Title: Reproductive behavior and early development in cotton mice, Peromyscus gossypinus, and white-footed mice, Peromyscus leucopus /
CITATION PDF VIEWER THUMBNAILS PAGE IMAGE ZOOMABLE
Full Citation
STANDARD VIEW MARC VIEW
Permanent Link: http://ufdc.ufl.edu/UF00098668/00001
 Material Information
Title: Reproductive behavior and early development in cotton mice, Peromyscus gossypinus, and white-footed mice, Peromyscus leucopus /
Physical Description: xiv, 183 leaves : ill. ; 28 cm.
Language: English
Creator: Lovecky, Deirdre Vally, 1945-
Publisher: Deirdre Vally Lovecky
Place of Publication: Gainesville, Fla.
Publication Date: 1976
Copyright Date: 1976
 Subjects
Subject: Mice -- Behavior   ( lcsh )
Mice   ( lcsh )
Psychology thesis Ph. D   ( lcsh )
Dissertations, Academic -- Psychology -- UF   ( lcsh )
Genre: bibliography   ( marcgt )
non-fiction   ( marcgt )
 Notes
Thesis: Thesis--University of Florida.
Bibliography: Bibliography: leaves 163-170.
General Note: Typescript.
General Note: Vita.
Statement of Responsibility: by Deirdre Vally Lovecky.
 Record Information
Bibliographic ID: UF00098668
Volume ID: VID00001
Source Institution: University of Florida
Holding Location: University of Florida
Rights Management: All rights reserved by the source institution and holding location.
Resource Identifier: alephbibnum - 000179298
oclc - 03162622
notis - AAU5814

Downloads

This item has the following downloads:

PDF ( 7 MBs ) ( PDF )


Full Text














REPRODUCTIVE BEHAVIOR AND EARLY DEVELOPMENT IN COTTON MICE,
Peromyscus gossypinus, AND WHITE-FOOTED MICE,
Peromyscus leucopus














By

DEIRDRE VALLEY LOVECKY


A DISSERTATION PRESENTED TO THE GRADUATE COUNCIL OF
THE UNIVERSITY OF FLORIDA
IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE
DEGREE OF DOCTOR OF PHILOSOPHY






UNIVERSITY OF FLORIDA


1976
































To Harold Musiker
with high regard,
deep respect,
and great fondness.
















ACKNOWLEDGEMENTS

I would like to thank my chairman, Donald A. Dewsbury, for

all his efforts, persuasions, comments, encouragements, and especially

his patience in helping me to conceive, organize, and write this

dissertation. His helpfulness was/is greatly appreciated. I would

like to thank the members of my committee, Louis Cohen, Paul Satz,

Carol Van Hartesveldt, Charles Vierck, Jr., and Wilse B. Webb for

their support, encouragement, and for their efforts in helping me

with the writing of this dissertation.

In addition, I would like to thank Antonio Puente for his

help in the collection of the maternal behavior and early development

data, and also, Harry Davis, Daniel Estep, Gary Gray, and David

Lanier for their technical advice and for their help in collecting

the mice.

I would also like to gratefully acknowledge the statistical

advice received from Morton Elfinbein, of the Department of

Psychology, and William Verenese, of the Computer Center of

Southeastern Massachusetts University, Dartmouth, Massachusetts.

















TABLE OF CONTENTS


Page

ACKNOWLEDGEMENTS ... . . . . . . .... iii

LIST OF TABLES . . . . . . . . ... . . viii

LIST OF FIGURES . . . . . . . ... .. . xi

ABSTRACT. .. . . . . . . .. ...... xii

CHAPTER I: INTRODUCTION . . . . . . . 1
Collection of Normative Data . . . 1
Intraspecific Comparisons . . . . 3
Interspecific Comparisons . . . . 4
Species Information . . . . . . 6

CHAPTER II: GENERAL METHODOLOGY . . . . . .. 8
Subjects . . . . . . . . 8
Maintenance . . . . . . . . 9

CHAPTER III: COPULATORY BEHAVIOR . . . . . .. 11

Introduction . . . . . . . 11
Patterns of Copulatory Behavior . 11
Quantitative Analyses of
Copulatory Patterns . . . 12
Behaviors Accompanying Copulation 13

Methods . . . . . . . .. 15
Subjects . . . . . . .. 15
Procedure . . . . . . .. 15
Measures . . . . . . ... 17
Behavioral Categorizations . . .. 18

Results . . . . . . . ... 21
Patterns of Copulatory Behavior . 21
Quantitative Analyses of
Copulatory Patterns . . .. 22
Behaviors Accompanying Copulation . 33

Discussion . . . . . . . . 40
Patterns of Copulatory Behavior . 40
Quantitative Analyses of
Copulatory Patterns . . .. 41
Behaviors Accompanying Copulation . 47













CHAPTER IV: MATERNAL BEHAVIOR . . . . . . . .

Introduction . . . . . . . . .
Descriptive Study of Maternal Behaviors .
Defense of the Young . . . . . .
Retrieval Responses . . . . . .

Methods . . . . . . . . . .
Subjects . . . . . . . . .
Procedure . . . . . . . . .
Descriptive study of maternal behaviors .
Defense of the young . . . . .
Retrieval responses . . . . . .

Results . . . . . . . . . .
Survival . . . . . . . . .
Litter survival . . . . . . .
Number of pups per litter . . . .
Descriptive Study of Maternal Behaviors .
Nursing pups . . . . . . .
Grooming pups . . . . . . .
Retrieval of pups . . . . . .
Nest building . . . . . . .
Maternal and nonmaternal behaviors . .
Intra- and interspecific comparisons .
Paternal behavior . . . . . .
Defense of the Young . . . . . .
Retrieval Responses . . . . . .
Retrieval latency . . . . . .
Per cent of pups retrieved . . . .
Total retrieval time . . . . .

Discussion . . . . . . . . .
Survival . . . . . . . .
Descriptive Study of Maternal Behaviors .
Defense of the Young . . . . . .
Retrieval Responses . . . . . .

CHAPTER V: SOMATIC AND BEHAVIORAL DEVELOPMENT . . .

Introduction . . . . . . . . .
Observations of Somatic and
Behavioral Development . . . .
Nipple Clinging . . . . . . .
Home Cage Observations . . . . .


Page

S 50

S 50
S 51
S 52
S 53

S 55
S 55
S 55
S 56
S 57
S 58

S 60
S 60
S 60
S 65
S 67
S 67
S 71
S 75
S 76
S 77
S 80
S 80
82
S 85
S 85
S 87
S 89

S 92
S 92
95
S 98
. 101

. 106

. 106

. 107
. 109
. 110












Methods . . . . . . . .
Subjects . . . . . . .
Apparatus . . . . . . .
Procedure . . . . . . .
Observations and tests of somatic


and behavioral development . .
Nipple clinging . . . .
Home cage observations . . .


Results . . . . . . .
Observations of Somatic and
Behavioral Development
Nipple Clinging . . . .
Home Cage Observations . .


Page

. . 112
. . 112
. . 112
. . 113

. . 114
. . 116
. . 117

S. 120


Discussion . . . . . . . . ... 144
Observations of Somatic and
Behavioral Development . . . ... 144
Nipple Clinging . . . . . . ... 151
Home Cage Observations . . . . .. 153

CHAPTER VI: DISCUSSION . . . . . . . . ... 156

The Study of Wild Species in the Laboratory . 156
The Adaptive Significance of Behavior . . .. 160

LIST OF REFERENCES . . . . . . . . ... . . 163

APPENDICES . . . . . . . . ... . . . . 171

Appendix A: Mean Copulatory Behavior Scores for
Individual Pairs of Peromyscus gossypinus
(Gainesville Population) . . . .. 172
Appendix B: Mean Copulatory Behavior Scores for
Individual Pairs of Peromyscus gossypinus
(Palmdale Population) . . . ... 176

Appendix C: Individual Subject Data for Total Time
Spent in Maternal Behaviors During
15-Minute (900 sec) Observation Periods
for Peromyscus gossypinus (Gainesville
Population) . . . . . . . 179
Appendix D: Individual Subject Data for Total Time
Spent in Maternal Behaviors During
15-Minute (900 sec) Observation Periods
for Peromyscus gossypinus (Palmdale
Population) . . . . . . ... 180

Appendix E: Individual Subject Data for Total Time
Spent in Maternal Behaviors During
15-Minute (900 sec) Observation Periods
for Peromyscus leucopus . . . ... 181


. .


I I I I I f I I










Page

Appendix F: Development of Pups: Mean First Day of
Occurrence of Somatic and Behavioral
Characteristics for Peromyscus gossypinus
(Palmdale Population) . . . . . 182

BIOGRAPHICAL SKETCH ............ . . . .. 183
















LIST OF TABLES


Page

Table
1. Measures (Weighted Means and Standard Errors) of
Copulatory Behavior in Peromyscus gossypinus
(Gainesville Population) . . . . . . ... 24

2. Measures (Weighted Means and Standard Errors) of
Copulatory Behavior in Peromyscus gossypinus
(Palmdale Population) . . . . . . ... 26

3. The Number of Intromissions in Each Quarter of the
Ejaculation Latency for Peromyscus gossypinus
(Gainesville Population) . . . . . . . .. 29

4. The Number of Intromissions in Each Quarter of the
Ejaculation Latency for Peromyscus gossypinus
(Palmdale Population) . . . . . . . ... 30

5. Measures of Copulation During the Incomplete Satiety
Series for Peromyscus gossypinus (Both Populations) . 32

6. Measures of Copulatory Behavior in Several Populations
and Species of Rodents . . . . . . . ... 43

7. Per Cent of Litters Surviving Prior to and During an
11-Month Experimental Period.for Peromyscus gossypinus
and Peromyscus leucopus . . . . . . ... 61

8. Per Cent Survival of Litters During Various Stages of an
11-Month Experimental Period for Peromyscus gossypinus
and Peromyscus leucopus . . . . . . ... 62

9. Intra- and Interspecific Comparisons of Litter Survival
for Peromyscus gossypinus and Peromyscus leucopus . 64

10. Comparisons of the Proportion of Time Spent by
Peromyscus gossypinus (Gainesville Population) and
Peromyscus leucopus Mothers in Maternal and Nonmaternal
Behaviors During the First Three Observation Periods
(Days 1, 3, and 5) and the Last Three Observation
Periods (Days 17, 19, and 21) . . . . . .. 70









Page


Table
11. Data from Individual Female Peromyscus gossypinus
(Palmdale Population): Time Spent in Each of Four
Categories of Maternal Behaviors During 15-Minute
(900 sec) Observation Periods on Alternate Days
Following Birth . . . . . . . .... . 72

12. Mean Amount of Time (in Seconds) Spent in Maternal and
Nonmaternal Behaviors by Peromyscus gossypinus and
Peromyscus leucopus Mothers . . . . . ... 81

13. Defense of the Young: Mean Scores Obtained by
Peromyscus gossypinus (Gainesville Population) and
Peromyscus leucopus Mothers . . . . . ... 83

14. Retrieval Latency: The Number of Seconds Required by
Peromyscus gossypinus and Peromyscus leucopus
Mothers to Initiate Retrieval of Pups . . . ... 86

15. The Mean Percentage of Pups Retrieved by Peromyscus
gossypinus and Peromyscus leucopus Mothers During
15-Minute (900 sec) Testing Periods . . . ... 88

16. The Mean Amount of Time (in Seconds) Spent by Mothers
of Two Species of Peromyscus in Retrieving Pups
During 15-Minute (900 sec) Testing Periods ...... 90

17. Defense of the Young: Mean Scores Obtained by Mothers
of Six Species of Muroid Rodents . . . . . .. 100

18. Development of Pups: Mean Day of First Occurrence of
Somatic and Behavioral Characteristics in
Peromyscus gossypinus (Gainesville) Males and
Females . . . . . . . . . . ... 124

19. Development of Pups: Mean Day of First Occurrence of
Somatic and Behavioral Characteristics in
Peromyscus leucopus Males and Females . . . ... .127

20. Development of Pups: Mean Day of First Occurrence of
Various Somatic and Behavioral Traits in Peromyscus
gossypinus (Gainesville Population) and Peromyscus
leucopus . . . . . . . ... . . . 129

21. Nipple Clinging: Per Cent of Peromyscus gossypinus
(Gainesville) and Peromyscus leucopus Pups Not
Attached at the Time of Testing, and the Per Cent
of Pups Clinging for 30 Seconds . . . . ... .132










Page


Table
22. Home Cage Observations: Per Cent of Peromyscus gossypinus
(Gainesville) and Peromyscus leucopus Pups Which Sit by
the Mother and Father During Each 15-Minute (900 sec)
Observation period . . . . . . . . ... 134

23. Home Cage Observations: Per Cent of Peromyscus gossypinus
(Gainesville) and Peromyscus leucopus Pups Which Follow
the Mother and Father During Each 15-Minute (900 sec)
Observation Period. . . . . . . . 135

24. Home Cage Observations: Per Cent of Peromyscus gossypinus
(Gainesville) and Peromyscus leucopus Pups Which Cling
to the Mother's Nipples as She Walks 15 or More
Centimeters During Each 15-Minute (900 sec) Observation
Period . . . . . . . . . . . . 137

25. Home Cage Observations: Per Cent of Peromyscus gossypinus
(Gainesville) and Peromyscus leucopus Pups Which Avoid
the Mother's Retrieval During Each 15-Minute (900 sec)
Observation Period. ... . . . . . . . 139

26. Home Cage Observations: Per Cent of Peromyscus gossypinus
(Gainesville) and Peromyscus leucopus Pups Which Scratch
and Play-Fight During Each 15-Minute (900 sec)
Observation Period . . . . . . . . .. 140

27. Home Cage Observations: Per Cent of Peromyscus gossypinus
(Gainesville) and Peromyscus leucopus Pups Which
Eat/Chew or Drink During Each 15-Minute (900 sec)
Observation Period. . . . . . . . . . 141

28. Home Cage Observations: Per Cent of Peromyscus gossypinus
(Gainesville) and Peromyscus leucopus Pups Which
Nest Build (When Given Gresh Nesting Material) During
Each 15-Minute (900 sec) Observation Period . . .. 143

29. Mean Day of First Occurrence of Somatic and Behavioral
Characteristics in Six Species and Subspecies of
Peromyscus . . . . . . . . ... . . 146

30. Mean Day of First Occurrence of Somatic and Behavioral
Characteristics in Peromyscus gossypinus . . . .. 149

31. Mean Day of First Occurrence of Somatic and Behavioral
Traits in Peromyscus leucopus . . . . . ... 150

















LIST OF FIGURES


Page

Figure
1. The categorization of the behavioral patterns
accompanying copulation in two populations of
male Peromyscus gossypinus . . . . . . .. 35

2. The categorization of the behavioral patterns accompanying
copulation in two populations of female Peromyscus
gossypinus ........... ..... . . 37

3. The mean percentage of time spent in maternal behaviors
(nursing of pups, grooming of pups, retrieval and nest
building) by mothers from two populations of Peromyscus
gossypinus and Peromvscus leucopus during 15-minute
observation periods on alternate days following birth 69

4. Mean percentage of time spent in combined maternal
behaviors (nursing of pups, grooming of pups,
retrieval, and nest building) by mothers from two
populations of Peromyscus gossypinus and Peromyscus
leucopus during 15-minute (900 sec) observation periods
on alternate days following birth . . . . . 79

5. Summary diagram of the first occurrence of various
somatic and behavioral characteristics for Peromyscus
gossypinus (Gainesville population) and Peromyscus
leucopus pups. . . . . . . . . ... .. . 122
















Abstract of Dissertation Presented to the Graduate Council
of the University of Florida in Partial Fulfillment of the Requirements for
the Degree of Doctor of Philosophy



REPRODUCTIVE BEHAVIOR AND EARLY DEVELOPMENT IN COTTON MICE,
Peromyscus gossypinus, AND WHITE-FOOTED MICE,
Peromyscus leucopus

By

Deirdre Vally Lovecky

December, 1976

Chairman: Donald A. Dewsbury
Major Department: Psychology

Normative data were collected on the reproductive behavior

and early development of two populations of Peromyscus gossypinus

and of Peromyscus leucopus. The two populations of P. gossypinus

were collected from the vicinity of Gainesville, Florida, and from

the vicinity of Palmdale, Florida. The P. leucopus were laboratory-

bred descendents of wild-trapped mice from the vicinity of Lansing,

Michigan.

Three main areas, copulatory behavior, maternal behavior,

and somatic and behavioral development were investigated. Differ-

ences among taxa were quantified and an attempt was made to relate

these differences to contrasting features of the environment.

Copulatory behavior was studied only in the two populations

of P. gossypinus. Data on the qualitative form of male copulation,









the quantitative measures of each behavior, and description of other

behaviors associated with copulation were collected. P. gossypinus

was found to exhibit a stereotyped pattern consisting of no locks,

no intravaginal thrusting, and multiple ejaculations. Ejaculation on

one insertion occasionally occurred although several intromissions

usually preceded ejaculation. The two populations were very similar

in both qualitative and quantitative measures. Data were compared

with those of P. leucopus, and other Peromyscus species.

Maternal behavior was studied in two populations of

P. gossypinus and in P. leucopus. Quantitative descriptions of

nursing, grooming of pups, retrieval and nest building were provided.

Data were also collected on the percentage of litters surviving

during the experimental period, and on the number of pups in each

litter at birth. For the Gainesville population and for P. leucopus,

data were collected on maternal defensiveness and retrieval

responses. For the Gainesville population, differences were found

in the amount of time spent by mothers during the first few days

following birth and the last few days prior to weaning in nursing of

pups, in nonmaternal behaviors, in the latency to initiate retrieval,

in the number of pups retrieved, and in the total amount of time

spent retrieving pups. Intra- and interspecific variation

were found in the percentage of litters surviving during the

experimental period. A handling effect was found for the Palmdale

population. The Gainesville population spent a greater proportion

of time retrieving pups and exhibited higher maternal defensiveness

scores than did P. leucopus.


xiii









Somatic and behavioral development was studied in preweaned

pups from the Gainesville population and in P. leucopus. Data

showed species differences in the development of six somatic and nine

behavioral traits. In general, P. gossypinus pups developed at a

faster rate than P. leucopus pups. There were no species differences

found in the sequence of development of the various traits.

Overall, few population differences were found. The intra-

specific variation that was found was hypothesized to have been

either the result of chance or of genetic changes in response to

differences in particular features of the micro-habitat, such as

availability of safe nest sites.

Many species differences were found. In general, P. leucopus

was the temperamentally less reactive and less aggressive species.

Interspecific variation was discussed in relation to the long

laboratory tenure of P. leucopus, as well as in relation to specific

adaptations to the natural environment. It was hypothesized that

the habitats of the two species might differ in the amount of

predation to which animals were subject, or in the availability of

safe nest sites.















CHAPTER I
INTRODUCTION


The present research was designed to provide normative data

on reproductive behavior and early development in two species of the

muroid rodent genus, Peromyscus. Reproductive behaviors (copulatory

behavior and maternal behavior) and the development of somatic and

behavioral traits in the offspring are biologically significant

events. The study of these events provides not only new data on the

individual species, but also provides a basis for meaningful

comparisons within and between species. Such comparisons are

essential for understanding the evolutionary history and the adaptive

significance of behavior.




Collection of Normative Data

The collection of normative data for individual species

provides information on the natural history, the life cycle and the

particular characteristics of a species. This is important when

generalizations regarding the relationship of species characteristics

to the natural environment are to be made. The behavior patterns of

reproduction and early life of species form a logical unit for study

because they are of critical importance in insuring the survival of

the genotype. It is in such behavior patterns that the operation of

natural selection may be expressed as the environment changes.

1









2

Reproductive behaviors and the developing characteristics of

the offspring form a continuing life cycle that proceeds from genera-

tion to generation. The life cycle begins with the selection of a

mate. For the genes of each individual to be passed on, it must

copulate successfully with a conspecific of the opposite sex. Once

mating has successfully occurred and gestation has been completed,

parturition ensues. The adult female, and often the male, must care

for the young. The parents must help insure survival of the genes

by aiding in survival of the young.

At birth, most species of rodents are helpless and dependent

on the mother for food, warmth and protection. The pups respond to

the mother's behavior with a few behaviors necessary for survival,

such as suckling when the mother assumes a nursing position, or

vocalizing when separated from her. Newborn pups are relatively

passive recipients of maternal attention. Without it, they soon die.

As the young grow, new behaviors develop, and the form and

frequency of old behaviors change. The pups begin to move about,

to groom, to eat solid food, and to defend themselves. At the same

time, the mother's behavior becomes more dependent on pup solicita-

tion. Both sets of behaviors change as a function of increasing

pup maturity (Rosenblatt, 1965, 1970).

If the young do survive and mature, they will subsequently

copulate, bear young and care for them, thus passing on the genes

to successive generations.

Study of the life cycle of a species can provide information

on the specific adaptations to the requirements of the environment.











In the present research, the life cycles (copulatory behavior,

maternal behavior, and early development) of Peromyscus gossypinus

and Peromyscus leucopus are described.




Intraspecific Comparisons

Many species are polytypic; that is, they are composed of

populations that are more or less different from each other (Mayr,

1970). The different populations may develop particular characteris-

tics determined by the type of environment inhabited. Populations

inhabiting different ecotypes may be very dissimilar in somatic and

behavioral traits. For example, King's (1958, 1963) investigations

of the maternal behavior and early development of two subspecies of

Peromyscus maniculatus demonstrated many physical and behavioral

differences related to specific adaptations to differing habitats.

Since behavior patterns may have diverged as populations adapted to

differing habitats, some generalizations about the behavior of a

"species" may not be valid. All too often, stereotyped behavior

patterns have been described as "species typical" without sufficient

evidence that the species as a whole exhibited these behaviors.

Bruell (1970) recognized that comparisons of behavior among

populations might show as much diversity as comparisons among species.

He suggested that studies of species that do not include data from

at least two populations might better describe data as "population

typical" rather than "species typical". Another alternative might

be to study populations that inhabit the most frequently preferred

ecotype. For example, if most populations of a species are known to








4

live in pastoral habitats, then description of the behavior of these

populations might more likely be "species typical" than that of a

population living in a xeric habitat.

In the present research, data from two populations of

P. gossypinus were collected. Both populations inhabited similar

sylvan environments. The data collected on copulatory behavior and

maternal behavior were compared and the hypothesis of "species-

typical" behavior evaluated.




Interspecific Comparisons

Comparisons among species are essential in understanding the

evolutionary history and the adaptive significance of behavior. The

study of species similarities focuses on the degree of genetic

relatedness among species. The concept of homology emphasizes the

tracing of behavior patterns to recent common ancestors (Dewsbury,

1973). The frequency and similarity of homologous behaviors

decreases as the degree of relatedness among the species decreases

(King, 1963). Thus, the study of the evolutionary history of

species through homologous traits necessitates the examination of a

sequence of closely related species inhabiting differing environ-

ments. This is difficult to accomplish, and few successful studies

exist (Dewsbury, 1973). Alternatively, most species comparisons have

focused on the relationship of species somatic and behavioral traits

to environmental factors.

The investigation of the adaptive significance of behaviors

has focused on differences among species that can be correlated with









5

the types of environments inhabited. The species compared have not

necessarily been closely related. For example, Allison and Van Twyver

(1970) studied sleep patterns in a variety of mammalian species.

"Good sleepers" tended to either rest in secure sleeping places, or

to be predators, while "poor sleepers" tended to be subject to

predation. Dewsbury (1973) reviewed this and several other investi-

gations of the adaptive significance of behavior.

A number of investigations of different species of Peromyscus

have related behavioral differences to habitat requirements. Layne

(1969, 1970, 1971) and Layne and Ehrhart (1970) studied differences

between P. gossypinus and P. floridanus in digging, climbing, and

nest building behavior and in rhythmicity of the diurnal cycle.

P. floridanus was a poorer digger, climber, and nest builder than

P. gossypinus. These behavioral differences were adaptive for a

species like P. floridanus, that lived in a warm, xeric climate and

nested in relatively safe places. P. floridanus occupies the

burrows of other species.

The comparison among Peromyscus species of reproductive

behaviors and the development of various traits has led to

hypotheses of the adaptive significance of these behaviors. For

example, Dewsbury (1975d) summarized data on copulatory behavior in

eight species of Peromyscus. He correlated the tendency of the

species to lock with the relative thickness of the glans penis. He

also hypothesized that species which lock nest in relatively secure

places. Layne (1968) reviewed investigations of the development

of 'matic and behavioral traits of 25 species and subspecies of








6

Peromyscus. He related the size of litters and the size of neonates

to adult body size, and to rate of development. King (1958, 1963) re-

lated differences in maternal behaviors between two subspecies of

P. maniculatus to differences in types of nest sites and amount of

predation pressure.

In the present research, data collected on the maternal behavior

and early development of P. gossypinus and P. leucopus were compared,

and hypotheses regarding the adaptive significance of behavior were

formulated.




Species Information

The cotton mouse (P. gossypinus) and the white-footed mouse

(P. leucopus) both can be classified as belonging to the genus,

Peromyscus, to the subgenus, Peromyscus, and to the species group,

leucopus. While there are many Peromyscus species sharing the genus

and subgenus, P. gossypinus and P. leucopus are the only two members

of the species group (Hall and Kelson, 1959; Hooper, 1968; Hooper and

Musser, 1964). The two species are very closely related, and apparent-

ly only recently diverged from common ancestral stock (Hooper, 1968).

The two species show some evidence of very similar gene pools

since hybridization has been reported to occur in the laboratory

(Dice, 1937, 1968; Wilson, Vacek, Lanier and Dewsbury, 1976), and

may occur rarely in the field (McCarley, 1954).

Both species share many traits in common. They are similar in

physical appearance. Both inhabit deciduous forests and are semi-











arboreal. Both have differentiated geographically. There are 15

geographic races of P. leucopus and 7 of P. gossypinus (Hooper, 1968).

The two species are, for the most part, allopatric.

P. gossypinus inhabits the smaller range, being restricted to the

southeastern part of the United States. P. leucopus is much more

widespread in range and inhabits much of the eastern and midwestern

part of the United States. The species are sympatric in the northern

part of the P. gossypinus range (parts of Alabama, Arkansas,

Louisiana, Mississippi, Oklahoma, Tennessee, and Texas). Where over-

lap of the range occurs, there is evidence that the two species may

separate ecologically. McCarley (1954, 1963) indicated that in

eastern Texas, P. gossypinus inhabited the preferred lowland forests

while P. leucopusinhabited the less preferred uplands.

The P. gossypinus selected for study in the present research

were collected from two naturally occurring populations. Both

populations inhabited similar sylvan environments, but were from

different locations (Gainesville, Florida and Palmdale, Florida).

The P. leucopus studied were derived from original stock from a

sylvan environment in Michigan (see Chapter II).

















CHAPTER II
GENERAL METHODOLOGY




Subjects

P. gossypinus (Gainesville population). Subjects were wild-

trapped mice and their laboratory-reared offspring. Wild-trapped

mice were caught in Sherman live traps using peanut butter and oatmeal

as bait. Trapping sites were located in lowland pine and hardwood

forests with moderate undergrowth to the west and north of Gainesville,

Florida (Alachua County).

P. gossypinus (Palmdale population). Subjects were wild-

trapped mice and their laboratory-reared offspring. Trapping sites

were located in the area around Lake Okeechobee near Palmdale,

Florida (Glades County).

P. leucopus. Subjects were descendants of wild-trapped mice

bred in the laboratory at Michigan State Univeristy since 1961-1962.

The original stock was trapped near Lansing, Michigan (see King,

Price, and Weber, 1968). Mice were shipped by air to Florida in

early 1972.

Mice were habituated to the laboratory for several weeks, and

then some were mated. Paired males and females which reproduced were

permanently mated. These animals produced offspring which comprised

the laboratory colony.











Maintenance

The mice were housed in plastic cages with wire tops. Small

cages, measuring 29 x 19 x 13 cm, were used to house males and females

from the age of weaning (approximately 23 days) until they were

paired or used in observations requiring larger cages. Small cages

were also used to house females which were subjects in experiments on

copulatory behavior.

Large cages, measuring 48 x 37 x 13 cm, were used to house

mated pairs, or mated pairs with litters, and males which were

subjects in experiments on copulatory behavior.

Cages were lined with a commercial litter material (San-i-cel).

Food (Purina Lab Chow) and water were available at all times.

Mothers with litters were given additional suuplements of sunflower

seeds. Cages were also supplied with nesting material (Nestlets,

Ancare, Corp.).

Cages were placed on seven-shelf racks in two colony rooms.

P. gossypinus (both populations) were housed together in one room, and

P. leucopus was housed in the second room. Other species of mice,

not included in this study, were housed in both rooms.

Cages were cleaned and maintained by the laboratory animal

care staff. Fresh cages, litter material, nesting material, food,

and water were provided on a regular basis. Used cages were washed

in a steam cleaner. In the case of mothers with litters, cages were

changed only when pups were old enough to be weaned (approximately 23

days of age).









10

The colony rooms were maintained at a constant temperature

(approximately 720F) year round. Fluorescent lights were auto-

matically controlled. There was a reversed light-dark cycle of 14

hours of light and 10 hours of darkness. Times of light and dark

were slightly different in the two colony rooms. In the P. gossypinus

colony room, the lights were out from 13:30 hr to 23:30 hr, while in

the P. leucopus colony room the lights were out from 12:30 hr to

22:30 hr. Two 25-watt red lights were left on in each colony room at

all times.

Prior to December, 1971, some P. gossypinus were housed in the

single story, cement block Psychology "surge" building. In December,

1971, the mice were transferred to the rooms described above. After

moving, the mice were allowed several weeks to adapt to the new

conditions. Some tests of copulatory behavior, including the pilot

tests of P. gossypinus were conducted in the old building. Most

tests of copulatory behavior, and all other tests and observations

were conducted in the rooms described above.















CHAPTER III
COPULATORY BEHAVIOR

Introduction


Patterns of copulatory behavior vary greatly among different

mammalian species (see Dewsbury, 1972). Variation is found not only

in the qualitative aspects of copulatory patterns, but also in the

quantitative aspects (i.e., in the spacing and frequency of the

various elements of the stereotyped pattern). Variation may also be

found in the behaviors that occur in association with copulatory

behavior, but which are not part of the copulatory acts themselves

(e.g., grooming the partner).



Patterns of Copulatory Behavior

The stereotyped copulation pattern exhibited by males can be

classified according to four characteristics proposed by Dewsbury

(1972). These include: (1) whether or not males exhibit multiple

ejaculations; (2) whether or not multiple intromissions are pre-

requisite to ejaculation; (3) whether there is a lock (i.e., a mechan-

ical tie between penis and vagina); and (4) whether there is intra-

4
vaginal thrusting. According to this proposal, there are 2 or 16

possible patterns of copulatory behavior. For example, pattern 13,

in which there are multiple ejaculations, multiple intromissions, no

thrusting during intromission, and no locking, is common to many rodent

species. Of 31 rodent species described by Dewsbury (1975d), pattern











13 occurred in 10.

A number of species of Peromyscus have been classified according

to this system. To date no species of Peromyscus has been observed to

lock, and all have attained multiple ejaculations (Dewsbury, 1975d).

While many species exhibit pattern 13 (e.g., P. crinitus, P. leucopus,

P. maniculatus, P. polionotus and P. truei, Dewsbury, 1975d), others

differ in that they exhibit intravaginal thrusting (P. californicus,

P. eremicus, Dewsbury, 1975d) or require only one insertion to attain

ejaculation (P. californicus, P. floridanus, and P. melanophrys,

Dewsbury, 1975d).


Quantitative Analyses of Copulatory Patterns

For many rodent species it is important to distinguish among

mounts, intromissions and ejaculations. Ejaculations are scored when

the male mounts the female, gains vaginal penetration and transfers

sperm. On intromission, there is mounting and insertion, but no sperm

transfer. Mounts are scored when the male's mounting activity fails

to result in vaginal penetration.

The frequency of mounts, intromissions and ejaculations as well

as the temporal spacing between them are important in delineating the

particular characteristics of species-typical copulatory patterns.

Different species may exhibit the same type of stereotyped pattern

(e.g., pattern 13), but differ with respect to the frequency and

spacing of the elements of which the pattern is comprised.

Variation among species on quantitative measures of copulatory

behavior can be quite substantial. For example, among species that

display multiple ejaculations, some exhibit many ejaculations prior










13

to the attainment of satiety (e.g., 5.4 for P. polionotus, Dewsbury,

1971), while others exhibit few (e.g., 1.7 for p. californicus,

Dewsbury, 1974a). Among species that exhibit multiple intromissions

prior to ejaculation, some show a decrease in the number of intro-

missions required to achieve ejaculation during successive series

(e.g., P. polionotus, Dewsbury, 1971), while others show an increase

(e.g., P. crinitus, P. eremicus, and P. leucopus, Dewsbury, 1975d).

Quantitative differences also can be found among different

strains, subspecies and populations of a given species. Typically,

such groups do not show qualitative variation in their copulatory

patterns. For example, different strains of laboratory rats all have

been shown to exhibit pattern 13, but they differ in the frequency

of intromissions required for ejaculation (Dewsbury, 1975c). Such

comparative studies have been completed for various strains of

laboratory mice (McGill, 1965, 1970) as well as for various populations

of wild rodents (P. eremicus, Dewsbury, 1974b; P. polionotus,

Dewsbury and Lovecky, 1974). Significant differences generally are

found among the various strains and populations for the quantitative

measures despite the great constancy of qualitative aspects of the

pattern within species.




Behaviors Accompanying Copulation

The behaviors directly related to copulation (i.e., mounts,

intromissions, and ejaculations) occupy only a small percentage of

the total test time. Dewsbury (1967) estimated for the laboratory

rat that less than 1% of the total test time is spent in actual


~









14

copulation. The remaining 99% of time is spent in other behaviors.

When these other behaviors were sampled for both males and females,

changes in the percentage of time devoted to each behavior were noted

as successive ejaculations occurred.

As with copulatory behaviors, variation among species in the

percentage of time spent in each of these other behaviors has been

demonstrated. For example, allo-grooming (grooming the partner)

during the period prior to the first intromission has been analyzed

for 12 species of rodents (Dewsbury, 1975d). Some species spent a

large percentage of the intromission latency period engaged in allo-

grooming (e.g., P. eremicus, Onychomys torridus) while others spent

little time (e.g., Rattus norvegicus, Oryzomys palustris). Dewsbury

(1975d) noted that where allo-grooming occurred, males always groomed

females for more time than females groomed males.

In the present research, data on the pattern of copulation

exhibited, the quantitative aspects of the behavioral elements which

comprise the pattern, and the behaviors accompanying copulation were

collected for P. gossypinus. Two populations were used to provide

data on intraspecific variation and to evaluate for this species the

hypothesis that qualitative characteristics of copulatory behavior

do not vary across different populations of the same species. Because

the copulatory pattern of P. leucopus has been previously described

(Dewsbury, 1975b), this species was used as a model from which the

present procedure was adapted.









15

Methods

Subjects

P. gossypinus (Gainesville population). Sixteen males and 16

females served as subjects. All were laboratory-born offspring of

either wild-trapped or first generation laboratory-reared parents.

Pairs were chosen so that no two animals from the same litter were

mated. All animals were at least 90 days old at the start of testing.

P. gossypinus (Palmdale population). Fifteen males and 15

females served as subjects from this population. They were selected

from the laboratory colony on the same basis as the animals from the

Gainesville population, and were at least 90 days old at the start of

testing.



Procedure

In preliminary tests with these and with other Peromyscus

species, copulation was not initiated reliably during the early part

of the dark phase of the light-dark cycle. Consequently, tests were

begun between 19:00 and 19:30 hrs (about six hours after the lights

went off). Tests were initiated when the female was placed into the

male's home cage. Testing continued until the males reached a

satiety criterion of 30 minutes without an intromission.

Three days prior to testing, females received subcutaneous

injections of 0.06 mg of estradiol benzoate. On the day of testing,

within six hours prior to the planned start of testing, the females

received 0.6 mg of progesterone. These injections produced fairly

reliable estrous behavior in the females at the time of testing. The











16

estrous behavior of the females did not appear to differ qualitatively

from the naturally occurring estrous behavior observed in females

which were used in maternal behavior experiments. However, females

in hormone-induced estrus did not conceive. The mortality rate of

repeatedly injected females appeared to be higher than that of non-

injected females in the colony.

On the morning of testing the males, still in their home cages,

were placed on a designated shelf in the colony room. Cages were

placed so that the inside of the cage could be easily observed from

a short distance. Cages were lit from behind by red light. The wire

mesh cage covers and the water bottles were removed. The cages were

covered by sheet metal lids in which air holes had been punched. These

lids eliminated climbing by the animals during tests.

After the completion of testing, the pairs of mice were allowed

to remain together until the following morning when the females were

returned to their individual cages.

A total of 63 successful tests were completed for the Gainesville

population and 60 tests for the Palmdale population, with a range of

two to six tests per male. Tests were spaced at least two weeks apart.

Although attempts were made to obtain five successful tests per

male, this did not prove feasible in all cases because of the death

of one of the partners. Tests completed prior to the death of a

partner, or to the development of swelling in the females were used

in data analyses.












Measures

Preliminary observations of P. gossypinus indicated that they

display a pattern of copulatory behavior similar to that seen in

laboratory rats and some other species of Peromyscus. Therefore,

behavioral measures commonly used in investigations of copulatory

behavior in these species were adopted for use (Beach and Jordan,

1956; Dewsbury, 1967, 1968, 1971, 1975b). A "series" is defined to

include all mounts and intromissions leading up to and including an

ejaculation. Measures taken from different series are numbered

sequentially. For example, IF-1 refers to the intromission frequency

of the first series of mounts and intromissions.

The measures used were as follows:

Mount latency (ML). Latency from the introduction of the

female to the first mount or intromission.

Intromission latency (IL). Latency from the introduction

of the female to the first intromission.

Ejaculation latency (EL). Latency from the first intro-

mission of a series to its ejaculation.

Intromission frequency (IF). The number of intromissions

in a series.

Mount frequency (MF). The number of mounts without vaginal

penetration in a series.

Mean inter-intromission interval (MIII). The mean interval

separating successive intromissions within a series. MIII is

calculated by dividing the ejaculation latency of a series by the

intromission frequency.










18

Postejaculatory interval (PEI). Latency from the

occurrence of an ejaculation to the next intromission.

Ejaculation frequency (EF). The number of ejaculations

(complete series) attained before the achievement of the satiety

criterion.

The occurrence of these behavioral patterns was recorded on

an Esterline Angus operations recorder.



Behavioral Categorizations

Copulatory behavior consists of a number of behavior patterns

in addition to mounts, intromissions, and ejaculations. These have

been described and tabulated for laboratory rats and species of

Peromyscus (Dewsbury, 1967, 1971, 1975b). The behaviors described

consisted of 10 to 20 mutually exclusive behaviors performed by both

males and females. In the present study, 14 mutually exclusive

categories of behavior for males and females were selected. Eight

mice of each sex were tested from each population. Categorization

of behavior began with the introduction of the female to the male's

cage and continued through the second postejaculatory interval. In

addition, data from the 30 minute satiety criterion period were

collected.

Behavioral categories were defined as follows:

Pursuit-mount. Behavior in which the male followed the

female. Often this behavior culminated in a mount (males only).

Running. Behavior in which the female ran from the male

which was pursuing her (females only).










19

Locomotor-exploratory behavior. Behavior which included

any motor activity not involving orientation toward the partner.

Examples of locomotor-exploratory behavior were digging, rooting,

rearing, chewing on food or cage litter, walking, jumping, and turn-

ing over in the air (males and females).

Genital grooming. Grooming and manipulation of the

genital region with the forepaws or mouth (males and females).

General grooming. All grooming and scratching not

pertaining to the genital region (males and females).

Sniff partner. Behavior in which the animal oriented to

its partner, and appeared to sniff it. Animals were within

approximately 2 cm of each other (males and females).

Ear wiggle. Behavior in which the male stood still and

wiggled his ears rapidly forward and backward several times (males

only).

Darting. Behavior in which the female approached the

male, paused and then ran away. The female often circled the cage

(females only).

Lordosis. Behavior in which the female stood still, rear

legs spread, and head pointing straight. As the males placed his

paws on her back, and mounted, the female elevated her rump and

arched her back in a concave curve (females only).

Standing still. Behavior in which the animal remained

motionless except for chewing movements or other slight head move-

ments (males and females).









20

Upright sparring. Behavior in which the animal adopted

an upright position oriented towards its partner with sparring move-

ments of the forepaws. This occasionally led to a biting attack

(males and females).

Upright resistant posture. Similar to the sparring

position, but without the sparring movements of the forepaws.

Females remained oriented toward males (females only).

Approach resistant female. Behavior in which the male

approached the female and circled her (males only).

Run away from female. Behavior in which the male moved

to a different location when approached by the female (males only).

Chase male away. Behavior in which the female rushed at

the male when approached by him (females only).

Genital groom partner. Behavior in which one partner

nosed, licked or manipulated the genitals of the other (males and

females) .

General groom partner. -Grooming of the partner with

mouth or forepaws in any body location other than the genital area

(males and females).

Receive grooming. Standing still while being groomed

by the partner (males and females).

Lie down. Behavior in which the animal reclined on the

cage floor, head on or under paws, and was motionless (males and

females).










21

Results

Patterns of Copulatory Behavior

The basic motor patterns of male copulatory behavior were

observed to be identical for both populations of P. gossypinus. The

interaction began with pursuit of the female by the male. As the

female ran from the male, the male mounted. Mounting was accompanied

by the assumption of a mating position lordosiss) by the female in

which the head was held straight, the back arched in a concave upward

posture, the rump held high, and the tail moved to one side. Mounts

which did not result in vaginal penetration were distinguished from

intromissions by the lack of a rapid stereotyped dismount. Rather,

the male seemed to slide off the female's rump. This sequence of

behavior (mount plus dismount) was easily discriminable from the

intromission which was brief, contained a single thrust, and was

terminated by a rapid dismount and subsequent genital grooming.

Ejaculations consisted of deeper and longer vaginal penetrations, with

a more vertical dismount that entailed several short backward steps.

At first, it was difficult to distinguish intromissions from ejacula-

tions because, in this species, differences between intromissions

and ejaculations are somewhat less clear than in other species such

as laboratory rats. With practice, however, the discrimination

became clearer. After ejaculation males often dashed around the

cage before beginning genital grooming. This behavior was not

observed following intromissions. Vaginal smears revealed the

presence of sperm following an ejaculation, but not after only intro-

missions.











22

Following the completion of one or more series, males sometimes

shifted to a pattern of "long" intromissions. These generally occur-

red as part of a last long incomplete "series". Instead of ending in

an ejaculation, intervals between these long intromissions became

longer until the 30 minute satiety criterion was attained. Once the

male shifted to a pattern of long intromissions, he never shifted

back to an organized series with the same female. The long intro-

missions resembled ejaculations, but were of longer duration. There

was a tendency for males to ride females part way across the cage,

and the dismount was different from an ejaculatory dismount. Males

did not dash about the cage.

An attempt was made to determine whether sperm were transferred

during long intromissions. Vaginal smears of females that received

one or more long intromissions, but no ejaculations,were to be

examined. However, placing a fresh female with a male which was dis-

playing long intromissions resulted in a return to organized series

in the two tests in which this was attempted. Consequently, no

direct evidence of sperm absence in long intromissions in this

species is available. Indirect evidence, based on similar patterns

in other closely related species (P. polionotus, Dewsbury, 1971;

P. leucopus, Dewsbury, 1975b), indicates that sperm may not be

transferred during long intromissions.




Quantitative Analyses of Copulatory Patterns

P. gossypinus (Gainesville population). The mean latency from

the introduction of the female to the first mount and intromission









23

was 1100 sec (SE188-sec) and 1138 (SE-193 sec) respectively. A mean

of 2.4 (SE0.2) ejaculations preceded the attainment of the satiety

criterion. No more than five ejaculatory series occurred on any test;

just six tests contained more than four series.

Mean scores and standard errors for the remaining measures are

presented in Table 1. Means and standard errors were calculated so

that data from all males were equally weighted. Standard errors

were calculated by Marks (1947) formula. Data from individual subjects

are presented in the appendices (Appendix A).

The first ejaculatory series occupied a mean of about 2-1/2

minutes. It contained about three intromissions spaced less than a

minute apart. In later series the ejaculatory latency decreased while

intromission frequency increased (until series 4), and mean inter-

intromission intervals decreased. There were few mounts without

vaginal penetration, although mount frequency increased through the

third series. A postejaculatory interval of about five minutes

followed the first ejaculation. Postejaculatory intervals increased

progressively throughout testing.

Most ejaculations were preceded by several intromissions.

Occasionally, males ejaculated on a single insertion. These ejacula-

tions, without preceding intromissions, occurred in a limited number

of tests (3.8%) and were confined to the first series. Because this

type of ejaculations occurred so rarely (in 2 of 63 tests), data

verifying the presence of sperm were not collected.











24









O O



OH 0 H



co co
0 c3 Cr N c o
"U O-, -




eI nH m
Q) 0 n 1


o 4N -N -i i











0






*4
r- 0 -,1
o O dd








4- -HE >

) a a H en






0 0 0 0 e H I
o 3-H-H Z 0 N N H
U



0( 0)
P43

















U) o ko 1
o





0 0











4o W m co a
CJO Ln r



14









04


0)
S3 0
PI C . .




d4 a) 0 N e en
-end r- o m en H










H -rl
S CP P^ Q] c~n ^ l o r
2 ? ^ nc










25

P. gossypinus (Palmdale population). The mean latency from the

introduction of the female to the first mount and intromission was

1062.1 sec (SE242.7 sec) and 1085.3 sec (SEt243.1 sec) respectively.

A mean of 1.9 (SE0.2) ejaculations preceded the attainment of the

satiety criterion. No more than four ejaculatory series occurred on

any test; four series occurred in just 4 of 60 tests.

Mean scores and standard errors for the remaining measures are

presented in Table 2. Means and standard errors were calculated so

that data from all subjects were equally weighted (as for the Gaines-

ville population). Data from individual subjects are included in the

appendices (Appendix B).

The first ejaculatory series occupied a mean of about three

minutes. It included about three intromissions separated by intervals

of about one minute. In later series the ejaculation latency decreased

while the intromission frequency increased, and the mean inter-

intromission intervals decreased. Few mounts without vaginal penetra-

tion occurred. A postejaculatory interval of about seven minutes

followed the first ejaculation; subsequent postejaculatory intervals

were longer than the first. Ejaculation on the first insertion, with-

out prior intromissions, occurred in 3.2% of tests.

Inferential statistics. In order to assess the statistical

significance of differences resulting from population, successive

tests and series, three-way analyses of variance with repeated

measures were conducted for all quantitative measures of the first

two series. Analyses were conducted on a sample of data taken from

the total sample. The sample consisted of subjects which had

























0 0


fa3 e
0,-i


0 0
Or 0
t *t S O
CS


-Ul






7o
0
U

0
4)













04
fti
0
-4



0























rrd
0
c---







4-E
0

0













U)





'4
S---i









0
0
*l




0 o
0m






- a)


10















w r
n









a m


n co
fl 04


M Ln IT


- 0 CN




e In coa
04 u) 0


S-A -i
Coc

O -
C -Hl 0b
-H E

44
00c
CO 4-1 4-







c,
14






<
0
-H

-H C
















m
-4
Cc,
01
4-1










-n 0
1')- <
0


-- N n) .,


o 4 10
O0 4 r-4









27

achieved at least two complete series in at least three complete

tests. If more than three tests were completed by a pair, only the

first three tests were included in the analyses. Twelve pairs (75%

of the total sample) from the Gainesville population and nine pairs

(60%) from the Palmdale population fulfilled these criteria. Where

significant main effects were found, individual comparisons among

means were made using the Newman-Keuls procedure (Winer, 1971). In

the case of significant interactions, tests for simple effects were

conducted (Winer, 1971).

Significant differences were found for 8 of 35 comparisons.

There were effects of population on PEI, F (1, 19)=5.56, p <.05.

Significant effects of series were found for EL, F (1, 19)=8.64,

p <.01; MIII, F (1, 19)=7.26, p <.05; and PEI, F (1, 19)=16.08,

p <.001. Significant changes across tests were found for MF,

F (2, 38)=5.01, p <.05; and PEI, F (2, 38)=6.17, p <.01. A

significant series by test interaction was found for IF, F (2, 38)=

4.41, p <.05; and PEI, F (2, 38)=3227, p <.05.

Individual comparison among means indicated that the Palmdale

population showed significantly longer PEIs than the Gainesville

population (< <.05). Series 2 was significantly longer than series

1 for PEI; series 2 was significantly shorter than series 1 for EL

and MIII. For tests, the first test was significantly lower in MF

and longer in PEI than were the second (p <.01, p <.01) or third

tests (p <.01, p <.01).

Tests for simple effects indicated that for IF there was a

significant difference between the two series during the second test.









28

IF was larger during the second series (F (2, 38)=4.67, p <.05).

Also, IF was larger during test 2 than during tests 1 (F (2, 38)=

3.81, p <.05); or 3 (F (2, 38)=6.27, p <.01). For PEI, test 1

showed a longer interval than tests 2 (F (2, 38)=13.3, p <.001); or

3 (F (2, 38)=13.9, p <.001), during the second series.

Spacing of intromissions. Data were collected on the spacing

of intromissions within each quarter of the ejaculation latency.

Tables 3 and 4 present these data for both populations of P.

gossypinus. Means and standard errors were calculated as described

for the other quantitative measures.

To assess the significance of population, successive quarters

and tests, three-way analyses of variance with repeated measures

were employed. Separate analyses were conducted for each of the

first two ejaculation latencies. In the case of significant main

effects, individual comparisons among means were made using the

Newman-Keuls procedure (Winer, 1971).

Significant differences were found for 3 of 14 comparisons.

There were effects of successive quarters in both ELs (EL-1:

F (3, 57)=4.91, p <.01; EL-2: F (3, 57)=7.97, p <.01). Significant

changes over tests were found for EL-2, F (2, 38)=3.73, p <.05.

Individual comparisons among means for quarters of EL-1

indicated that more intromissions were found in the first quarter

than in the second (p <.01) or third (p <.01). Significantly fewer

intromissions were found in the second quarter than in the third

(p <.01) or fourth (p <.01). The third quarter contained fewer

intromissions than the fourth (p <.01). For EL-2, significantly










29

Table 3. The Number of Intromissions in Each Quarter of the
Ejaculation Latency for P. gossypinus (Gainesville
Population).

Number of intromissions
Quarter of Weighted Weighted standard
Series ejaculation latency mean error

1 1 1.09 0.04

2 0.28 0.11

3 0.60 0.13

4 1.08 0.29


2 1 1.27 0.06

2 0.54 0.16

3 0.65 0.10

4 0.82 0.14


3 1 1.50 0.14

2 0.85 0.13

3 0.93 0.20

4 1.11 0.34


4 1 1.00 0.00

2 0.33 0.38

3 0.10 0.12

4 1.11 0.95


5 1 1.70 ---

2 0.33

3 0.70 ----

4 0.00











30

Table 4. The Number of Intromissions in Each Quarter of the
Ejaculation Latency for P. gossypinus (Palmdale
Population).

Number of intromissions

Quarter of Weighted Weighted standard
Series ejaculation latency mean error

1 1 1.17 0.13

2 0.56 0.24

3 0.73 0.26

4 0.85 0.21


2 1 1.53 0.35

2 1.14 0.47

3 0.88 0.31

4 0.76 0.21


3 1 1.73 0.40

2 1.45 0.14

3 1.63 0.33

4 1.13 0.37


4 1 2.35 0.48

2 2.15 0.64

3 2.35 0.48

4 1.15 0.64











more intromissions were found in the first quarter than in the second

(p <.01), third (p <.01), or fourth (p <.01) quarters. The second

quarter contained significantly more intromissions than the third

(p <.05) or fourth (p <.05) quarters.

Analysis of the tests factor indicated that for EL-2, during the

second test, significantly more intromissions occurred than during the

first (p <.01) or third (p <.01) tests. The first test contained

significantly more intromissions than the third test (p <.05).

Satiety. Satiety could be achieved in either of two ways. Pairs

could discontinue copulation immediately following a final ejaculation.

For the Gainesville population, 12.7% of tests ended in this manner.

For the Palmdale population, 16.7% ended with a final ejaculation. Al-

ternatively, mice could discontinue copulation following an intromission,

i.e., as part of an incomplete "series". The majority of tests ended

in this manner. Most of the tests ending as part of an incomplete

series contained at least one "long" intromission. There was only one

test (per population) ending with an intromission that contained no

long intromissions. Some tests involved shifting, during an incomplete

series, from regular to long intromissions (44% for the Gainesville

population and 16.7% for the Palmdale population). The remaining tests

contained only long intromissions as part of the incomplete series.

Weighted means and standard errors for measures within the final

incomplete series are presented in Table 5. Data from the two popula-

tions were very similar. Analyses using Mann-Whitney U tests (Siegel,

1956) indicated no significant differences between populations for "EL",

"LIF", "MF", "MIII", or "IF". (Measures are similar to those used for

complete series. "EL" is the time from the first intromission of the











32

Table 5. Measures of Copulation During the Incomplete Satiety Series for
Peromyscus gossypinus (Both Populations).


Gainesville population
Weighted Weighted standard
mean error


Palmdale population
Weighted Weighted standard
mean error


"Ejaculation
latency" (sec)

"Long intromission
frequency"

"Mount frequency"

"Intromission
frequency"

"Mean inter-
intromission
interval" (sec)


Measures


2076.7



10.1

1.3



4.6




170. 7


340.4



1.7

0.4



0.6




16.0


2076.7



8.3

2.0



6.3




224.8


390.5



1.6

0.4



1.3




23.7








33

incomplete series to the attainment of the last intromission before

the 30 minute satiety criterion. "LIF" refers to the frequency of long

intromissions, and "IF' to regular intromissions. "MIII" included the

intervals between both regular and long intromissions.)



Behavior Accompanying Copulation

Results of the categorization of the behavior patterns accompany-

ing copulation for males and females from both populations of P. gossy-

pinus are presented in Figures 1 and 2. Intromission latency periods,

ejaculation latencies for the first two series, postejaculatory inter-

vals for the first two series, and the 30 minute satiety period were

each divided into quarters and analyzed. Means and standard errors

(where equal to one or more) are presented for each of the 14 behavior-

al categories in which data were collected.

Males. It was evident from the data for both populations that

several behaviors were associated with copulation. Pursuit-mount in-

creased during the IL, during the ELs, and during the last quarter of

the PEIs. Genital grooming was associated with the ELs, and the first

quarter of the PEIs. Approach to resistant females occurred during

ELs, and during the latter part of PEIs. Grooming and genital grooming

of females was associated with the last quarters of the IL and with

the ELs.

Behaviors that decreased during the ELs included locomotor-

exploratory behavior, general grooming, standing still, running from

the female and lying down.

Approach to resistant females, locomotor-exploratory behavior,

and lying down increased as the test progressed.































Figure 1. Graph of the categorization of the behavioral patterns
accompanying copulation in two populations of male Peromyscus
gossypinus. Means and standard errors (where equal to or greater
than one) are shown for the 14 most frequent categories. Each
intromission latency (IL), ejaculation latency for the first two
series (EL-1, EL-2), postejaculatory interval following the first
two ejaculations (PEI-1, PEI-2), and 30 minute period in which the
satiety criterion was attained (SAT) was divided into quarters.
Data are from eight males in each population.















100 PURSUIT-MOUNT LOCOMOTOR- GENITAL GROOMING GENERAL GROOMING
EXPLORATORY
BEHAVIOR
80-


60








40-
20-








z
a- 0



UPRIGHT SPARRING RUN AWAY FROM ( GENERAL GROOM 0 GENITAL GROOM

40-


20-


0 ... .. .. .. .. .. .. ... ... ... ...
RECEIVE GROOMING LIE DOWN
40- Peomyscu

(oainesv/Ie)
20- s
Pegrai7yscus

2 0 - I.J----- gPo/mssyes
































Figure 2. Graph of the categorization of the behavioral patterns
accompanying copulation in two populations of female Peromyscus
gossypinus. Means and standard errors (where equal to or greater
than one) are shown for the 14 most frequent categories. Each
intromission latency (IL), ejaculation latency for the first two
series (EL-1, EL-2), postejaculatory interval following the first
two ejaculations (PEI-1, PEI-2), and 30 minute period in which the
satiety criterion was attained (SAT) was divided into quarters.
Data are from eight females in each population.

































UPRIGHT SPARRING


DARTING


ii v


STAND STILL


UPRIGHT RESISTANT
POSTURE


1,1:II










38

The IL was characterized in males by locomotor-exploratory

behavior, general grooming, sniffing the female, standing still,

grooming the female and pursuit-mount.

The ELs were characterized by pursuit-mount, genital groom-

ing, sniffing the female and approach to resistant females.

Behaviors most reliably associated with the PEIs were

genital grooming, general grooming, grooming the female, running

from the female, locomotor-exploratory behavior, standing still,

and lying down.

Behaviors prominently associated with the 30 minute satiety

period were locomotor-exploratory behavior, standing still, lying

down and general grooming.

SThe two populations differed little in the types or

percentages of time spent in performing the 14 behaviors accompany-

ing copulation. The only differences between populations were

that the Gainesville population spent a greater percentage of time

in grooming and genital grooming of the female, and in receiving

grooming than the Palmdale population.

Females. Several behaviors of females were closely

associated with copulation. Running from the male increased during

the IL, the ELs, and the final quarter of the PEIs. Genital groom-

ing was associated with the ELs and the first quarter of the PEIs.

Sniffing and grooming the male were associated with the ELs.

Behaviors that were depressed during the ELs included

locomotor-exploratory behavior, general grooming, and lying

down.










39

Standing still, receiving grooming, and locomotor-exploratory

behavior increased as the test progressed.

The IL was characterized by running from the male, general

grooming, grooming the male, receiving grooming, standing still, and

locomotor-exploratory behavior.

The ELs were characterized by running, genital grooming,

sniffing and grooming the male.

Behaviors associated with the PEIs were receiving grooming,

genital grooming, general grooming, grooming the partner, darting,

standing still and locomotor-exploratory behavior.

Behaviors associated with the 30 minute satiety period were

receiving grooming, general grooming, standing still, lying down and

locomotor-exploratory behavior.

Data were collected for the behavioral category lordosiss".

However, due to inconsistencies in collecting these data, complete

data were collected for only a few females (two Gainesville and

three Palmdale females). For these animals it was found that

lordosis was associated with the ELs.

The two populations differed very little with respect to the

type or percentage of time spent in these behaviors. The only

population differences were that the Gainesville population spent

a greater percentage of time in receiving grooming and in lying

down.









40

Discussion

Patterns of Copulatory Behavior

Copulation, in both populations of P. gossypinus, involves

neither a lock nor intravaginal thrusting. Males attain multiple

ejaculations, and are capable of ejaculation on a single insertion.

Therefore, according to Dewsbury's (1972) classification system,

P. gossypinus displays copulatory pattern 15, a pattern shared with

P. floridanus (Dewsbury, 1975a) and Reithrodontomys megalotis

(Davis and Dewsbury, 1974). Pattern 15 has been found to occur

infrequently among species of Peromyscus. Most of the species of

the subgenus Peromyscus, to which P. gossypinus belongs, display

copulatory pattern 13. This pattern requires that multiple intro-

missions precede ejaculation. However, it should be remembered

that not all ejaculations in species with pattern 15 occur on a

single insertion. Furthermore, species which display pattern 15

differ in the percentage of ejaculations that occur on a single

insertion. R. megalotis for example, ejaculated on a single

insertion in approximately 21% of tests (Davis and Dewsbury, 1974),

while P. gossypinus ejaculated on a single insertion in only about

3% of tests. By contrast, some species virtually always ejaculate

on a single insertion (rabbits, cats, some ungulates). P.

gossypinus seems intermediate between species always ejaculating

on a single insertion and those never ejaculating on a single

insertion.

Davis and Dewsbury (1974) and Dewsbury (1975d) have

hypothesized that rapid copulation involving ejaculation on a









41

single insertion might be associated with an environment exposed to

heavy predation. Rapid copulation would serve as an adaptation for

a species unable to protect itself by copulating under dense cover,

or in burrows. Western harvest mice (R. megalotis) and P. floridanus,

both of which inhabit regions with little ground cover, also display

pattern 15 (Dewsbury, 1975d). P. gossypinus has been trapped in

regions where P. floridanus was also trapped and they generally nest

in areas where predation might be expected to be more intense than

for many other species of muroid rodents.



Quantitative Analyses of Copulatory Patterns

The latency to the first mount or intromission in P. gossypinus

exceeded 18 minutes. Among species of muroid rodents for which

comparable data are available, P. leucopus, P. californicus,

P. eremicus, P. polionotus, and Onychomys torridus were slower to

initiate copulation (Dewsbury, 1975d). Dewsbury (1971) and Dewsbury

and Jansen (1972) hypothesized that long ILs might be associated

with a tendency to form pair bonds in the wild. While this

hypothesis may be valid for P. polionotus (Dewsbury, 1971),

P. californicus (Dewsbury, 1974a) and 0. torridus (Dewsbury and

Jansen, 1972), it does not seem to hold for P. leucopus (Dewsbury,

1975b) as P. leucopus appears not to form pair bonds in the wild.

P. gossypinus, like P. leucopus, has a relatively long intromission

latency and appears not to form stable pair bonds in the natural

habitat.









42

P. gossypinus tended to display lower values for mean

ejaculation frequency, intromission frequency, and mount frequency

than many other species, but were generally comparable with other

species displaying pattern 15 (e.g, R. megalotis and P. floridanus).

Scores were also similar to those found for P. leucopus.

Both ejaculation latency and mean inter-intromission interval

were shorter during the second series than during the first. Intro-

mission frequency increased within tests. It would appear that as

tests progressed, copulation occurred at a faster pace, but with

increasing amounts of stimulation required to produce ejaculation.

Comparison with other species (see Table 6) presents no consistent

picture. P. leucopus appears to display slower copulation as

tests progress, while P. maniculatus and P. polionotus display

faster copulation. As in virtually all other species for which

data were available, P. gossypinus displays increasingly longer PEIs

over series.

There were changes in MF and PEI across tests. As males

became more experienced and older, the number of mounts without

intromission increased. The length of the PEI decreased after the

first test. This was most evident in the second series. There

appeared to be two different processes acting on PEI. That is,

PEI would tend to decrease as the animals became more experienced

and older, but it tended to increase during the course of each test.

The spacing of intromissions within each quarter of the EL

differed for the first and second series. In interpreting these

results, however, care must be taken. Since by definition the EL


















H H


0 N r0


a a a (N ~D 9 (Z (N

Go 00 O; m ,
Q-r Nr


H H Hdd d m d



d C C C C C c'


4 o
(d (S>


o i

r, r


0 I
o
rl
0
0




U





u













0 H
(Jl




















nE ,-



ul o
C-
(C


N a d .N
H C C C


N N (N 'C u)




















o; o C d d o
o 0 H- 0 0 0












O H O O O 0




r m c (







0 a a a 0 a


HH N


Sr-
N










44





S4 -I Ln uo o o o

U)1 ai



e 0 0 0 0 4
S 0 0 0 0 U )
n in m p n 0

4a .o

0 .-H 0
Bo 3 c m < n


0 (O 0 0H



0 U) o U ) O U)
r- cr v io r C 0O

-S m i la

41 4j


ow 0
o O
:.. .- 4 ) 0




1 U)I. .0 L 0 0 U
a g ~ N O O C4 rd W
V U Vr (N HH U)



o0 C m H r'- o


0i 0 0
T I 0 in a H)


U o LA CD 40 U r
M H (N OD VO ) i M

al 04 P
.14 W n -1 r-o t o co a%


1 H CO 0 40 N r- am
m H i N r 1 m LO LO m


0U in o

m ) U) H C)
O ~ n 10 o n o c 10 u


W W




w) 4 .0
H 1$N ( H cCC o



U) I I( i H ( ( U)
o H H H I H U U a) U
rd H H H H
a W OD ml 0 n 9:i 0 4







H U) H U-4 En )U
v ,, ico m M 4J 4J M




a) 2) JL d ) 'Am

U) r s
ko a) a3 a om









45

is initiated with an intromission, the first quarter must contain at

least one intromission. This may weight the first quarter most

heavily. The required placing of the first intromission in the

first quarter might not be significant if the frequency of intro-

missions is high because some might be expected to occur within

each quarter. With P. polionotus, for example, data indicate that

the number of intromissions increased from the first through the

fourth quarters. The mean IF for this species was 10.1 (Dewsbury,

1971). Where the mean IF is low, however, the placement of the

first intromission in the first quarter may be more important,

because the other quarters might not have many intromissions. A

conservative interpretation of the data for P. gossypinus, a

species with a low IF, might be that with the exception of the first

quarter, the frequency of intromissions in each quarter of the first

series increased. In the second series, the frequency of intro-

missions decreased through the third quarter. The difference between

the two series could be the result of a warmup effect in the first

series (as occurs in P. maniculatus, Clemens, 1969). Investigation

of later series in P. gossypinus and other species might be valuable.

This, and other hypotheses would require further evidence for

support.

In over 80% of tests, the satiety criterion was achieved

within an incomplete series rather than between series. If P.

gossypinus is compared with other species of Peromyscus in this

regard, much variation will be found. Most species for which data

are available appear to achieve satiety within an incomplete series.









46

However, within a species, the percentage of males achieving satiety

in this manner varies. For example, in P. eremicus, 96% of tests

ended within a series (Dewsbury, 1974b), while in P. leucopus, only

46% of tests ended within a series (Dewsbury, 1975b).

The function of these postejaculatory copulations is yet to

be determined. Dewsbury (1974b, 1975d) has offered two nonexclusive

hypotheses. These additional copulations may aid in the triggering

of critical neuroendocrine reflexes. Evidence for this hypothesis

may be found in the requirement of multiple series for successful

pregnancy in montane voles (Davis, Gray, Zerylnick and Dewsbury,

1974) and golden hamsters (Lanier, Estep, and Dewsbury, 1975).

Postejaculatory copulations were found to function in facilitating

successful pregnancy in cactus mice (Dewsbury and Estep, 1975).

Alternatively, postejaculatory copulations may have evolved

in males as a means of decreasing female receptivity. There is

evidence that multiple insemination by several males may occur in

the wild in Peromyscus (Birdsall and Nash, 1973). Multiple

inseminations could reduce individual reproductive success through

sperm competition (Parker, 1970). Postejaculatory copulation could

reduce female receptivity and thus increase male reproductive

success. Evidence for increase in female resistance through the

PEIs, satiety series and 30 minute satiety period was found in

P. eremicus (Dewsbury, 1974b). Since P. gossypinus are not known

to form pair bonds in the wild, such a mechanism would appear to

be advantageous.












Behaviors Accompanying Copulation

For P. gossypinus, the general pattern of behaviors accompanying

copulation was similar to that of other Peromyscus species exhibiting

pattern 13 (Behavior of other pattern 15 species has not yet been

categorized). As with other species, males confine pursuit of the

female to the intromission latency and ejaculation latencies, with

little pursuit during the postejaculatory intervals. The pattern of

running from males, which is displayed by females, shows a related

temporal pattern. The percentage of time spent by females in running

during the IL varies across species. Dewsbury (1975d) noted that

some species do much running during the IL, while others, particularly

locking species, do little. Dewsbury attempted to associate the

amount of running with the relative safety of the copulation site.

Locking species, which live in burrows, spent almost no time running.

In contrast, P. gossypinus spent about 5% of the IL running.

Certain patterns of grooming are associated with the IL and

the ELs. Both self-grooming, particularly of the genital region,

and grooming of the partner are important during these periods.

Time spent in genital grooming increased throughout the test for

females, but decreased for males. The amount of grooming of the

partner tended to decrease somewhat as tests progressed. During the

IL the per cent of time spent in grooming the partner was 12% for

the Gainesville males and 6% for the Palmdale males. As in many

other species, females groomed males much less than males groomed

females. Gainesville females groomed males for 3% of the IL; Palmdale

females groomed males for 1% of the IL. Dewsbury (1975d) compared

the time spent grooming the partner (allo-grooming) for 12 species










48

of rodents. In some species there was very little mutual grooming

(e.g. R. norvegicus), while in others there was a great deal (e.g.

P. eremicus, and 0. torridus). P. leucopus (Dewsbury, 1975b) was

similar to P. gossypinus in per cent of time spent in allo-grooming,

though P. leucopus males seemed to allo-groom for a greater

percentage of time (20%).

There was an increase in resistance by the female as the test

progressed. This was shown in increases in the per cent of time

females spent in chasing males, in increases in per cent of time

males spent in approach to a resistant female, and in increases in

receiving grooming for both sexes. Termination of copulation may

be partially dependent on the behavioral responses of the females,

as well as on the satiation of the male.



There were relatively few differences between the two

populations of P. gossypinus. The only differences were in the

length of the PEI and in some of the behaviors accompanying

copulation. Gainesville mice spent more time in mutual grooming

activities. The significance of these population differences is

yet to be determined, as they do not appear to limit copulation.

It is possible that they are chance occurrences. Within the con-

fines of the sample tested, it would appear that copulatory

behavior is "species typical". Essentially equivalent copulatory

behavior among populations has been previously determined for three

populations of P. polionotus (Dewsbury and Lovecky, 1974) and two

populations of P. eremicus (Dewsbury, 1974b).









49

Comparison of P. gossypinus with the very closely related

P. leucopus demonstrates species differences in copulatory behavior.

The two species differ in pattern type (pattern 15 versus pattern 13),

and in some quantitative measures, notably the greater frequency of

mounts, the longer ejaculation latencies, and postejaculatory inter-

vals in the second and third series, and in the larger mean inter-

intromission intervals in the third series for P. leucopus. Also,

P. leucopus males are more variable in the manner in which they

achieve satiety, and they appear to engage in more allo-grooming.

These differences in copulatory behavior might be hypothesized to be

the result of previous geographic isolation and speciation. In the

wild, such behavioral differences might serve, with other differences,

as isolating mechanisms. However, despite these behavioral differ-

ences, hybridization has been successfully induced in the laboratory

(Dice, 1937, 1968; Wilson, Vacek, Lanier, and Dewsbury, 1976).

Consequently, the behavioral differences described here may not be

sufficient to prevent copulation in the absence of other variables

(e.g., presence of a mate of the same species in a free choice

situation).

















CHAPTER IV
MATERNAL BEHAVIOR

Introduction


Successful reproduction depends not only on mating with an

appropriate partner under conditions likely to maximize the probabil-

ity of pregnancy, but also on the development of the subsequent

offspring to sexual maturity and successful reproduction. For

mammals, parental care is important to the successful development

of offspring. Maternal care provides the offspring with nourishment,

warmth, protection, etc. until they are developed enough to provide

for themselves. The duration of the period of dependency varies

greatly across species; but, in general, independence is attained

prior to sexual maturity of the young.

Because the dependence of the young on the mother is so

important, maternal behavior is likely to have been molded by the

forces of natural selection acting in the natural habitat of the

species. Natural selection favors behavioral adaptations that insure

successful offspring survival. For example, in environments where

predation pressure is great, natural selection might favor develop-

ment of rapid and persistent retrieval of the young.

In the present research, data were collected on the amount

of time mothers spent in nursing, grooming pups, retrieving pups and

nest building. Also collected were data on defense of pups and on

retrieval responses. Data were collected for two populations of

50









51

P. gossypinus and for P. leucopus. Two populations and two species

were used to provide data on intra- and interspecific variation.



Descriptive Study of Maternal Behaviors

Variation can be found among mammalian species in the

percentage of time mothers spend in the nest with the young. For

example, rabbits may spend only five minutes per day in the nest

(King, 1963), while laboratory rats and deermice may spend up to 21

hours with neonatal pups (Ader and Grota, 1970; Grota and Ader, 1969;

King, 1963). Hill (1972) indicated that for P. leucopus, the 17 to

21 hours spent by mothers on the nest were critical in maintaining

a constant temperature since the young could not yet regulate body

temperature. Grota and Ader (1969) found that for laboratory rats

the time spent with the litter in the nest was spent in nursing,

nest building and retrieving activities. As the pups matured, time

spent with the litter decreased substantially, as did the percentages

of mothers observed nursing, building nests and retrieving.

Nursing, licking or grooming the young, retrieving the young,

and nest building are patterns of maternal behavior that have been

commonly studied in domesticated rodents (e.g. Daly, 1972; Elwood,

1975; Noirot, 1964 a, b, 1969b; Rosenblatt, 1969; Rosenblatt and

Lehrman, 1963; Seitz, 1958; and Slotnick, 1967). Although dif-

ferences in maternal behaviors toward neonatal and more mature pups

have been described for laboratory rats and mice, few such studies

have been completed for other rodent species (deermice, King, 1958,

1963; flying squirrels, Muul, 1970; ground squirrels, Michener,









52

1971; white-footed mice, Hill, 1972; Layne, 1968). Also few

attempts have been made to compare maternal behaviors across

naturally occurring populations and species. Hartung, Looney and

Dewsbury (1976), compared the time parents of five species of

muroid rodents spent sitting on the nest, and in licking and

manipulating pups. King (1958, 1963) compared the time mothers

of two subspecies of P. maniculatus spent on the nest. The time

taken by mothers to initiate retrieval of pups was also studied.



Defense of the Young

Protection or defense of the young has been considered

essential to survival in most mammalian species (Ross, Sawin,

Zarrow and Denenberg, 1963). Defense of the young may involve

interactions with either conspecifics or predators. For example,

King (1958, 1963) introduced strange conspecific males into the

cages of female P. maniculatus with litters. In most cases this

procedure produced attack by the female. Such attack also occurred

when the resident animal was a nonlactating female or a male, but

the results were more variable, and often the strange male was

accepted without attack. Attack towards a strange male by a

lactating female has been described for a variety of Peromyscus

species (e.g. P. crinitus, Eisenberg, 1963; P. floridanus, Layne,

1966; P. gossypinus, Pournelle, 1950, 1952; P. leucopus, Nicholson,

1941; and P. maniculatus, King, 1958, 1963).

Females with litters have also been found to defend the young

against human investigators. Seitz (1958) reported the responses










53

of female laboratory rats to the opening of the cage. Seitz reported

that "maximally maternal" rats hid the young in the nest, and then

advanced towards the cage door. "Minimally maternal" rats fled the

nest. Female rabbits were found to defend the young in a similar

manner. Ross, Sawin, Zarrow and Denenberg (1963) rated females with

litters on a scale of 1 to 4 ranging from a "timid" response to

"biting and kicking". Females of different races were found to dif-

fer in the amount of aggression shown upon cage opening. King

(1958, 1963) reported the reactions of female P. maniculatus with

litters to removal of the young with a forceps. Reactions ranged

from ignoring the presence of the forceps to biting the forceps.

Differences in the average amount of aggression displayed by the

females with litters were found for two subspecies of P. maniculatus.

Mothers of both subspecies were also found to decrease in the average

amount of aggression displayed to the forceps as the pups matured.

Similar intra- and interspecific variation in the responses of

mothers of four genera and several strains of mice to forceps was

found by Scudder, Karczmar and Lockett (1967).



Retrieval Responses

Mothers of many mammalian species will return young to the

nest if they stray, or if they are removed by the investigator.

This retrieval response is difficult to exhaust. Michener (1971)

reported that ground squirrels retrieved up to 200 young, before

retrieval tests were terminated. Mothers had not stopped retrieving

young at this point. Muul (1970) reported similar behavior for









54

flying squirrels, as did Wilsoncroft (1969) for laboratory rats.

The retrieval response has been one of the most frequently

studied maternal behaviors among rodents. Generally, in tests of

retrieval, young are removed from the mother and placed some dis-

tance away from her in the cage. The time the mother takes to

pick up the first pup and return with it to the nest has consti-

tuted the primary dependent variable. The responses of experienced

mothers have been compared with those of inexperienced virgins

(Noirot, 1964a, b, 1969a, 1972; Rowell, 1960). The effects of

practice (Beach and Jaynes, 1956; Carlier and Noirot, 1965) and of

increasing pup maturity (King, 1958, 1963; Noirot, 1964b;

Rosenblatt, 1969; Rosenblatt and Lehrman, 1963; Rowell, 1960) have

been investigated.

Retrieval has been studied in a wider variety of rodents

than have some other maternal behaviors. Laboratory rodents in-

cluding rats (Beach and Jaynes, 1956; Rosenblatt, 1969; Rosenblatt

and Lehrman, 1963; wilsoncroft, 1969), mice (Carlier and Noirot,

1965; Noirot, 1964a, b, 1969a, b, 1972), hamsters (Daly, 1972;

Rowell, 1960) and gerbils (Vick, 1972) have been studied. A variety

of wild rodents have also been investigated. In some wild species

such as P. floridanus (Layne, 1966), P. gossypinus (Pournelle,

1950, 1952) and P. polionotus (Laffoday, 1957) only anecdotal

accounts of the occurrence of retrieval behavior in mothers have

been reported. In other wild species such as P. maniculatus (King,

1958, 1963), ground squirrels (Michener, 1971) and flying squirrels

(Muul, 1970) more detailed investigations have been reported.









55

Methods

Subjects

P. gossypinus (Gainesville and Palmdale population). The

mated pairs studied were either wild-caught or first generation off-

spring of wild-trapped animals. All mice were at least 90 days old

at the time of mating. Pairs were permanently mated. If one

partner died, it was replaced with a new animal in a fresh cage.

Fifteen pairs of Gainesville mice and 10 pairs of Palmdale

mice were mated. Of these, eight Gainesville mothers and three

Palmdale mothers produced offspring and were used for the maternal

behavior observations. The other females either did not produce more

than one litter, or regularly killed their litters. Only mothers

which produced two or more surviving litters were studied.

P. leucopus. Mated pairs were obtained from the laboratory

colony. Animals were at least 90 days old at the time of mating.

Nine pairs were mated. Of these, seven mothers produced more than

two viable litters, and were consequently used as subjects.



Procedure

During observations, because of concern about the number of

offspring failing to survive, particularly from the Palmdale

population, care was taken to minimize the amount of disturbance

and handling to which animals were subjected. To minimize

disturbance, pups were handled as little as possible, and observa-

tions were made only on alternate days following birth.










56

Frequency counts were made both of the number of litters born

in each population and species, and of the number of litters sur-

viving until weaning (the 21st day following birth). The number of

pups in each litter was counted on the day following birth.



Descriptive study of maternal behaviors

Mated pairs were checked daily for litters. When newborn

litters were found, observations of the percentage of time the

mothers spent in nursing, grooming pups, retrieving pups, and nest

building were made. The observations started on the day following

birth (Day 1) and continued on alternate days until weaning (11

observation periods).

During observations, males remained in the cages, and the

cages were left undisturbed. Observations were conducted in the

colony rooms during the first four hours of darkness. Cages were

lit from behind by 25-watt red light bulbs. Two red lights were

left on at all times.

Observation periods were 15 minutes (900 sec) long. Data

were collected using an Esterline Angus operations recorder.

Four mutually exclusive categories of maternal behaviors were

scored. These were adapted from the procedures of other investiga-

tors (Noirot, 1964a, b, 1969b; Seitz, 1958). The categories

selected consisted of the following:

Nursing. Behavior in which the mother rested on top of

the pups with her body arched over them. The pups appeared to be

nursing (i.e., they were attached to the mother's nipples).









57

Grooming pups. Behavior in which the mother licked,

sniffed, or manipulated a pup with forepaws.

Retrieval of pups. Behavior in which the mother carried

a pup in her mouth.

Nest building. Behavior in which the mother rearranged

nesting material without transporting it to a new site, covered

the pups with nesting material, or rebuilt the nest on a new site.

Since males remained in the cages throughout observation

periods, any parental behaviors in which males engaged were recorded.

These observations were not recorded on the event recorder; only

presence or absence of the behavior was noted. Behaviors designated

as paternal were the behaviors designated maternal (i.e., nursing

position, grooming pups, retrieving pups and nest building).



Defense of the young

Mothers tested for defense of pups were those which had

raised at least two litters to the age of weaning. When the mother

was found to have delivered her next (usually third) litter,

observations were made of her reactions to the intrusion of a pair

of forceps into the cage. Observations started on the day follow-

ing birth and continued on alternate days until weaning. Each

mother was tested for one set of 11 observation periods.

Males were removed prior to the start of observations and

were placed in separate cages. Observations were conducted in the

colony rooms during the first four hours of darkness. Cages were

lit from behind by 25-watt red light bulbs.





. . .. .









58

Defense of pups was tested in the following manner: A pair of

25-cm-long forceps was inserted under the cage cover into the nest.

The reaction of the mother to the intrusion of the forceps was noted.

Five possible reactions were scored. These had been adapted from

other investigators (King, 1958; Scudder, Karczmar, and Lockett,

1967). Each possible reaction was given a numerical score ranging

from -1 to +3. The possible reactions of the mothers to the forceps,

with assigned numerical scores, were as follows:

Retreat. The mother actively avoided the forceps (-1).

Indifference. The mother ignored the forceps (0).

Nosing. The mother put her nose near the forceps (+1).

Mild biting. The mother nibbled the forceps with

infrequent bites (+2).

Savage biting. The mother bit the forceps intensely and

frequently (+3).

Six mothers from the Gainesville population, one mother from

the Palmdale population, and four P. leucopus mothers were tested.

Mean scores for each population and species were calculated for each

observation period.



Retrieval responses

The same mothers and offspring studied in the tests for

defense of pups were also studied for retrieval responses. Observa-

tions were made starting on the day following birth, and continued










59

on alternate days until weaning. Conditions in the colony room

(i.e., the phase of the light-dark cycle when observations took

place, the lighting, etc.) were the same as for testing defense of

pups.

Males were removed from the cages and placed in separate cages

prior to the start of observations. Pups were also removed and the

various observations and tests performed in Chapter V were made.

The pups were returned to their home cages and placed on the cage

floor about 2 cm apart. Pups were placed at the greatest possible

distance within the cage from the mother. Data were collected on an

Esterline Angus operations recorder, and included 1) the latency

from the return of the last pup of the litter to the first contact

of a pup by the mother, 2) the per cent of pups in each litter

retrieved by the mother, and 3) the total amount of time taken by

the mother to retrieve all the pups.

Observations continued until all pups had been retrieved or

until 15 minutes (900 sec) had elapsed. At the end of this time,

recording was discontinued, and the male was returned to the

cage.

The measures used in these observations of retrieval

responses (latency, per cent of litter retrieved, and total time

taken by the mother) were adapted from Beach and Jaynes (1956) and

Noirot (1964a, b, 1972).










60

Results

Survival

Litter survival

Data were collected on the number of litters surviving to the

age of weaning. Data were collected from three time periods, 1) from

1970 to 1972 (i.e., prior to the initiation of studies of maternal

behavior and early development), 2) from the first five months of

observations, and 3) from the second six months of observations.

Prior to data analyses, frequency data were converted to percentages

to account for the differences resulting the inequality in the length

of time periods. Also, differentiation was made between the litters

which had been studied as part of the retrieval response and early

development observations, and those which had not been subjects

(i.e., between handled and relatively nonhandled litters).

P. gossypinus (Gainesville population). The per cent of

litters surviving prior to the beginning of the study and during the

first and second parts of experimental observations is shown in

Table 7. Also shown is the per cent of handled and nonhandled

litters surviving. Results of data analyses are shown in Table 8.

Comparisons of the per cent of litters surviving during the

various phases of the study indicated that a larger percentage of

litters survived prior to the start of the study than during

experimental observations. A larger percentage of litters failed

to survive during the first five months of observations than during

the second six months. Both handled and nonhandled litters showed

















'-

4-C)

I) 4?
e '
r4
8. E


'H

dP dP p dP

ITr [- C14 rq
I N N4 N
ILN 'H ~


V0O 2


dP dP dP dP
nr-o 0
m)OO

n0 on 00
i-tb~


-NN0








dP IO dP dP
























T In m




















0000






I I



C' C^) H 0
H4 4- '
-Hv
























M00
cr>












C rH






a, c)



JI1 S -iQ
*d r-
'-I Ul'^ T
n LJc
1-1(U d r
f(3 4-)Tl
*U 1-' C


S 0
0>
C) -H



aC >
0> >
*H H4
> >0j
34 0)4

0 C0) -H
04 'C 4-1 -





OM~lCC
41 '
f 4c ]

0 (3


C

0 >
0 -H
C > 0-
*H -*d
> > 0)
> 00
C) ) C
:3 Q
m 'c) IC 4j
00 'H C0) *4
id 4J
4- 0 *




'H 4I1 V r_
E-I ) S


dP oP dp dP
0000

N N 0(











N 'H iD r
'T fC r-' in












Table 8. Per Cent Survival of Litters During Various Stages of an
11 Month Experimental Period in Peromyscus gossypinus and
Peromyscus leucopus.


Subjects surviving


Species and population


P. gossypinus
(Gainesville)
z score p <


P. gossypinus
(Palmdale)
z score p <


P. leucopus

z score p <


All litters
Prior to start of experiment
vs. total experiment time.

First part vs. second parta
of experiment.

Handled litters
Total litters vs. total
handled litters.

Handled litters from first
vs. second parts of
experiment.

Nonhandled litters
Total litters vs. total
nonhandled litters.

Nonhandled litters from
first vs. second parts
of experiment.

Handled vs. nonhandled
litters.
First part of experiment

Second part of experiment

Total experiment


.001 4.31 .001


n.a.


.01 3.12 .01 n.a.


0.82 n.s. 3.11 .01


0.47 n.s.


.025 2.40 .025 n.a.




n.s. 1.19 n.s. 0.95 n.s.



.001 4.07 .001 n.a.




.05 5.25 .001 n.a.


0.93 n.s. 3.54 .001


n.a.


1.61 n.s. 4.30 .001 0.32 n.s.


Note: The statistic, z, was calculated from the test for
differences between percentages for correlated data
(Downie and Heath, 1970).


aFirst part of the experiment included five months of observations.
The second part of the experiment included six months.








63

this effect. However, fewer nonhandled than handled litters survived

during the first five months of the study.

P. gossypinus (Palmdale population). The per cent of litters

surviving during the various phases of the experiment is shown in Table

7. The results of data analyses are shown in Table 8. A larger

percentage of litters survived prior to the start of the study than

during the study. A larger percentage of litters failed to survive

during the second part of the study than during the first part. Com-

parison of data from handled litters with data from all litters indi-

cated that fewer handled litters survived, particularly during the

second part of the study, when no handled litters survived. Comparison

of nonhandled litters with all litters indicated no significant dif-

ferences in the per cent of litters surviving. However, a larger

percentage of nonhandled litters survived during the first five months

than during the second six months of the study. Comparison of the

percentages of handled and nonhandled litters surviving indicated that

a larger percentage of nonhandled litters survived.

Intraspecific comparisons. The per cent of litters surviving

during the various phases of the experiment, and the per cent of

handled and nonhandled litters surviving in the Gainesville and Palm-

dale populations are shown in Table 7. Inferential statistics are

shown in Table 9. The two populations significantly differed in the

per cent of litters which survived during the second part of the study.

Generally, more Gainesville litters survived. Comparison of the sur-

vival of handled litters indicated that significantly more Gainesville

litters survived.













Table 9. Intra- and Interspecific Comparisons of Litter Survival for
Peromyscus gossypinus and Peromyscus leucopus.


Litter survive
during
experiment


Gainesville


Species and population


al vs. Palmdale Gainesville vs.
populations. P. leucopus.
Statistic df p< Statistic df


Palmdale vs.
P. leucopus.
p< Statistic df


Prior to
start

First part
(5 months)

Second part
(6 months)


z=1.57 -- n.s. n.a.



z=0.28 -- n.s. n.a.


z=2.04


-- .05 n.a.


Total time z=1.75 -- .08 z=2.48

Per cent
litters
handled z=1.20 -- n.s. z=2.10

Handled
litters
(total) t=2.34 31 .05 z=1.32


Nonhandled
litters
(total)


z=0.14


-- n.s. t=2.16


n.a.



n.a.



n.a.

- .025 z=3.39




- .05 z=2.18




-- n.s. t=3.21




37 .05 t=2.64


Note: Statistics (z and


t) were calculated from the test for


differences between two proportions for uncorrelated data
(Downie and Heath, 1970). Where the number of subjects
was too small to calculate z, t scores were calculated.


S.001




-.05




16 .01




26 .025


--









65

P. leucopus. Data on litter survival were collected for a

shorter period of time for P. leucopus (6 months) than for P.

gossypinus (11 months). Consequently, data were not available on

the per cent of P. leucopus litters surviving prior to the start of

experimental observations, or on the per cent of litters surviving

during the first five months of the experiment. Such data as are

available are shown in Tables 7 and 8. In this species most litters

survived, including both handled and nonhandled animals. There were

no significant differences in per cent of handled and nonhandled

litters surviving.

Interspecific comparisons. Gainesville population and

P. leucopus. Significantly fewer Gainesville litters than P.

leucopus litters survived (see Table 9). Significantly fewer of the

nonhandled Gainesville litters survived, but there were no differ-

ences between the two species in the per cent of handled litters

surviving, even though significantly more P. leucopus litters were

handled.

Palmdale population and P. leucopus. Significantly fewer

Palmdale litters survived (see Table 9). A significantly larger

percentage of P. leucopus litters were handled, but a significantly

smaller percentage of both handled and nonhandled Palmdale litters

failed to survive.



Number of pups per litter

P. gossypinus (Gainesville population). Fifty-nine litters

were born during the 11 months of observations of maternal behavior










66

and early development. Litter sizes ranged from one to six with a

mean of 3.14 and a standard error of 0.27. The mode was three.

Mean litter size of a similar population of P. gossypinus was reported

by Pournelle (1950, 1952) to be 3.7 with a standard error of 0.14.

The mode obtained by Pournelle was four.

P. gossypinus (Palmdale population). Twenty-nine litters

were born during the 11 months of the study. Litter sizes ranged

from one to four with a mean of 2.3 and a standard error of + 0.32.

The mode was two.

Intraspecific comparison. No significant differences were

found in the size of litters between the two populations of

P. gossypinus (Mann-Whitney U test, Siegel, 1956, U=74, z=1.46, n.s.).

P. leucopus. Fifteen litters were born during the six months

for which data were collected. Litter sizes ranged from one to

seven with a mean of 4.73 and a standard error of 0.39. The mode

was both four and six. These data were comparable to those reported

by Svihla (1932) for P. 1. noveboracensis (mean 4.36).

Interspecific comparisons. Gainesville population and

P. leucopus. Comparison of the size of litters for these two

species indicated that P. leucopus had significantly larger litters

(Mann-Whitney U test, U=90, z=3.06, p <.002).

Palmdale population and P. leucopus. Comparison of the size

of litters for these two species indicated that P. leucopus had

significantly larger litters (Mann-Whitney U test, u=12, z=3.10, p <

.002).









67

Descriptive Study of Maternal Behaviors

Nursing pups

Behavioral description. Nursing occurred when the mother was

in the nest, with her body arched over the pups. The pups were

attached to the mother's nipples. The mothers were usually quiet,

and often appeared to be asleep.

P. gossypinus (Gainesville population). Of the four maternal

activities observed, nursing behavior occupied the greatest proportion

of time. Figure 3 shows that nursing occupied more than 50% of the

total observation period (15 min) until the 13th day following birth.

After that, the per cent of time spent in nursing slowly decreased.

By Day 21, nursing occupied less than 35% of the observation period.

The mean percentage of time spent in nursing pups (i.e., the mean

of all 11 observation periods) was 53.5%.

The amount of time spent nursing pups during the first few

days following birth was compared with the amount of time spent

during the last few days (i.e., the mean proportion of time for the

first three observation periods was compared with that of the last

three observation periods using the Sign test, Siegel, 1956).

Mothers spent more time nursing pups during the first few days fol-

lowing birth than just prior to weaning. Results approached

significance (see Table 10).

P. gossypinus (Palmdale population). The percentage of time

spent by mothers in nursing pups is shown in Figure 3. Much

variability can be seen. An examination of the data from individual

subjects showed that the behavior of the three mothers which were

















y0l Un



O acm



0 CM 0a

CO 4 C)

C) Sc .-4
M 0 j42 M
0 3 '0 -H

A) 0 H -0
0044



O) ) Q) .




)H 40. 0
c: 2 o a) W











C CJC)C H
0C 01 0CC
0 4J ,7 -I
CaC ) >


0 M U
cp e o o
















. -1 0 4-)
-C U) CC








) '0 A 0 (*
Q)q a0 n id













POl 4 HI4 0



:1 ). 0 )M
e ( C)0' Vo





_0 I C )
C -H -H 0



-g E 'n

(0 (n C

mC Ell u *001
m r, w 0



-M 40 0 c
c r 4a0 (

v H -1 -H x

0 m I d1
r. C7 4
Stnc C 1 )

(u -ja o-

L4rOJ I,"o mr















.0)




*r-
-,
*In"












re)


o^ JL

<" y)


0O 00 0o Oo 0 LO
I o- o0 oN I- lo N o I-

31AII_ A0 IN3]d13d


0in 0











Table 10. Comparisons of the Proportion of Time Spent by
Peromyscus gossypinus (Gainesville Population)
and Peromyscus leucopus Mothers in Maternal and
Nonmaternal Behaviors During the First Three
Observation Periods (Days 1, 3, and 5) and the
Last Three Observation Periods (Days 17, 19, and 21).


Species and Population


P. gossypinus
(N=8)
x P<


(Gainesville)


Nursing 0 0.06 1 n.s.

Grooming of pups 1 n.s. 2 n.s.

Retrieval of pups 1 n.s. 1 n.s.

Nest Building 1 n.s. 1 n.s.

Nonmaternal behaviors 0 0.03 2 n.s.


Note: The statistic,
(Siegel, 1956).


X, was calculated from the Sign test


P. leucopus
(N=6)
x p <


Behaviors










71

subjects differed greatly. Table 11 indicates that one mother (#1)

spent approximately 73% of the observation period nursing pups

while another mother (#34) spent approximately 15% of the time

nursing. The mean percentage of time spent in nursing for all

three subjects was 46.2%.

Because of the small number of subjects from this population,

comparisons of the mean amount of time spent by mothers in nursing

during the first three and last three observation periods could not

be made.

P. leucopus. Of the four maternal activities observed,

nursing occupied the greatest proportion of the total observation

period (15 min). Figure 3 shows that for the first 17 days follow-

ing birth, mothers spent more than 60% of the observation period

nursing. After that, nursing decreased slightly, but even during

the last observation period more than 50% of the time was spent in

nursing. The mean percentage of time spent in nursing pups (i.e.,

the mean of all 11 observation periods) was 67.8%.

Comparison of the proportion of time spent nursing pups

during the first and last three observation periods is shown in

Table 10. The difference was not significant.



Grooming pups

Behavioral description. Mothers were observed to groom pups

by licking and cleaning them with their mouths while manipulating

them with forepaws. Pups were groomed by the mother when she

returned to the nest, or when she returned pups to the nest after




















1 0
"-4

C -d
0) 4J
o >
2S 1
















0'm








Em


oa,
-0
0a)

-4













0O
1 0
S0)



1-4
H4 H








Sa0






4 c







0rr
0o n




C 0'C









PL2 a) Q
"-4
H44sCi










4 Cl-C 01

12 .1 *4
PC))-.^
121212a






Hd


4J

0)


c N m 1
.5 12
121 0. d
H- M m y r
3 a1212


o a
0
a-


0-.
N
0 12'


00


N r (



oo




o r-I o r
m








o-)
o c o L ,
C O



0
0 0 m
QD N








o o o






o o M



0 Cn 0 Co
r-














H o o m
o














a o N


o 12 0
,-4


o a


0
in 1N
in 4n


H (N Hi




n -N Co r-



0
0 0 0 o
clj


0 N
rH


r- I0


a
NH N i H
(N

rQ

P0 (N 412 <
r- m
m

kf r r

0 40 N N
Hri


En

4J

0 1


LI 12C
0 .

0 M
12


a { o a o0
0) -1 C



















O 0
0 0 0 0 0



m r-

o o o
<^ r-l

o 0
o 0 0 0 0




o 0 un H H


M
0 0 r
O HOH

^ r-


O O
0 0 0 0 0






o o o 0 0





0 0










0 0 0 0 0
0- r
o o (




OON
N ID N





o


4)1




H
















CC


00
















'U
i l a

































Q)
A















'U




U .
E-) a)


0.0
4.) 4I)
04 0. 4
j1C c
0 Qa) I )
u 3


4 M 4-3

4-3 CN m U) CN m
aJ~ ~~i (Ua U uQ
f" S









74

retrieving them. Mothers usually returned young pups to the nest

before grooming them. When pups were older and outside the nest

with the mother, her self-grooming episodes often included periods

of pup grooming.

P. gossypinus (Gainesville population). Mothers generally

spent at least 5% of the time during observation periods grooming

pups. Figure 3 indicates that grooming pups occupied an increasing

amount of time through the first 17 days following birth. The mean

percentage of time spent in grooming pups was 10.9%.

Comparison of the proportion of time spent in grooming pups

during the first three observation periods and the last three

observation periods is shown in Table 10. Differences between the

time periods were not significant.

P. gossypinus (Palmdale population). The amount of time spent

in grooming pups is shown in Figure 3. Data were quite variable (see

Table 11), but grooming appeared to reach a peak on Day 11. The

mean percentage of time spent by mothers in grooming pups was 7.1%.

Because of the small number of subjects, no comparisons of

the proportion of time spent grooming pups during the first three

and last three observation periods were attempted.

P. leucopus. Grooming of pups reached a peak at about Days

13 through 17 (see Figure 3). The mean percentage of time spent in

grooming pups was 7.9%.

Comparison of the proportion of time spent in grooming pups

during the first three and last three observation periods is shown

in Table 10. No significant differences were found.












Retrieval of pups

Behavioral description. Mothers usually retrieved pups

which had wandered from the nest. Mothers also carried pups to

other locations within the cage. Retrieval was accomplished by

mothers grasping pups by the skin of the abdomen, back, or back of

the neck. Occasionally, pups were seized by a limb or other part

of the body. After being grasped in a carrying position by mothers,

pups were lifted from the floor of the cage and transported to the

nest.

P. gossypinus (Gainesville population). Little time was

spent by mothers in retrieving pups during observation periods.

Figure 3 indicates that retrieval occurred at a very low rate through-

out all 21 days following birth. The mean percentage of time spent

in retrieval of pups was 0.5%.

No significant differences were found when the proportion of

time spent by mothers in retrieval during the first three and last

three observation periods was compared (see Table 10).

P. gossypinus (Palmdale population). Much variability was

shown by mothers in the percentage of time spent in retrieving pups

(see Figure 3 and Table 11). Some mothers spent most of the

observation period carrying pups around the cage. For example,

Table 11 indicates that as much as 82% of the observation period was

spent in retrieving pups (Subject #34, Day 13). The mean percentage

of time spent in retrieving pups was 8.6%.

Because of the small number of subjects, no comparisons were

made of the differences in the proportion of time spent in retrieving









76

pups during the first three and last three observation periods.

P. leucopus. Very little retrieval of pups occurred during

observations (see Figure 3). The mean percentage of time spent by

mothers in retrieving pups was 0.04%.

Comparisons of the proportion of time spent by mothers in

retrieval of pups during the first three and last three observation

periods were not significant (see Table 10).



Nest building

Behavioral description. Mothers were observed to build several

different types of nests. Some mothers completely covered themselves

and the pups with nesting material, leaving only a very small opening.

Males were often found in these nests. Other nests consisted of only

a ground covering of nesting material with a shallow indentation

where pups were located. Mothers built or added to nests if they

were given new nesting material. They also rebuilt nests if nesting

material became disarranged because of the activity of the pups.

P. gossypinus (Gainesville population). The mean percentage

of time spent by mothers in building nests is shown in Figure 3.

Nest building occurred at a fairly steady rate through Day 21. The

mean percentage of time spent in building nests was 2.3%.

No significant differences were found in the proportion of

time spent in building nests when the first three and last three

observations were compared (see Table 10).

P. gossypinus (Palmdale population). Mothers spent little

time in nest building after Day 7 (see Figure 3). Data from









77

individual subjects (Table 11) indicate that only one mother built

nests during more than one observation period. The mean percentage

of time spent in building nests was 0.5%.

No comparisons of the proportion of time spent in building

nests during the first three and last three observation periods were

made due to the small number of subjects.

P. leucopus. Mothers engaged in a steady, but low, rate of

building nests throughout most of the observation periods. The

greatest percentage of an observation period devoted to building

nests occurred on Day 1. The mean percentage of time spent in

building nests was 1.6% (see Figure 3).

No significant differences were found when comparisons of

the proportion of time spent by mothers in building nests during

the first three and last three observation periods were made (see

Table 10).



Maternal and nonmaternal behaviors

The amount of time mothers of both populations of P. gossypinus

and P. leucopus spent in all maternal behaviors combined is shown in

Figure 4. In general, the percentage of time spent in all maternal

behaviors decreased during the last few observation periods. As the

amount of time spent in maternal behaviors decreased, the amount of

time spent by mothers in behaviors such as eating, self-grooming,

and exploration increased. (Data from individual subjects for the

amount of time spent in maternal behaviors have been placed in

Appendices C, D and E.)




















0 Q-
00i4-
o Hu




0 0
,n . 4.
14 C 0 Hj





0H H 10
0000
0 H -4 H
0- 0 00i

S0 >
O> j r




0 0c


4.J 0







D O~H
S (1) 0 4


0 0


-H& NO 1
) a



-4 4 -J 1



On tc V,
0 *H 0
4-1 I 00






00 -




0
0 04
0o Ct 0 ia






















It to c i) 0
a4 Z 4-1 4. )








U 00 0
S-H od M










r0 m '- a
0 CH I H
cH 0 0 (1
14 a0- C












W\\\\\\\\\\\\\\\\\\\






--^^w#\\\\\\\\\\\\\\\\\\\\



N0)













3111l dO lN3Dd3d









80

Comparisons of the time spent by mothers of each group in

nonmaternal behaviors during the first three and last three observa-

tion periods indicated that significantly more time was spent by

Gainesville mothers in nonmaternal behaviors during the last three

observation periods. Comparisons of time spent by P. leucopus

mothers were not significant. Comparisons using Palmdale mothers

were not made because of the small number of subjects (see Table 10).



Intra- and interspecific comparisons

The mean percentage of time spent by mothers of all three

groups in the four maternal activities and in nonmaternal activities

is shown in Table 12. The mean percentage of time in each activity

was compared by means of the Kruskal-Wallis one-way analysis of

variance (Siegel, 1956) for the three groups. Only in the mean

percentage of time spent in retrieving pups did the three groups

differ. Further analyses via the Mann-Whitney U test indicated that

mothers from the Gainesville population spent a significantly

greater amount of time retrieving pups than did P. leucopus mothers,

p=6.5, p < .05.



Paternal behavior

P. gossypinus (both populations). Males were observed to

perform some of the same activities engaged in by mothers. Males

were observed to groom pups and to build nests over pups. Males

also assumed a crouched position over very young pups, similar to

the mother's nursing position. When pups were older, they were


















V

m ) ,
S al


a)

Ca m






4-1
U











a]
'I a ,)
a s a
a]a


ao
C o

;a, ) ) U







C n u
a) Q





aa


T a)
a] U

miJ c u :
a] >-4







a ma
0- rz- C





a)
:RD a)
'^ -I a] a]
>- a
a) a]

0OH- C' U
a] 'da]
*a U )M
a- S -


* a

a) a) a) a)


(N C
0 0
a N


- 08
r-


0 0

S ,-H


aD a aQ
i a
a N


r- a
N H a


a A

a


in or
a o
L a






co
a a


a





o.
OA


4,4
0
C a
04 t?


.c n o
a] a] 0
C 3
a] a) U-


am a a

0 a a
ID


a

La 0
a


0 4
a) 0
H >




0 )



4-1 I> a



< z


co
4-J
4
0) .0





lcl
D nl








I--
a)









ml.
>0 o



i
4-4
0C:


Sco









0
Q a]









im
mai
4J







OU





7mun




Ga
>-, IM







cm
ma]
H 0

ma]








'4-







a]a~
t4J





ui 11l





ra]
aH
'n

o C

I 4-
a] ma









])










a]
U H
4-4
a].-a
-H a a



H I
a]- O *-4-
>Ha
a ]>




E-'-- 0









82

often observed sitting or lying near males outside of the nest.

P. leucopus. Males were observed to perform the same

behaviors as the P. gossypinus males, with the addition of retrieval

of pups.



Defense of the Young

The mean defense scores obtained by Gainesville and P.

leucopus mothers in response to an attempt to remove their offspring

with a forceps are shown in Table 13.

P. gossypinus (Gainesville population). The mean defense

scores of mothers gradually decreased as the pups matured. Calcula-

tion of the mean of all 11 observation periods indicated a defense

score of 0.98. This mean score corresponded to the scores assigned

to exploratory or mild contact behavior (+1 nosing). The most

frequent response seen during observations was sniffing or nosing

the forceps.

The mean defense score decreased from a mean of 1.67 on

Day 1 to 0.67 by Day 21. Scores were highest from Day 1 through

Day 11.

P. gossypinus (Palmdale population). Only one mother was

included in this test. No other mother met the criterion for

inclusion (i.e., a third viable litter). Data from the one avail-

able subject were not included in Table 13. The most frequent

response of this subject was retreat. Of the 11 trials given,

retreat occurred in 9 of them. One instance each of sniff and

ignore were also recorded.










83

Table 13. Defense of the Young: Mean Scores Obtained by Peromyscus
gossypinus (Gainesville Population) and Peromyscus leucopus
Mothers.


Days

Following


Species

P. gossypinus (Gainesville)


Mean


P. leucopus


Mean


1.67 0.42

1.00 0.52

0.83 0.40

1.17 0.54

1.17 0.17

1.33 0.33

0.66 0.56

1.00 0.00

1.00 0.52

0.33 0.42

0.67 0.33 .


-0.50 0.50

-0.75 0.25

-0.50 0.50

0.00 0.58

-0.25 0.48

-0.50 0.50

0.50 0.50

0.50 0.50

0.00 0.58

0.00 0.58

1.00 0.00


Birth


Note: Scores are the responses of mothers to intrusion of a pair of
forceps into the nest. Mothers' responses were rated as:
retreat (-1), indifference (0), nosing (+1), mild biting (+2),
and savage biting (+3).










84

P. leucopus. The mean defense scores of mothers increased as

pups matured. The mean score calculated as the average of the mean

scores of all 11 observation periods was -0.45. This score corre-

sponded to the scores assigned to indifference or retreat from the

forceps entering the nest. The most frequent responses of mothers

were nosing and retreat from the forceps.

The mean defense score increased from a mean of -0.50 on Day 1

to +1.00 by Day 21. Mean scores were highest from Day 13 through

Day 21.

Interspecific comparisons. Since there was only one Palmdale

mother included in the study, data from this subject were not in-

cluded in comparisons. Comparison of the mean defense scores for

each day for the Gainesville and P. leucopus mothers indicated that

P. leucopus mothers showed significantly lower defense scores (Mann-

Whitney U test, U=10.5, p <.001). P. leucopus mothers tended to

retreat rather than to explore or to attack the forceps. The

Gainesville mothers tended to explore rather than to ignore or

retreat from the forceps. The behavior of both groups of mothers

changed over time. Mean scores of zero or less were shown by

P. leucopus mothers prior to Day 13, and higher scores thereafter.

Gainesville mothers initially showed much higher scores than P.

leucopus mothers, but after Day 13, scores of both groups were

similar.












Retrieval Responses -

Retrieval latency

P. gossypinus (Gainesville population). The mean number of

seconds taken by mothers to initiate retrieval of the first pup of

their litters is shown in Table 14. Also shown are the median number

of seconds. Mean latencies prior to Day 15 generally were less than

120 sec. Mean latencies increased after Day 15, until by Day 21,

they reached 600 sec. This increase in mean latencies included the

occurrence of some intervals greater than 900 sec, the total testing

time. None of these 900 sec intervals occurred prior to Day 15.

Comparison of the mean latencies from the first three and

last three observation periods indicated that there was a significant

increase in mean latencies for the last three observation periods

(Wilcoxon matched pairs signed-ranks test, Siegel, 1956, =-0, p
P. gossypinus (Palmdale population). Since only one mother

was available for testing, data were not included in analyses.

However, the mean latencies to initiate retrieval for this mother

were longer than the latencies of the Gainesville mothers. Pups

were retrieved until Day 19.

Many mothers from this population destroyed their litters.

Observations indicated that mothers which destroyed litters either

did not retrieve pups at all, or retrieved one pup and carried it

around the cage for most of the observation period. Males also ran

around the cage during these periods, but without contacting pups.

Some litters, when returned to the cage, were promptly eaten by the

mothers.












Table 14. Retrieval Latency: The Number of Seconds Required by
Peromyscus gossypinus and Peromyscus leucopus Mothers
to Initiate Retrieval of Pups.


Species
P. gossypinus (Gainesville)
(N=6)
Mean SE Median


(sec)


17.2

11.8

108.0

3.7

3.2

9.2

5.3

305.8

316. 7

484.7


(sec)


9.5

9.4

103.8

1.1

1.1

3.2

2.6

187.8

184.6

188.0


7.0

2.5

3.0

2.8

2.5

6.0

3.0

14.0

42.5

550.0


P. leucopus
(N=4)
Mean SE Median
(sec) (sec)


25.8 10.2


31.0

99.5

116.5

53.0

49.2

13.5

26.2

454.8

460.0


26.0

60.3

90.9

46.7

43.6

12.0

13.8

257.1

254.0


22.5

5.5

65.0

38.5

8.5

7.5

9.5

16.0

457.0

463.5


607.7 184.9 899.8


Days

Following


Birth


681.5 218.5 899.8


L




University of Florida Home Page
© 2004 - 2010 University of Florida George A. Smathers Libraries.
All rights reserved.

Acceptable Use, Copyright, and Disclaimer Statement
Last updated October 10, 2010 - - mvs