - 27 NS

RA - CH + 90 NS

Sa - CH +1 17 NS

Ml. Comparisons of Percentages of PS

MO - HA - 13.81 v-

MO - RA - 9-95 "

MO - sa - 15.09 "

MO - CH - 2.78 NS

HA - RA + 3.86 NS

HA - SQ + 1.28 NS

HA - CH + 11 . 03 "

RA - Sa - 5. 1^ NS

RA - CH + 7. 17 NS

SQ - CH + 12.31 "

IV. Comparisons of Total Sleep Period (Sleep Episode) Lengths

MO - HA - 3-56 NS

MO - RA + 1.33 NS

MO - SQ - 4.37 NS

MO - CH + 2. 10 NS

HA - RA +4.89 "

124

APPENDIX L (Continued)

Conpa ri son
HA - Sd
HA - CH

RA - sa

RA - CH
SQ - CH

Difference Between Means
+ 0.81 NS
+ 5.66 -.'.
- 5.70 ---
+ 0.77 NS
+ 6.47 '■<

Comparisons of Slow Wave Sleep Period Lengths

MO - HA + 0.2 1 NS

MO - RA + 1 . 45 -"

MO - sa

MO - CH

HA - RA

HA - SQ.

HA - CH

RA - sa

RA - CH

Sd - CH

.10 NS

+ 1.29 ■'■

+ 1.34 -•-

.11 NS

+ 1.08 V-

- 1.45 --■--

.26 NS

+ 1.19 "

Comparisons of Paradoxical Sleep Period Lengths.

MO - HA - 1.53 "

MO - RA - 0. 1 1 NS

MO - Sa - 1.33 "

MO - CH + 0. 14 NS

HA - RA +1.42 ■■--

125

APPENDIX L (Continued)
Compar i son

HA - sa

HA - CH

RA - sa

RA - CH
SQ - CH

Difference Between Means
+ 0.20 NS
+ 1 . 68 V.
- 1.22 '-'r
+ 0.25 NS
+ 1.48 vv

VII. Comparisons of "sleep cycle" lengths

MO - sa + 0.2 NS

MO - HA + 0.7 NS

MO - RA +3-8 --'.-

MO - CH +5.8 -'•-

sa - HA + 0.5 NS

Sa - RA +8.5 "

sa - CH +5.6 -'-

HA - RA + 3 . 1 NS

HA - CH +5.1 "

RA - CH + 2.0 NS

APPENDIX M
SUMMED FREQUENCIES OF STAGE CHANGES
FOR ALL SPECIES

These tables show the frequency of change from each EEG
stage to the one which followed. The values are summed across
animals for both days for each species. Diagonal values from
upper left to lower right are the total minutes spent in each
stage. These data indicate that the animals studied slept in
no fixed pattern of stage sequence as described in the text.
The non-zero frequencies of changes from stage 1 to stage 3
are due to short periods of arousal within periods of PS or
else PS episodes preceded by less than 30 seconds of SWS .

126

127
APPENDIX M (Continued)

MICE

FROM
T(] 1 2

6705 1082 137

121C 6 98 2 253

8 382 508

Stage 1 - Awake

Stage 2 - SWS

Stage 3 - PS

Stage 4 - Drowsy-alert pattern (squirrels)

123
APPENDIX M (Continued)

RATS DAYS i AND 2
TO 1 2

6256 1079 398

1453 5853 362

23 7 36 1106

0

Stage 1 - Awake

Stage 2 - SWS

Stage 3 - PS

Stage 4 - Drowsy-alert pattern (squirrels)

129
APPENDIX M (Continued)

.<ATS DAYS 3 AND 4
FROM

TQ 1 2 3

1 4808 648 37C

2 LOOT 3056 187

3 11 546 868

4 0 0 0

Stage 1 - Awake

Stage 2 - SWS

Stage 3 - PS

Stage 4 - Drowsy-alert pattern (squirrels)

130
APPENDIX M (Continued)

HAMSTERS

FROM
TO 1 2 3

1 5926 826 137

2 952 6505 501

3 13 626 1788

4 0 0 0

Stage 1 - Awake

Stage 2 - SWS

Stage 3 - PS

Stage 4 - Drowsy-alert pattern (squirrels)

131
APPENDIX M (Continued)

GRGL.NO SQUIRRELS

FROM
TO 1 2 3 4

1 4325 541 150 480

2 440 6159 479 407
.3 14 631 1741 6
4 716 153 23 1009

Stage 1 - Awake

Stage 2 - SWS

Stage 3 - PS

Stage 4 - Drowsy-alert pattern (squirrels)

132

APPENDIX M (Continued)

CHI'ICHILLAS

FRUM
TO 1 2

7619 1173 274

1396 5572 237

53 423 552

Stage 1 - Awake

Stage 2 - SWS

Stage 3 - PS

Stage 4 - Drowsy-alert pattern (squirrels)

133
APPENDIX M (Continued)

HAMSTERS

FRCM
TO 1 2

5926 826 137

952 6505 501

13 626 1738

Stage 1 - Awake

Stage 2 - SWS

Stage 3 - PS

Stage 4 - Drox'jsy-alert pattern (squirrels)

134
APPENDIX M (Continued)

CHINCHILLAS

FRUM
TO 1 2

7619 1173 274

1396 5572 207

53 423 552

Stage 1 - Awake

Stage 2 - SWS

Stage 3 - PS

Stage 4 - Drowsy-alert pattern (squirrels)

APPENDIX N
RELATIONSHIP BET/7EEN LSNGITI OF SWS
AND SUCCEEDING PS PERIODS

In the tables of this appendix, the frequencies of SWS
periods are plotted against length. In columns 1 to 11, fre-
quencies of PS periods which follov;ed SWS periods of given
lengths are presented. The zero column indicates the number
of times that the sleep period was followed by an awake period,
while col'j:nns 1 through 10 give frequencies of PS periods of
1 through 10 minutes duration. Column 11 gives frequencies of
periods 11 or more minutes long.

In general, as sleep periods increased in length, the
proportion of them which were followed by PS periods increased
also. This indicates that the occurrence of a PS episode is
dependent upon the length of the SWS period which precedes it.

135

APPENDIX N (Continued)

M I C I

136

KIN HIST

10 11

J

8

Q

4

4

c

0

0

0

0

0

G

0

I

303

285

7

3

3

0

0

<J

0

0

0

0

0

2

208

181

3

9

9

1

0

0

0

0

0

0

0

3

163

130

10

11

5

^

0

3

1

1

c

0

0

4

135

9S

10

11

12

z.

1

■J

1

X

n

0

0

0

5

109

77

9

12

10

1

0

0

0

0

0

6

36

61

11

10

2

0

1

1

0

0

0

r-.

0

7

79

55

5

y

7

3

1

0

0

r\

0

0

d

61

31

10

7

9

3

1

0

0

0

0

0

0

9

^9

21

6

7

3

3

0

1

0

0

0

0

0

iO

4B

26

6

9

4

2

0

0

1

0

0

0

0

11

39

19

9

4

5

1

1

0

0

0

0

0

12

25

17

I

1

2

2

1

0

1

0

0

0

0

1.3

26

16

■J

3

3

1

0

0

0

0

0

0

0

14

24

11

3

4

5

1

0

0

0

0

0

0

15

13

6

1

2

3

1

0

0

0

0

0

0

0

16

19

11

3

1

L

2

1

0

0

0

0

0

0

I 7

12

1

2

3

1

c

0

0

G

0

0

13

9

3

1

3

1

1

n

0

o

w

0

0

0

0

19

9

3

1

2

2

]_

n

r

0

c

0

0

0

20

1

C

1

Q

0

0

0

0

0

0

G

0

0

21

11

2

1

2

3

1

2

0

0

G

0

0

0

22

7

2

2

1

1

1

0

c

0

0

0

0

G

23

3

2

1

0

0

G

0

0

c

0

c

0

24

4

3

0

1

0

0

0

0

0

G

0

25

2

2

0

0

0

0

0

0

0

0

0

0

0

26

4

3

0

1

0

0

0

0

0

0

0

0

0

27

5

2

1

0

1

1

0

0

0

0

0

c

0

28

3

1

1

0

0

1

0

0

0

0

c

0

0

29

4

3

1

0

0

0

0

0

0

0

0

n

^

0

30

0

0

0

c

0

0

0

0

0

0

0

0

0

^l

5

2

0

2

1

c

0

G

0

0

0

0

CROSS HIST 10S4 119 12C 100 32

137

APPENDIX N (Continued)

RAfS DAYS 1 AN3 2

' I \

HIST

p

1

l~

'/

4

5

6

7

0

9

10

11

J

24

o

9

7

2

4

1

1

0

0

0

0

0

i

431

360

23

21

11

11

3

<J

1

0

1

0

2

2 96

2iS

26

25

13

8

4

0

1

0

0

'0

0

3

2 24

131

27

29

22

7

6

1

1

0

0

0

0

'+

l^JO

85

30

40

14

13

4

3

1

0

c

0

S

149

74

2 5

22

16

7

2

0

2

1

0

0

0

S

144

6 3

19

25

19

9

5

4

w

0

0

0

0

7

102

47

13

22

13

2

2

3

0

0

0

c

0

B

f3 2

33

11

2C

7

5

4

2

0

0

0

0

■9

6 5

23

10

14

8

8

1

1.

1

0

0

0

0

10

36

12

6

3

8

1

1
i.

0

Q

0

0

0

0

11

3 3

i 3

3

3

6

3

3

2

0

0

0

0

0

12

30

9

3

3

5

3

2

'J

0

0

0

0

0

13

12

1

3

2

3

0

1

2

0

c

0

0

I'-i

8

3

1

2

1

0

1

0

Pi

0

0

0

1 'j

7

5

0

1

0

1

0

0

0

0

0

0

0

16

3

1^

1

C

1

0

0

0

c

0

c

0

0

17

3

2

0

1

0

0

0

0

G

0

0

0

0

18

1

1

0

0

0

0

0

0

0

0

Q

G

0

V)

2

2

0

c

c

0

0

0

0

0

0

G

2 0

Q

0

0

0

0

0

0

0

0

0

0

,--»
w

2 1

1

0

0

c

0

1

0

0

0

0

0

0

0

22

1

Q

0

1

0

0

0

0

0

0

0

0

0

2^

G

0

0

0

0

0

0

G

0

0

0

0

0

24

0

0

0

r

0

0

0

0

0

c

0

0

2S

0

0

c

0

0

0

0

c

0

0

0

0

2 6

0

0

0

0

0

0

0

0

Q

0

0

0

27

0

n

0

0

0

0

0

l'^,

0

G

0

23

0

0

r

0

0

0

c

0

0

0

0

0

2 9

0

0

0

Q

0

0

0

G

0

0

0

0

iD

0

c

c

0

0

0

0

0

0

0

0

0

31

0

0

0

c

c

0

C)

0

0

0

0

0

0

CROSS MIST 1034 210 251 149 83 40 24

138

APPENDIX N (Continued)

i^ATS DAYS 3 AND ^
KI'Ni HIST 0 1 2 3 4 5 6 7 8 9 IC 11

0

11

r

2

2

6

,.->

i

0

0

0

/-■,

0

0

1

3 13

213

32

24

19

13

13

1

1

0

r

0

G

2

221

135

22

2 5

2 0

7

9

1

2

G

.^
O

0

0

3

168

89

27

22

13

6

3

1

2

G

G

0

0

/
■+

148

72

28

12

13

8

0

3

3

0

0

G

'S

115

49

23

19

12

8

4

c

0

C

0

G

0

6

7 8

4 0

16

5

6

9

2

0

0

0

0

G

0

7

47

19

9

6

4

2

7

0

G

0

Q

0

c

8

39

14

6

7

5

3

2

2

Q

0

G

0

G

9

23

9

1

4

4

3

1

1

0

G

C

G

C

' 0

14

2

5

1

3

2

0

i

0

0

0

0

0

11

16

3

2

2

4

3

1

1

c

0

0

G

G

12

5

1

0

r>

3

1

0

0

0

r\
U

G

G

0

13

6

2

3

c

1

0

0

0

0

0

0

0

G

14

3

2

0

0

0

n

1

0

0

0

G

0

Q

15

1

1

0

o

n

0

0

0

G

0

0

G

16

0

0

0

0

C

0

c

G

0

0

0

0

17

0

0

0

c

0

0

0

0

0

G

0

0

0

13

0

0

0

n
U

0

G

0

0

0

Q

Q

0

Q

19

0

c

0

0

0

0

0

0

0

0

G

G

0

20

0

0

Q

0

c

0

0

0

0

0

0

0

G

2 1

0

0

0

c

0

0

0

c

0

r.

U

0

G

22

0

'J

0

0

0

C

0

0

G

0

0

0

0

23

0

c

0

a

0

0

n

0

0

G

c

0

G

2 4

0

c

0

0

0

0

0

0

0

0

0

0

G

23

0

0

0

0

0

G

0

G

G

0

0

0

0

26

0

0

0

Q

0

0

0

0

C

0

0

0

0

2 7

0

G

0

0

0

0

0

c

c

c

0

G

0

28

0

n

0

c

n

0

0

G

G

c

G

29

0

0

n

0

0

0

G

0

0

0

G

G

0

33

0

0

0

0

0

0

r>^

0

0

G

G

0

31

0

c

0

0

0

0

0

0

0

Q

0

G

0

CftUSS HIST 651 176 129 118 66 49 13 11

139
APPENDIX N (Continued)

HAMSTERS

I ^

HIST

0

1

2

3

4

5

6

7

8

9

10

11

3

13

0

3

2

2

0

3

1.

1

1

0

0

0

1

2 41

188

19

12

9

7

3

2

1

0

0

0

0

2

180

13-3

10

7

9

S

6

2

1

1

i

0

0

146

94

11

O

13

a

2

2

3

2

0

0

2

A

149

97

9

3

6

12

9

3

4

1

0

0

0

5

137

73

7

11

11

10

16

4

3

2

0

0

0

6

132

72

4

12

10

13

12

2

4

2

1

0

0

7

GO

3 5

3

B

9

9

4

7

2

1

L

0

1

3

95

34

4

9

15

13

9

3

2

1

0

0

9

80

28

1

6

6

13

13

8

2

2

0

0

1

10

53

17

5

6

7

7

3

4

3

1

c

-^
^

0

11

41

12

2

3

3

9

10

2

0

G

0

0

0

i;^

34

11

1

5

4

3

2

6

0

1

1

0

0

13

17

4

2

2

2

1

4

1

0

0

o

0

1

14

17

6

0

3

1

3

2

1

1

0

r^

0

'J

15

11

7

0

0

0

1

0

1

0

2

0

0

r

16

10

3

0

1

0

2

2

1

0

1

0

0

v>

17

10

5

1

1.

0

0

3

0

0

1

0

0

0

V.'

18

4

0

0

3

0

1

0

0

0

0

0

0

19

4

1

0

0

0

0

2

1

0

0

0

G

20

5

3

1

0

c

0

0

1

Q

0

0

0

0

Zl

3

1

0

1

1

0

0

0

0

0

0

0

22

2

2

0

0

0

0

0

0

c

0

0

23

1

0

0

0

0

0

0

1

0

0

0

0

2 4

1

1

0

0

0

0

0

0

0

0

0

0

0

25

0

,o

0

0

C

0

0

0

0

0

0

0

2^1

0

c

.■•^

G

0

0

0

0

0

0

0

0

r.

27

1

0

0

0

0

1

0

0

0

Q

0

0

0

2B

0

c

0

0

c

0

0

0

0

0

0

0

0

29

0

r

0

G

0

0

0

0

0

0

0

0

0

-JO

0

0

0

0

0

0

0

0

0

0

0

0

;n

1

0

0

1

0

0

r.

0

0

0

0

0

0

GSS

HIST

8 29

83

1 0 8

107

125

103

5 8

28

18

4

0

5

140
APPENDIX N (Continued)

GRQU;\D SQUIRRELS

IN

HIST

C

1

2

3

4

D

6

7

8

9

10

11

J

21

0

0

4

0

3

2

2

1

1

0

n

L

257

200

21

11

/

3

10

5

0

0

'■J

0

0

2

198

137

20

c

12

4

4

4

5

2

2

0

Q

3

159

9C

10

12

17

9

8

c

3

r,

1

0

1

4

113

62

12

9

12

6

2

5

3

0

2

0

0

5

10 1

54

3

10

7

7

5

4

4

1

1

0

0

6

75

3 0

11

5

4

3

12

7

3

0

0

0

c

7

72

27

3

9

7

12

9

J.

X

1

■-1

2

0

8

64

14

4

&

10

16

6

3

1

0

2

0

c

9

45

15

2

5

6

5

4

5

1

2

^

0

10

56

15

•^.

6

10

4

6

7

1

2

2

0

0

11

2 9

7

1

1

6

8

4

2

0

0

0

0

12

36

7

3

3

8

4

6

3

1

1

0

0

0

13

20

5

0

5

6

"I
i

0

1

0

0

1

0

1

14

19

3

1

2

3

2

4

■J

1

3

0

0

0

15

13

6

3

2

2

2

0

1

2

0

n

0

0

16

12

4

0

1

1

3

1

1

0

0

c ■

0

L7

8

2

0

1

2

1

1

J.

1

n

0

0

c

13

10

4

0

2

2

1

1

0

0

0

0

0

0

19

10

3

0

3

1

1

1

0

0

0

'^

u

1

2 0

5

G

0

G

2

1

■1

r,

1

0

0

0

0

21

2

2

n
y

0

0

0

0

0

0

0

0

0

c

2?

1

0

0

0

0

0

0

0

0

0

0

0

1

2 3

4

2

1

c

0

0

0

1

0

0

0

0

24

2

1

0

1

0

0

0

0

0

0

0

0

0

25

0

0

0

0

0

0

0

0

c

0

0

0

c

26

4

1

0

0

1

0

1

1

0

0

0

0

0

2 7

1

1

0

c

0

0

0

0

0

0

0

0

0

23

1

1

0

0

0

Q

c

0

0

n

29

0

c

0

c

0

0

0

0

0

0

0

0

30

1

0

0

0

c

0

0

1

0

0

0

0

31

6

5

0

c

0

0

0

1

0

c

0

0

CROSS HIST 698 HI 106 126 98 88 63 30 13 11

141

APPENDIX N (Continued)

CHINCHILLAS

HIST

IG li

■J

33

2 7

7

11

6

T

r-1

c ■

0

0

0

c

I

3S'3

327

2 8

16

7

2

3

0

2

0

0

0

0

2

272

221

23

13

7

3

1

p

3

o
^

1

0

0

3

210

155

2 6

14

10

1

2

1

0

0

1

0

0

4

156

109

19

16

5

4

0

2

0

0

1

0

0

5

105

7 6

13

3

3

1

3

0

1

0

0

/>,

0

6

102

6 5

16

9

5

6

1

0

c

0

0

0

7

79

48

14

10

4

1

1

1

0

0

0

0

0

a

68

51

9

5

2

0

T

0

0

0

0

0

0

9

r>6

32

11

5

4

1

]_

2

0

n

0

0

r,

10

41

2 9

2

4

J>

2

'^'

X

0

0

G

u

0

11

24

10

6

5

2

1

0

J

0

0

0

0

1.2

25

11

5

2

6

1

0

u

0

0

0

0

0

13

2 1

13

3

3

1

0

0

_»

o

n

1

0

0

14

16

6

4

3

1

2

0

0

0

n

0

0

0

15

12

5

5

1

0

0

0

1

0

C

0

0

0

lA

6

•J

1

1

o

1

0

0

c

0

0

0

n

17

7

4

1

1

0

0

1

1.

0

0

c

0

0

18

6

2

0

1

2

'J

0

0

0

0

0

0

19

2

T

r,

0

1

Q

Q

0

0

0

0

0

Q

:^o

2

c

0

1

1

0

0

'-I

0

0

0

0

0

21

6

^

0

c

1

J.

0

0

0

0

0

0

0

2 2

0

0

0

c

0

0

0

0

c

0

0

0

n

2 3

2

2

0

0

0

0

0

0

0

vJ

0

o

24

0

0

0

c

0

0

Q

0

0

0

0

0

0

2"?

0

0

0

o

v

0

0

\1

p.

0

0

0

0

26

0

c

0

c

c

0

0

0

0

0

0

0

0

27

0

0

0

0

0

0

0

0

0

0

0

0

0

2^3

0

c

0

0

0

0

0

c

n

0

0

0

0

29

0

r;

0

I-'

0

0

0

0

0

0

0

0

0

3 0

0

0

0

c

0

0

0

0

0

0

0

0

0

31

1

1

0

0

0

0

0

0

0

0

0

0

0

CRoSS HIST 1176 213 124 76 34 14 IC

4

REFERENCES

Adey, W. R., Kado, R. T. and Rhodes, J. M. Sleep: cortical and

subcortical recordings in the chimpanzee. Science , 1963,
JM> 932-933.

Affanni, J. and Vaccarezza, 0. Observations on the sleep of the
marsupial Didelphis azarae ("comadreja" or zarigueya).
Rev. Soc. Argent. Biol . , 1964, kO_, 2-8.

Agnew, H. and Webb. W. The sleep patterns of long and short sleepers.
Paper presented at the Assoc. Psychophysiol . Study Sleep,
April, 1967, Los Angeles.

Aserinsky, E. and Kleitman, N. Two types of ocular motility occurring
during sleep. J. Appl . Physiol . , 1955, 8, 1-10.

Batini, C., Radulovacki, M., Kado, R. and Adey W. Effects of inter-
hemispheric transection on the EEG patterns in sleep and
wakefulness in monkeys. E lectroenceph . Clin. Neurophys io 1 .
1967, 21, 101-1 11.

Berger, R. J. and Meier, G. W. The effects of selective deprivation

of states of sleep in the developing monkey. Psychophys io I . ,
1966, 2, 354-371.

Bert, J., Collomb, H. and Martino, A. De I e lectroencepha logramme

du sommeil chez les Primates. Compt. Rend. Acad. Sci. Paris,
1966, 262, 1006-1009.

Bonamini, F., DeCarolis, V., Pastorino, P. and Rossi, G. Light sleep
and deep sleep in the dog (e lectroencepha lograph i c , electro-
myographic, and pressor study). Boll. Soc. I ta 1 . Biol. Sper.,
1962, 38_, 1298-1299. (NIH translation no. 3-26-67)

C louds ley-Thompson , J. Rhythmic Activity in Animal Physiology and Behavior.
New York: Academic Press, I96I.

Dement, W. C. The occurrence of low voltage, fast electroencephalogram

patterns during behavioral sleep in the cat. E lectroencepha log .
Clin. Neurophysiol. , 1958, JO, 291-296.

Dement, W. and Kleitman, N. Cyclic variations in EEG during sleep and
their relation to eye movements, body inotility and dreaming.
Electroencephalog. Clin. Neurophysiol.. 1957, 9., 673-690. "(a)

142

143

Dement, W. and Kleitman, N. The relation of eye movements during
sleep to dream activity: an objective method for the
study of dreaming. J. Exp. Psychol . , 1957, 53_, 339-346. (b)

Dillon, R. and Webb, W. Threshold of arousal from "activated" sleep
in the rat. J. Comp. Physiol. Psychol., 1965, £9, kkG-hky .

Duncan, R., Henry, P., Karadzic, V . , Mitchell, G., Pivik, T. , Cohen, H.
and Dement, W. Baseline sleep and REM deprivation in tne
rat. Paper presented at the Assoc. Psychophys io 1 . Study
S leep, April, 1967, Los Angeles.

Ephron, H. and Carrington, P. Rapid eye movement sleep and cortical
homeostasis. Psychol . Rev. , 1965, 73, 500-526.

Ephron, H. and Carrington, M. Ego-functioning in rapid eye movement
sleep: implications for dream theory. Science and Psycho-
analysis, 1967, li, 75-102.

Faure, J. Role du mesencephale dans lo phase "paradoxale" du sommei 1
Chez le Lapin. J. de Physiol . , I962, £4, 333-334-

Faure, J., Bensch, C. and Vincent, D. An sujet des mecanismes

responsables du compartement a 1 facto-bucco-ano-gen i to-sexue 1
du Lapin: ses rapports aves le sommei 1, Compt. Rend. Soc.
Biol . , 1962, j[56, 629-631. (a)

Faure, J., Bensch, C. and Vincent, D. Role d un systeme mesencepha lo-
limbique dans la "phase paradoxale" du sommei I chez le Lapin.
Compt. Rend. Soc. Biol. , 1962, _156, 70-73- (b)

Faure, J., Vincent, D., LeNovene, J. and Geissmann, P. Sommei 1 lent
et stade paradoxal chez le Lapin des deux sexes: role du
mi 11 leu. Compt. Rend. Soc. Biol . , I963, 157, 799-804.

Hartmann, E. The D-state: a review and discussion of studies on the
physiological state concomitant with dreaming. New Enq. J.
Med., 1965, 221, 30-35.

Hartmann, E. Sleep and dreaming in the elephant. Paper presented at

the Assoc. Psychophysio 1 . Study Sleep, April, 1967, Los Angeles.

Hermann, H., Jouvet, M., and Klein, M. Etude polygraph ique du sommei 1

chez la Tortue. Compt. Rend. Soc. Biol . , 1964, 258, 2175-2178.

Hobson , J. E lectroc raph ! c correlates of behavior in tne frog with

special reference to sleep. E iectroenceph . Clin. Neuropnys i o I . ,
1967, 22, 1 13-121.

Jouvet, D. and Valatx, J. Etude po lygraph ique du sommei 1 chez 1 agneau.
Compt. Rend. Soc. olol . , 1962, 156, I4ll-l4l4.

Jouvet, D., Valatx, J. L. and Jouvet, M. Etude po lygraph i que du
sommei 1 du chaton. Compt. Rend. Soc. Biol., 1961, 155 .
1660-1664.

Jouvet, D., Monnier, M. and Astic, L. Study of sleep in the adult and
newborn guinea pig. Compt. Rend. Soc. Biol., 1966, 160 ,
1453-1^57. (NJH translation).

Jouvet, M. Recherches sur les structures nerveuses et les

mechanismes responsables des differentes phases du sommei 1
physiologique. Arch. I ta 1 . Biol., 1962, JOO, 125-206.

Jouvet, M. Neurophysiology of the states of sleep. Phys iol . Rev. ,
1967, iiZ, 117-177. (a)

Jouvet, M. The states of sleep. Sci . Amer . , 1967, 2 16, 62-72. (b)

Khazan, N. and S> yer, C. H. "Rebound" recovery from deprivation of

paradoxical sleep in the rabbit. Proc. Soc. Exp. Biol. Med.,
1963, M, 536-539.

Klein, M. , Michel, F. and Jouvet, M. Polygraphic study of sleep in birds.
Compt. Rend. Soc. Biol . , 1964, JSS, 99-103.

Kleitman, N. Sleep and Wakefulness. Chicago, University of Chicago
Press, 1963.

Konig, F. and Klippel, R. The Rat Brain, a Stereotaxic Atlas.
Baltimore, Williams and Wilkins, I963.

Lommis, A. L. , Harvey, E. N. and Hobart, G. Cerebral states during

sleep as studied by human brain potentials. J . Exp. Psycho 1 . ,
1937, 2_L> 127-144.

Meier, G. W. and Berger, R. J. Development of sleep and wakefulness
patterns in the infant Rhesus monkey. Exp. Neuro 1 . , 1965,
12, 257-277.

Oswald, I. Sleeping and Waking. New York, Elsevier, 1962.

Redding, R., Prynn, B. and Colwell, R. The phenomenon of alternate
sleep and wakefulness in the young dog. J. Am. Vet. Med.
Assoc. , 1964, _I44, 605-606.

Reite, M. L. , Rhodes, J. M. , Kavan, E. and Adey, W. R. Normal sleep

patterns .n Macaque monkey. Arch. Neurol . , 1965, _L2., 133-144.

145

Roldan, E. and Weiss, T. The cycle of sleep in the rat (preliminary
report). Bol . Inst. Estud. Med. Biol . . 1962, 20, 155-164.

Roldan, E. and Weiss, T. Excitability changes of the reticular
ascendent activating system during sleep cycle in the
rabbit. Arch. Int. Physiol . Biochem. , I963, 7±, 518-527.

Roldan, E. and Weiss, T. Comparative study of sleep cycles in rodents.
Experientia, 196-^-, 20, 280-281.

Roldan, E., Weiss, T. and Fifkova, E. Excitability changes during

the sleep cycle of the rat. E lectroenceph . Clin. Neurophysi o 1 .
1963, ii, 775-785.

Ruckebusch, Y. Activite cortlcale au cours du sommei 1 chez la

chevre. Compt. Rend. Soc. Biol . , I962, _I56,, 867-87O. (a)

Ruckebusch, Y. Evolution post-natale des sommei 1 chez les

ruminants. Compt. Rend. Soc. Biol . , 1962, J^, 1869-1873- (b)

Ruckebusch, Y. Etude polygraphique et compartementa le de 1 evolution
post-natale du sommei 1 phys iologique chez 1 agneau. Archi v.
I tal. de Biol ■ , I963 , JOl, 111-132. (a)

Ruckebusch, Y. EEG and behavioral studies on the wakef ulness-s leep

alternations in the donkey. Compt. Rend. Soc. Biol., 1963,
157, 8k0-8kk. (b) (NIH translation No. 3-19-67)

Ruckebusch, Y. and Bost. J. Activite corticale an cours de la somnalence
et de la rumination chez la chevre. J . Physi o 1 . (Pari s) ,
1962, 5ii, 409-'4l0.

Snyder, F. In quest of dreaming. In H. A. Witkin and H. B. Lewis (Eds.)
Experimental Studies of Dreaming. New York. Random House,
1967, P. 3-75.

Soulairac, A., Gottesman, C. and Thangapregassam, J. Analyse du sommei 1
profond chez le rat. Compt. Rend. Acad. Sci., 1963, 257 ,
3040-3041.

Soulairac, A., Gottesman, C. and Thangapregassam, M. Etude electro-
physio logl que des differentes phases de sommei 1 chez le
rat. Arch. I ta 1 . de Biol. , I965, 103., 469-482.

Spector, W. Handbook of Biological Data. London, Saunders, 1956.

Sterman, M. Paper presented at the Assoc. Psychophys io1 . Study
S leep. April, 1967, Los Angeles.

146

Sterman, M. B., Knauss, T., Lehmann, D. and Clemente, C. D. Circadian
sleep and waking patterns in the laboratory cat. E lectro-
enceph. Clin. Neurophys i o 1 . , 1965, 19, 509-517-

Swisher, J. E. Manifestations of "activated" sleep in the rat.
Science, 1962, _138, 1110.

Tauber, E., Roffwarg, H. and Weitzman, E. Eye movements and

electroencephalogram activity during sleep in diurnal
lizards. Nature, I966, 212, 1612-1613-

Tauber, E., Rojas-Rami rez , J., and Hernandez-Peon, R. REM-sleep in the
lizard. Ctenosaura Pectinata. Paper presented at the
Assoc. Psychophys iol ■ Study Sleep. April, 1967, Los Angeles.

Ursin, R. Two stages of slow wave sleep in the cat and their relation
to REM-sleep. Paper presented at the Assoc. Psychophysiol .
Study S leep, April, 1967, Los Angeles.

Valatx, J. L. , Jouvet, D. and Jouvet, M. Evolution e lectroencepha lo-
graphique des differents etats de sommei 1 chez le chaton.
Electroenceph. Clin. Neurophys io 1 . , 1964, _1_7. 2 18-233-

Van Twyver, H. D., Levitt, R. A. and Dunn, R. S. The effects of

high intensity white noise on the sleep pattern of the rat.
Psychon. Sci . . I966, 6, 355-356.

Van Twyver, H. and Webb, W. Modification of a sleep pattern in the
rat. Paper presented at the Assoc. Psychophysiol. Study
S leep, April, I967, Los Angeles.

Walker, E. Mammals of the V/orld. Baltimore, Johns Hopkins, 1964.

Weiss, T. and Fifkova, E. Sleep cycles in mice. Physiol. Bohemoslov.
1964, _I3., 242-245.

Weiss, T. and Roldan, E. Comparative study of sleep cycles in
rodents. Experientia, 1964, _20, 280-281.

Weitzman, E. D. A note on the EEC and eye movements during behavioral
sleep in monkeys. E lectroencepha log . Clin. Neurophysi o 1 . ,
1961, il, 790-794.

Williams, R. L. , Agnew, H. W. and Webb, W. B. Sleep patterns in young
adults: an EEC study. Electroenceph. Clin. Neurophys i o 1 .
1964, 17, 376-381.

Winer, B. Statistical Principles in Experimental Design. New York,
McGraw-Hill, 1962.

BIOGRAPHICAL SKETCH

Henry Bauman Van Twyver was born in Quincy, Massachu-
setts on April 8, 1939. He graduated from S, W. Kearny High
School in San Diego, California in June, 1956, During his
student years he attended the California Maritime Academy for
two years and San Diego City College for one year. He re-
ceived the B- A. degree from San Diego State in 1963, and the
M, A, degree at the University of Florida in 1965. From August,
1964, until August, 1965, he held the position of research
assistant. From that time until the end of his training,
he was a Center for Neurobiological Sciences trainee. Mr.
Van Twyver is married and has two children, Kathleen and
Robert.

147

juttJ.3 u

iiisnTtczLon \7zz p"rc?G-cd under the direction o;

h£.a been approved by all meabcrs of tha. con-^itcee. 'Iz we:;
subu.itiLec '_o •eiiiS Deciv. of ihe College of .-.rto and Sciences and
to ti'.c Graduate Council, and v;aa approved aj pareial ful-
filliueni; of tne rcquirenicnts for the degree of Doctor of
PLiloiophy .

December 19, 1967

C^«— **-— «,>t /(^ ^-♦v-^

'> V-W^^v.^^ <JJ

4ia tJ'w. ▲e fc>v^ s

Supervisory Cosi^ittee:

K/

IA:?U^-

iyCt^Ais 0«.U.UUv<.wC OCwv^Oj

/^;

(

OlLiciA.^4^iJL*l

w

^^^s>:>o/^ i.N^v:^,^

U. c, cL'A ^ e-^-L ^^----;-