STIMULUS DETERMINANTS OF SEXUAL
BEHAVIOR IN THE MALE JAPANESE QUAIL
(COTURNIX COTURNIX .IAPDNICA) AND
THE INFLUENCE OF EARLY ISOLATION
FROM FEMALES ON THE SEXUAL-OBJECT
CHOICE OF THE ADULT MALE

Thesis for the Degree of M. A.
MICHIGAN STATE UNIVERSITY
ROBERT E. OTIS
1968

THESIS

    
  

LID;~ 1.",-
MIC“ I I'C

..
4-)“... ,_ . ‘_&

Unive mi Cy

   
  

   

IIDAII & SONS'
BOOK BINDERY INC.

LIBRARY amozns
may. mama;

 
  

 

BINDING BY g

ABSTRACT
STIMULUS DETERMINANTS OF SEXUAL BEHAVIOR IN THE MALE
JAPANESE QUAIL (COTURNIX COTURNIX JAPONICA) AND

THE INFLUENCE OF EARLY ISOLATION FROM FEMALES
ON THE SEXUAL-OBJECT CHOICE OF THE ADULT MALE

BY
Robert E. Otis

The purpose of EXperiment I was to eXpose those fea-
tures of the female Japanese quail which are important for
the elicitation of male sexual behavior. Experiment II
sought to determine what effect isolation from females,
prior to sexual maturation, would have on the stimuli which
elicit sexual responses from the adult male. Several females
and a male were taxidermically stuffed with various parts of
their bodies either missing or modified. These stimuli were
eXperimentally presented to two groups of males; one group
(Experiment I) was normally reared with other females while
the second group (Experiment II) was separated from females
just prior to sexual maturation and raised in all—male
groups.

Sexual responses were elicited from only a few quail
in Experiment I and no preference for a particular model was
shown. The results of Experiment II indicated that social

eXperience with females, up to the time of sexual maturation,

Robert E. Otis

is sufficient to ensure female-elicited sexual behavior from
adult males. The quail in this study responded sexually
more often and to a wider range of stimuli than did the
normally reared quail. There was also evidence of an
enhanced general activity level. Stimulus specificity was
also more apparent in these quail; a preference for a female
model with missing tail feathers was shown.

Treatment effects in both studies may have become
contaminated with the unknown effects of head height. The
occurrence of homosexual behavior indicated that the quail
were responding in an abnormal manner, perhaps due to the
experimental design. It was suggested that the eXperimental
design did not provide optimal conditions for expression of
sexual behavior in either study. Further work is needed to
assess other ways of investigating the stimulus control of

this bird's sexual behavior.

I7 ,7 " ,/ L p. .
Approved 3 Ii Oak/1 (L j (/1; Mic“,

Date: OYQIM 192, ff‘LE

STIMULUS DETERMINANTS OF SEXUAL BEHAVIOR IN THE MALE
JAPANESE QUAIL (COTURNIX COTURNIX JAPONICA) AND
THE INFLUENCE OF EARLY ISOLATION FROM FEMALES

ON THE SEXUAL-OBJECT CHOICE OF THE ADULT MALE

BY

Robert E. Otis

A THESIS

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

MASTER OF ARTS

Department of Psychology

1968

To my parents
Robert and Veda Otis
and to my grandparents
Ed and May Sills

who have made this possible.

ii

ACKNOWLEDGMENTS

The author wishes to express his gratitude to Dr.
Ralph Levine for serving as chairman of the thesis committee
and for his support and encouragement. Special thanks are
also given to Dr. Stanley C. Ratner and to Dr. Robert L.
Raisler for their criticisms and helpful suggestions and for
serving on the thesis committee.

The author is indebted to Mr. Charles E. Smith of
the museum staff at Michigan State University for his prep-
aration of the quail models.

And to my wife, Barbara, for giving her time and
sharing both the many hours of tedious work and the final

accomplishment, a very special thanks.

*****

iii

LIST OF TABLES

Page
Sexual and Approach behavior of the Normal
sexual responders O O O O O O O O O O O O O O O 15
The number of Normal sexual responders
giving component sexual and Approach responses
to live stimuli and the mean duration of their
sexual behavior . . . . . . . . . . . . . . . . 18
Other behavior given by the Normal sexual
responders in the test situation . . . . . . . 20
Spearman rank correlation coefficients
between sexual behavior, Approach, and
General Activity in Experiment I . . . . . . . 23
Sexual and Approach behavior of the Isolate
sexual responders . . . . . . . . . . . . . . . 33
The number of Isolate sexual responders
giving component sexual and Approach
responses to live stimuli and the mean
duration of their sexual behavior . . . . . . . 35
Other behavior given by the Isolate sexual
responders in the test situation . . . . . . . 38
Spearman rank correlation coefficients
between sexual behavior, Approach, and
General Activity in Experiment II . . . . . . . 4O

INTRODUCTION

One of the more fascinating problems in the study of
animal behavior concerns the nature of the stimuli which con-
trol reproductive behavior in birds. Insemination of the
female by the male usually is the culminating phase of an
orderly series of displays and actions by both partners.
Interest has centered on the chain sequence nature of the
behavior pattern (Schein and Hale, 1965), particularly as it
relates to the elicitation and coordination of the male's
courtship responses. Although researchers have shown that \
(internal) hormonal changes influence the male's responses I
(Beach, 1948), many are convinced that these endocrine
changes are consequences of feedback from (external) stim—
ulus-induced behavior and that this behavior depends for its
appearance upon definite patterns of stimulation from the
female mate (Hinde, 1965; Ratner and Denny, 1964).

The fact that many domesticated male birds will
court and copulate with models of their mates has made it
possible to study these sexual stimuli under controlled
laboratory conditions. Both chicken and turkey males, for
example, respond to crouching female models in a manner
indistinguishable from their reactions to receptive females

(Domm and Davis, 1948; Schein and Hale, 1965).

Fisher and Hale (1957) found that the posture of a I
female plays an important role in the type of response (i.e., '
sexual or aggressive) elicited from chicken males. These
investigators changed the posture of the hen model from a
squatting to a raised stance and Observed aggressive rather
than sexual responses in the cock.

Both chicken and turkey males show greater sexual
activity to female models which have the head in a lowered
rather than a raised position (Carbaugh, Schein, and Hale,

1962; Hale, 1960), suggesting that the effectiveness of a
crouching posture may be determined by the position of the
head.

Various parts of the body also play crucial roles as
sexual stimuli. Carbaugh, Schein, and Hale (1962) found
that hen models without bodies were the least effective
sexual stimuli for cocks, indicating the importance of the
body component. Later, Schein and Hale (1965) concluded
that both the head and body together are necessary for
sexual activity in cocks.

Turkey males, on the other hand, require only the I
female's head for complete sexual responding. Schein and 3
Hale (1965) have described how a bodyless female turkey head,
supported 12 to 15 inches above the floor, elicited courting
displays, approach, and pr0perly oriented mounting movements

in male turkeys. But when headless bodies were presented I

alone, these same turkey males would court and approach the

model but rarely mount it. Hale (1960) found that this
female head model was courted by young (4 weeks old) andro-
gen—injected males when it was raised 5 inches off the floor
and attacked when it was raised 12 inches off the floor.
Scjoettle and Schein (1959) further found that the factors
of texture, color, size, shape, and detailed features all
contributed materially and equally toward making the female
turkey head an effective sexual stimulus.

The "stillness" of the female may also be a determin-t
ing factor for male sexual behavior (Lack, 1941). Tinbergen I
(1948) reported that "keeping quiet" was a stimulus which 1
demonstrated the female's readiness to mate. Fisher and
Hale (1957) noted that male chickens responded sexually to
both standing and squatting models and suggested that both
the posture and the stationary quality of the model were
important factors.

Investigators have also been concerned with the prob-
lem of early social experience as it relates to the stimuli
which elicit sexual behavior in adulthood. Lorenz (1935)
theorized that juvenile experience with an object leads to
a preference for that object in adulthood, but research
evidence has been contradictory. Positive evidence for this
hypothesis can be found in the studies of Heinroth (1910) and
Lorenz (1935) with ducks, Raber (1963), Schein (1963), and
Schein and Hale (1959) with turkeys, and Guiton (1961, 1962)

with chickens, all of whom found that male nedifugous fowl,

raised in isolation from all but their human caretakers,
showed courting responses to humans in preference to females
of their species. Likewise, Warriner, Lemmon, and Ray (1963)
reported that male pigeons mated with birds of the same color
as the adults that reared them. On the other hand, some
investigators have reported instances of domestic fowl fail-
ing to show a preference for objects towards which they had
previously directed their early filial responses (Goodwin,
1948; Wood-Gush, 1958). Fabricus (1962) mentioned the fact
that hens are often used as foster parents for a wide variety
of species, but abnormalities in mating activities are rarely
Observed.

Schutz (1965) raised ducks for 1-3 weeks with their
own species and then with another species for 5-6 weeks. He A
found that about one-third of his subjects showed a sexual I
preference for the other species when they were adults.
Schutz also reared bantam cocks with others of their species
until three weeks of age, then shifted them to a pen which
contained only a mallard. In adulthood, these bantam cocks
directed their sexual responding to mallards even though hens
of their own species were also present. These results indi-
cated that the time at which early social contact was given
was an important factor (see also Kruijt, 1964).

It thus seems that mating responses in the male bird
may be elicited by a relatively few visual features of the

female and that social eXperiences during certain periods

early in the male's life may determine his responsiveness
towards these stimuli. Research in this area, however, has
not been without its limitations. Most studies have focused
on only a few of the domesticated Species, and when early
experience has been the independent variable, researchers
have only been concerned with the gross nature of the stim-
uli to which the adult bird responds. As a result, the data
on species other than the chicken and turkey are fragmentary,
the precise nature of early experiential effects upon the
specific stimuli which elicit sexual behavior is unclear,

and the degree of generality among domesticated birds remains
undefined.

The present studies served to add some comparative
data to this area of investigation through the use of
Japanese quail (Coturnix coturnix japonica) as experimental
subjects. Very little behavioral information is available
for this Species. Farris (1964) studied the development of
the quail's sexual behavior and later (1967) showed that the
male's courtship display could be classicially conditioned
to a buzzer. Beach and Inman (1965) reported that "all
sexual activity disappeared (in the male) within eight days:
after removal of the testes and returned to normal within I
eight days after the implantation of an androgen pellet" I
(p. 1428). No research, however, has been directed at defin-;

ing the stimulus features of the female which control the

male's mating responses. This problem was investigated in

EXperiment I. The importance of the male's early social
experience, as it relates to the stimulus control of his
sexual behavior in adulthood, has, likewise, not been
investigated. This problem was studied in EXperiment II.
Since the Japanese quail is being used more extensively in
behavioral research (Padgett and Ivey, 1959; Reese and Reese,
1962; Fidura and Gray, 1966), it becomes imperative that we
understand the conditions which control its characteristic

(species-specific) behaviors (Scott, 1958).

EXPERIMENT I: MORPHOLOGICAL FEATURES OF THE FEMALE

WHICH ELICIT SEXUAL RESPONSES IN THE MALE QUAIL

Experiment I sought to determine those features of
the female Japanese quail's body which elicit sexual re-
sponses in the male. Preliminary investigation by this
author showed that a male quail, when housed separately from
other quail, would readily court and copulate with taxider—
mically stuffed female models. A number of female models
were then taxidermically constructed, differing with respect
to the presence or absence of various body parts. Experiment
I was designed to observe sexual responses of males in the

presence of these models.

Method

Subjects.--The gs were 14 male Japanese quail,

 

designated the "Normals" in this study. These birds were
hatched on May 10, 1967, under the care of the Poultry
Science Department at Michigan State University. When two-
and-a-half months old, these §s were transported to the
Psychology Department and kept in commercial stock sheet
metal pens with an approximately equal number of females.
Room temperature was kept at 750 F, and the lights remained

on continually.

During the experiment, two of these quail were
slightly injured and their data were discarded. Both of
these birds were tested in the same manner as the other §s
so as to maintain standard eXperimental conditions, such as
total test times, for the remaining 55.

Apparatus.--The apparatus used in this study con-
sisted of a sheet metal brooder and a wooden superstructure.
The superstructure was placed in front of the brooder and
served two functions: (1) as a shield through which observa—
tions were made and (2) as a platform for the boom and
pulley mechanism which controlled the eXposure of the test
models (see Appendix A).

The brooder was<mfcommercial stock and measured 23
by 36 by 10 inches. Its floor was of one-half inch hardware
cloth. A sheet of one-eighth inch masonite served as the
roof of the brooder. Extending below the roof in the far
right-hand corner of the brooder (i.e., with respect to the
observer's position) was a white seven—and-a-half watt bulb
(GE 115-125 volt).

The superstructure was essentially half of a wooden
box, measuring 24% by 48% by 24 inches. Observations were
made through a window, measuring 8 by 8 inches, in front of
this superstructure. This window was covered with red
cellophane paper to minimize, if not eliminate, visual
stimuli that could be produced by E, A distance of two feet

separated an exposed model from the observing window.

A shroud, raised and lowered through the roof of the
brooder, served to cover or expose the test models on the
floor of the brooder (see Appendix A). It measured 3 7/8
by 6 7/8 by 11 1/4 inches and was constructed of one-eighth
inch masonite. One end of the shroud was open and the other
was covered by a hinged top. The shroud was controlled by'E
from his observing position by way of a pulley system. A
wooden boom (42 inches long) extended obliquely upwards from
the superstructure to a point above the brooder. A cord was
attached to the shroud and run through an eye bolt at the
top of the boom down the front of the superstructure (i.e.,
to the observer's position). E was able to hoist and set
the shroud at the height of the brooder by hooking the cord
to the supefstructure. This could be done without looking
so that behavioral observation was not interrupted.

Behavioral measures were taken on an Esterline Angus
event recorder. The recorder was placed in an adjacent room
and controlled via microswitches attached to a wooden frame.
The frame could be held in the observer's lap and controlled
without interrupting Observation (see Appendix A).

Temperature in the experimental room was thermostat—
ically controlled and was kept as near as possible to 720 F.
During the eXperiment, however, malfunction of this thermo-
stat caused temperatures to fluctuate from 710 to 750 F on
several days. To determine the effect of such fluctuation,

temperature gradients were determined within the brooder

10

under hot and cool room conditions. When room temperature
was 700 F, temperature fluctuated between 720 and 730 F with-
in the brooder, being warmer near the bulb. Under 760 F

room conditions, brooder temperatures ranged from 750 to

760 F. These measures were taken under simulated eXperimen-
tal conditions. Measures were again recorded 60 minutes
later and brooder temperatures were found not to have

changed.

Test stimuli.--The test stimuli included seven taxi-

 

dermically stuffed quail (see Appendix B) and live quail of
both sexes.

The models were six female and one male quail,
between three and four months old when sacrificed and taxi-
dermically stuffed. These models represented six combina-
tions of three body components--the head, the anterior fea-'
tures of the body (the female's spotted chest, in partic-
ular), and the posterior half of the body (the tail). The
seven models were characterized as follows:

females--head only

tailless body

full bird

headless body

headless body with a black rather

than a spotted chest

body without head or tail.
male-—full bird.

The height of the head above the floor for the "head
only," "tailless body," and “full bird" models was 4, 4,

and 5 inches respectively.

11

In addition to the models, six live males and females
served as stimuli. These birds were the same age as the gs
and were raised under the same conditions as the gs.

Procedure.--Testing was begun in early August when
the quail were three months old. The 14 Se were separated
from the females and placed together in the test brooder for
three days prior to testing. During this adaptation period
all lights in the room remained off, the shroud was on the
brooder floor, and food and water were always available. On
day 4, all birds were transported to a nearby room and
placed in retaining cages in pairs. Only food was available
to the gs in these retaining cages. The gs were then brought
back to the experimental room individually for testing. The
order in which the §s were tested was randomly determined.

Each bird received two trials per day. One trial
was given in the morning, between 8:00 and 12:00 A.M. and
the second was given at night, between 7:00 and 11:00 P.M.,
resulting in an intertrial interval of approximately 12
hours.

A test trial lasted 13 minutes and was begun by
placing the §_into the brooder through a side opening. An
adaptation period of 10 minutes was allowed during which
time the shroud rested on the brooder floor. A test stimulus
was then eXposed, and all behavior of interest was recorded.
The shroud was then lowered to the brooder floor, covering

the test stimulus, and the §_was returned to its retaining

12

cage in the nearby room. After all gs had been tested in
this manner, they were placed, as a group, back into the
testing brooder until the next test session. All lights
remained off, and the shroud rested in its down position
during the non-testing periods.

The testing procedure beyond day 4 was divided into
three phases (I, II, and III) according to the stimuli
tested.

Phase I-—On day 4, following the three day adapta-
tion period, the gs were habituated to the raising
shroud. This procedure involved raising the shroud
without eXposing any of the models.

Phase II--From days 5 to 12 the eight stimulus con-
ditions (the seven models plus a shroud conditioning
control event identical to the habituation event given
in Phase I) were randomly presented to the gs in a man-
ner such that each bird was exposed to each stimulus
twice during this eight day period. No two birds had
the same stimulus presentation schedule.

Phase III—-On day 13,§§ were randomly divided into
two groups of even size. One group was eXposed to live
male quail in the same manner as they were to the other
eight stimulus conditions, once in the morning and once
at night. The second group was exposed to a live female
quail, once during the morning session and once during
the night session.

The original design of this study had a fourteenth
day of testing. On this day the gs were exposed to live
quail of the sex opposite to that given on day 13. Construc-
tion in a nearby room interfered with testing on this day,

and consequently, these data were not reported.

13

Classification Q£_behavior.-—The components of the
male quail's sexual behavior were classified as follows:

1. Neck and body tonus--head is extended forward,
slightly downward, cocked toward the female. The neck
appears as slightly bow-shaped with a hump behind the
head. Body is raised up on the legs and slightly tipped
forward. Legs are stiff. Large primaries are fanned,
almost touching the floor.

2. Strut--stiff legged walk while body is raised up

on its toes, showing full body display as described in
1. Male usually strutted short distances in arcs around
the female when in the brooder. Distance and frequency
of the strut are probably influenced by the size of the
enclosure and, of course, the behavior of the female.
When the female runs away, the strut is abandoned in the
chase.

3. Vocalization--typically occurs while strutting. A
low, hoarse call described by Farris (1967) as a two-
syllable squawk sound lasting several seconds.

4. Body Orient--the male cranes its head over the head
and neck of the female while orienting its body to the
axis of female's body.

5. Neck Grab, Mount—-ma1e grabs skin on back of female's
head or neck in his beak and mounts by placing both feet
on female's back. The male lowers his posterior, raises
his tail and vent feathers, spreads his wings for balance

while pulling backwards on the female‘s neck. Copulation
follows.

For a carefully detailed description of the quail's
sexual behavior, the reader should refer to Farris (1964).
In the present study a somewhat modified version of Farris's
classification system was used. Several components of his
system, such as "feather puffing" during courtship, were
extremely difficult to measure by visual observation alone,
and no attempt was made to do so. The "body orient" was

added as a behavior class because it appeared as a distinct

14

component and was capable of being measured. The classifica-
tion system used was based upon easily identified body move-
ments.
Other non-sexual responses were recorded as follows:
6. Approach--movement within 1% inches of the model.
This response was determined by noting the position of
the §_in relation to the number of % inch "squares" (in
the hardware cloth) between it and the model.

7. Crowing

8. Wing Flap Display--head and body raised to full
height, wings flapped several times, often while running.

9. Defecation
10. Aggressive Pecking--pecking at the test stimuli

11. Preening, eating, and drinking.

Results

The stuffed models elicited sexual responses from
ten of the twelve quail gs. Table 1 shows the number of
quail giving each of the defined sexual responses to the
various models and to the control event. The two sessions
each §_had with each model have been combined. A §_respond-
ing to a model during one or both sessions was reported as
responding once to that model. The sexual responses have
been listed in their order of natural occurrence and the
Approach response was included because it was obviously
necessary for copulation to occur. Although the response
classes were mutually exclusive, the reported reSponse fre—

quencies represent repeated measures on similar gs.

15

.OHQHmmomEH mm3 mmaommOM*

 

o.o+ o.H+ H.n+ o.m+ o.m+ e.m+ 6.0+ 6+ Om mco.H
o.H m.H o.n o.m o.m o.m o.m o 20H»
Imnso cam:
H e m e m m H o uoH>mamn

Hmsxmm mad
1:6 2 H33.

 

 

o o o o o o o I.. Hugo:
o o o o o o o e.. ammo somz
o o H m o o o o HamHHo seem
6 e m A e s e m nomoumma
o o o o o o .o o Hmoop
o o o o H o o o mcHuusuum
H e H m m N H o mscos zoom

muflm x002
mam: mamamm mHmEmm mamamm mamfimm ummnu xumam nmwm Honucoo mmcommmm

Hasm Hasm mmmHHHmB mwmawmmm mmmaawma mHmem mamfimm

mmmapmmm mmmapmmm

 

HHSEHum Hmucmfiwummxm

 

 

 

.m mcHUcommou Hmm HoH>m£mQ Hmsxmm Ham mo Amosoomm Gav coHumuso
some mnu Ucm mmmcommmu nomoummfi cam Hmsxmm ucmsomeou msH>Hm WW mo Hmnfisc Hmuou
ms» OCHOOHUCH .mumpcommmn Hmsxmm HmEHoz OH m3» m0 H0H>m£mn numoummm cam Hmsxmm .a magma

16

Individual quail often gave more than one component of the
mating pattern to a particular model.

The total number of quail responding sexually to
each model and control event, irregardless of the response
types, and the mean duration (in seconds) of their sexual
behavior is also reported in Table 1. Approach responses
were not considered to be sexual responses in the computa—
tion of these two measures.

As can be seen, all seven models elicited sexual
behavior from at least one S, Since there were no sexual
responses observed during the shroud Control event nor dur-
ing the Day 4 habituation phase, we can be sure that the
models alone were eliciting these behaviors. Complete sex-
ual patterns, i.e., courting followed by mounting and copula-
tion, were never elicited. Two models (Headless Female,
Tailless Female) elicited Body Orienting movements but none
of the §s actually made body contact with these two models.

The Cochran Q test (Siegel, 1956) was used to deter-
mine whether the seven models differed in terms of the total
number of gs giving sexual responses to them. The resulting
Q score of 5.51 (p<1.50) indicated that differences were not
statistically significant.

It was of interest that two §s gave Body Orienting
movements to the Headless Female and Tailless Female models
without first giving any Neck and Body Tonus or Strutting

behavior. gs frequently abbreviated their responding at the

1?

Neck and Body Tonus stage. This latter behavior occurred
during 12 test sessions and at least once with each of the
seven models.

It should be remembered that each §_was eXposed to
each experimental stimulus twice, a total time of six min—
utes. However, the longest mean duration of sexual behavior
given to any model was only seven seconds. This particular
model was the Tailless Female, the model with the four inch
head height.

Only three of the ten sexual responders gave any
sexual responses to the live stimuli on day 13. Table 2
shows how these three gs responded. Six gs were tested with
the Live Male stimulus and four others were tested with the
Live Female. This discrepancy in N tested was due to the
fact that two gs, originally assigned to the Live Female
group prior to testing, did not give any sexual responses to
the models. Only the behavior of the model reactors was
reported. One of these two non-reactors did show brief
(approximately one second) Neck and Body Tonus to the Live
Female.

Table 2 reveals that two gs responded to the male
stimulus and only one to the female stimulus. This frequency
difference was not statistically significant (p<:.05, Fisher
Exact Probability test; Siegel, 1956). The Live Male elic-

ited much different durations of sexual behavior from two gs.

18

Table 2. The number of Normal sexual responders giving
component sexual and Approach responses to the
Live Male and Live Female stimuli and the mean
duration (in seconds) of all sexual behavior per
responding §_

 

 

 

 

Live Male Live Female
Response (N=6) (N=4)
Neck and Body Tonus 2 0
Strutting 1 0
Vocal 1 0
Approach 3 2
Body Orient 1 1
Neck Grab 1 1
Mount 1 1
Total N giving sexual behavior 2 1
Mean duration 107.5 65.0
i_one SD 1147.7 +0

 

One §_re5ponded for three seconds, the other for 212 seconds.
This explains the large standard deviation reported for this
stimulus.

Approach behavior.--Approach responses were recorded ,K
regardless of whether they were a part of a longer sequence
of sexual activity or not. The fact that some quail
Approached a model without giving any of the defined sexual
reSponses to it is clear from Table l. The occurrence of
Approach behavior was found to correlate -.23 (p2>.05) with
the occurrence of defined sexual responses (see Table 4).
This was determined by first ranking the experimental stim—

uli according to the frequency of quail Approaching them and

19

according to the frequency of quail giving sexual responses
to them, and then computing the Spearman rank correlation
coefficient (Siegel, 1956) between these two rank orders.
Since Approach was assumed to be a component of the mating
pattern, this very low correlation was uneXpected.

The Cochran Q test was used to determine if the ten
model reactors showed any preference towards certain models
in terms of their Approach behavior. The resulting Q score
of 12.7 (p<2.10) was large but a .05 level of significance
was not reached. Inspection of these data in Table 1 points
out the fact that twice as many quail Approached the Control
stimulus event as Approached the Full Female model. It was
also found that the same eight quail that Approached the
Control event also Approached during the day 4 habituation
phase. It should be remembered that the shroud Control event
was the same stimulus condition given during the day 4 habit—
uation phase.

Other behavior Observed in EXperiment4;.--Table 3
reports the number of gs giving crowing, wing flap, defecat—
ing, aggressive pecking, preening, eating, and drinking
responses in the presence of the various eXperimental stimuli.
Included in Table 3 is the behavior of the six gs which were
given the Live Male and the four §s which were given the Live
Female on the final day of the study. Only the behavior of

the 10 sexual responders is reported.

.mHnHmmOQEH mm3 mmzommmme

 

20

 

 

 

 

N N e m m N m N 6 N e mcchHun
N m m 6 m o o e e m 6 mcHumm
N N m e m e 6 e q q o mchmmum

o o o o H o o o o e.. o maqumm
m>HmmmHmm<
H H m N H N H o m m e coHumommmm

H o H o N N o H H N H NmHmmHa
OmHm mcHz
o o N N N o o N H H H maHgouo
mamfimm mHmz OHM: mamamm mamfimm mHmEmm mHmEmm ummno xomam pmmm Houusoo g mmsommmm

w>HH O>HH Hadm Hasm mmmH mmmH mmwaaame .mHmEmm mamfimm awn
Henzv Amaze IHHme Inmmm mmchmmm mmmHemmm
HHDEHum Hmucmawummxm
HHsaHum

amusmefiummxm mcu on cam w amp so mmmnm soHumsuHan may msHHsp mcHocommmH am no Hon
lads Hmuou mnu mchsaocH .mnmpcommmu Hmsxmm Hmfiuoz cm» on» an cm>Hm H0H>m£mn Monuo .m mHQmB

21

The Cochran Q test was again used to determine
whether the experimental stimuli could be differentiated by
the number of §s giving each of these behaviors to them.

The data from the live stimuli and from day 4 were not
included in these Q tests because these stimuli were pre—
sented separate from the models. Thus, procedural differ—
ences would have confounded any effects had this data been
included. The Q scores and probability levels, all of which
were not significant at a .05 level of significance for each
response tested, were as follows: crowing, Q =10.36, p <.20;
Wing Flap Display, Q==5.79, p<I.70; defecation, Q==9.31,
p<<.30; aggressive pecking, Q==6.00, p<1.70; preening,
Q=3.93, p<.80; eating, Q=7.50, p<.50; drinking, Q=7.60,
p<<.50.

As can be seen in Table 3, crowing was elicited in
the presence of all but two of the models but not at all in
the presence of the live stimuli. Three gs, however,
accounted for all of this crowing behavior. Wetherbee (1961)
stated that this vocal response is usually associated with
the absence of a female quail and is heard from mid—March to
August. He further noted that a captive coturnix will only
crow in a darkened room. Observations showed that the quail
in the present study crowed in both lighted and darkened
rooms and even while being handled in a lighted room.

The patterning effect of crowing responses was very

interesting and should be further investigated. Crows

22

initially heard in a three minute test session were typi-
cally single responses. Later in a session the crows were
typically heard as a burst of two responses and still later
three or four crows would often be heard in close temporal
association. It was felt that longer test sessions might
have allowed even longer bursts of crowing responses.

Although the crowing response has been associated
with the quails' sexual behavior (Wetherbee, 1961), the sig—
nificance of the Wing Flap DisPlay is not known. But this
latter response, like the defecating and aggressive pecking
responses, was infrequently elicited in this study.

The Wing Flap DiSplay was a dramatic response, often
carrying the bird completely across the brooder before the
response was completed, This response was observed in both
standing and running birds. On one occasion a male seemingly
attacked the model by running up to it and displaying the
violent movements of the wings.

Preening, eating, and drinking responses were clearly
the most frequently elicited of the seven responses reported
in Table 3. The models which frequently elicited these
three responses also elicited sexual behavior more frequently.
This association was determined by first combining the fre-
quency of gs giving preening, eating, and drinking responses
to each model and ranking the models according to this index

frequency. This rank order correlated .81 (p<1.05, Spearman

23

rank correlation) with a rank ordering of these models
according to frequency of sexual responders (see Table 4).

Qualitatively, the reaction of a quail to a stimulus
situation was generally consistent from session to session.
Several gs could be reliably identified by their character-
istic responses. For instance, one §_would run over beneath
the light bulb in one corner of the brooder whenever a stim-
ulus was uncovered. In this location § would perform the
characteristic "orienting" movements of the head and neck.
This behavior pattern appeared early (by the third day of
testing) in the study and could be reliably elicited by
raising the shroud. Other birds could be easily identified
by the frequency of crowing or by the manner in which they
ate, drank, or preened.

Relationship between GeneralgActivity, Approach,

and sexual behavior.--The Spearman rank correlations between

 

these three behaviors have been summarized in Table 4.

Table 4. Spearman rank correlation coefficients between
sexual behavior, Approach, and General Activity

 

 

 

 

 

 

Approach General Activity
Sex Behavior -.23 .81**
Approach .64*
*p< .05.

**p< .01.

24

As can be seen, there was a tendency for those models elic-
iting greater amounts of preening, eating, and drinking
behavior to also elicit more sexual and Approach behavior.
There was, as reported earlier in the Results, essentially
no relationship between the elicited amounts of sexual and
Approach behavior.

Other behavior observed outside of eXperimental
situation.--Of particular interest in this study was the
appearance of homosexual activity among the quail during
inter-test periods. It was not unusual for a §_to give
little or no sexual behavior during a test session with a
model and then show immediate and fervent copulatory activ-
ity with other males when placed back into its retaining
cage. When gs were placed together in the brooder, follow-
ing the completion of the morning and evening sessions, the
homosexual activity elicited was of such high magnitude that
injury to the quail was a primary concern. The flexible
masonite roof on the brooder probably reduced the number of
injuries incurred from popping (i.e., upward leaps). When
the lights in the room were turned off, there was an immedi-
ate reduction, and often complete cessation, of this homo-
sexual activity. Turning the lights back on resulted in an
immediate resumption of this copulatory behavior. This
behavior was not noticed during the first two or three days
of the study, suggesting that adaptation to the test condi—

tions may have been a critical factor.

25

For several gs the test brooder was an average
situation which elicited active attempts to escape. This
behavior consisted of "popping," rapid pacing back and forth
along one wall, and attempted escape through the water and
feed troughs. There was a noticeable increase in such
behavior upon stimulus presentation and over the test ses-
sions. Observations showed that §s which consistently gave
such avoidance behavior were those which were consistently
COpulated with by other males during inter-test periods. No
Objective measures of avoidance were used in this study,
unfortunately. One of the two gs which were discarded would
have been classified under the category of active "avoider."
This §_apparent1y injured its neck during a session by
“pOpping" into the roof of the brooder upon stimulus presen-
tation. The other discarded §_showed continued abnormal
behavior during the entire study, resting its head on the

floor of the brooder during most of the test sessions.

Discussion

The male quail in this study did not show a clear
preference for any of the models nor for either of the live
stimuli. The number of gs resPonding sexually to each model
was very low, never exceeding 40% of the total N tested for
any one model. The fact that the female head, presented
alone without the rest of the body, elicited sexual responses

from only one §_does not parallel the research with turkeys

26

(Schein and Hale, 1965). The fact that the Full Female and
the Headless Female models elicited sexual behavior more
often than did the other models, and since these two models
were the most "complete" in terms of body surface suggests
that completeness of the female body may be an important
factor. However, these differences were low and not statis-
tically significant.

It would be premature to conclude that stuffed
females are not adequate sexual stimuli for male quail,
especially since the live quail presented on the final day
of testing elicited sexual responses from only three §s.
During preliminary investigation (i.e., prior to the present
study) the Full Female model was used reliably to elicit
courting, mounting, and copulation from males. This same
female model did not elicit any copulation when it was used
as a stimulus in the present study. The reason for these
contrasting effects is most likely to be found in the differ-
ing experimental procedures. During pilot work the male gs
were maintained alone in the test brooder for several days
at a time and the model was presented once a day, usually
in the afternoon. The SS in the present study were kept
with other males when not being tested so that opportunity
for sexual behavior (homosexual) was not prevented. Further-
more, these §s were not maintained in the brooder for con—
tinuously long periods of time. Perhaps isolation from all

other quail and familiarity with the environment lowers the

27

threshold for sexual behavior by reducing the probability of
avoidance or other competing responses. This question must
be answered experimentally.

Both the Headless Female and the Tailless Female
models elicited Body Orienting responses in male quail. The
fact that body contact was never made with these models indi-
cates that the stimuli eliciting Neck Grab and Mounting
responses were missing. During normal mating patterns the
female either runs away or crouches when the male orients
his body above hers. Since these movement stimuli were
absent in the passive model, the prObability of Neck Grab
and Mounting may have been reduced. This argument does not,
however, explain why a passive model was able to elicit the
full copulatory pattern during the pilot investigations.
Furthermore, the repeated testing of §s should have removed
any surprise or novelty element; nevertheless, the full
pattern was not displayed to the models.

The fact that none of the gs showed any preference
for the stuffed male model was surprising since fervent
homosexual activities were observed during inter-test peri-
ods. This observation further suggests that sexual behavior
in male quail may be under the control of moving stimuli of
a visual nature. The importance of visual stimuli was indi—
cated by the fact that darkness completely eliminated homo-

sexual behavior in the interval between test sessions.

28

The live birds presented as stimuli on day 13
remained unexpectedly quiescent during the three minute
exposure period. Unless sexual behavior was elicited from
the S, it was not unusual to be able to lower the shroud
back down over the live stimulus at the end of the three
minute test period. This quiescence was most likely due to
an apparent "blinding" effect of the 25 watt lamp on the
outside of the brooder; this due to the fact that the bird
was kept within the darkened shroud during the entire ten
minute adaptation period prior to exposure.

Preening, eating, and drinking were the most fre-
quently observed nonsexual behaviors. Since the retaining
cages held only food containers, those gs tested near the
end of a test sessions were deprived of water for approxi-
mately three and one—half hours. These gs were expected to
show considerably more drinking behavior during the ten
minute adaptation period prior to a stimulus's exposure.
Observations showed that ill.§§: upon placement in the test
brooder prior to testing, would either eat or drink within
one minute's time. In general, after ten minutes time, eat-
ing, drinking, and preening responses were reduced to a min-
imum in all gs, irregardless of whether they were tested
early or late in the morning (or evening) session.

The results of the present study were difficult to

assess because of the small amount of sexual behavior

29

actually given in the presence of the models. The experi—
mental design has been suggested as the possible reason for
this low response level. Another interpretation of these
data is that normally reared quail do not respond sexually
to biologically inappropriate objects because of their past
sexual experiences with appropriate females. If this early
eXperience were a factor in the present study, then it would
be of interest to test a population of quail which did not
have past heterosexual experiences. This problem was inves-

tigated further in Experiment II.

EXPERIMENT II: MORPHOLOGICAL FEATURES OF THE FEMALE
WHICH ELICIT SEXUAL RESPONSES FROM THREE MONTH

OLD, SEXUALLY NAIVE, MALE QUAIL

In Experiment II, male quail, which were originally
reared with females, were separated from them on day 28.
Since the male quail does not reach sexual maturity until
day 28 (Farris, 1964), these §s were sexually naive in the
sense they had never copulated with a female. The Object of
the present study was to determine the effect of such isola—
tion on the stimulus features of the female which elicit
sexual behavior from the male. Also of interest was the
general question of whether early social contact with females,
up until the time of sexual maturation, is sufficient to
ensure female-elicited sexual behavior from males when tests

are given at three months of age.
Method

Subjects.--The §s were 14 male Japanese quail of the
same stock used in Experiment I. These birds were hatched
on March 17, 1967, under the care of the Zoology Department
at Michigan State University and were reared communally with

approximately 100 other quail of both sexes. When 14 days

30

31

old, these chicks were transported to the Psychology Depart—
ment and kept in commercial stock sheet metal pens with an
equal number of females from the same brood. Room tempera-
tures and light conditions were the same as for the Normals
in Experiment I. When these chicks were 28 days old, the
females were removed from the pen, leaving the males visually
but not auditorally isolated from females in the room. No
courting or copulatory responses were ever observed among
these male §s. For purposes of comparison with Experiment I,
these §_were designated the Isolates.

Apparatus.--The apparatus was the same as that used
in Experiment I. Temperature fluctuation in the experimental
room was again an uncontrollable factor. Temperature gradi-
ents in the test brooder under hot and cool room conditions
were found to be the same as those reported for Experiment I.
Attempts were made to maintain a 720 F room temperature.

Test stimuli,--The test stimuli were the same as
were used in Experiment I.

Procedure.--Testing was begun in mid-June when the
gs were three months old. Other than the time of the year,
these §s were studied under the same conditions and by the
same procedure as were the Normals in Experiment I. Again,
half the group was given a live female, the other half a
live male, after being tested with the models.

Classification of behavior.--Behavioral measures were

the same as those recorded in Experiment I.

32

Results

The stuffed models elicited sexual responses from 13
of the 14 Isolated males. Table 5 shows the number of gs
giving component sexual and Approach responses to the vari-
ous stimuli, the total number of gs responding sexually to
each model, and the duration of sexual behavior given by
those responding gs. The duration of the vocal response was
not measured and thus did not enter into the computation of
the mean duration of sexual behavior. Table 5 is analogous
to Table l in Experiment I.

No sexual responses were elicited during the day 4
habituation phase. As can be seen in Table 5, one § gave a
Sex Vocal response to the shroud Control event but none of
the other components of the mating pattern were elicited.
Each of the seven models, however, elicited sexual responses
from at least four Isolated males.

There was an apparent preference shown for the
Tailless Female model, both in terms of the frequency of
courting behaviors, and in terms of the total number of gs
giving sexual behavior of any type to it. The ten quail
giving Neck and Body Tonus and the six quail giving Strut-
ting responses to this model equaled, in each case, 3T%
greater frequency of such behavior than was elicited by the
next most effective models. Overall, 23% greater frequency
of sexual responding was shown to this model. This differ-

ence in total number of sexual responders was not statistically

33

coHumusp muH uoc .wmusmmme mm3 mmcommmu Hmoo> mnu mo mosmsvmum Nascee

.OHQHmmomEH mm3 mmcommmme

 

 

 

m.nH. O.OH. N.O+ O.NH. m.H+ O.H+ H.O+ OH Om mco.H

%¥O.© %¥N.OH ¢.m *«h.m m.m N.m N.¢ xso COHHMHSU COOS
m O OH N O O m H HoH>mnmn
Hmsxmm OCH

:23 2 H309

O O O O O O O I.. pesos

O O O O O O O I.. Hmuw Homz

O O N N O O O O HcmHuo seem

O O O OH O OH O OH eumoummm

N H O H O O O H .;. Hmoo>

H O O H O O N O . sIIOfifisfi m
m O OH O O O m O mscoe Noam

paw Homz

mam: mamfiwm mamfiwm mHmEmm mHmEOm ummsu xomam pmmm Houucoo mmcommmm

Adam Hasm mmOHHHmB mmmapmmm mmmHHHmB .mHmEmm mHmEmm
mmmapmmm mmmavmmm

 

HHOEHum HmucmEHHmmxm

 

 

 

.w mzapcommmu Hmm Hofl>m£mn Hmsxmm Ham mo Ampcoomm GHV coHumusp smmE

on» can mmmcommmu nomoumma cam Hmsxmm ucmcomeoo mcH>Hm am no Aeneas Hmuou esp

mchsHucH

.mumpsommmn Hmsxmm mumaowH ma any mo Meabmcmn nomoummd 0cm Hmsxmm .m magma

34

significant, however, This was determined with the Cochran
Q test, which yielded a Q score of 8.42 (p<:.30).

Neither the Normals nor the Isolated quail gave
mating responses beyond the Body Orient stage. It was
interesting that the Headless Female and the Tailless Female
models elicited Body Orienting behavior in both studies.

Mean duration of sexual behavior in the present
study was again very low and the standard deviation was
again very high in comparison with the mean. None of the
models elicited sexual behavior lasting longer than 10.2
seconds on the average.

The difference in a stimulus's effectiveness across
the two studies was determined by comparing the total fre-
quency of Isolate and Normal sexual responders in a Fisher
exact test (Siegel, 1956). Only one significant difference
was detected, that being for the Tailless Female model. A
greater proportion of Isolate quail responded sexually to
this model than did Normals.

Of note was the fact that the Isolated males gave
Sex Vocal responses to four different stimuli in the present
study. Vocalization was never heard during Experiment I.

There was other evidence of enhanced sexual respon-
siveness of Isolated quail in the present study. Sixty—two
percent of the Isolates responded to three or more different
models, the range being from one to six models and the

median was 3.2 models. On the other hand, 70% of the

35

normally raised qual responded to only pgg model, the range
being from one to five models and the median number of
models responded to was 1.2.

Eleven of the 13 Isolate sexual responders gave
sexual responses to the live stimuli on day 13. Table 6
shows how these eleven quail responded. The discrepancy in
N tested for each sex (seven with the Live Male, six with
the Live Female) was due to the fact that one of the gs,
originally assigned to the Live Female group prior to the
start of this study, did not give any sexual responses to

any of the models. Only the behavior of those quail that

Table 6. The number of Isolate sexual responders giving
component sexual and Approach responses to the
Live Male and Live Female stimuli and the mean
duration (in seconds) of all sexual behavior per
responding S

 

 

 

 

Live Male Live Female
Response (N=7) (N=6)
Neck and Body Tonus 6 4
Strutting 3 4
Vocal 2 1
Approach 6 5
Body Orient 4 3
Neck Grab 4 3
Mount 4 3
Total N giving sexual behavior 6 5
Mean duration 87.3 99.2

i_one SD “1103.2 .195-9

 

36

responded sexually to the models was reported. This one
quail §_that failed to respond to a single model did give
4 seconds of Neck and Body Tonus to the Live Female.

Table 6 reveals that one more §_responded sexually
to the Live Male than to the Live Female. This difference
was not statistically significant (p>'.05, Fisher exact
test). Differences in the duration of sexual behavior given
to each live stimulus were likewise found to be insignifi—
cant (t==.18, p>>.05). It is of note that both the Live
Male and Live Female stimuli elicited complete mating
patterns.

Approach behavior.-—Frequency of Approach behavior
was found to correlate .10 (Spearman rank correlation) with
frequency of sexual behavior in the present study (see Table
8). This low correlation, plus a similar low correlation
found in Experiment I between these two behaviors (rs==-.23),
was unexpected, since Approach was assumed to be a component
of the full mating pattern.

The Cochran Q test failed to detect any significant
differences in frequencies of SS giving Approach responses
to the various models (Q==3.4, p>’.90). Inspection of these
data (Table 5) pointed out the interesting fact that, as in
Experiment I, more §s Approached the shroud Control event
than the Full Female model.

Differences between frequencies of Normal and Iso—

lated quail Approaching a particular model were analyzed

37

with the Fisher exact probabilities test. Only one signif—
icant difference was detected, that being for the Tailless
Female model which elicited a greater number of Approachers
from the Isolate group (p<:.05).

Other behavior Observed in Experiment II.—-Table 7
reports the number of Isolated quail giving crowing, Wing
Flap, defecating, aggressive pecking, preening, eating, and
drinking responses in the presence of the various experimen-
tal stimuli. Only the behavior of the 13 sexual responders
is reported.

Differences between the experimental stimuli, in
terms of the number of §s giving each of the above responses
to them, were analyzed with the Cochran Q test. These
analyses did not include the day 4 data nor the data for the
live stimuli. Results of these analyses for each response
measure were as follows: crowing, Q==12.85, p<<.05; Wing
Flap Display, Q= 5.12, p>’.05; defecation, Q==5.19, p>>.05;
aggressive pecking, Q==6.00, p >.05; preening, Q==l.00,
p>.05; eating, Q=4.54, p>.05; drinking, Q=6.55, p> .05.

The only significant difference was found with the
crowing response. Inspection of the crowing data in Table 7
reveals the high response rate given to the Headless Female
with the darkened chest. It was also interesting that all

but the Live Male stimulus elicited crowing.

38

.OHQHmmomEH mm3 mmcommmms

 

 

O O HH O O O O HH O O O OOHHOHHO
m N HH HH HH O OH OH O NH OH OcHumm
O O O O O O O O N O O OOHOOOHO

H N H O H H O O O O.. O OOHHOOO
O>Hmmmumm¢
O O O N O O O O O O H OoHumomOmo

H H O O O O O O O O O NOHOOHO
mmam mcflz
H O m N N N N O O N H OcHzouo
mHmEmm mamz OHM: magfimm mHmEmm mamfimm OHmEmm ummnu xUmHm pmmm Houucou O mmsommmm

O>HH
HOuzv

0>HH HHsm Hasm mmma mmma mmmHHHmB .mHmEmm manfimm awn

:uzv

IHHOB Iwmmm mmmapmmm mmmapmmm

 

HHOEHHO HmucmsHumOxm

 

 

HHOEHum HmucmEHHmmxm

mnu 6cm O amp so mmmnm GOHUOOUHQOS map mcflusp mcHUcommmH aw mo HOQEOQ Hmuou

OOH OOHOsHoaH

.mumccommmn Hmsxmm mumaomH ammuuflnu man an cm>Hm Hofl>m£mn Hmnuo

.h magma

39

The preening, eating, and drinking responses were
clearly the most frequently elicited of the seven responses
listed in Table 7. This was also true in Experiment I. How-
ever, in study I there was a very high correlation between
the frequency of gs giving sexual responses and the fre-
quency giving combined preening, eating, and drinking,
responses. In the present study these two behaviors corre-
lated -.l4 (Spearman rank correlation, see Table 8).

Frequency of preening, eating, and drinking behavior
is indicative of general activity level. It was Efs evalua-
tion that not only did the Isolated quail give more sexual
behavior during the study but they were also more active in
nonsexual ways. This hypothesis was tested statistically by
comparing the combined frequency of Se giving preening, eat—
ing, and drinking responses to each stimulus across the two
studies. Three significant differences were found, all
indicating a greater level of activity among the Isolated
quail. The three models were the Headless Female with the
darkened chest (x2: 4.20), the Tailless Female (X2=4.06),
and the Full Male (X2==5.80). All chi squares were signif-
icant at the .05 level with one degree of freedom. No dif-
ferences were found between groups on day 4 nor with the
live stimuli.

Relationship between General Activity, Approach,
and sexual behavior.—-The Spearman rank correlation between

these three behaviors have been summarized in Table 8.

40

Table 8. Spearman rank correlation coefficients between
sexual behavior, Approach, and General Activity

 

 

Approach General Activity
Sex Behavior .10 -.14
Approach -.74*

 

*p< .05.

As can be seen, models eliciting greater amounts of preening,
eating, and drinking behavior had a tendency app to elicit
Approach responses. This finding is opposite to that found
in Experiment I (see Table 4). It is noteworthy that analo-
gous correlations in Tables 4 and 8 have opposite signs.
Other behavior Observed outside of experimental
situation.-~Homosexual responses were again Observed in the
present study. These behaviors occurred during inter—test
periods, both in the brooder and in the retaining cages.
There were no apparent differences between the Isolates and
Normals in terms of the form and frequency of these responses.
This was a subjective evaluation since there were no objec-

tive measures of homosexual activity used in either study.

Discussion

Although further evidence is needed, the results of
the present study suggest that social contact with females,
up until the time of sexual maturation, is sufficient to

ensure female-elicited sexual behavior from male quail at

41

three months of age. This finding parallels the report by
Kruijt (1962) with junglefowl.

There was evidence of enhanced sexual activity among
the Isolate quail in that they responded more often and to a
wider range of stimuli than did the normally reared birds.
However, the quail in the present study also showed an
enhanced general activity level as well. This was denoted
by frequency of SS giving preening, eating, and drinking
responses. The validity of these responses as representa-
tive of general activity is questionable. It was interest-
ing, however, that this measure of general activity corre—
lated highly with sexual activity in Experiment I, but not
in Experiment II. This is puzzling and cannot be explained
with the data available.

The Isolate quail showed an apparent preference for
the Tailless Female model. This preference may have been
due to the model's lower head height since head height is
a critical factor for sexual behavior in both turkeys (Hale,
1960) and chickens (Carbaugh, Schein, and Hale, 1962). It
was interesting that the normally reared quail did not show
a similar preference for this model.

It is difficult to explain why Se in both studies
directed sexual responses to stuffed models of males and to
other live males. It should be said that such homosexual
behavior is not an unusual occurrence among all-male groups

in our laboratory (Fidura, 1967). Schein (1963) theorized

42

that young birds develop preferences for secondary sexual
objects early in life. Whereas imprinting determines the
primary sexual preference for the female, other males or
Objects may take on secondary sexual identity through the
process of familiarization. The literature contains many
instances in which male birds have directed sexual responses
to biologically inappropriate stimulus objects (cf., Ficken
and Dilger, 1960; Lack, 1941). Schein and Hale (1965) cau-
tion investigators not to mistakenly interpret such abnormal
responses to secondary preferences as "hypersexuality."

It is not known how the male discriminates the sexes,
whether on the basis of behavioral responses or morphological
features. The most reasonable hypothesis is that both are
important factors. Since all of the models elicited Neck
and Body Tonus, presumably the adequate stimulus characters
were present on the models. It follows that the female must
respond to the male's courting display in some characteristic
manner in order that the male reciprocate with Mounting and
Copulatory responses. Since the models could not provide
this necessary female behavior, the male quail did not pro—
cede further in their mating activities. Further studies
are necessary to substantiate these ideas but it is note-
worthy that the data fit such a conceptualization.

It was mentioned earlier (Discussion, Experiment I)
that the manner in which the quail were repeatedly tested

with the models may have had marked effects on their sexual

43

behavior. A repeated measures design was used in the present
studies because only a few quail were available for testing
with the seven models. The major limitation of a repeated
measures design is the possibility that treatment effects
may not dissipate before the next treatment is administered
(Lindquist, 1953). In the present studies the quails'
responses may have become conditioned through practice with
the various models. Some examples of apparently conditioned
behavior have already been mentioned (e.g., avoidance re-
sponses). A shroud control event was included in both
studies to determine if sexual responses would become condi-
tioned to the rising shroud, and they proved not to be.
Possible effects of practice or conditioning due to the
order and sequence in which the models were presented were
ruled out by randomization techniques. But there was no way
of defining to what extent practice with the models influ-
enced the Ss' sexual behavior (cf., Tinbergen, 1942). And
one is still unable to ascertain how quail might respond to
any one model if it were presented alone.

A model's effectiveness as a sexual stimulus was
determined by the total number of §s directing sexual re-
sponses of any type to it. Completeness of the elicited
mating patterns were also reported but deemphasized by this
author. Components of the mating pattern were, on several
occasions, observed to occur out of normal sequence, followed

by responses which normally occur earlier in the chain. The

44

factors governing this out-of—sequence behavior are unknown.
The duration of elicited sexual behavior was also deempha-
sized because of the possibility of human error in measure-
ment. lNo reliability checks were possible with only one §_
recording behavior.

Perhaps some of the contradictory evidence on early
experience effects is due to differences among investigators
in the response measures they use. Different investigators
have used different measures to illustrate the effectiveness
of a particular stimulus, including the percentage of sexu-
ally reactive §s giving sexual responses to the stimulus
(Kruijt, 1962; Schein and Hale, 1959), frequency with which
component responses are elicited by this stimulus (Fisher
and Hale, 1957; Guiton, 1962), percentage of copulators
(Andrew, 1966), and sexual behavior scores based on level
and latency of the elicited responses (Carbaugh, Schein, and
Hale, 1962). Further work is necessary to understand rela-
tionships between these various measures.

Note of caution when classifying sexual behavior.--
The high degree of similarity between the aggressive, sexual,
and what appears as "orienting" responses in the male quail
creates a problem in response classification when such clas-
sification is based on visual observation of responses with-
out full knowledge of the effective eliciting stimuli. The
vocal call heard during courtship is also commonly heard dur-

ing fighting behavior. During pilot work, a live male was

45

often substituted for a female model and the behavior of the
gs was highly predictable. A g would move rapidly towards
the newly introduced male, spread its wings in a "fanning"
motion, and give a vocal call which was similar, but dis—
criminably different from the call given during courtship.
The characteristic tonus of the head, neck, and body, which
is observed during courtship, was not observed during these
apparently aggressive activities. The "orienting" response,
commonly observed in all §s during the first few seconds
after a model was exposed, could also be mistaken for the
neck stretch given during the courtship display (i.e., the
Neck and Body Tonus). The head is raised higher, however,
and it is not cocked to one side as it is during the mating
reaction. It appears as if the S were "looking over" the
stimulus situation. Aggressive, sexual, and orienting
behavior in the quail must eventually be defined operation-
ally in terms of the sets of stimuli that lead to their
occurrence (Selinger and Bermant, 1967). It is apparent
that future work of this nature may find it necessary to
rely on more objective means of observing the quails'

behavior, such as high-speed photography.

SUMMARY

The purpose of Experiment I was to expose those
features of the female Japanese quail which are important
for the elicitation of male sexual behavior. Experiment II
sought to determine what effect isolation from females,
prior to sexual maturation, would have on the stimuli which
elicit sexual responses from the adult male. Several
females and a male were taxidermically stuffed with various
parts of their bodies either missing or modified. These
stimuli were experimentally presented to two groups of males;
one group (Experiment I) was normally reared with other
females while the second group (Experiment II) was separated
from females just prior to sexual maturation and raised in
all-male groups.

Sexual responses were elicited from only a few quail
in Experiment I and no preference for a particular model was
shown. The results of Experiment II indicated that social
experience with females, up to the time of sexual maturation,
is sufficient to ensure female-elicited sexual behavior from
adult males. The quail in this study responded sexually more
often and to a wider range of stimuli than did the normally

reared quail. There was also evidence of an enhanced general

46

47

activity level. Stimulus specificity was also more apparent
in these quail; a preference for a female model with missing
tail feathers was shown.

Treatment effects in both studies may have become
contaminated with the unknown effects of head height. The
occurrence of homosexual behavior indicated that the quail
were responding in an abnormal manner, perhaps due to the
experimental design. It was suggested that the experimental
design did not provide optimal conditions for expression of
sexual behavior in either study. Further work is needed to
assess other ways of investigating the stimulus control of

this bird's sexual behavior.

REFERENCES

REFERENCES

Andrew, R. J. (1966) Precocious adult behavior in the
young chick. Anim. Behav., 14, 485-500.

Beach, F. A. (1948) Hormones and Behavior, Hoeber, New
York.

Beach, F. A., and Inman, N. G. (1965) Effects of castra- D?
tion and androgen replacement on mating in male quail. ‘
Proc. Nat. Acad. Sci., 54, 1426- 1431.

Carbaugh, B. T., Schein, M. W., and Hale, E. B. (1962) \/I
Effects of morphological variations of chicken models on
sexual responses of cocks. Anim. Behav., 10, 235-238.

Domm, L. V., and.Davis, D. E. (1948) Sexual behavior in \///
intersexual domestic fowl. Physiol. 2001., 21, 14—31.

Fabricus, F. (1962) Some aSpects of imprinting in birds.
Symp. 2001. Soc., Lond., No. 8, 139-148.

Farris, H. E. (1964) Behavioral development, social orga—
nization, and conditioning of courting behavior in
Japanese quail, Coturnix coturnix japonica. Unpublished
doctoral dissertation, Michigan State University.

Farris, H. E. (1967) Classical conditioning of courting, //
behavior in the Japanese quail, coturnix coturnix '
japonica. g, exp. Anal. Behav., 10(2), 213— 218.

Ficken, M. S., and Dilger, W. C. (1960) Comments on
redirection with examples of avian copulations with
substitute objects. Apim. Behav., 8, 219—222.

Fidura, F. (1967) Personal communication. Dr. Fidura
was the caretaker of the quail colony at the time of
these studies.

Fidura, F. G., and Gray, J. A. (1966) Visual discrimina-
tion of color, pattern, and form in the Japanese quail
(Coturnix coturnix japonica). Psychon. Sci., 5(11,
427-428.

48

49

Fisher, A. E., and Hale, E. B. (1957) Stimulus determinants
of sexual and aggressive behavior in the male domestic
fowl. Behavior, 10, 309-323.

Goodwin, D. (1948) Some abnormal fixations in birds.
Ibis, 90, 45-48.

Guiton, P. (1961) The influence of imprinting on the
agonistic and courtship responses of the Brown Leghorn
cock. Anim. Behav., 9, 167-177.

Guiton, P. (1962) The development of sexual responses in
the domestic fowl in relation to the concept of imprint-
ing. Symp. 2001. Soc., Lond., 8, 227-234.

Hale, E. B. (1960) Role of head height in releasing sexual
versus fighting behavior in turkeys. Anat. Rec., 138,
354-355. (Abstract)

Heinroth, D. (1910) Beitrage zur Biologie, namentlich
Ethologie und Psychologie der Anatiden. Verh. 5th Int.
Orn. Kongr., Berlin, 589-702. (Cited in Fabricus, 1962)

 

Hinde, R. A. (1965) The integration of the reproductive
behavior of female canaries. In Beach, F. A. (Ed.)
Sex and Behavior, John Wiley, New York.

Hinde, R. A. (1966) Animal Behaviour: A_Synthesis pf
Etholggy and Comparative Psychology. McGraw-Hill,
New YOrk.

Kruijt, J. P. (1962) Imprinting in relation to drive
interactions in Burmese Red Junglefowl. Sy p. 2001. Soc.,
Lond., No. 8, 219-226.

Kruijt, J. P. (1964) Ontogeny of social behavior in
Burmese Red Junglefowl. Behaviour, Supplement 12.

Lack, D. (1941) Some aspects of instinctive behavior and
display in birds. Ibis, Ser. 14, 5, 407-441.

Lindquist, E. F. (1953) Design and Analysis 2f_Experiments
ip_Psychology and Education. Houghton Mifflin Company,
Boston.

Lorenz, K. Z. (1935) Der Kumpan in der Umvelt des Vogels.
g, Ornitbol., Leipzig., 83, 137-213, 289—413. (Cited by
Marler and Hamilton, 1966.)

50

Marler, P., and Hamilton, W. J., III. (1966) Mechanisms
‘g§_Animal Behavior. John Wiley and Sons, Inc., New
York.

 

Padgett, C. S., and Ivey, W. D. (1959) Coturnix quail as
a laboratory research animal. (abstract) Science, 129,
267-268.

Rfiber, H. (1948) Analyse des Balzverhaltens eines
domestizierten Truthahns (Meleagris). Behavior 1,
237-266. (Cited in Fisher and Hale, 1957.)

Ratner, S. C., and Denny, M. R. (1967) Comparative
Psychology: Research ip Animal Behavior. The Dorsey
Press, Homewood, Illinois.

Reese, E. P., and Reese, T. W. (1962) The quail, Coturnix
coturnix, as a laboratory animal, J, exp. Anal. Behav.,
5, 265-270.

Schein, M. W. (1963) On the irreversibility of imprinting.
Zeits f, Tierpsychol., 20, 462—467.

Schein, M. W., and Hale, E. B. (1959) The effects of early
social experience on male sexual behavior of androgen
injected turkeys. Anim. Behav., 7, 189-200.

Schein, M. W., and Hale, E. B. (1965) Stimuli eliciting ./
sexual behavior. In Beach, F. A. (Ed.), §§§_QQQ
Behavior. John Wiley and Sons, Inc., New York,
Chapter 18, 440-482.

Schoettle, H. E. T., and Schein, M. W. (1959) Sexual
reactions of male turkeys to deviations from a normal
female head model. Anat..B§£,. 134, 635. (Abstract)

Schutz, F. (1965) Sexuelle Pragung Bei Anatiden. Z,
Tierpsychol., 22, 50-103.

Scott, J. P. (1958) Animal Behavior. Univ. Chicago Press,
Chicago. ,

Selinger, H. E., and Bermant, G. (1967) Hormonal control
of aggressive behavior in japanese quail (Coturnis
coturnis japonica). Behav., 28, 255-268.

Siegel, S. (1956) Nonparametric Statistics, McGraw-Hill,
New York.

Tinbergen, N. (1942) An objectivistic study of the innate
behavior of animals. Biblioth. biotheor., I, 39-98.
(Cited in Fisher and Hale, 1957.)

51

Warriner, C. C., Lemmon, W. B., and Ray, T. S. (1963)
Early experience as a variable in mate selection. Anim.
Behav., 11, 221—224.

Wetherbee, D. K. (1961) Investigation of the life history
of the common Coturnix. The American Midland Naturalist,
65, 168-186.

Wood-Gush, D. G. M. (1958) The effect of experience on the
mating behavior of the domestic cock. Anim. Behav., 6,
68-71.

APPENDIX A

APPARATUS

APPENDIX A

APPARATUS

  

Observer's view of model

 

Pulley and boom system above
the superstructure

52

“w..." .' .vJ

Raising shroud above model ,

 

 

§_with recording equipment

APPENDIX B

MODELS

APPENDIX B

MODELS

 

 

 

 

 

 

Headless Female, Black Chest Headless Female

 

 

 

 

 

’__ ._ _—-— _

Full Male Full Female

53

 

 

Tailless Female

54

 

 

 

' ,

Headless-Tailless Female (below)
and Female Head (above)