Calhoun, John B. - A Comparative Study of The Social Behavior of Two Inbred Strains of House Mice

A Comparative Study of the Social Behavior of Two Inbred Strains of House Mice
Author(s): John B. Calhoun

Source: Ecological Monographs, Vol. 26, No. 1 (Jan., 1956), pp. 81-103
Published by: Ecological Society of America
Stable URL: http://www.jstor.org/stable/1943578
Accessed: 05-07-2015 01:50 UTC
Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at http://www.jstor.org/page/
info/about/policies/terms.jsp
JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of content
in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new forms of scholarship.
For more information about JSTOR, please contact support@jstor.org.
Ecological Society of America is collaborating with JSTOR to digitize, preserve and extend access to Ecological Monographs.
http://www.jstor.org
This content downloaded from 134.114.138.130 on Sun, 05 Jul 2015 01:50:10 UTC
All use subject to JSTOR Terms and Conditions
A CO.MPARATIVE STUDY OF THE SOCIAL BEHAVIOR OF
TWO INBRED STRAINS OF HOUSE MICE'
JOHN B. CALHOUN
National Institute of Mental Health, Bethesda 14, Maryland
TABLE OF CONTENTS
PAGE PAGE
INTRODUCTION ........... ...................... 81 Intensityof activity........................... 91

MATERIALS AND METHODS ........................ 82 Quantitative aspects of hierarchial organization . 91
C57 neutral aggregates, Colony 2............. 92
THE STANDARDIZED ENVIRONMENT .............. 82
DBA neutral aggregates, Colony 2.............. 93
GENERAL METHOD OF ESTABLISHING COLONIES .... 82
Behavioral aspects of social interaction......... 93
BRIEF HISTORY OF COLONIES ................... 83
C57 Colony 2................................ 93
DBA /2 Colony 1 A ....... .................. 83
DBA Colony 2............................. 94
DBA /2 Colony 1 B ....... .................. 83
Places of aggression and retreat . . 94
DBA /2 Colony 2 ........ .................... 83
C57 Colony 2.......................... 94
C57BL /10 Colony 1 A ...... ................ 83
DBA Colony 2.............................. 95
C57BL/1O Colony 1 B ....................... 83
C57BL /10 Colony 2......................... 83 DISCUSSION ................................... 95
Previous pattern of living ...................... 84 Physiological stability .................... 95
Behavioral comparison,C57 with DBA .......... 96
RESULTS ........................... 84 97
Consequencesof social interaction...............
Initial history of DBA Colony 1 A ............. 84 97
Change in mood..............................
Initial history of C57 Colony 1 A .............. 85 Activityand emotion.......................... 98
Other strain specific behaviors . . 85 Behavioral changes and physiological homeostasis 99
General account of DBA Colony 1 A ............ 86 Concluding assumptionsand observations........ 99
General account of C57 Colony 1 A ............. 87 Homeostasis and behavior, a theory............ 100
General account of C57 Colony 1 B .............. 87 General implications ........... ............... 101
General account of DBA Colony 1 B ............ 88
SUMMARY ..................................... 102
Size and dispersion of groups . . 88
Colony 2 interactions; quantitative aspects ...... 90 LITERATURE CITED ............................ 102
INTRODUCTION ruentsprovidedthat the nmemibers withina particular

The pattern of social organization,exhibitedby population have similar heredity,and that there
any group of organismsat some specifiedtime, is are marked differencesin hereditarycharacteristics
presumedto have resulted from the interactionof betweenpopulations. In such cases the interference
three sets of factors: (1) the heredityof the indi- of cultural factors may be minimizedby initiating
viduals composingthe group; (2) the conditionsof the population with a single pregnantfemale. This
the externalenvironment whichhave impingedupon female's prior social experience can be limited to
the membersof the society during their life-time; living under caged conditionswith her motherand
and (3) the natureof the culture,throughwhichthe sibs to weaning,and cohabitationwith a male for
experiencesof prior generationsaffectthe contem- timeto insureconception.Where such pre-
sufficient
cautions are taken regardingthe environmental and
porary one. Any study of the ecology or sociology
cultural variables, most differences,which may de-
of natural populations, including that of man, is
beset with the difficulttask of assessing the role of velop betweentwo such populations,may be attrib-
and utedto theirdifferences in heredity.
thesethreedeterminants(heredity,environment,
culture). The objective of the present study was to deter-
The role of hereditymay be studied in experi- mine the extentto whichhereditymightmodifyso-
mental populations establishedin identical environ- cial behavior. A clear cut demonstrationof this
1 This study was conducted at the Roscoe B. Jackson Me-
objectivedemandedthat two strainsof a single spe-
morial Laboratory. It was made possible by the facilities of the cies be selectedfor use. These strains should differ
Jackson Laboratory, the advice of its staff, and by a Special
Fellowship from the National Institute of Mental Health. in certain characteristicssuspected of relevancyin
82 JOHN B. CALHOUN Ecological Monographs
Vol. 26, No. 1
the expression of social behavior. House mice (Mins four nest boxes were placed on each shelf. Access to
musculus) were selected as subjects. Many inbred each shelf from the floor was by way of a narrow
strains of this species are available at the Jackson ramp. Thus there was choice of 32 nest boxes scat-
Laboratory. A survey was made of the characteristics tered over eight shelves. This provided wide lati-
of these strains. Two strains, DBA/2 and C57BL/ tude of choice of place of residence, although some
10, differedmarkedly in two characteristics, suscepti- nest boxes were farther from the food pen than
bility to audiogenic seizure (Fuller & Williams 1951) others.
and incidence of inammary tuinors (Griineberg 1952). Four pens were constructed as nearly identical as
DBA/2 mice have a high incidence of mammary the available laboratory room permitted. Each was
tumors, and a marked susceptibility to lethal audio- isolated by a meter high fence over which the mice
genic seizures. In contrast C57BL/10 mice have could not climb. Details of this environment are
both a low incidence of mammary tumors, and a shown in Figs. 1 and 2. Other than the braces sup-
low susceptibility to audiogenic seizures. This sug- porting the harborage stand, the floor of the larger
gested that DBA s are physiologically unstable and pen was unstructured. The temperature of the rooms
C57s are physiologically stable. Exposure to a more were under thermostatic control at either 65 or 72
complex physical and social milieu, than previously F. The light-darkness cycle was reversed. Thus the
experienced in many prior generations of rearing in major period of activity of these nocturnal animals
small cages, might be expected to provide a marked was between 6 :00 A.M. and 6 :00 P.M. Dim lights
taxing of the physiological homeostatic mechanisms approximating moon light were on between 6:00
available to the mice. In a complex environment the A.M. and 6 :00 P.M. while several 100 watt lamps
homeostatic mechanism of the physiologically unstable were on during the other twelve hours. These brighter
strain should be upset more readily than that of the lights were also turned on during the day hours
physiologically stable strain. Alterations of motiva- when observations were recorded.
tion and emotion might be anticipated from aberra- GENERAL METHOD OF ESTABLISHING COLONIES
tions of physiological homeostasis. If motivation and
An attempt was made to make the introduction into
emotion are altered one might further anticipate
the more complex environment a gradual one. One
changes in the social relations between individuals.
end was removed from the wooden breeding cage
This rather loose theoretical conceptualization served
as the basis of the decision to use DBA /2 and C57-
BL/10 mice as subjects.
Since the main objective was to investigate the
role of heredity on the expression of social behavior, 4
physical environment and culture had to be con-

trolled. Culture was minimized by initiating colonies
with mice whose past experience was confined to lB INCHES4
living in small cages with their parents and sibs. The

physical environment as a variable was controlled to
the extent that members of each strain were intro-
duced into identically structured environments.
MATERIALS AND METHODS I~~~~~~ 2
THE STANDARDIZED PHYSICAL ENVIRONMENT

Two criteria served as a basis for structuring the
environment. First, there should be a single location
to which all individuals must go from time to time.
Here contacts between mice should be maximized.
HARBORAGE
STAND
Second, a large number of places of abode should be
FIG. 1. Harborage stand. The shelves at the four
provided. This would enable choice of companions. levels were numberedfrom 1 to 4; the lowest shelf was
The first criterion was realized by constructing a No. 1 and the highest No. 4. A 37 mm wide, screen
small pen in which containers for food, water, and covered, wooden ramp connected the inner end of each
nesting material were located. Contact at this loca- shelf with the floor. Four wooden nest boxes, 150 mm
on a side with a 25 mm opening on one side, were
tion was further accentuated by providing a single
equally spaced along each shelf. The 16 boxes on the
passage through the screen wall of this pen. This four left hand shelves were removedleaving these shelves
passage was large enough for only one mouse to pass bare during the history of DBA /2 Colony 2 and C57-
at a time. Thus the latter was one point to which BL /10 Colony 2.. Each nest box on the end of a shelf
nearest its ramp was designated as No. 1. The other
all individuals must come. This pen, which shall be three nest boxes were numbered 2, 3, and 4, with No.
designated for brevity as the food pen, was located 4 being at the end of the shelf farthest from its junc-
in a much larger pen. The second criterion was met ture with a ramp. Each box could be designated accord-
by placing within this larger pen a set of shelves on ing to the level of the shelf, and its location along the
shelf from the ramp. Thus 1-1 was the box nearest
which nest boxes were placed. There were four the floor,and 4-4 was farthestfrom the floor. Each of
levels of shelves; two shelves were at each level; and the 16 positions was duplicated.
January, 1956 THE SOCIAL BEHAVIOR OF Two INBRED STRAINS OF HOUSE MICE 83
SOUTHWEST PEN was available at a hopper througha 20x150 mm slot.
SOUTHEAST
PEN
[SF A
4
X
3
DBA12 Colony 1 B. Since populationincreasewas
not proceedingthroughreproductiona larger colony
N
4-FENCE
was establishedby the followingmethod. Four pairs
no
of DBA mice gave birthto youngon March 20, 1949,
175 F
in breedingcages. One each was suspendedon March
X g if A
PASSAGE
21 betweenthe two shelvesat each level on the stand
in the,northeastpen so that a 25 mm door bared
-FENCE
175 FT_
throughone end of each box providedaccess to an
adjoining shelf. On April 13, 1950,the door connect-
ing the southeast pen and the northeastpen was
BRACE
SUPPORTIN
'/- IS
2SX
opened. This permittedthe minglingof the nine Col-
ony 1 A micein the southeastpen and the eightadults
STANGD5
and theirthirteenyoung of the northeastpen. This
S 2 combinedgroup of 29 micewas designatedas Colony
1 B. Observationswere continuedto July 17, 1950.
NORTHEAST OBSERVATION 2
PEN BOOTH 3S S
4
All food and water afterApril 20 was available only
4S WALL+
in the food pen of the southeastpen.
4FENCE WALL [El DBA 2 Colony 2. This colony was establishedin
the northeastpen. Two litterswithonlytheirmothers
NORTHWEST PEN present were born on October 12, and 13, 1950.
FIG. 2. Diagram of the floorplan of the pens in which These litters consistedof nine males and four fe-
the harborage stands were placed. The east pens were males. They were providedaccess to the larger pen
mirrorimages of the west pens. N=hopper of nesting on November15. Nest boxes were located only on
material. F=hopper of food. X's represent the points
at which ramps to the shelves on the harborage stand the four shelves on the rightside of the harborage
were attached to the floor. They are numbered in ac- stand; the left shelves were bare. This procedure
cordance with the level above the floor to which they fostered more frequent interaction. The mothers
led. Those X 's numbered 1 to 5 are for the position were removedtwo days later. Observationswere con-
of ramp bases in a set of experimentsin which the
stands consisted of five nine-foot long shelves. These tinued to January 26, 1951. A food hopper with a
will be reportedin another paper. Note their proximity 300 x 300 mm vertical feeding surface was used.
to the wall. The cross hatched area representsthat por- C57BLI10 Colony 1 A. This colony was estab-
tion of the floorshaded by the shelves above. The black lished in the southwestpen. A litter of three S S
rectangles are at the positions where the 150 x 150 x
300 mm nesting boxes, which also contained food and and three ?? was born on December 1, 1949. On
water, were placed when mice were allowed to explore January5, 1950, at 35 days of age the six miceborn
the 5 x 9 stand. Note that the floor of the pens are in the food pen were provided access to the larger
relatively unstructured. The food pen was removed pen. The motherwas removedJanuary 12. Ob-ser-
from the northwestpen for C57 Colony 1B, and from
the northeastpen for DBA Colony 1B. vations were continuedto April 13, 1950. Fourteen
young of two litterswere born on March 17 and 20
in which the pregnant female or the adult had been and all survived for 60 or more days. Powdered
living. This cage with its inhabitants was placed in Purina chow was available at a hopper througha
the food pen. The door to the larger pen was kept 20 x 159 mm slot.
plugged until after the litter was born and had C57BLI10 Colony 1 B. Four pairs of mice with
reached nearly weaning age. Thus the initial experi- littersborn on March 20, 1949, were introducedon
ence of the mother, or both parents if the male was to the harborage stand of the northwestpen on
included, was only with the contents of the food pen. March 21 by an identicalproceduredescribedabove
The father was usually removed shortly after the for DBA/2 Colony1 B. On April 13, 1950, the door
birth of the litter. The mother remained with the connectingthe southwestpen and the northwestpen
young for the first few days they were given access was opened. This permittedthe minglingof the 20
to the harborage stand in the larger pen. Colony 1 A mice in the southwestpen and the eight
adults and their 25 young of the northwestpen.
BRIEF HISTORY OF THE COLONIES This combinedgroup of 53 mice was designatedas
DBA12 Colony 1 A. This colony was established Colony 1 B. Observationswere continuedto July
in the southeast pen. A litter of three S S and three 17, 1950. All food and water after April 20 was
if was born on October 10, 1949. On November 1, available only in the food pen of the southwestpen.
1949, the mother and her six young were released C57BL/10 Colony 2. This colony was established
from the food pen, and on November 19, 1949, the in the northwestpen. Two litters with only their
mother was removed. Between November 19, 1949, motherspresentwere born October8 and 16, 1950.
and April 13, 1950, this colony essentially consisted These littersconsistedof eleven males and four fe-
of these six mice. Although ten litters were born males. They were providedaccess to the larger pen
only three members of the seventh litter, born on on November15. The motherswere removedtwo
March 22, 1950, survived. Powdered Purina chow davs later., There were no nest boxes on the four
8484 B. CALHOUN
JOHN B. CALHOUN Ecological Monographs
JOHN Vol. 26, No. 1
shelveson the left side of the harboragestand. Ob- On the following,the 12th day, a nest had been built
servationswere continuedto March22, 1951. A food in the box in whichthe food had been placed. This
hopper witha 300 x 300 mm verticalfeedingsurface nestwas composedonlyof thosewood shavingswhich
was used. the mice had previously deposited in the corners
of the food pen, despite the fact that each trip with
PREVIOUS PATTERN OF LIVING
a wood shavingrequiredpassing around the pile of
For over 100 generationsprior to the present shavingsin the centerof the pen. No feces or shav-
studies the ancestorsof each of the inbred strains ings were in the othernest box in whichno food had
had experienced a very simple environment.At previouslybeen placed. On the basis of all later
weaning,a brotherand a sister were placed in an evidence,there should have been feces in this box
open topped woodenbox 150 x 150 x 300 mm. Wood if the mice had entered and remained even mo-
shavingscoveredthe floor,and a screencoverformed mentarily.It was not until15 days afterthisthatthe
theroof. A portionof the screencoverwas depressed fear of this second box had amelioratedsufficiently
to forma food hopper. A glass tube froma water for the remainingfemale to enter and remain long
bottle protrudedthroughthis lid. Both parents re- enough to defecate. Once this change in behavior
mained with the young until they were removedat occurred,this second bQx became the principleplace
weaning. This practice curtailedthe opportunityof of defecation.A litterof threemales and threefe-
the behaviorexhibitedby membersof one generation males was born on the 15th day of confinement in
affectingthat of succeedinggenerations. Artificial the food pen. The adult male was removedon this
selection,beginningbetween1909 and 1920 by C. C. date. When these young were 33 days old their
Little, on the progenitorsof these two strains,was motherhad a second litter in-the nest box in the
directedtoward developing a strain with a higher food pen with her first litter. On this date the
incidenceof mammarytumors(the DBAs), and one motherwas removed. It must be rememberedthat
with a low incidence of these tumors (the C57s). the adult male was removeda few hours after the
Little natural selectioncould have taken place under birth of the firstlitter. Thus under the conditions
these conditionsof rearing. On the otherhand, one of lactationand augmentedstressfroma more com-
mightanticipatethat there would be chance loss or plex environment, gestationwas prolongedfrom 11
fixationof genes not concernedwith productionof to 14 days beyond the normal period. GrUneberg
mammarytumors. Also mutationshad the oppor- (1952) reportssimilarprolongationof gestationdur-
tunityof being preserved. Thus the genotypesof ing lactation. All membersof this second litterwere
these two strainsmay well have lost genes requisite partiallyeaten beforethemotherwas removed.Three
for survivalin a morecomplexenvironment; i.e., one days prior to the birthof the second litterthe pas-
whichrequiresgreatercomplexitiesor intensitiesof sage throughthe food pen was opened. Ten days
social interaction. Likewise some genes may have later many feces were spread over all eight shelves
been fixed which are detrimentalto survival in a of the harborage stand. From the lowest to the
more complex environment. highestshelfthe count was 297, 117, 160, 165. Each
of the two nest boxes nearesta ramp on the lowest
RESULTS
shelf level containeda single fecal bolus. No other
INITIAL HISTORY OF DBA COLONY 1 A boxes contained feces. On this date the two nest
Two wooden nest boxes similar to those on the boxes in the food pen were removed. That night
harboragestand were also placed in the food pen. wood shavingsfromthe food pen were transported
One was to the leftand the otherto the rightof the into the box nearestthe ramp on one of the two low-
open-endedwoodenbreedingcage in whichthe colo- est shelves. Despite frequent visitations to both
nizing pair was introduced. During the firstnine shelves at each of the four levels, as indicated by
days manyfecesweredepositedabout the pen, mostly defecation,no feces were deposited in any but the
at the periphery,but none in eitherof the nest boxes. one box containingthe nest duringthe next 14 days.
The mice continuedto reside in their open-ended From the-lower throughthe upper shelf the total
breedingcage exceptfor periodsof wanderingabout. feces were 212, 103, 154 and 155. Thus there were
On the 10thday this breedingcage was removed;the approximately9 feces per mouse per day deposited
containedwood shavingswere placed in a pile in the on the stand duringthe first24 days of exposure. On
centerof the food pen; the two wooden nest boxes this 24th day the six mice were captured, and
were moved250 mm apart and the lid of the breed- weighed. A stopper was placed in the door of the
ing box placed across them. On the followingday box containingtheirnest. Each mouse was replaced
therewerestill no feces in eitherof the boxes. How- in this box followingweighing. The stopper was
ever,nestshad been built in two cornersof the food then removed. Two mice immediatelyleft this box
pen witha few of the wood shavings. A ball of food and ran up and downthe shelfon whichthisbox was
composedof equal parts of suet, dry bread, peanut located. Each repeatedlypoked its head in the door
butter,and raisins was placed in one of the nest to the adjoining box, but did not enteruntil several
boxes. The two mice were then caught and placed minuteslater. Once these mice experiencedentering
in this box. Later in the day each of the DBAs anotherbox withoutdeleteriouseffects,the behavior
were seen to enter this box after having left it. of theentiregroup towardthe otherboxes drastically
changed. During the ensuingseven days 3672 feces Within a week after release into the larger pens
(87 per mouseper day), were depositedon the stand. the food pen for both groups was restructuredby
Over 95% of these were in the boxes. Thus at 61 introducingtwo one gallon chickenwater fountains,
days of age following31 days of exposure to the a 150 x 150 x 300 mm box containingwood shavings,
harborage stand, adjustmentto it apparently was a food hopper containingground Purina laboratory
completed. chow.
INITIAL HISTORYop C57 COLONY 1 A OTHER STRAIN SPECIFIC BEHAVIORS
Two wooden nest boxes similar to those on the Several otherdifferencesin behaviorbetweenthese
harboragestand were also placed in the food pen. strainswere observed.
One was to the left and the otherto the rightof the 1. DBAs in the open colonysituationof Colony 1 A
open-endedwoodenbreedingcage in whichthe colo- exhibited poor success in rearing young. It was
nizing pair was introduced.By the end of the first thoughtthat the wood shavingsmightnot providean
day a few feces were in each of the nest boxes, and optimummaterial for heat conservation.Therefore
by the ninthday large depositsof feces were in both cottontow was mixed throughthe shavings in the
boxes. Feces were also depositedabout the periphery nesting material hoppers for both DBA and C57
of the pen. Thus in contrastto the DBAs, the C57s mice. The C57s immediatelymade use of both ma-
showed little avoidance or fear of these two nest terials. The DBAs selectedout and used only wood
boxes. However,except for enteringand defecating shavings. On one occasion three paper towels were
in the two boxes the C57s continuedto residein their placed on the floorof each pen. The C57s utilized
open-endedbreedingcage. On the 10th day altera- thesefor nestingmaterial,but the DBAs did not.
tions were made as for the DBAs on this date. In 2. In the Colony 2 for both DBA and C57, nesting
contrastto the DBAs, the C57s immediatelytrans- materialwas providedin the formof stripsof paper
portedwood shavingsinto one of the nest boxes. A towelingin an open box. For the 28 days following
ball of food was placed in this box as a controlpro- weaning at 33 days of age the two 150 x 150 x 300
cedureto the DBAs. These two micecontinuedto use mm breedingboxes in each food pen were left in
these two boxes during the next 60 days, one as a place while the young were adjusting to the har-
nest box and the other as a place for defecation. borage stand. Each box had a 25 mm diameterdoor
Since the female bore no young, the pair was re- and the floorof each was coveredwithwood shavings.
moved and replaced by a single near term C57 fe- During the 28 days the DBAs carriedin 20 stripsto
male. She built a nest in one of the two nest boxes one box and threeto the other. On the other hand
and used the otheras a place of defecation.Six days the C57s made large bulkynests of the paper strips
after introductiona litter of three males and three in both boxes. However,when these breedingboxes,
femaleswas born. When 35 days of age the young in whichthe mice were born,were removed,mice of
withtheirmotherwere given access to the harborage both strainstransportedpaper strips and made ade-
stand by unsealing the door from the food pen. quate nests in the boxes on the harboragestand.
During the firstfour hours followingaccess a total 3' When the 1 A Colonies were enlarged to form
of 53 feces had been deposited in four boxes, and the 1 B Colonies,a breedingcage was suspendedbe-
10 feces were scatteredover four shelves. By the tweenthe two shelvesat each level. Each cage con-
end of three days, 21 of the 32 boxes containeda tained an adult pair of mice with a one day old
total of 769 feces. Seventy seven other feces were litter. Just prior to suspendingeach cage a 25 mm
scatteredover the shelves. This gave an average of diameteropening was bored throughone end. This
40 feces per mouse per day deposited on the har- was to provideaccess to the adjoining shelf. Within
borage stand. In order-toinducethe mice to live on 15 minutesall four pairs of C57s had plugged these
the harboragestand, the lids of the two nest boxes holes with the shavings and tow coveringthe floor
in the food pen were removed. During the next of their cages. The DBAs also exhibitedthis hole
threedays nestswere built in two boxes on the har- plugging behavior. However, it took 3-4 days to
borage stand. These were in boxes nearestthe ramp reach the fullnessof expressionachievedby C57s in
on shelves1 and 2. Furthermore31 boxes contained 15 minutes.
1267 feces and 98 were scatteredover 7 of the 8 4. Casual observations were made during these
shelves. This gave 65 fecesper mouseper day on the periods of handlingon the frequencyof squealingor
stand. On the 6th day of exposureto thestand when the position followinganesthesia.
the youngwere 41 days of age, the two nest boxes in TABLE 1. Strain characteristics noted at time of han-
the food pen and the motherwere removed. During dling and anesthesia.
the followingweek 2898 feces were depositedin the
32 boxes, and 260 other on the shelves. This gave Did not Out on Out on side
an average of 75 feces per mouse per day on the Squealed Squeal Belly or back
stand. They were then 48 days of age. Thus it is DBA ....... 16 20 25 24
quite apparent that C57 mice made a much more
immediateadjustmentwhen exposed to a new and C57........ 11 57 81 18
more complex environment than did the DBA mice.
5. Vocalization: Whilefree rangingDBAs frequently is attributableto two conditions.First, neitherusu-

chirped. C57s nevervocalized. ally proceededto the food pen fromthe harborage
6. Tail rattling: only during fightsdid DBAs beat stand if the dominantmale was already in the food
their tails against the floor. C57s did likewise,but pen. Second, shortlyafterone of themdid go to the
they also exhibitedthis behaviorat any place where food pen, the dominantmale usually emergedfrom
theymightbe, even when alone. his nest box, proceededto the food pen, and chased
7. Exploratorybehavior: C57s frequentlyremained the subordinatemale back to the harborage stand.
in theirnest boxes for severalminutesat a timewith This difficultyin remainingin the food pen did not
their heads protrudingout the door. DBAs never affectgrowth. Between54 and 184 days of age the
did this. Periodicallyall feceswereremovedfromthe dominantmale was consistentlythe lightermouse;
shelves and nest boxes. The mice were usually 1407 was heaviest,and 1405 just slightlyheavierthan
weighedat this time and released after the cleaning the dominantmouse. Incrementsin weight during
was accomplished.Following release C57 mice usu- this period were respectively6.6, 6.7, and 6.0 gin.
ally spent several minutesinvestigatingall boxes on No antagonism,otherthan mild pushingwith the
all shelf levels. On the other hand the DBA mice forepawswhile at the food hopper, was ever noted
usually immediately retiredto a nest box. among the three females. Nor did they evince any
8. Position at food hopper: DBAs usually remained antagonismtoward any of the males which joined
withtheirhead facing the food hopper duringeach themin the food pen.
period of eating. On the otherhand the C57s usually The behaviorof the dominantmale 1409 to either
got a mouthfulof food and turned around, facing of the two subordinateswas quite otherwise.Upon
away fromthe hopper whileeating. reachingthe door of the food pen in which a sub-
9. Transportationof food in the 1 A Colonies: Occa- ordinate was present, he would stalk throughthe
sionallypellets of Purina laboratorychow was made door with legs extended,a tense tip-toeingstance,
available in the food pens. The DBAs only scattered and withhis tail arched stifflyupward. Immediately
themabout the food pen. In additionto thisthe C57s on enteringhe would dash at the subordinateand
carried many pellets up into the nest boxes on the engage in a tumblingcombat. Very soon the sub-
harboragestand. ordinate would break away, dash out of the food
pen with 1409 in close pursuit. The subordinate
GENERAL ACCOUNTOFpDBA COLONY 1 A
male was usually not able to go up a ramp imme-
Frequentlyand detailed observationswere made diately. Instead the wild chases continuedabout the
of the six mice of this colony when they were be- main floorof the pen. To anyone who has played
tween 94 and 184 days of age. Althoughtherewas hand ball, the actionsof the two mice,resemblenoth-
some antagonismamong the threemales between94 ing so much as two hand balls bouncingback and
and 108 days of age, the conflictswere not very forthbetweenthe four walls. Their paths of travel
markedin intensity.In fact all six mice mightbe in appeared to bear no relation to each other,except
the food pen simultaneously.After108 days of age that when,by chance,theycame withina few inches
tolerance betweenmales during their hours of ac- of one another,approach and brief combat ensued,
tivity completelyceased. The pattern established prior to resumingthe wild running. Eventuallythe
was that of completedominanceby one male, 1409, subordinatemouse escaped up on to the harborage
and subordinatebehaviorby the othertwo, 1407 and stand. After this escape the dominantmouse would
1405. Afterthe establishment of this patternthe two occasionallyexhibitthe motionsof fighting;that is
subordinates,1407 and 1405, were rarelyout of the is would move about on its hind legs thrustingwith
nest boxes at the same time. 1407 appeared to be its forepaws and shoulders. Subordinatemales also
dominantover 1405, but this judgmentwas derived exhibitedphantomfightingupon losing contactwith
more fromtheirrelationswith 1409 than with each the dominantmouse. After such periods the domi-
other. There was a greater elementof challengeto nant mouse occasionally attacked a female, even
1409 by 1407 than by 1405. Furthermore,the rela- pregnantones. These attackson femalesusually ter-
tivelyhigherrank of 1407 was indicatedby the fact minated rather rapidly withoutthe female fighting
that he resided more frequentlyon a lower shelf back. This was probably the origin of wounds ob-
near the food pen than did 1405 (Table 2). The servedon females.
lack of association betweenthese two subordinates Both sound and scentservedas stimuliin directing
TABLE 2. Number of times DBA Colony 1A mice were the actions of the dominantmale toward the two
found residing in the boxes at each shelf level. subordinates.The clatterof a mouse's toe nails on
the screencoveringof the ramp was even audible to
Level of shelf from floor... . 1 2 3 4 the observer. Such a sound elicited the appearance
of the dominantmale. Furthermoreif a subordinate
MOUSE
Dominant male 1409.... 32 5 0 0
mouse did make the trip to the food pen and back
Mid-ranking male 1407 ........ 26 10 1 2 to his nest box withoutbeing attacked,he was often
Low-ranking male 1405 ....... 16 10 8 11 soon afterwardattacked in his nest box by 1409.
The three females. 122 17 2 1 Within two to five minutes after the subordinate
mousehad reenteredhis nest box, 1409 would emerge
and go to the bases of the eight ramps. He would vidual or, at most,a few steps takentowardit. Corre-
investigate about these for a moment before pro- sponding movementsaway were made by the sub-
ceeding up the appropriate one and into the correct ordinatemouse. Even when a female was in estrous
box where a fightsometimes ensued. therewas only an occasional fight. On only two oc-
The dominant mouse always went directly from the casions were femalesseen in shortchases, and never
door of the food pen to the food hopper. On the did a male attack a female. Lashing of the tail
other hand, the two subordinate mice frequently against the flooror shelves was a commonbehavior
circled along the inner wall of the food pen, and cau-
of all mice, but I never detectedany regular asso-
tiously approached the hopper from the back. ciation of events to indicatethis behaviorwas asso-
During the twelve hours of light all six mice fre-ciated withan alteredemotionalstate.
quently slept together even though this period may Female 11 was noted to be receptiveto sexual ad-
have followed this intense fighting. However, there vances on two dates, once when she was 86 days
was some dispersion of the males in which the sub- old and again at 106 days, directlyafter birth of
ordinate one did not sleep with the females as did the litterconceived20 days before. On the firstdate,
male 1409. 1405 rather frequently lived alone on the lowest ranking male 1 was observed to mount
the top shelf farthest removed from the food pen, female 11 18 times during a span of 35 minutes.
and from the other mice. The other subordinate male, Duration of the first17 unions was 0.5 to 5.0 sec-
1407, more frequently resided nearer the floor, and onds, while the terminalunion lasted 10 seconds.
thus nearer the food, than did 1405 (Table 2). How- Preceding each of these the female usually ap-
ever, when he resided on the lowest level, he was proached the male. Particularlyduring the last of
always on the opposite shelf from that inhabited by this series,the femaleturnedand ran as soon as the
the three females and the dominant male, 1409. male noted her approach and turned toward her.
The reproductive performance of this colony was During the mountthe male grasped the female by
interesting. Ten litters were born, five to 1403, three
the nape of her neck with his teeth, and he also
to 1411, and two to 1401. The members of nine of grasped her firmlyabout the body withhis forelegs.
these survived for only a few hours. By the time Otherunionswerenotedto last as long as 60 seconds.
they were found, part would be dead, and partially Occasionally,while still united,the pair fell over on
eaten. Only three males of one litter survived. This theirsides.
was the third litter born to 1403. Both 1403 and The dominantmale 7 was not observed to ac-
1411 nursed them. However, at adulthood they only complishmountingsuccessfully.Female 11 resisted
attained 70% of the adult weight of 25 gm. It was his more vigorousand aggressiveapproaches. How-
my impression that the newly born young of these ever,she toleratedthe approaches of female 13, who
females were paler than is characteristic for newbornwas twiceobservedto mountfemale11 and maintain
miee. the graspingunion for several seconds. After these,
GENERAL ACCOUNT OF C57 COLONY 1 A female 13 chased male 1 away when he again ap-
Because of the later date of birth of the litter, proached female 11. Despite her homosexual be-
which initiated this colony, observations continuing havior,female13 conceivedthreedays later.
to April 13, 1950 covered the age span of only 56
Female 11 gave birthto a litterof 10 in the nest-
ing materialbox in the food pen at 106 days of age.
to 133 days. At 56 days of age the social hierarchy
Eight of these survived to adulthood. -Female 13
among the three males was already established. Male
gave birthto a litterof 6 in a nestingbox next to
7 was dominant to 4, while both were dominant to
the ramp on a lowest shelf level. All of these sur-
male 1. Expression of status consisted of leisurely
vived to adulthood. Each female assisted in the
chases which very rarely terminated in fights. This
nursingof the other'syoung. The other adult mice
slowness of pace was not due to lack of ability to
were oftenobservedin the nest withone or the other
run rapidly. As will be described later in regard to
of these litters. By 19 days of age the young were
the formation of Colony 1 B, rapid movements were
movingback and forthbetweenthe two nestlocations.
observed when strange mice initially contacted each
other. Occasionally chases were rapid, but never with GENERAL ACCOUNT OF C57 COLONY 1 B
the wild abandon and long duration of the DBAs. Colony1 B was formedon April 13, 1950 by open-
Even within a few seconds after a chase, both par- ing the small door betweenthe southwestpen, which
ticipants were likely to go to the food pen and eat contained20 mice previouslydesignatedas Colony
side by side. So little further intensification of ag- 1 A, and the northwestpen, which contained four
gressiveness took place between 56 and 133 days of littersand their parents. At the time of the union
age, that it is impossible to demonstrate any such and for a week thereafterfood and water was left
change. It will be recalled that after 108 days of available in the breeding box in which each pair
age intense strife developed among DBA males. The and their litter had been introducedon each shelf
relative social rank of the three C57 males persisted, level of the northwestpen. For the firsttwo hours
but its expression was primarily in the form of mild after union the six adults from the southwestpep
threats or mild avoidances. By mild threats is meant (the formerColony 1 A) roamed the floor of the
a mere rotation of the body toward another indi- northwestpen. Gradually partial trips were made
up the ramps,beforeturningaround and going back food was procured. Apparentlywhat happened was
to the floor. Such hesitant,partial, trips up ramps that the bottommembersof a feedingaggregatewere
had neverbeen observedin theirhomesouthwestpen. forced into the slot and suffocated.Although the
Finally some of the southwestmales got up to the survivorspartially ate these unfortunateassociates
shelvesof the northwest stand,or some of the north- the slot was never completelyunplugged,until re-
west males got down to the floor. As soon as males movedby Hinckley,and this was not everyday.
from the two pens encounteredeach other,intense GENERAL ACCOUNT OF DBA COLONY 1 B
fighting ensued. The intensity,duration,and pattern
Colony 1 B was formedon April 13, 1950 by open-
of fightingbecame extremelysimilarto that already
ing the small door betweenthe southeastpen, which
describedas typical for DBA Colony 1 A males.
contained the nine mice previously designated as
The southwestmales were always the winners. These
Colony 1 A, and the northeastpen, which contained
males had more experiencein both social interaction
four littersand theirparents. The moreexperienced
and in orientation of movements through this type
southeast adult mice (formerlyColony 1 A) in-
of environment.After a few such encounters,the
vestigatedthe floorof the northeastpen duringthe
threesouthwest males began similar intense fighting
firsthours after union but did not climb up the
among themselves.Such fightingamong these three
ramps of the northeaststand. No fightswere ob-
sibs had never been noted previously nor did it
served. On April 14 male 1407 was seen on the
occur on later dates. The southwestadult males still
northeaststand with one of its residentadult males,
attacked the northwestadult males on the day fol-
34. 1407 lost each of several consecutiveencounters.
lowingthe union. Althoughintensityof fighting per-
Four of thesix adult males died betweenApril14 and
sisted on this date its frequencydeclined. Some con-
June 8, 1950. Each of these four males during the
tacts did not result in fights. No furtherobserva-
few days before death was characterizedby many
tions were made until a week later. By this time
wounds and loss of weight. After the death of his
fightingwas greatlyreduced. Even the adult males
two brothers,the formerlysubmissive1405 became a
from the two pens could feed simultaneouslywith
highlyaggressiveindividualover the smalleryounger
littleshow of antagonism.Male 1 had becomesome-
males. By the timethe latterwere 53 days old, they
what of a tyrant and occasionally chased other mice
also were quite aggressive. Multiple fightsensued
fromthe food pen.
among themand the older mice in whichfiveto ten
There was no effective reproduction by the seven
males were simultaneously engagedin combat. There
adult femalesbetweenApril 13 and June 8. Two of
appeared to be no directionof aggressionof one indi-
the northwestfemales were near term on April 14
vidual towardanothermouse as a specificindividual.
and female 11 conceivedon this date. Though even All merelydashed wildlyabout the floorof the pen.
the latter carried her litter to term at least, none
When two contactedeach other there was a brief
of thesethreefemalesreared theselitters. The other
fight,followedby a simultaneousjumping up in the
fourfemales did not conceive.
air, and then separation and continuationof the
Some of the fightsbetweenmales on the day of running. Because of the jumpingup in the air upon
and the day followingunion of the pens took place contact the appearance of such a group resembled
in the suspended breeding cages. When this hap- popcorn popping.
pened both the young and the adult female were
Feeding aggregateswere smallerthan for the C57
sometimesforcedto abandon theirnest box to escape
mice. On only one occasion did a mouse die in the
the clatterof bodies bouncingback and forthinside
slot of the food hopper. It was removedbefore it
these cages. Fighting soon diminishedabout places
could contributemuch to reductionof food intake
of repose and membersof the two pens intermingled
by the other mice. None of the 25 mice alive on
in nest boxes. However,there was one situationin
June 8 survivedto June 28. At death they were
whichthe individualsprobablyexperiencedconsider-
usually not emaciatedbut precedingdeath theywere
able stress. This was at the foodhopper. Here it was
quite lethargic. In fact Hinckley reportedthat the
possible for only fiveor six mice to feed simultane- general state of lethargy was such that marked
ously. Yet 10 to 20 would crowd in at once. There fighting, apparent throughMay, was then absent.
was muchpushing,kicking,and piling on top of one
another. Althoughovertfightingrarelyoccurred,the SIZE AND DISPERSION OF GROUPS
young between30 and 60 days of age frequentlyre- At irregularintervalsthe boxes on the harborage
ceived ratherhard slaps fromthe forepawsof their stand were opened and the inhabitantsof each re-
elders. corded. The data forthe 1 A and 1 B coloniesof both
BetweenJune 9 and July 17 I was away fromthe strainswill be examinedfor thosedates duringwhich
laboratory. General observationswere made during the inhabitantswere 60 days of age or older. Some
thistimeby A. Dexter Hinckley. Thirtyfour of the of the mice were not includedin some of the counts
mice alive on June 9 succumbedduringthe next five since thev wereroamingabout the pen.
weeks. The remainderwere quite emaciated. Starva- The first analysis concerns sexual associations
tionappeared to be the major cause of death. One or (Table 3). Males or femalesmay be -alonein a box.
two mice at a timebecamewedgedin the narrowslot When two or more individualswere in a box, each
of the food hopper through which the powdered was consideredas having an association with each
other individual. There are N (N-1) /2 associations The second analysis concerns the size of groups
per group. These will be MM, MF, or FF. If the formed by these two strains of mice (Table 4). These
aggregations form at random without regard to sex data include a few aggregations formed in the nest-
the frequency of the three types of sexual associa- ing material box in the food pen as well as in the
tions should be proportional to a2 + 2ab + b2 when boxes on the harborage stand. Tests of significance
a and b are the proportions of males and females in of differences between these distributions of fre-
the population of which the groups are a part. For quencies of sizes of groups can not appropriately be
each colony a :b approximated 0.5:0.5. Therefore made because of differencesin the size of the popula-
on a random association of sexes a2 = b2 and 2ab = tions or because of difference in frequency of sam-
a2 + b2. For both C57 colonies the observed does pling them. In the equivalent sized 1 A populations
not differ significantly from the expected. What the average greater size of C57 groups is probably
little irregularities are evident indicate a slight at- biologically real. In the 1 B populations, C57s form
traction of the two sexes for each other. However, larger groups, both actually and relative to population
the observed frequencies for both DBA colonies differ size.
significantly from the expected. There are propor-
tionately too few MM and too many FF associa- TABLE 4. The frequencyof differentsized groups in
tions, and a slight deficit in MF association. There nest boxes and in the nesting material box.
remains the question of interpretation of the cause
of this difference. The key of this interpretation Strain...... DBA C57
lies in the greater number of DBAs which live alone
in comparison with the C57s. The DBA males in Colony ... | 1A lB 1A 1B
particular tend to live alone. This greater antagonism No. of Mice 6 24 6 51
among DBA mice, which will be further documented,
must give rise to sexual association other than on a Size of group NUMBER OF GROUPS
random basis. 1............. 104 22 22 16

2............. 28 11 0 6
TABLE 3. Frequency of sexual associations in boxes on 3............ 13 6 2 1
harborage stand. All mice 60 days of age or over. 4............ 7 1 7 1
5............ 0 1 9 2
6............ 3 2 1 0
Strain...... DBA C57 7.... ....... ..0o. ...
.. 3
.......
8. ........... .. 0 1
Colonv ... . 1A 1B 1A IB 9......... .. 1 .. 1
10...... .. .. .. 1
No. of Mice.. . 6 24 6 51 11....... .. .. .. 5
Per centFemale.. .5 .5 .5 .51 12.... .. . ... 1
Sexual 19 . .___ .. . 1_
Association
Total groups... 155 44 41 40
M only.......... 67 19 12 8
F only. . 30 10 10 7 Mean . 1.58 2.1 2.61 5.0
MM . 21 18 35 94
F F.... 52 42 34 112
M F.60 47 91 224
The third analysis concerns dispersion of groups.
EXPECTED SEXUAL ASSOCIATION Table 5 is based on all the data for 1 A and 1 B
colonies of both strains of mice. It is immediately
MM ........... 33.25 26.75 41.25 107.5 obvious from this table that all nest boxes were not
FF .. 33.25 26.75 41.25 107.5
MF 66.5 53.5 82.5 215.0 equally utilized. In general, the lower the shelf, or
the nearer the box to the ramp end of the shelf, the
Chi Square. 15.7 13.0 2.1 2.3 greater were the average size of groups, and the
greater were the number of groups found in each
P of x2 (Approx.) .001 .001 0.3 0.3
location. In other words, more and larger groups
formed nearer the food pen. This differentialusuage
of *space will be discussed in detail elsewhere (Cal-
The (O-e)2/e, Chi Square test, with two degrees houn 1956). As a result, the probability of one
of freedom used above is not exactly appropriate, mouse encountering another was greater in those
since the populations were sampled more than once. boxes nearer the food pen. However, this was ob-
The exact number of degrees of freedom could not viated in part by the fact that there were two shelves
be ascertained since the entire population was not at each level. Each box position was represented
included in all periods of sampling. Even so the twice. Thus it was possible for two mice to be to-
magnitude of the Chi Squares for the DBAs argues gether or separate at the same time and yet each be
for validity of the observed differences from the in the most favored position, that is in the box
expected. nearest the ramp on one of the two shelves at the
9090 JOH~iNB. CALHOUN
~~~~~~~~~~~~~~~~~~
26, No. 1
lowest level. This position is the critical one for a chance phenomenon, or that each member of the
judging attraction or dispersion. The six adults in population was equally attracted or repelled by each
each of the 1 A colonies provide the most accurate other individual. Under such conditions it still would
data. For C57s the average size of groups in this be anticipated that some pairs would have more
position was 3.95, but for DBAs it was only 2.06. associations than others. This hypothesis was ex-
amined with reference to the observed data. Let N
TABLE 5. Distribution of 179 groups of mice on the
= number of individuals in the population,
harborage stand.*
a = N (N-1) /2 = number of different kinds of
POSITION OF Box ON SHELVES FROM RAMP paired associations.
Level of _ Mean per box
accordingto
n = total paired associations observed.
shelffrom
the floor 1st 2nd 3rd 4th levelof shelf P = 1/a = probability of observing any two indi-
4th 26 6 5 1 9.5 viduals together.
1.54 1.83 1.00 2.00 1.53
Then:
3rd 16 2 2 0 5.0 PO = (i-p)n = probability of not observing pairs.
1.00 .00 1.15
1.19 1.00 P1 = P0 (n a.-i) = probability of observing pairs
2nd 27 7 4 3 10.3 once.
1.26 1.14 1.25 1.67 1.22
1st 96 32 20 9 39.3 P2 = P1 (ni) (1/a-1) = probability of observ-

3.11 2.25 2.80 1.44 2.67
ing pairs twice.
Mean pernest
box posi ion 41.3 11.8 7.8 3.3
alongshelf 2.37 1.98 1.55 1.33 P3 = P2( n ) (i/a-1) probability of observ-
*DBA colonies1A and 1B; C57 colonies1A and 1B. ing pairs three times.
All mice60 days of age or over.
For each of the 16 nest box positionsthe total groupsare listed in italics.
The mean size of the groupis listedbelow and to the rightof the italicised And the general case up to Pn is:
number.
Another indication of degree of attraction or dis- Pi Pi-- ( j ) (1/a -1)

persion was the use made of the nesting material
box in the food pen as a place of resting. For C57s: These equations were developed for me by Mr..
7 (mean size, 2.4) of the 22 groups recorded for the James U. Casby of the Army Medical Service Gradu-
1 A population were here; 4 (mean size, 26.7) of ate School. Utilizing these equations the expected pro--
the 40 groups recorded for the 1 B population were portion of N(N-1) /2 pairings observed 0, i, 2
here. For DBAs: None of the 100 groups for the 3 . . . 16 times out of n, observations, was calculated
1 A population were here; only one group of three for Colony 2 males of both C57 and DBA. The re-
mice of the 44 groups recorded for the 1 B popula- sults are shown in Figures 3 and 4.
tion was here. It will be noted from later remarks
that the food pen was the focal point for antagonistic .30 .
relations. The greater number of larger groups of C 57
C57 mice residing in the food pen is further evidence .25 -
N II
of the greater pacificity of this strain. With such a
....
~~~~~~~~N
(N -I) /2=55
n=383
differenceit might be anticipated that a greater pro- . 20
portion of the DBA groups would be elsewhere than
in the box position nearest the food pen. Such was
the case.
.10
TABLE6. Comparativedistributionof groups in boxes 0
near and furtherremovedfrom the food pen. cr05~~~~~~~~~~~~~~~~~~~~~~~C

Z5 10 1 08adoe
NUMBER OF GROUPS 0 5. 1015.8an.oe
Box 1 position Other15 positions

on lowestshelf of boxes
NUMBER OF ASSOCIATIONS PER PAIR OF MICE
C57... 18 16
FIG. 3. Frequency of association of male 057 mice
56 90 of Colony 2 outside of nest boxes. When there are 11
DBA . individuals, as in the present case, there are 55 different
combinations involving two individuals. If association
between any two individuals is purely by chance, there
COLONY 2 INTERACTIONS; QUANTITATIVE ASPECTS will be more associations between members of some
pairs than of others. The curve is based upon such an
The most parsimonious hypothesis concerning the assumption. Its poor fit to the observed data, shown
frequency with which two mice come in contact with by the histogram, indicates that associations between
each other outside of the nest boxes is that this was individuals.is not a chance phenomenon.
Results are shown in Tables 7 and 8. For both C57s
.30 and DBAs there was a single dominant mouse, whom
DBA
the others were never successful in challenging.
.25N9 Among the remainder there was a marked difference
N (N-I)/2 36
n =221 between C57s and DBAs. Among these C57s 50%
.20 of the dominant actions were by subordinates over
.. . . ..l
,l..'- .
superiors, while it was only 9% among these DBAs.
In other words, among the C57s there was consider-
..... ... ........
c , . .::::::::::
. .
.15
:;:: . .
A . ;; ... .. _
7:::::X. ,.
: . . . . . . . ......, able give-and-take with regard to status; while among
- He
~.
0 6
,. . . . . . . . . .
.
. . :::.
cr the DBAs there was not only a lack of challenging a

cr~~~~~~~~~~~~~~~~~~~~~~~~C superior, but also more of them failed to exhibit any
.', .'.'"..,. '. . . w
. '. :'..
' " ,. .'.'
Ia:: : ; - . . .:::::
::::: dominant reactions.
~05
0 0
The three types of actions, dominant, submissive,
and neutral, were taken into consideration in the
0 5 10 15 18 and over calculation of an index of social status, which was
NUMBER OF ASSOCIATIONS PER PAIR OF MICE designated as the DSN Score. Dominant reactions
FIG. 4. Frequency of association of male DBA mice were rated as 3 each, submissive ones as 2, and neu-
of Colony 2 outside of nest boxes. See legend of Figure tral ones as 1. A submissive reaction was rated
3 for additional remarks. higher than a neutral one, because the behavior of the
mice frequently indicated an element of challenge by
It is quite obvious that the observed data do not the subordinate mouse. This index, as shown in
approximate that based on the above hypothesis.
There were too many pairs with too few associations, TABLE 7. Social hierarchyamongC57 mice of Colony2.*
as well as too many pairs with too many associations.
This indicates that the associations among male mice SUBMISSTVE
MOUSE
are structured in some other way than that dictated Domi-
nant DSN
__ Total
Domi-
by the above hypothesis. There are three factors Mouse Score 315 311 282 286 319 303 317 321 290 288 301 nant
which may have contributed to the observed struc- 315..299 -14 17 13 9 5 11 9 4 2 4 88
turing of frequency of association of pairs:
282. . 175 1.. - 2 1 2 7 ... .. ..1 14
1. Some mice are more active. than others. If this 286.. 144 ...4 2 - 1 2 ...2 ....... 11
is the case the more active mice should have more 319Sb 4
....14 3 2 3 5.... 8
associations with each other than with the less 303. 97 .... 1 3 1 1
64 - .17
I 31 ... . 7
317....85 .....6 .... 1 2- 1........10
active ones. 321. 78 1.. 1 1 .... 3...- 1 .....7
2. Certain mice mutually attract each' other, or 290. 44 ... 1 1 .... ....... .... . . 2
288....31 .. .... . ...I . ...... 1..
certain mice are attracted by certain other ones.
301....27 ..1.. .1I... .
.... .... ....
3. Certain mice mutually avoid each other, or cer-
tain mice are avoided by others. Total
Submissive.... 0 25 38 29 13 18 19 13 5 3 5
These three possibilities will be examined.
Total Neutral.. 35 72 57 53 64 40 17 31 28 22 14
INTENSITY OF ACT'IVITY
No accurate measures of spontaneous activity, that *Data listedin mainportionof table are numbersof interactions
betweeneach
pair of mice.
were independent of the social setting, were made.
However, there was wide variability in the number
of interactions engaged in by differentmice (Tables TABLE 8. Social hierarchyamong DBA mice of Col-
ony 2.*
7 and 8). No decision can be made as to the possible
effect of differential levels of activity on the fre- SUBMISSIVE MOUSE
quency with which pairs were observed together. A Dominant DSN Total
relevant datum, whose importance I failed to realize Mouse Score 257 275 267 278 255 280 273 259 261 Dominant
and to obtain, was a time sample of the frequency 257....... 284 26 9 22 8 9 5 5 3
- 87
each mouse was out of the nest boxes. Such data 275 .
..... 3 - 10 4 1 4 4 3 26
would have answered the question of activity level, 267.......76 .....- 1 ....1.... 1 3
278T......l 70 .... 1....-... .. ..... 1
and would have been of considerable help in analyz-
255......51 1 1....-I.............1 3
ing the social hierarchy. 280......49 .. ......-.... ... ...... 0
The questions of avoidance and attraction will be 273......49 .1. .... .. ........- 1
treated in several of the following sections. 259......35.... .... .... ...... 0
261......28....... .... .... ...... - 0
QUANTITATIVE, ASPECTS OF HIERARCHIAL
Total Submissive... 1 27 20 28 9 13 9 9 5
ORGANIZATION
Whenever two mice were observed to approach Total Neutral....... 21 31 27 11 24 23 28 17 18
within a few inches of each other, the action of each
was recorded as neutral, dominant or submissive. *Data listedin mainportionof table are numbersof interactions
pair of mice.
betweeneach
Vol. 26, No. 1
Tables 7 and 8, indicates rank better than does con- C57 NEUTRAL AGGREGATES, COLONY 2
sideration of dominant actions alone. An analysis of Neutral aggregates at or near the food hopper
each day's observations indicated that there was no were analyzed (Tables 10 to 12) to determinethe in-
change in social rank of the C57 Colony 2 or the fluenceexertedupon themby the two mostdominant
DBA Colony 2 from the time observations were ini- mice Nos. 315 and 311. A neutralaggregatewas one
tiated until they were terminated: 91 to 106 days of in whichtwo or more mice were in close association
age for DBA; 122 to 169 for C57s. near the food hopper for a minuteor more with no
more show of animositythan that of mild mutual
TABLE 9. Strain differencein aggressiveness between pushing with the forefeet. The dominant315 was
the DBA Colony 2, and the C57 Colony 2. much less likely to be a memberof neutral aggre-
gates (Table 10) than was the secondrankingmouse,
Neutral Non-neutral 311, who frequentlychallengedhim unsuccessfully.
ions
Interact' Interactions
TABLE 10. 34 C57 neutralaggregatesin thefoodpen,
DBA . . 100 121 Colony 2.
C57..... 217 168
No. 315 No. 311
Present
... 10 20
The difference between C57 and DBA mice is
Absent.... 24 14
further revealed by comparing the neutral actions to
the more dynamic dominant or submissive ones. The
number of neutral actions by a mouse may be taken
TABLE 11. The social controlof neutralaggregatesin
as a Passiveness Index, while the DS component of thefoodpen amongC57 mice,Colony2.
the DSN Score may be taken as an Aggressiveness
Index. These are plotted for each mouse in Figure AgtMALPRESENCES
(No. of Aggregates)x
5. For C57 mice there are corresponding increases in (Mean No. per Aggregate)
Dominants Mean
these two indices as the number of interactions per Present No. of size of
mouse increases. However, with DBA mice the Pas- Aggregates Aggregates Observed Expected*
siveness Index does not show a corresponding in- 311 only...... 15 4.0 60 30
crease. That is, as the frequency of the interacting 311 & 315.... 7 3.0 21 33
315 only...... 3 4.3 13 30
increases, the DBA mouse is relatively more aggres- Neither 8 3.3 26 27
sive. Table 9 points out this differencein aggressive-
Total ...... 43 ... 120 120
ness.
*Based on chanceencountersdevoidofattractionor repulsion.
I I I I
C 57_- Neutral aggregates may be characterizedby the

D BA U-
75 - presenceof both of these mice, neitherof them,or
60~~~~~~ only one of them. Several aggregates of each of
these four categorieswere observed (Table 11). If
there was neitherattractionnor repulsion between
X 45 - individuals,the total numberof timesmice were re-
w
a corded in each of these four categories should be
z
- proportionalto_the numberof different individuals
em3 315
on in the respectivecategories. However, the observed
data differstrikinglyfromthe expected,except when
5 15 257
neitherof the two more aggressivemice were pres-
> 0
ent. 315 appears to act as a repulsionand 311 as an
0 50 100 150 200 250 attractionin the formationof these neutralgroups.
AGGRESSIVENESS INDEX When both were presentthe repulsioncharacteristic
FIG. 5. Strain differencesin relative aggressiveness of 315 still dominatesthe relationship,but the fact,
between males of RBA Colony 2 and males of C57 that the observeddoes not differso markedlyfrom
Colony 2. Passiveness Index=I X number of neutral the expected,suggeststhat the attractiontoward311
interactions. AggressivenessIndex=2 X numberof sub- in part cancels the repulsionfrom315.
ordinate interactionsplus 3 X number of dominant in-
teractions. For DBA 's each individual becomes rela- Five of the nine lower rankingmice (Table 12),
tively more aggressive as it engages in more interac- not only avoided 315, but were much more likely to
tions. With the exception of the dominant male 315, be presentin neutralaggregatesif 311 was present.
whose behavior more nearly approximates that of the The reality of this aggregationis attested by the
dominant DBA male 257, C57 males show no decline
in relative passiveness regardless of the number of in- fact that four of them (311, 303, 286 and 288) were
teractions they engage in. commonly, observedhuddled togetheron one of the
TABLE 12. Numberof times membersof two ' groups" next 37 days of intensiveobservation(13 periods of
were found in association with the two most dominant observationeach lasting several hours). Therefore
C57 mice, of Colony 2.
it was possible only to analyze the static pattern
ratherthan its development.
Dependent* Tndependentt
Group Group No object such as a pellet of food could be pos-
sessed by a single mouse. The term possession is
Absent... . 44 28 only applicable in the sense that a mousemay utilize
315
Present .... 16 8 a particulargoal, such as the food hopper or a nest
box, withoutthe distractionof the presence of an-
Absent 16 19 othermouse. The surfaceof the food hopper or the
311 size of the nest box was sufficient to permit10 mice
Present.... 38 17
at least to react with it satisfactorily.Aggression,
*Mice: 286, 288, 303, 319, 321.
however,was maximizedin the food pen where it
tMice: 282, 290, 301, 317. was possible to satisfythe hungerdrive. What then
was the nature of elicitationof fightinghere?
bare shelves of the ha.rborage stand. The presence Three to six mice at a time were observedto eat
of the other four in neutral aggregates appears to simultaneously.Such a feeding aggregatemightbe
have been controlled largely by the absence of the initiatedby one of the more dominantmice, 315 or
dominant 315. 311, or they might join other mice already there.
DBA NEUTRAL: AGGREGATES, COLONY 2 Even thoughtherewas ample space for 10 mice to
eat simultaneously,mutualpushingwiththe forepaws
A similar analysis to the above was made for the
would begin after a few secondsor minutesor pas-
DBAs (Tables 13 and 14). It is evident that the
sive association. This pushing became more intense
seven lower ranking mice not only avoid both of the
until the most dominantmouse present turned on,
two highest ranking ones (257 and 275), but that
attacked,and chased from the food pen one of his
this avoidance was particularly exaggerated when
nearest feeding associates. The remainder of the
both of these mice were present. The reason for this
feeding aggregate also scattered and sought tem-
was that a marked fightnearly always (26 out of 29
porary residenceat favored places of retreat. The
observed encounters) resulted when these two mice
dominantmouseusuallyreturnedto the food pen, but
met each other in or near the food pen.
rarely proceeded immediatelyto eat. Instead it
vacillated back and forthbetweenthe food pen and
TABLE 13. 40 DBA neutral aggregates in the food pen,
of Colony 2.
those shelvescontainingnest boxes. Shelf 1 received
most visitations. Any mouse it encounteredduring
these travels was attacked. When the dominant
No. 257 No. 275 mouse resumed feeding, without having entered a
Present. 13 11 nest box for a period of rest,he usually would not
toleratebeing joined by anothermousefor sometime.
Absent...... 27 29 No measurementof the required time intervalwas
made, but the reality of time lapse before suppres-
sion of aggressionseemswithoutquestion. If a domi-
TABLE 14. The social controlof neutral aggregates in nant mouse enteredthe food pen after a period of
the food pen among DBA mice, of Colony 2. rest and found subordinatesalready eating he was
not so likelyto attack them,but instead would join
ANIMAL PRESENCES
(No. of Aggregates)x them. However,any mousemovingeitheraway from
Dominants Mean
(Mean No. per Aggregate)
__ _ _ _ _ _ _ _ _ _ _ _ _ _
the feeding aggregate or approachingit was much
Present No. of size of morelikelyto be attacked. Motionper se appears to
Aggregates Aggregates Observed Expected*
be a stimulusinitiatingattack. Likewise,a subordi-
275 only ...... I1 3.18 35 25.5 nate mouse in a motionlesspositionwas quite likely
275 & 257 .... 3 2.67 8 28.7
257 only...... 13 2.08 27 25.5
to attack anothermouse in retreatpast it from a
Neither....... 13 2.46 32 22.3 stillmoredominantmouse,even thoughthe retreating
mouse usually dominatedthe motionlessone. This is
Total ...... 40 .... 102 102 one type of social chain reaction. A second, and
more frequentsocial chain reaction,is one in which
*Based on chanceencountersdevoidof attractionor repulsion.
a defeated mouse attacked the next mouse he en-
countered.These two types of social chain reactions
BEHAVIORAL ASPECTS OF SOCIAL INTERACTION
account for the frequentreversals in social status
C57 Colony 2. By the time observations on be- observedamongmostof the C57 mice except in rela-
havior were begun on February 13, 1951 the mice tionto the most dominantone, 315 (Table 7).
were 124 days old. They were sexually mature and Recognitionand avoidance contributeto reduction
the social system already established by that time in overt conflict.All subordinatesat timesretreated
showed no indication of becoming altered during the from the dominant315 withoutbeing attacked di-
9494 JOHN B. CALHOUN
~~~~~~~~~~~~~~~~~~
26, No-.-1
rectly by him. These include retreat from the locality eluded the dominant mouse and several subordinates.
where 315 is attacking another mouse. However, Whatever the reason for the formation of these pas-
they also appeared to recognize 315 as an individual sive aggregates, the nesting box itself probably be-
or at least as a dominant mouse, even when 315 had came a sign eliciting passive behavior or inhibiting
his back turned. Similar avoidance, however, oc- aggressive behavior. Lack of aggression about the
curred among the other mice of each other. One boxes on the lowest shelf of the harborage stand may
general characteristic of these avoidances was that be equally, or even more so, a result of its frequent
the stationary mouse being avoided occupied a posi- avoidance by all but the dominant mouse, 315. An
tion where conflicts were frequent. It may well be understanding of the frequency with which aggres-
that it is the situation, which includes a. mouse, that sions occur at particular places (Table 15) requires
is avoided, rather than another mouse as a specific consideration of the places to which the subordinate
individual. mice retreat following aggressions (Table 16).
Posture and movements frequently become altered
following a series of conflict situations. Approaches TABLE 15. Frequency of aggressive actions according
are made with slow deliberate motions and may be to the place of their occurrence,C57 Colony 2.
followed by rapid withdrawal. Even when no other
mouse is in the vicinity the deliberate movements Place Percent
may be accompanied by a marked extension of all
Food penfloor......... .. 74.0
four limbs. In some cases the tail is arched stiffly Nestingmaterialbox .... . 9.4
over the back. At other times the tail may be rattled Near bases of ramps..... . 6.5
against the floor with unusual vigor, the mouse may Under harborage stand .. 4.4
Fromharborageboxes.. 4.4
prance about with extended limbs, and the entire Bare shelves.1.... . . . . .... .4
body may tremble. Even the dominant mouse occa-
sionally behaved in this latter fashion.
DBA Colony 2. The over-all pattern of behavior
TABLE 16. Frequency and place of retreat from domi-
described for C57 mice also applies to the DBAs. nant mice, Colony 2.
What is differentis the greater intensity of inter-
actions by the DBAs. For example, when the domi-
Place of retreat C57 DBA
nant C57 male, 315 entered the food pen containing
a feeding aggregate, he would most likely walk all Shelveslackingnestingboxes....... 45 4
of the way to the food hopper before initiating an Outsidecornersof foodpen....... . 0 34
Underor behindharboragestand.... 17 12
aggressive action, if at all. However, the more typi- Shelveswithnestingboxes........ . 4 5
cal action of the dominant DBA male, 257, was to Behindfoodhopperor in nesting
enter the food pen, pause just inside the door, become materialbox. 10 0
tense as indicated by extended legs, and then dash Roofof foodpen. 0 11
Justoutsidefoodpen door.. . 3 0
to the food hopper where he would throw himself at
a mouse, knock it down, and then chase it away. Total. 79 66
Fights, particularly between the dominant, 257, and
the contender, 275, were of greater duration, and
chases continued over a greater distance. Following
Only five percent of retreats were to the lowest
a sequence of chases and fights, the dominant 257
shelf, although 76% of records of mice resting in
would even attack a female. This was never observed
boxes was for this shelf (Table 5). The majority of
among C57s. Following fights quivering of the body
retreats were to the set of shelves lacking nesting
was frequently noted. On two occasions this agita-
boxes. Most (28) of these were to shelf 2. One to
tion became so accentuated in the contender, 275,
five subordinates would huddle together on the edge
that even after 257 had left him he continued to run
of the shelf where they either slept or peered down
about on his hind legs, striking out with his forefeet,
the ramp. Since only 5 of 45 retreats to the bare
and thrusting with his head and shoulders, just as if
shelves were to shelf 1, it is concluded that there is
he was still engaging an opponent in fierce combat.
a generalization of the avoidance of the lowest shelf
PLACES OFFAGGRESSIONAND RETREAT with boxes to include the lowest shelf without boxes.
C57 Colony 2. There were two locations in which The dominant mouse, 315, was never seen to go to
mice most frequently encountered each other. One the bare shelves. He spent his periods of activity
of these was within or at the entrance of the food going back and forth between the food pen, and the
pen. Here three-fourths of the aggressions took set of shelves containing boxes, particularly to shelf
place (Table 15). The other place was in and about 1. Places of retreat lay peripheral to this route.
the nesting boxes on the lowest shelf. Less than five The second such major position of retreats was under
percent of the aggressions took place here. Lack or behind the harborage stand. Even within the food
of fightingabout these nesting boxes were associated pen, retreat into the nesting material box or behind
with two other sets of circumstances. Aggregates the food hopper served to avoid this major route
of sleeping and resting mice formed here which in- of the dominant mouse.
With this background we may now return to the protruded through the floor near the wall about 45 cm
question of places where aggressions occur (Table in front and to the right of the entrance of the door
15). Aggressive actions take place either along the from the food pen. This was to one side of the usual
main route of travel by the dominant mouse or at route of travel of the dominant 257 from the food
those places where retreats terminate. The most pen to the base of the ramp to shelf 1 with boxes.
dominant mouse, 315, rarely followed subordinates to When attacked by 257 the pipe served as a barrier
the places where retreats terminated. Actions by which enabled 275 to dodge around and escape.
him do not account for aggressions in the peripheral 275 frequently tolerated 280, and less frequently one
positions. As may be seen from (Table 17) the ag- of the other mice at this position. They were further
gressions of lower ranking mice were more frequently protected here since the aggression by 257 was usu-
at these peripheral locations. Places of security from ally directed toward 275. Aggressive actions by 275
*dominant mice became contended goals. This con- were directed toward the seven other still lower rank-
ing mice, but they generally took place outside the
TABLE 17. C57, Colony 2. Places of aggressive actions food pen at the favored peripheral places of retreat.
according to social rank (DSN score). Even more precisely than with the C57 mice, prox-
imity to the food pen was inversely related to so-
Other9 cial rank. Number of mouse, social rank, DSN
Mouse 315 311 mice score in parenthesis, and usual location during the
DSN score. 299 189 88 (mean) periods of activity were: 257 (284) in the food pen;
Total aggressiveactions .... 76 17 45 275 (163) and its parasite 280 (49) near entrance
to food pen; 267 (76) on top of food pen; 278 (70)
Percentat locations
Food penfloor.... 90.8 82.3 42.3 under or behind harborage stand; 273, 255, 259 and
Elsewhere.... 9.2 17.7 57.7 261 (mean DSN score of 41) in boxes on harborage
stand.
DISCUSSION
tention caused further dispersal. Usual positions
taken: 311, 286 and 303 on bare shelf 2; 282 and PHYSIOLOGICAL STABILITY
317 under and behind the harborage stand; 288 on It has already been stated that the two strains were
bare shelf 1; and 290 in one of the boxes on the selected for use in the present study solely on their
fourth shelf. On the basis of the social rank of the differential susceptibility to audiogenic seizures and
mice, the places of retreat are of decreasing value on their differential incidence of mammary tumors.
in the order listed. The ramp to shelf 2 was a part However, there are many studies which report differ-
of shelf 2 in so far as its use by mice was concerned. ences, in the general characteristic which I am desig-
Thus the advantages of shelf 2 over under-and-be- nating, physiological stability. By physiological sta-
hind-the-harborage-standas a place of retreat appear bility I mean any condition, such as enzyme activity,
to have been: a. Mice on it were better able to detect hormonal balance, or reaction to antigens, which per-
whether or not the dominant 315 was in or near the mits the individual to make those responses or ad ,ust-
food pen; b. Mice on it were more effectively re- ment which are of survival value. In nearly every
moved from the dominant mouse's route of travel. instance the data indicate a relative physiological
DBA Colony 2. The general pattern of places of stability for C57 Blacks and instability for DBAs.
aggression and places of retreat characteristic of These differenceswill be used in the development of
DBA mice are similar to those of C57 mice except an hypothesis concerning the origin of the different
that aggressions were essentially limited to the domi- behaviors observed for these two strains. Unless
nant 257 and his contender, 275 (Tables 16 and 18). otherwise mentioned, the following data, or references
Places of retreat were somewhat differentfrom the to the appropriate literature pertaining to it, may be
C57 mice. 267 when fleeing from either 257 or 275 found in Griineberg (1952). The comparisons listed
usually climbed up on top of the food pen. The con- below primarily concern C57BL sublines 6 and 10,
tender, 275, usually ran behind a 40 mm pipe which and DBA sublines 1 and 2.
DBA produce 8.20 eggs per ovulation of which
TABLE 18. Aggressiveaction by the two most dominant only 58% survive through fertilization to birth, while
mice, DBA Colony 2. C57 produce only 6.65 eggs per ovulation but 84% of
these survive to birth. DBA ova introduced into
Place No. 257 No. 275 C57 uteri do just as well as do C57 ova, while C57
Withinor nearentranceof foodpen. .. 50 5 (257 ova introduced into DBA uteri do just as poorly as
absent) do DBA ova. Thus it may be seen that the deficiency
Lowestshelfwithnestboxes .... 5
Outsidecornersor roofof foodpen. lies in the intrauterine environmentof DBAs. Gesta-
9
Underharboragestand. 2 tion is on the average 12 hours longer for DBAs.
More DBA than C57 are stillborn and more die be-
Total. 55 16 tween birth and weaning. Fekete (1946) found a
similar but lower consistent strain differencein ova
9696 JOHEN B.
JOHN B. CALHOUN
CALHOUN Ecological Monographs
~~~~~~~~~~~~~~~~~~~
26, No. 1
per ovulation (5.2 for DBA, and 4.2 for C57). DBA develop leukemia. Several lines of evidence indicate a
are presumed to produce more estrogen because of generally lowered resistance to infection on the part
the greater number of growing follicles. Similarly of DBAs. Their ability to produce antibodies is ex-
DBA are presumed to produce more progesterone, ceptionally poorly developed in comparison with
since there are not only more corpora lutea from C57s. In other strains of mice, which were particu-
each ovulation, but each set persists longer, so that larly selected for differential resistance to disease,
in DBAs 7 or more sets may simultaneously be pres- high resistance strains had high leucocyte counts and
ent, but never more than 3 for C57s; and the meas- low resistance strains had low counts. Although there
ured volume of DBA ovaries was 2.3 x that of C57. are exceptions, most workers find that C57 of both
The estrogenic and luteal hormones stimulate cell di- sexes outlive DBA. Dr. H. B. Andervont (personal
vision in the manmmary~ glands. As the corpora lutea communication) findsthis true of his colonies although
regress, and the hormonal balance is toward more both strains lack the extra-chromosomal mammary tu-
control by anterior pituitary prolactin, cell division mor exciting factor. Most DBA die by 24 months
ceases and elaboration of milk begins. This shift while many C57 live longer. DBAs have a much lower
takes place before the termination of pregnancy in resistance to malaria infection than do C57 (Nadel
C57s. However, due to the relatively greater 1954). In this process C57 show a greater enlarge-
amounts of progesterone present in DBAs some por- ment of the spleen (Dr. E. M. Nadel, personal com-
tions of the mammary glands persist in cell division munication).
right through termination of pregnancy. This per- Morris & Dunn (1951) found that all DBA on
sistence of cell division due to hormonal imbalance pyridoxine deficient diets were dead by 38 weeks but
is presumed to be a major predisposing factor in that some C57 were still alive at 56 weeks.
development of mammary tumors. Furthermore 30% There is normally a sex difference in size. This
of DBAs develop ovarian cysts but only 13% of develops by four weeks in C57, but not until six
C57s. weeks in DBA. Retardation of growth of male DBAs
Gonadectomy of either sex leads to extensive nodu- is a maternal effect. When newborn DBAs are cross
lar hyperplasia and hypertrophy of the adrenal cor- fostered on C57 mothers the males become larger
tex of DBAs but such changes are rare in C57s. At than females at four weeks. Similarly cross foster-
34 days of age, and presumably also at later ages, ing of C57 young on DBA mothers, causes retarda-
both the diameter of the cortex and the per cent tion of their growth.
of it consisting of Sudanophil cells in C57 is twice When injected with the barbiturate, hexobarbital
that in DBA. Furthermore C57s frequently have sodium, DBA sleep much longer than C57 (Dr.
accessory adrenal glands. The greater size of adre- George Jay, personal communication). The enzyme
nals in C57 should enable them to adjust to those system, which metabolizes this substance, is located
stresses on the physiology falling under Seley's "Gen- in the liver. It is concluded that some aspect of it
eral Adaptation Syndrome." Development of adeno- is less efficientin the DBA.
carcinomas in gonadectomized DBAs perhaps results No DBA are resistant to implantation of "sar-
as a byproduct of the greater demand placed on its coma 37" while 99% of C57 are resistant. On the
smaller adrenals to synthesize sex hormones. other hand, DBA are less likely to develop sarcoma
The thyroid of DBA is more active than that of of subcutaneous tissue following subcutaneous in-
C57. In fact Silberberg & Silberberg (1954) inter- jection of carcinogenic hydrocarbons.
pret this as hyperactivityon the part of DBA. When The above data strongly indicate that DBA mice
they radiothyroidectomized these strains at 6 months are much more prone to develop imbalances in their
of age with 1l31 differential strain results were ob- maintenance of physiological homeostasis than are
tained. More DBAs developed joint lesions and these C57BL mice. There was certainly no conscious se-
were developed at an earlier age. These authors in- lection for many of these traits, and yet the end
terpret this as developing from a relatively greater result of inbreeding has been to develop two strik-
deficiency of thyroid hormone. On the other hand, ingly different strains. Although good information
more C57s developed tumors of the pituitary. These is lacking on most of the genes involved, it is quite
authors suggest that this is not because of a greater likely that some genes affect the expression of more
hormonal imbalance of C57, but rather that more
than one of the above characteristics.
113l is free to attack the pituitary since less of it is
taken up by the thyroid. Furthermore C57 on the BEHAVIORAL COMPARISON, C57 WITH DBA
average lived 1.8 months longer, although still rarely All data recorded in this paper involving differ-
living longer than two years. ences in behavior between these two strains appear
The normal level of leucocytes in the blood is to be derivatives of a differential ability to adjust
higher for C57, yet DBA commonly develop lymph- to altered situations. The first area of difference is
oid leukemia in later life. Again there is the possi- that revealed in the accommodation to a new and
bility of a causal relationship between normal hy- more complex environment. Without exception the
pofunction of the lyiniphoid tissue, the relatively DBA mice exhibited a marked tendency either to
greater demand placed on it at times when the body avoid all situations requiring a response not previ-
is invaded by disease organisms, and the tendency to ously expressed, or to develop a more stereotyped
behavior. This avoidance led to a long lapse of time the DBA mice, the female's physiologywas appar-
prior to full utilization of the harborage stand and ently under considerablestress. The major stressor
its contained nest boxes. It also led to a failure to was attackby males. During the hyper-excited state,
utilize solid pellets of food or other types of nestingcharacteristicof fightingDBA males, a male fre-
material than that with which they had experience quentlymade the errorof attackinga female,even
prior to weaning. Their failure to peer out the door a pregnantone, which happened to be in his route
of nest boxes indicates a withdrawal in the sense of travel. Adjustmentto a more complex environ-
that by not doing so they avoided perceiving the mentis apparentlyalso a sufficient stressorto dis-
actions of their associates. These avoidances, pro- rupt maternalbehavior of DBA mice. The second
duced the stereotyped behavior of persisting, wher- litter born to the female which initiated colony
ever possible, in prior activities. Stereotyping was 1 A had undergonefoetal developmentwhile their
also apparent in the infrequent reversals of social motherwas undergoingadjustmentto the food pen.
rank among males. With reference to each of these However,the criticalvariable was the fact that for
types of behavior the C57 exhibited marked plastic- the threedays precedingparturitionthe motherwas
ity and rapidity of adjustment. Nest boxes and the undergoingthe much more stressfuladjustmentto
harborage stand were explored very soon after they the harboragestand. This litterwas eaten. At the
had the opportunity. New forms of nesting material lower density of Colony 1 A, C57 females reared
or food were immediately utilized. There were fre- theiryoung. However,theywere not exposed to in-
quent reversals of social rank involving all but the tense fightingamong males or attacks by them. At
dominant male. the higherdensitiesof Colonies1 B and 2 therewas
The second area concerns the intensity of fighting. both failure to conceive and destructionof litters
Upon the attainment of sexual maturity, and inferen- by both DBA and C57, even though conflictinter-
tially a heightened androgen level, fighting among actions were not so intenseamong C57. Thus it is
males ensued. Among the DBA mice the dominant evidentthat both strains are susceptibleto disrup-
male for practical purposes never made any further tion of reproductionthroughsocial interaction,but
alteration of his behavior; he always attacked every that theseeffectsappear at a muchlower population
subordinate mouse each time it was encountered. densityamong DBA mice. In fact it is highlyun-
Even though a. neutral relationship might exist mo- likely that DBA mice could successfullymaintain
mentarily it was soon disrupted by an attack on the themselvesin any environment other than in small
part of the dominant male. In Colony 2 the second cages, where density is maintainedlow by human
ranking mouse assumed much of the character of interference.
the dominant mouse in his relationship to all the The second differentialconsequenceof social in-
other subordinate mice. After the first few fightsthe teractionrelates to hasteneddeath. Within 75 days
subordinate mice developed the undeviating behavior of the formationof DBA Colony 1 B, consistingof
of submission and avoidance. Most fights were in- 14 adults and 15 23-day-oldyoung,all the mice were
tense and prolonged. The fightingand social behavior dead. Some mice did receive small nicks during
was quite otherwise among the C57 mice. With the fights,but these could hardly have been considered
one exception, to be emphasized later, aggressive be- the cause of death. Furthermorelack of availability
havior, even by the dominant male, was less intense of food was not a factor. It is, therefore,concluded
and the fightsor shorter duration than for the DBA that death was an indirectconsequenceof fighting,
mice. All individuals developed adjustments involv- throughsome physiologicalalteration,possibly re-
ing, cessation of aggression, replacement by the milderlating to the adrenal cortexas discussedby Christian
forms of pushing characteristic of adolescence, or (1950). C57 Colony 2, thoughstartingwith nearly
mild threats of short approaches or motions of the twice the population (53 vs. 29) did not die offso
body toward another individual who in return usually rapidly. Furthermore,a major contributingfactor
exhibited withdrawals of shorter distances. to death was starvation,since the food hopper fre-
CONSEQUENCES OF SOCIAL INTERACTION
quently became jammed with the bodies of mice
suffocatedin the food slot. It thereforeappears that
There are three major consequences of social inter- C57 mice are more resistantto the stress of social
action in which these two strains differ. Reproduc- interactionthan are DBAs.
tive performance may be altered. Deleterious effects The thirdconsequenceof social interactionis dis-
always accompanied reproduction of DBAs when the persal. As a consequenceof the more intensesocial
mother had been ranging through the more complex interactionsof DBA mice, they formed relatively
standardized environment in the presence of three or smaller aggregatesboth in the food pen and in the
more males. Disruption of maternal behavior, in nest boxes.
which the mother ate her young shortly after birth,
was the major effect. Rosvold (1949, 1953) has CHANGEIN MOOD
shown that such disruptions ill maternal behavior During the resting hours mice were much more
will arise in a laboratory strain of Norway rat, when tolerantof each other. Even DBA males, who regu-
the mother is exposed to chronic disturbance suffi- larly foughtintenselyon the floorof the food pen,
cient to produce hypertrophy of the adrenal cortex. oftenslept in the same nestbox. This greateramica-
Although no histological examinations were made of bilityat timesof reducedlocomotoractivitywas even
Ecological Monographs
9898 JOHN B. CALHOUN
JOHEN ~~~~~~~~~~~~~~~~~~
26, No. 1
more marked among C57 mice. Similar changes conditions which aggravated both locomotor activity
from passivity towards aggressiveness took place and emotion.
among males at the food hopper. Initially each mem- ACTIVITY AND EMOTION
ber of a feeding aggregate confined its activity to
From the above discussion there are indications
eating, even though in direct contact with a neighbor.
of an interdependent relationship between level of
Gradually pushing with the forepaws or with the
activity and emotion, where the latter is judged by the
whole body ensued. This pushing became intensified
behavior of the mice. Certain data regarding other
and regularly terminated in a sufficientlyaggressive animals, particularly domestic Norway rats (Munn
action by one mouse to disrupt the feeding aggregate.
1950), provides insight into this relationship, which
That is, aggressive action followed an increase in
provides the key to the interpretation of intrastrain
level of general activity. The type of action, by which
behavior as well as differencesbetween strains.
mood is inferred, exhibited by the dominant mouse
Secretions of the anterior pituitary, adrenals, thy-
of a colony as he entered the food pen and ap-
roid and gonads maintain a level of activity whieh
proached a passive feeding aggregate was related
is diminished upon their removal, or which can be
to his prior state of activity. If he had just emerged
returned to normal by injections of the hormones.
from a nest box after a period of rest he was much
Hunger, thirst, and vitamin deficiency increased ac-
more likely to passively join the aggregate. How-
tivity. Where activity relating to that associated with
ever a preceding period of activity (exploring the
the nest, food source, and spontaneous running were
harborage stand, etc.) predisposed him to initiate an
recorded separately (for the cotton rat, Sigmodon;
immediate attack. Such attack by a dominant DBA
Calhoun 1945) a short period of feeding usually pre-
male arose after a shorter preceding period of ac-
ceded running, although for the remaining portion
tivity, than by a dominant C57 male. For the DBAs
of each period of activity, these two types of activity
it is quite apparent that both auditory and visual
alternated. These data indicate that physiological
stimuli arising from other mice serve to alter the
alterations can modify activity level, which in the
mood from passivity toward aggressiveness. Changes
case of food consumption produces an increase within
in stance and direction of motion-indicate the imme-
a few minutes. The question that remains unsatis-
diate effect of such stimuli. Such stimuli, are usu-
factorily answered is: what is the function of spon-
ally less effectivein inducing mood changes in C57s,
taneous activity? Inherent in most of these studies is
probably because of their greater facility in develop-
the implication that modification of spontaneous ac-
ing adjustive behaviors which reduce marked con-
tivity assists in maintenance of physiological ho-
flict. This latter, at least, makes more difficultthe
meostasis. This assumption will be adhered to in
observational identification of the effect of such
the development of a rationale to explain the ob-
stimuli. When a subordinate mouse, particularly
served behavior of house mice. Studies that are
among C57s, was attacked by a more dominant one,
greatly needed are those in which the time course of
it was in turn much more likely to attack another
change in activity is followed from the moment when
subordinate mouse. The level of activity of this sec-
physiological homeostasis is altered by administra-
ondary attacker, and presumably also his emotional
tion of appropriate substances.
level, was heightened by being attacked. A final and
pertinent incident of mood change was that which Emotional state also alters activity level. Begin-
arose at the time of formation of C57 Colony 1 B ning with Hall (1934) there have been many studies
with domestic Norway rats which show that introduc-
by the union of two adjoining pens. Members of the
longer established Colony 1 A invaded the adjoining tion into a strange situation or presentation of un-
pen. Locomotor activity of these mice was definitely accustomed stimuli results in a temporary elevated
increased as they initiated a period of intense ex- emotional state as judged by such outward mani-
festations as defecation, urination, or excessive
ploration of the new environment. After spending
some time on the floor each mouse finally reached grooming. Similarly it is a common experience that
one of the shelves of the stand which was new to rats, which have access to an activity wheel, engage
them. However, this was not accomplished until each in intense periods of running, immediately following
mouse had gone part way up several ramps, before unaccustomed loud noises in the laboratory.
returning to the floor each time. This vacillating be- I am at present engaged in experimental studies
havior, which had never been exhibited on their own (unpublished) of exploratory behavior and home
harborage stand, indicated that exposure to this new range. In one series of experiments, rats are placed
environment, though similar to their own, was an in an enclosed alley 3.5 in long with a nest compart-
emotional experience. Sight of strange mice similarly ment at one end. The time course of excursions
must have been an emotional experience. Fights with through 72 hours are automatically recorded. Rats
the strange mice immediately began. However, the are placed in the alley during the middle of the day.
most interesting phase of their behavior was that The nearly invariable immediate response is to defe-
extremnel-vsevere attacks, in which wounds were re- cate. Defecation is immediately followed by a 30-120
ceive(l, began among the Colony 1 A males. Such minute period of more intense activity than takes
fightingamong these males was never seen prior to or place during the remaining 70 hours. Essentially no
followinig this (late. This fighting was preceded by running takes place during the diurnal period of the
January,1956 THE SOCIAL BEHAVIOR OF Two INBRED STRAINS OF HOUSE MICE 99
next two days. There is little doubt that heightened of homeostasis to learning. Another (Dr. D. H. K.
motor activity usually accompanies an elevated emo- Lee, personal communication) is that of an Aus-
tional state. tralian marsupial which was used in heat tolerance
In summary, undirected or spontaneous activity tests. Each day they were placed in cabinets at
nmay be altered by exteroceptive stimuli of emo- room temperature. Behavioral adjustment was to
tional import, or by permanent states or temporary turn over on the back and extend all four legs as the
changes of physiology. These latter must initiate the temperature was gradually elevated. This exposed
origin of afferent visceral impulses terminating at maximum surfaces for loss of body heat. Within a
some neural center which induces the spontaneous few days, each would turn over on its back as soon
activity, or the physiological changes must affect this as placed in the cabinet and extend its legs in antici-
center directly. Recent studies (French et al. 1952; pation of the forthcoming elevation in temperature.
Verney 1947) indicate the possibility of a system Thus there can be little doubt that animals can adjust
which can modify spontaneous activity by either ex- behavior in a direction that facilitates resumption oi
teroceptive stimuli or physiological changes. At the physiological homeostasis.
base of the brain there lies an area comprised of the CONCLUDING ASSUMPTIONS AND OBSERVATIONS
midbrain tegmentum, the subthalmus, the hypothal-
1. DBA mice are physiologically less stable than are
mus, and the medial portion of the thalmus. It is
C57 mice. That is homeostasis of many physio-
designated as the reticular activating system. It
logical systems of DBA mice are more readily
responds as a unit to stimuli of either visceral, visual,
placed in a state of imbalance.
auditory, or somatic origin. Conscious perception is
2. DBA mice exhibit more stereotyped behavior than
not involved in its activity, but rather firing of this
do C57 mice. In other words, C57 mice more
system continues long after termination of stimulus,
readily develop learned adjustments to new en-
and leads to a general activity of the cortex, which
vironmental situations.
in turn produces a state of increased activity, aware-
3. DBA mice exhibit malfunction of reproductive
ness of sensation, and arousal to wakefullness. If
and maternal behavior, and premature death
physiological changes can initiate afferentvisceral im-
takes place at lower population densities than
pulses this activating system may be placed in opera-
characterizes C57 mice.
tion. However, it is possible that functioning of this
4. Intensity of activity, including social interaction
activating system may be directly initiated by physi-
is much greater among DBA mice.
ological changes as is the excretion of antidiuretic
hormones from the neurohypophysis. Increases in 5. Groups of DBA mice are smaller and more dis-
osmotic pressure in the carotid arterial blood is de- persed through their environment than are C57
tected by osmoreceptors in the neurohypophysis mice.
which initiates release of the antidiuretic hormone. 6. Social interactions of C57 mice are more passive,
Just as with the reticular activating system, extero- labile, and of mutual benefit than are those of
ceptive stimuli of emotional import also activate the DBA mice.
release of antidiuretic hormone. Whether or not the 7. Frequency of association between pairs of mice
osmoreceptors involved in antidiuretic hormone re- of either strain are other than random.
lease also stimulate the reticular activating system is a. Non randomness of association among DBA
unknown. Suffice it to say, that the above indicates mice is primarily dependent upon avoidance of
the plausibility of existence of a system which can the dominant mouse.
mediate alterations in spontaneous activity from b. Non randomness of association among C57 mice
either exteroceptive stimuli or physiological changes. is influenced by both avoidance of the dominant
mouse and attraction between subordinate mice.
BEHAVIORALCHANGES AND PHYSIOLOGICAL
8. Subordinate mice of both strains engage in fewer
HOMEOSTASIS social interactions than do more dominant mice.
One further general concept is needed to construct 9. DBA males occasionally attack females, but C57
a formulation of the origin of behavioral differences males were never observed to do so.
between DBA and C57 mice. This is that an animal 10. Change of mood from pacificity toward aggres-
can make learned adjustments in behavior that result siveness follows increases in level of activity and
in restoration of physiological horneostasis. The tre- recency of experiencing situations of emotional
mendous literature on hunger and thirst motivation in import. These changes are more frequent and in-
learning problems by rats (Munn 1950) substantiates tense among DBA mice.
the reality of this ability. Also a rat may be deprived 11. Emotionality tends to place systems of physio-
of some essential chemical compound, which is lack- logical homeostasis in a state of imbalance.
ing in its normal diet. When exposed to a series of 12. Stimuli of emotional import increase activity.
containers, only one of which incorporates this lack- 13. Increase in activity is an adjustive behavior,
ing material in the contained food or water, the rat which, if not unduly aggravated, assists in the
develops the ability to select food or water from the return of physiological homeostasis.
one container in which the deprived substance is lo- 14. Modification of the level of undirected spontane-
cated. This is just a special case of the relationship ous activity may arise through exteroceptive
100 JOHN B.CALHOUN
JOHNB. CALHOUN
Ecological Monographs
~~~~~~~~~~Vol.
26, No.
stimuli, visceral afferent impulses, or physiologi- Other individuals or any new situation may be
cally induced alterations in osmolarity. avoided. Such avoidance prevents disruption of
15. Behavioral adjustments, which may develop into homeostasis. This leads to a generalized learned be-
learned or habitual patterns of action or inaction, havior of avoiding new situations. When a situation
develop along lines assuring prevention of dis- cannot always be completely avoided, such as meeting
turbance of physiological homeostasis, or a rapid another individual, persistence of physiological dis-
return to homeostasis following an imbalance in turbance will be proportional to probability of con-
it. tact. Persisting aggression or avoidance are logical
consequences of an unstable physiology. Dispersion
Item 1 is an inference drawn from the extensive
of individuals, and even further reduction of repro-
literature on these two strains. It is the critical as-
ductive performance and survival, from that seen in
sumption from which stems the present interpreta-
the small cages, is also a logical consequence of
tion of the observed behavioral differences between
phenotypic characteristics of persisting aggression or
these two strains. Items 2 through 10 are the major
avoidance engendered by the more complex environ-
observed behaviors and consequences thereof. Items
ment.
11 through 15 are concepts or inferences based upon
Now let us consider the C57 type with its more
studies reported in the literature. Items 1, and 11
stable physiology. Exposure to new situations are
through 15, will be utilized in formulating a con-
likely to be accompanied by slight alterations in
ceptual framework for gaining insight into the origin
homeostasis, and a lessened elevation of activity.
of differences in behavior exhibited by these two
There will be possible a wide range of extent of ho-
strains of mice.
meostatic disturbance. Thus it will be possible for
HOMEOSTASISAND BEHAVIOR,A THEORY the animal to develop those habits which reduce the
As the individual matures it begins to explore its degree of physiological disturbance, without having
environment. In doing so it is exposed to new stim- to resort so frequently to avoidance. Because activity
uli, new associates, or new interaction situations with level and emotion are not so greatly enhanced in a
former associates. Each such new experience is an new situation, such as when two individuals meet,
emotional one which places some aspects of its physi- fighting is less likely to begin, and the learning of
ology in imbalance. Increase in activity is a gen- more passive or cooperative behavior is possible.
eralized state which facilitates return to homeostasis. Satisfactory reproductive performance, longer sur-
However, while in the increased state of activity one vival, and greater aggregation of individuals are
of two events transpire. The individual by chance logical consequences of the amelioration of open
encounters situations or makes adjustments which conflict and a reduced prevalence of avoidances,
foster reduced emotion and return to homeostasis; or whose origin was possible in the physiologically
it encounters situations which further aggravate the stable strain.
state of physiological imbalance. Furthermore the One unsatisfactorily explained question is what
animal will by chance experience a variety of situa- underlies the first instance of fighting between two
tions or make a variety of adjustments which vary in individuals. Scott & Fredericson (1951) have dis-
the extent to which homeostasis is altered. Those cussed the "Causes of fighting in mice and rats".
behaviors will become fixed that tend to bring a Actually they take fighting as a given behavior and
rapid return to homeostatis, or which avoid its dis- discuss how physiological alterations and learning
ruption in the first place. All that has been said affect intensity or prevalence or fighting. However,
up to this point is an explicit component or implicit there are two concepts in the present paper, not ex-
inference from current concepts of physiology or plicitly stated by Scott and Fredericson, which help
learning theory. to predict the probability of fighting. These are
Animals do differin their ability to adjust physio- that the extent to which a given situation involving
logically when exposed to stressful situations. The two individuals elevates the emotional state is corre-
question is "How does the ability to make physio- lated with the degree of physiological instability,
logical adjustments determine behavior?" iLet us take and that augmentation of activity is correlated with
the DBA type first. Any new experience, particu- level of emotionality. The more activity is elevated,
larly those involving another individual, will likely the more likely two individuals will interact vigor-
produce a maximum disturbance of one or more as- ously with each other. There still remains the unex-
pects of physiology. Therefore, it will be impossible plained phenomenon of why two individuals inter-
for the development of learned behavioral adjust- acting vigorously tend to fight.
ments, whose origin requires differential alterations Fredericson (1951) later postulated the origin of
of homeostasis. Where adjustments do develop, as spontaneous fightingbetween mice as a result of the
in the present study in relation to entering nest attempt by each to restore perceptual homeostasis.
boxes, the learning process is impeded. Where an- He defined perceptual homeostasis as a "psycho-
other individual is involved, physiological disturbance logical need to predict and control environmental
is so great that there is never an opportunity to de- stimuli". Thus he postulates that two mice encoun-
velop any interaction other than that of fighting. tering each other for the firsttime, or in a new situa-
The one adjustment possible is that of avoidance. tion, engage in combat in order to control the be-
havior of the other and make it more predictable. tionality and thus greater ease of learning non-
This assumes, what appears to me, an untenable fighting adjustments. Frederieson (1952) has shown
teleological supposition, of cognizance that engaging that C57BL female mice can be induced to fight
in a fighting interaction for the first time will lead following conditions which foster elevation of emo-
to greater predictability of behavoir of the other tional level and activity. These conditions were food
mouse. Rather initial fighting needs no further as- deprivation terminated by placing two mice together
sumptions than that of increased activity, and thus in a small enclosure containing a single small piece
more vigorous bodily contacts, in which prior be- of food which could be possessed by a single mouse.
haviors such as the forepaw-pushing associated with In my studies of wild rats (Rattus norvegicus) fe-
nursing, and the biting associated with obtaining males lived longer and fought less than did males.
food, are by chance expressed in relation to another However, during lactation, when considerable de-
individual. Once individuals have experienced such mands are made on their physiology to supply nour-
interaction, then changes in pattern of fighting or ishment for their young, these females become ex-
other substitute social interaction, may be antici- tremely aggressive.
pated, in the direction of stereotyping, or predicta-
GENERAL IMPLICATIONS
bility according to Frederieson's concept of per-
ceptual homeostasis. Where a species has lived for many generations in
A slight redefinition of perceptual homeostasis is a particular habitat, its physiology has become
adapted to the chemical and meteorological charac-
required in order to maintain it as a special case of
the concept of physiological homeostasis, as this was teristics of the environment(Hesse et al. 1937). Physi-
ological adjustments have also probably evolved in
apparently Fredericson's intent. This is: Perceptual
relation to behavior, including social behavior, so
homeostasis is the state of maximum predictability
and control of environmental stimuli. that very few species exist in which hereditarily de-
termined constitution is so disharmonious with its
Changes in perceptual homeostasis will generally
environment as it was for the DBA mice in the pres-
be paralleled by similar changes in physiological
ent study. Where abrupt changes in the environment
homeostasis. That is, as the interaction becomes
lead to such disharmony the species stands small
more predictable, there will be fewer alterations of
chance of surviving in that locality. Physiologically
physiological homeostasis and less elevation of emo-
aberrant genotypes do exist in man and his domestic
tion. However, certain characteristics of the social
animals. Snyder (1954) wears a very rosy pair of
hierarchies indicate the possibility of their inde-
glasses in viewing the accommodation of these aber-
pendence. The conditions of C57 Colony 2 and DBA
rant human genotypes into society. He believes that
Colony 2 are particularly noteworthy. In each col-
medical and social science will continue making prog-
ony the second ranking mouse frequently challenged
ress toward more adequate survival and social accom-
the dominant mouse, yet he always lost. Behavior
modation of such individuals. Increasing life span
was highly predictable; perceptual homeostasis was
certainly indicates that this opinion is in part true.
achieved. Yet its very achievement, involving fight-
But what about behavioral disorders? Juvenile de-
ing, must have upset physiological homeostasis. Per-
linquency, neuroses, psychosomatic disorders, psy-
sistence of this masochistic behavior indicates the
value of perceptual homeostasis for the individual. choses, and senile dementia, despite the difficultyof
obtaining adequate data on trends, appear to be a
Among DBA mice at all population densities and
much greater problem in present day society. If
among C57 mice at higher densities, the dominant
this is really true, the important inference, in the
mouse, or despot, regularly achieved predictability of
light of the fact that such changes cannot be ac-
his associates' behavior, at the expense of imbalance
of his own physiology as well as that of the subordi- counted for in terms of the slow rate of accumulation
nate mouse. of genes with deleterious effects,is that the physical,
situational, and social aspects of man's environment
A generalization based upon DBA and C57 mice:
have become so altered that genotypes which formerly
Attainment of perceptual homeostasis, or the attempt
made adequate adjustments (i.e., physiological and
to do so, even though this simultaneously disrupts
behavioral phenotype) no longer are able to do so.
physiological homeostasis, forms a frequent char-
Population density is increasing and our technological
acteristic of phenotypically physiologically unstable
societv is becoming more complex. Per unit of effort
individuals in contrast to phenotypically more stable
the volume of food stuffs produced continues to in-
individuals, who adjust their social behavior in such
crease. There is a wide spread opinion that popula-
a way that there is a simultaneous achievement of
tion increase should be encouraged to keep pace with
predictability of their associate's actions, as well as
food production. The population of the United States
prevention of disruption of physiology.
is expanding, and this expansion appears to exert
There is also the question of why female mice usu- a favorable effect on the business economy. As a
ally do not fight. It is a general rule that females correlate to these circumstances one encounters the
have a longer life span than do males (Lansing 1952). opinion that if we can only manage to maintain an
If this means that they are physiologically more expanding population, this favorable state of the
stable, lack of fightingmay derive from lowered emo- business economy will continue.
102 JOHN B.B.CALHOUN
JOHN CALHOUN Ecological Monographs
~~~~~~~~~~Vol.
26, No.I
Perhaps this is so, but as biologists we need to c. These disturbances usually augment activity
pose the question: "Are all consequences of popula- and increase probability of intense interactions
tion growth equally favorable." We have seen that between animals.
a fairly physiologically and behaviorally stable ani- d. The degree of disturbance of physiological
mal such as the C57BL mouse can be altered into an homeostasis varies according to the situation
unstable one merely by increasing the size of the and the behavior it elicits.
group. There is little at present to anticipate other e. Individuals with a relatively stable physiology
than a continued increase in prevalence or intensity will learn to enter those situations and repeat
of human behavioral deviations or associated psy- those behaviors which were accompanied by
chosomnaticconditions as density and complexity of small changes in homeostasis. Groups consist-
society continues. Experimental studies of popula- ing of such individuals will exhibit marked tol-
tion dynamics and social behavior with infra-human erances between individuals. Such is the C57
animals promises a rich field for ecologists to eluci- type mouse.
date general principles applicable to the understand- f. Most any new situation will cause marked dis-
ing and control of such problems. ruption of the homeostasis of physiologically
unstable individuals. It will thus be impossible
SUMMARY for discrimination among gradations of be-
1. Colonies of inbred strains of house mice were havior, except that of avoidance of situations.
established in standardized environments. Most behaviors will involve avoidance or intense
2. This standardized environment consisted of a pen interactions. Groups consisting of such indi-
covering 17.5 sq m of floor space in which was viduals will exhibit marked antagonisms be-
located a harborage stand and a smaller pen. tween individuals. Such is the DBA type mouse.
The smaller pen, which had a single route of ac- g. As population density increases, continued dis-
cess from the larger pen, contained hoppers for ruptions of homeostasis can transform a geneti-
food, water, and nesting material. On each of cally determined stable physiology into a pheno-
eight shelves on the harborage stand there were typic unstable physiology with all the accom-
four nest boxes. A ramp connected each shelf panying alterations in behavior.
with the floor.
3. Each colony was initiated by introducing one or LITERATURE CITED
more pregnant females, or one or more pair Calhoun, John B. 1945. Diel activity rhythmsof the
of mice into a pen. rodents, Microtus ochrogaster and Sigmodon hispidus
Ecology 26: 251-273.
4. Strains studied: DBA /2, C57BL/10.
(1956). Behavior of house mice with refer-
5. DBAs fought more frequently and intensely than ence to fixedpoints of orientation. Ecology (In press).
did C57s. . (Mss.) An experimentalstudy of the ecology
6. On the other hand C57s developed more tolerant and sociology of the Norway rat.
and passive social adjustments. Christian,J. J. 1950. The adreno-pitutiarysystem and
7. DBAs formed smaller and more dispersed groups. population cycles in mammals. Jour. Mammal. 31:
247-259.
8. Reproduction beyond that of the initial litter of
Fekete, E. 1946. Comparativestudy of ovaries of virgin
each introduced female was unsuccessful for DBA
mice of DBA and C57 Black strains. Cancer Re-
mice. C57 mice were more successful. search, 6: 263-269.
9. At higher densities DBA mice began dying off Fredericson, Emile. 1951. Time and aggression. Psy-
more rapidly than C57 mice. chol. Rev. 58: 41-51.
10. On the basis of many reports in the literature . 1952. Aggressiveness in female mice. Jour.
DBA mice are less stable physiologically than are Comp. and Physiol. Psychol. 45: 254-257.
C57 mice. Criteria used: DBA have a shorter life French, J. D., F. K. Von Amerongen& H. W. Magoun.
span, greater intra-uterine mortality, greater sus- 1952. An activating systemin brain stem of monkey.
ceptibility to audiogenic seizures, increased inci- A.M.A. Arch. Neur. and Psychia. 68: 577-590.
dence of several tumors, lowered resistence to dis- Fuller, J. L., & E. Williams. 1951. Gene controlled
ease, etc. time constants in convulsive behavior. Nat. Acad.
11. A formulation of the relationship between physio- Sci. Proc. 37: 349-356.
logical stability and observed behavior is as fol- Griineberg,Hans. 1952. The genetics of the mouse. The
lows: Hague, Martinus Nijhoff, 650 pp.
a. Physiological homeostasis is more easily dis- Hall, C. S. 1934. Emotional behavior in the rat. Jour.
turbed, and is reinstated with greater difficulty, Comp. Psychol. 18: 385-403.
in those animals characterized by the above Hesse, Richard, W. C. Allee, & Karl P. Schmidt. 1937.
types of criteria as being physiologically un- Ecological animal geography. New York: John Wiley,
stable. 597 pp.
b. New situations, including relations with other Lansing, Albert I. 1952. Cowdry's problems of aging.
individuals, produce such disturbances. Baltimore: Williams & Wilkins Co., 1061 pp.
Morris, H. P., & T. B. Dunn. 1951. Pyridoxine defi- & J. R. Royce. 1953. Electroshock and the rat
ciency. Nutritional and histological observations on adrenal cortex. A.M.A. Archives of Neurol. and Psy-
five strains of mice. 12th Congr. Pure and Applied chia. 70: 1-12.
Chem. 13: 164-165. Scott, J. P. & Emile Fredericson. 1951. The causes of
Munn, Norman L. 1950. Handbook of psychological fightingin mice and rats. Physiol. Zool. 24: 273-309.
research on the rat. New York: Houghton MifflinCo. Silberberg, R., & M. Silberberg. 1954. Radio-iodine in-
598 pp. duced athyroid joint disease in mice of different
Nadel, E. M., J. Greenberg,G. E. Jay & G. R. Coatney. strains. Endocrinology55: 535-542.
1954. Increased resistanceto malaria of certain inbred Snyder, Laurence H. 1954. The effectsof selection and
strains of mice, their hybridsand backcrosses. Amer. domestication on man. Jour. Nat. Cancer Institute
J. Pathology 30: 658-659. 15: 759-769.
Rosvold, H. Enger. 1949. The effectsof electroconvul- Verney, E. B. 1947. The antidiuretic hormoneand the
sive shocks on gestation and maternal behavior. Jour. factors which determineits release. Roy. Soc. London
Comp. and Physiol. Psychol. 42: 118-136; 207-219. Proc. Ser. B. 135: 25-106.

Calhoun, John B. - A Comparative Study of The Social Behavior of Two Inbred Strains of House Mice

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Calhoun, John B. - A Comparative Study of The Social Behavior of Two Inbred Strains of House Mice

Uploaded by

Copyright:

Available Formats

A Comparative Study of the Social Behavior of Two Inbred Strains of House Mice

Author(s): John B. Calhoun

INTRODUCTION ........... ...................... 81 Intensityof activity........................... 91

INTRODUCTION ruentsprovidedthat the nmemibers withina particular

physical environment and culture had to be con-

living in small cages with their parents and sibs. The

MATERIALS AND METHODS I~~~~~~ 2

THE STANDARDIZED PHYSICAL ENVIRONMENT

5. Vocalization: Whilefree rangingDBAs frequently is attributableto two conditions.First, neitherusu-

random basis. 1............. 104 22 22 16

1st 96 32 20 9 39.3 P2 = P1 (ni) (1/a-1) = probability of observ-

Another indication of degree of attraction or dis- Pi Pi-- ( j ) (1/a -1)

near and furtherremovedfrom the food pen. cr05~~~~~~~~~~~~~~~~~~~~~~~C

NUMBER OF GROUPS 0 5. 1015.8an.oe

Box 1 position Other15 positions

cr the DBAs there was not only a lack of challenging a

C 57_- Neutral aggregates may be characterizedby the

You might also like