BIOLOGY OF REPRODUCTION 35, 592-598 (1986)

Food Restricting Young Hamsters (Mesocricetus auratus) Affects Sex Ratio

and Growth of Subsequent Offspring’

U. WILLIAM HUCK,3 JAY B. LABOV,2’4 and ROBERT D. LISK3

Department of Biology3
Princeton University
Princeton, New Jersey 08544
and
Department of Biology4
Colby College
Waterville, Maine 04901

ABSTRACT
Trivers and Willard (1973) predicted that stressed adult female mammals may enhance their fitness by
skewing offspring sex ratios and maternal investment to favor daughters. The present study investigated whether
stressing young hamsters might also influence sex ratio and growth of subsequent offspring. Control females
received food ad libitum (A) on Days I-SO postpartum (AA). Experimental females were food-restricted (R)
either on Days 1-25 (RA), Days 26-50 (AR), or Days 1-SO (RR) postpartum. Subjects were mated when
91-95 days old. Litter sizes and survivorship (=% litters within a treatment that contained at least one pup), sex
ratio (=% males), and pup weights in the next generation were recorded every fifth day from parturition until
Day 25 postpartum. Control litters contained significantly more offspring at birth than did RR litters. Sex ratio
was significantly higher at birth for AA litters than for the other treatments. Postpartum sex ratio within each
group remained similar to that recorded at birth. RR litters contained significantly fewer pups compared to the
other three treatments from Days 5-25. RR female pups weighed significantly more at birth than their counter-
parts in the other treatments. Weights of males at birth were similar in all treatments. By Day 25, both male and
female RR pups weighed significantly less than control, AR, and RA pups. Food restriction early in life may
have long-term consequences on sex ratio and pup growth in golden hamsters.

INTRODUCTION Trivers and Willard argued that daughters weaned in

For species in which females expend more effort less-than-optimum condition would more likely breed
and energy than do males in producing and rearing as adults than would sons.
young (e.g. mammals with polygynous or promiscuous Data from several studies suggest that when physi-
mating systems), the amount of investment available ologically or socially stressed during pregnancy (Lane
to offspring will be related to the physical condition and Hyde, 1973; Rivers and Crawford, 1974; Meikle
of the mother. Trivers and Willard (1973) hypo- et ab., 1984; Paul and Thommen, 1984; Labov et al.,
thesized that if the fitness of a female’s male offspring 1986) or lactation (McClure, 1981), female mammals
(e.g., the ability to hold and defend territories) is at may produce or maintain female-biased litters. Others
beast partially dependent on the quality and quantity have found no relationship between maternal con-
of maternal care which those males received prior to dition and offspring sex ratio (Myers, 1978; Clutton-
weaning, then females should skew the sex ratio of Brock et al., 1981; Small and Smith, 1985), or have
their progeny to favor sons when in good condition reported that females in poor condition produce
and to favor daughters when in poor condition. male-biased litters (Altmann, 1980; Michener, 1980;
Simpson and Simpson, 1982; Silk, 1983).
Prevailing environmental and social conditions
Accepted March 13, 1986. during reproduction can be important determinants
Received August 13, 1985.
‘This work was presented in part at the 65th Annual Meeting of of reproductive success (Bronson, 1985; Myers et al.,
the American Society of Mammalogists (1985), Orono, ME. Supported 1985; Wise et al., 1985). Labov et al. (1986) observed
by Grant BNS-8300892 from the National Science Foundation to
that female hamsters that were food-restricted during
R.D.L. and U.W.H.
2 Reprint requests. pregnancy delivered litters with sex ratios biased

592
 SEX RATIO AND PUP GROWTH IN FOOD-RESTRICTED HAMSTERS 593

significantly toward female offspring. By Day 25 F = 0.75 [(wdrd) + (wprp)}

postpartum, both male and female pups from ex-
perimental litters weighed significantly less than their where
control counterparts. Because the ability to produce F = weight of food ration,
and raise progeny can be influenced by an animal’s wd = body weight of dam,
developmental history (Massaro et al., 1974; Salas et Wp = total body weight of a dam’s pups,
al., 1984), the present study was undertaken to = weight of food/maternal body weight,
determine whether restricting the food intake of rp = weight of food/total body weight of pups on
hamsters early in life would also result in the birth of a given day (all weights in grams).
female-biased litters or in the postnatal adjustment of
Generation 2 pups were removed from their
offspring sex ratios to favor daughters.
mothers at 25 days of age. Generation 2 female pups
were divided further into four groups, and placed into
MATERIALS AND METHODS
individual 20 X 19 X 12 cm plastic cages with wood
shavings and a wire mesh top. Two groups received
Animals and Maintenance
food ad libitum (AA: N25;RA: N=17). Juveniles in
Subjects (Generation 1) were laboratory-born the other two treatments (AR: N17; RR: N17)
descendants of the randomly bred LVG strain obtained were fed a diet restricted to approximately 65% of
from Charles River, Inc. (Wilmington, MA). Seventy- that given to AA and RA subjects until 50 days of
two multiparous females aged 170-251 days (mean =
age. All Generation 2 females were fed ad libitum
199.3 days) and weighing 100-172 g (mean = 130.9 thereafter.
g) were used. After mating to satiety with a stud Generation 2 subjects were weighed every fifth day
male, subjects were placed into 30 X 36 X 16 cm until they were mated at 9 1-95 days of age to
plastic breeding cages with wire mesh tops. The sexually rested (2 wk) stud males in a circular Plexiglas
bedding of wood shavings was changed weekly until arena 35 cm in diameter. The number of intromissions
parturition and added to occasionally thereafter to and ejaculations was observed and recorded. All
minimize disturbance to the dams. Hamsters were females received at least nine ejaculatory series, a
maintained in a colony room at 23#{176}Cand 14L:1OD level of copulatory stimulation that resulted in 100%
photoperiod (bights on at 2200 h). Tap water was pregnancy rates among similarly aged females in
available ad libitum. Wayne Rodent Blox (Formula previous studies (Lanier et al., 1975; Huck and Lisk,
8604-00; Wayne Pet Food Division, Continental 1985). After mating, all Generation 2 subjects were
Grain Co., Chicago, IL) was provided as described placed into breeding boxes as previously described.
below. To minimize disturbance, Generation 2 females were
not weighed during pregnancy.
Procedure Beginning at parturition (Day 0), Generation 3
One to 12 hours after parturition, Generation 1 litter survivorship (the percentage of litters within a
females and their litters (Generation 2) were assigned treatment that contained at least one pup), litter size
randomly to one of two treatments. Dams and young (number of pups/litter), body weights of pups, and
of one group (A) received food ad libitum for the the sex of each pup (determined by examination of
first 25 days after parturition (Parturition = Day 0). ano-genital distances) were determined at five day
Dams and young of the other group (R) were placed intervals for 25 days. Body weights of Generation 2
on a diet restricted to approximately 75% of the females were also monitored every fifth day, from the
intake of control females with the same body weights birth of Generation 3 pups to Day 25 postpartum.
until Generation 2 pups were 25 days old. Based on
preliminary feeding studies, a formula for postpartum Statistical Analyses
food-restriction was developed that accounted for the The effects of treatment and time on weights of
amount of food consumed by dams of a given body Generation 2 dams, Generation 3 sex ratios (%
weight as well as the amount of food likely to be males/litter), and Generation 3 litter sizes were
consumed by her offspring (all dams and offspring analyzed using two-way analyses of variance
were weighed daily): (ANOVA). The effects of gender and treatment on
 594 HUCK ET AL.

weights of Generation 3 pups for each day of mea- one pup from birth through Day 25. In contrast, all
surement were assessed using ANOVAs with pups pups in three of 17 RR bitters (18%) died by Day 20.
nested within litters (Sokal and Rohlf, 1981). The
Newman-Keuls Multiple-Range test (corrected for Sex Ratios of Generation 3 Litters
unequal sample sizes) was employed to test for The percentage (mean ± SE) of male pups in
differences between any two treatments when an control litters at parturition was significantly greater
ANOVA indicated significant treatment effects than in the three experimental treatments (p<0.04),
(Bruning and Kintz, 1977). whose sex ratios at birth did not differ significantly
from each other (Fig. 3). This difference is attributed
RESULTS to the effects of treatment, as no significant cor-
relations were found within any treatment between
Weights of Generation 2 Females
sex ratio and either dam weights at birth (controls:
Body weights of Generation 2 females prior to r23 = 0.07, p = 0.75; AR: r15 = 0.07, p = 0.89; RA:
pregnancy and during lactation are illustrated in r15 = 0.05, p = 0.84; RR: r15 = -0.41, p = 0.10) or
Figure 1. There were no significant differences in litter size (controls: r23 = 0.20, p = 0.33; AR: r15 =
weights of Generation 2 females from birth through -0.06, p = 0.80; RA; r15 = -0.14, p = 0.60; RR: r15
Day 10, after which AA (control) females weighed = -O.33,p = 0.20).
significantly more than subjects in the other treat- Postpartum sex ratios within treatments remained
ments for the remainder of the experiment (p<0.01 similar to those at birth in control, AR, and RA
on Day 15, p<0.001 thereafter). Adult body weights litters. Mean sex ratios of RR litters increased to
of Generation 2 females were related to the total nearly 50% on Days 10 and 15, but this increase was
amount of food restriction as juveniles. Despite their due to the deaths of all females in two litters. By Day
receiving portions of food ad libitum after Day 20, all pups in those two litters had died, and RR sex
50, body weights of RR females remained significantly ratios once agains became similar to those observed at
less than those of females that were food-restricted birth. Because of the large variation in postpartum
for only 25 days. Weights of AR and RA females sex ratios of RR litters, there were no significant
were intermediate to the other two treatments, but effects of treatment on sex ratios during lactation.
were significantly greater than RR females after Day There were also no significant changes in sex ratios
30 (Figure 1). over time and no significant interaction of treatment
and time.
Generation 3 Litter Sizes and Survivorship
Growth of Generation 3 Pups
Significant differences were observed in the number
of pups per litter at birth among treatments (p=0.03) Male weighed pups significantly more than their
(Fig. 2). However, Newman-Keuls’ test indicated that sisters at birth
in control litters (p=0.016). Male and
this difference existed only between control and RR female pup weights did not differ at parturition in
litters. any experimental litters. No significant differences in
The typical pattern in hamsters of production of body weight were observed for males and females
large bitters and subsequent loss of offspring (Day and within any treatment from Days 5-25.
Galef, 1977) was observed in all treatments (p<0.025). Separate comparisons of male (Fig. 4a) and female
Loss of offspring was most acute in RR litters (Fig. pups (Fig. 4b) by treatment indicate a variable
2). No significant differences in litter size were pattern of weight gain. There were no significant
observed among control, AR, and RA litters, but RR effects of treatment on weights of male pups at birth.
litters always contained significantly fewer pups than In contrast, RR female pups weighed significantly
litters in the other three treatments (p=0.03 at birth more at birth than females in the other three treat-
and on Day 5, p<0.001 thereafter). There was also a ments. By Day 25 both RR males and females weighed
significant interaction between the effects of treat- significantly less than their counterparts in the other
ment and number of days postpartum on litter size three treatments (p<0.O01). No significant differences
(p< 0.00 1). in weights of control, AR, and RA males or females
All control, AR, and RA litters contained at least were observed during the experiment.
 SEX RATIO AND PUP GROWTH IN FOOD-RESTRICTED HAMSTERS 595

140

120

100

I-
I 80
C,
uJ

a
0
60

40

20

0 10 20 30 40 50 60 70 80 90 100 110 120 130

AGE (DAYS) Mail.g P&rtvrllloa
from da

FIG. 1. Body weights (mean ± SEM) of Generation 2 females from birth until Day 130. Days 105 to 130 represent body weights of Generation
2 females from the birth of their offspring to Day 25 postpartum. Broken lines indicate the period of pregnancy when body weights were not
ascertained. Removal of Generation 2 females from their dams, mating, and parturition are shown along the abscissa. AA: controls (fed ad libitum
from Days 1-50), RA: females whose dams were food-restricted from Days 1-25, AR: females food-restricted during Days 25-50, RR: females
whose dams were food-restricted from Days 1-25 and that were food-restricted from Days 26-50. Statistical analyses and details of protocols
for food restriction are in text.

DISCUSSION offspring, and, as hypothesized by Trivers and Williard

Data from the present experiments and those of (1973), the sex ratio of offspring. Labov et al. (1986)
Labov et al. (1986) demonstrate that food restriction stressed female hamsters during pregnancy and
in golden hamsters affects litter survival, growth of lactation. The present study indicates that con-
596 HUCK ET AL.

U) Murphy (1971) reported that pup cannibalism by

caged wild Syrian hamsters was low. Labov et al.
AA#{149}
AR 0 (1985) hypothesized that facultative manipulation of
Q.
RA A
offspring sex ratio may also result from this behavior,
but we found no significant changes in postpartum
cr sex ratio from that observed at birth. This lack of
RR

I
postnatal adjustment of sex ratio was also observed
6-
by Labov et al. (1986).
z
Although food-restriction for fifty days did not
4
affect sex ratios of litters more than did 25 days of
I I I - I I I
deprivation, there was a reduction in survivorship and
0 5 10 15 20 25 growth of both male and female offspring from RR
AGE (DAYS) litters. The number of pups within individual RR
FIG. 2. Number of pups (mean ± SEM) in Generation 3 litters from litters was culled to approximately 50% of that in the
birth (Day 0) to Day 25 postpartum. RR litters always contained other treatments by Day 15 (Fig. 2). All pups in one
significantly fewer pups than litters in the other three treatments
(p0.03 at birth and on Day 5, p<0.001 thereafter). AA: controls
of 17 RR litters were eliminated by Day 15, and two
(dams not food-restricted), RA dams food-restricted as juveniles from additional litters disappeared by Day 20, despite the
birth to Day 25 postpartum, AR: dams food-restricted as juveniles
from Days 26-50 postpartum, RR: dams food-restricted as juveniles
fact that these pups were most likely feeding de-
from birth to Day 50 postpartum. pendently of their dams (c.f., Labov et a!., 1986). In
contrast, no control, AR, or RA litters suffered total
mortality during the experiment. Weights of RR
male pups at birth were similar to, and those of RR
sequences similar to those of food restriction during female pups significantly exceeded the weights of
pregnancy may result when dams have received
males and females in the other treatments. However,
insufficient nutrition early in life. Because there was
by Day 25, RR males and females both weighed
no significant relationship between sex ratios of
significantly less than their counterparts in control,
litters at birth and either litter size (as postulated by
AR, and RA litters.
McGinley, 1984) or the weight of the dam, restricted
access to food for as little as 25 of the first 50 days
postpartum appears sufficient to cause females to
deliver significantly female-biased first litters 60-80
days later.
55-
Our data also suggest that this skewing of sex ratio
resulted
manipulation
from some
rather
unknown
than form
mechanism
greater
of maternal
vulnerability
w
50 + If

IF
AA.
AR0
RA
RR
A
±
of male embryos to stress in uterio (Clutton-Brock et -J

al., 1985). While the number of males/litter was 45-

I-
reduced in all treatments that were food-deprived z
w
early
only
litter
in life,
between
sizes were
mean
control
bitter
and
maintained
sizes
RR
similar
differed
litters.
to that
significantly
AR and RA
of controls
()
LU
0
40-

35 -
cf
by the birth of more females (Fig. 2).
In the laboratory, domesticated female hamsters
typically produce more offspring than they can raise, 0 5 10 15 20 25

and they cull their litters by cannibalizing some pups AGE (DAYS)
(Day and Galef, 1977; Siegel, 1985). Pup cannibalism FIG. 3. Sex ratios (% males) of Generation 3 litters from birth
(Day 0) to Day 25 postpartum. Values expressed as means ± SEM.
may be an artifact of laboratory caging. Cannibalism
Sex ratio of AA (control) litters was significantly greater than those
was reduced, though not eliminated, among female of the other three treatments at birth (p<0.04) but not thereafter
hamsters permitted to hoard food compared to (details of statistical analyses are in text). AA: dams not food-re-
stricted as juveniles from birth to Day 25 postpartum, AR: dams food-
controls (Miceli and Malsbury, 1982). Whether similar restricted as juveniles from Days 26-50 postpartum, RR: dams food-
behavior occurs in the wild is unknown, although restricted as juveniles from birth to Day 50 postpartum.
 SEX RATIO AND PUP GROWTH IN FOOD-RESTRICTED HAMSTERS 597

50 50
004 AA#{149}
AR 0

RA a

40
RR
40

04 04
30 0
CD
. LU

c30 20
820

10 10

401k

0 I I I I I I 0
I I I I I I

0 5 10 15 20 25 0 5 10 15 20 25
AGE (DAYS) AGE (DAYS)

FIG. 4. a) Body weights (mean ± SEM) of Generation 3 males from birth (Day 0) to Day 25 postpartum. AA: Controls (dams not food-restricted),
RA: dams food-restricted as juveniles from birth to Day 25 postpartum, AR: dams food-restricted as juveniles from Days 26-50 postpartum, RR:
dams food-restricted as juveniles from birth to Day 50 postpartum. Levels of statistical significance and Newman-Keuls’ analyses for differences
between any two treatments are as follows: Day 0: p0.079 (NS); Day 5: p<0.0O1, AR and RA> RR; Day 10: p 0.405 (NS); Day 15: p<0.010,
AA> AR, RA, and RR; Day 20: p0.228 (NS); and Day 25: p<0.001, AA, AR, and RA> RR.
b) Body weights (mean ± SEM) of Generation 3 females from birth (Day 0) to Day 25 postpartum. AA: Controls (dams not food-restricted), RA:
dams food-restricted as juveniles from birth to Day 25 postpartum, AR: dams food-restricted as juveniles from Days 26-50 postpartum, RR: dams
food-restricted as juveniles from birth to Day 50 postpartum. Levels of statistical significance and Newman-Keuls’ analyses for differences between
any two treatments are as follows: Day 0: p<0.001, RR > AA, AR, and RA, Day 5: p<0.001, AA, AR, and RA> RR; Day 10: p 0.85 (NS);
Day 15: p<0.03, AA and RR > AR and RA; Day 20: p<0.05, AA, AR, and RA> RR; and Day 25: p<.001, AA, AR, and RA> RR.

The pattern of growth observed in RR pups is forms of physiological stress. Similar studies using
similar to that reported by Salas et al. (1984), who wild or first generation laboratory-reared conspecifics
found that offspring of female rats which had been should also be undertaken. Whether additional litters
food-restricted during their first 23 days of life failed produced by Generation 2 females would exhibit
to gain weight as rapidly as did controls. They at- similar skewing of sex ratios has not been tested. We
tributed this difference to changes in behavior of are currently investigating whether Generation 3 fe-
experimental dams (e.g., reduced nursing, longer time males that have been allowed unrestricted access to
to retrieve pups, less maintenance of nests). These food will repeat these patterns of prenatal adjustment
deficits in maternal care may have been compounded of sex ratio and restricted postpartum growth of their
by the inability of underfed young to elicit adequate offspring (Huck, Labov, and Lisk, in prep).
levels of maternal care (Smart and Preece, 1973). This study focused on overt changes in production
To our knowledge, these data are the first to of offspring by females. However, it is unknown
demonstrate alteration of sex ratios in offspring of whether restricting the food intake of males as
dams that were food-restricted as neonates or juveniles. neonates or juveniles also influences their subsequent
While our results are consistent with the predictions reproductive success. The stunted growth observed in
of Trivers and Willard (1973), further research is RR male pups may be manifest by a diminished
needed to elucidate the mechanisms of sex ratio ability to compete with other males for social domin-
manipulation following food restriction or other ance and ultimately for mating opportunities.
598 HUCK ET AL.

ACKNOWLEDGMENTS Meikle DB, Tilford BL, Vessey SH, 1984. Dominance rank, secondary
sex ratio, and reproduction of offspring of polygynous primates.
We thank Prabha Vaswani, Mary V. Taylor, Barn L. Katz, Deborah Am Nat 124:173-88
E. Principato, and E. Lee Bruce for help with weighing and feeding Miceli MO, Maisbury CW, 1982. Availability of a food hoard facilitates
animals. The comments and suggestions offered by two anonymous maternal behaviour in virgin female hamsters. Physiol Behav
reviewers for improving the manuscript are appreciated. 28:855-56
Michener GR, 1980. Differential reproduction among female Richard-
son’s ground squirrels and its relation to sex ratio. Behav Ecol
Sociobiol 7:173-78
REFERENCES
Murphy MR, 1971. Natural history of the Syrian golden hamster- a
Altmann J, 1980. Baboon Mothers and Infants. Cambridge: Harvard reconnaissance expedition. Am zool 11(Abstr):632
University Press Myers JFI, 1978. Sex ratio adjustment under food stress: maximization
Bronson FH, 1985. Mammalian reproduction: an ecological perspective. of quality or numbers of offspring? Am Nat 112:381-88
Biol Reprod 32:1-26 Myers P, Master LL, Garrett RA, 1985. Ambient temperature and
Bruning JL, Kintz BL, 1977. Computational Handbook of Statistics, rainfall: an effect on sex ratio and litter size in deer mice. J Mam-
2nd Ed. Glenview: Scott Foresman and Co. mal 66:289-98
Clutton-Brock TH, Albon SD, Guinness FE, 1981. Parental investment Paul A, Thommen D, 1984. Timing of birth, female reproductive
in male and female offspring in polygynous mammals. Nature success and infant sex ratio in semifree-ranging Barbary macaques
289:487-89 (Macaca sylvanus). Folia Primatol 42:2-16
Clutton-Brock TH, Albon SD, Guinness FE, 1985. Parental investment Rivers JPW, Crawford MA, 1974. Maternal nutrition and the sex ratio
and sex differences in juvenile mortality in birds and mammals. at birth. Nature 252:297 -98
Nature 313:131-33 Salas M, Torrero C, Pulido S, 1984. Long-term alterations in the ma-
Day CSD, Galef BG, 1977. Pup cannibalism: one aspect of maternal ternal behavior of neonatally undernourished rats. Physiol Be-
behavior in golden hamsters. J Comp Physiol Psychol 91:1179-89 hay 3 3:273-78
Muck UW, Lisk RD, 1985. Determinants of mating success in the Siegel HI, 1985. Parental behavior. In: Siegel HI (ed.), The Hamster,
golden hamster (Mesocricetus auratus): II. Pregnancy initiation. Reproduction and Behavior. New York: Plenum Publishing Corp.,
J Comp Psychol 99:231-39 pp. 207-28
Labov JB, Muck UW, Elwood RW, Brooks RJ, 1985. Current problems Silk JB, 1983. Local resource competition and facultative adjustment
in the study of infanticidal behavior of rodents. Q Rev Biol of sex ratios in relation to competitive abilities. Am Nat 121:
60: 1-20 56-66
Labov JB, Huck UW, Vaswani P, Lisk RD, 1986. Sex ratio manipulation Simpson MFA, Simpson AE, 1982. Birth sex ratios and social rank in
and decreased growth of male offspring of undernourished ham- rhesus monkey mothers. Nature 300:440-41
sters (Mesocricetus auratus). Behav Ecol Sociobiol 18:241-49 Small MF, Smith DG, 1985. Sex of infants produced by male rhesus
Lane EA, Hyde TS, 1973. Effect of maternal stress on fertility and sex macaques. Am Nat 126:354-61
ratio: a pilot study with rats. J Abnorm Psychol 82:73-80 Smart JL, Preece J, 1973. Maternal behavior of undernourished mother
Lanier DL, Estep DQ, Dewsbury DA, 1975. Copulatory behavior of rats. Anim Behav 21 :613-19
golden hamsters: effects on pregnancy. Physiol Behav 1:209-12 Sokal RR, Rohlf J, 1981. Biometry, The Principles and Practice of
Massaro TF, Levitsky DA, Barnes RH, 1974. Protein malnutrition in Statistics in Biological Research, 2nd Ed. San Francisco: W. H.
the rat: its effects on maternal behavior and pup development. Freeman and Co.
Dcv Psychobiol 7:55 1-61 Trivers RL, Willard DE, 1973. Natural selection of parental ability to
McClure PA, 1981. Sex-biased litter reduction in food-restricted wood vary the sex ratio. Science 179:90-92
rats (Neotoma floridana). Science 211:1058-60 Wise DA, Eldred NL, McAfee J, Lauber A, 1985. Litter deficits of
McGinley MA, 1984. The adaptive value of male-biased sex ratios socially stressed and low status hamster dams. Physiol Behav
among stressed animals. Am Nat 124:597-99 35:775-77