Reproduction, Growth and
Development in Captive Beluga
(Delphinapterus leucas)
Todd R. Robeck,1n Steven L. Monfort,2 Paul P. Calle,3 J. Lawrence Dunn,4
Eric Jensen,5 Jeffrey R. Boehm,6 Skip Young,7 and Steven T. Clark8
1SeaWorld Texas, San Antonio, Texas
2Conservation and Research Center, National Zoological Park, Smithsonian Institution,
Front Royal, Virginia
3Wildlife Conservation Society, Bronx, New York
4Mystic Aquarium, Mystic, Connecticut
5U.S. Navy Marine Mammal Program, San Diego, California
6John G. Shedd Aquarium, Chicago, Illinois
7Vancouver Aquarium Marine Science Centre, Vancouver, British Columbia
8Corporate Zoological Operations, Orlando, Florida
Recent success propagating captive beluga has resulted from combined efforts by
North American zoos and aquariums to manage disparate collections as a single
population. This success has provided a tremendous opportunity to increase our
understanding of beluga reproductive biology. Blood samples were collected on a
weekly to biweekly basis from 23 female and 12 male beluga, ranging in age from
2?15 years, for analysis of serum progesterone (P) and testosterone (T),
respectively. Peri-parturient observational data, including food intake, duration
and signs of labor, and nursing patterns were collected from 15 days prepartum to
30 days postpartum during 21 births. Total body lengths and weights were
collected from 10 captive-born beluga. For female beluga, the mean (7SD) age,
body length, and weight at ?rst conceptions were 9.172.8 years, 318.079.1 cm,
and 519784 kg. Thirty-?ve luteal phases and 13 conceptions were detected from
January?June, and 70% of luteal phases and 80% conceptions occurred from
March?May. The mean luteal phase and total estrous cycle lengths were
30.076.5 days and 48.074.6 days, respectively. For male beluga, the mean age
that males sired their ?rst calf was 13.372.6 years. Compared to younger
males (o8 years of age, 0.95 ng/ml), levels of T secretion in older males (48 years
of age, 5.0 ng/ml) were elevated signi?cantly only during the interval from
Grant sponsor: Wildlife Conservation Society; Grant sponsor: Busch Entertainment Corporation.
nCorrespondence to: Todd R. Robeck, DVM, PhD; SeaWorld Texas, 10500 SeaWorld Drive, San
Antonio TX 78251. E-mail: Todd.Robeck@SeaWorld.com
Received 5 May 2004; Accepted 3 September 2004
DOI 10.1002/zoo.20037
Published online in Wiley InterScience (www.interscience.wiley.com).
Zoo Biology 24:29?49 (2005)

c 2005 Wiley-Liss, Inc.
January?April. Highest T concentrations (6.274.9 ng/ml) were recorded from
January?March, whereas nadir concentrations (1.171.0 ng/ml) were detected
from August?September. The mean gestation length was 475.0720.4 days (n? 9).
For parturition, the mean time from the ?rst appearance of ?uke or rostrum to
delivery, delivery to placental passage, and delivery to nursing were 4.472.9 hr,
7.671.8 hr, and 43745 hr, respectively. All cows had decreased food intake on
the day of delivery, with 44% having zero intake. Peak 24-hr nursing activity
occurred 3.972.7 days post-partum. Growth (i.e., body weight and length) as a
function of age were well described by the Gompertz model (r2? 0.91, 0.93).
Based on the model, growth in body weight and length were signi?cantly greater
in males compared to females. Predicted birth weight (88.9 kg) was similar for
both sexes, however, and male calves were predicted to be shorter (154.3 cm) than
female calves (160.7 cm). The results provide the ?rst descriptions of captive
beluga reproductive physiology, including endocrinology, peri-parturient beha-
vior, growth, and reproductive maturity. This knowledge is important for helping
to maintain genetically diverse, self-sustaining populations of captive beluga
whales. Zoo Biol 24:29?49, 2005. 
c 2005 Wiley-Liss, Inc.
Key words: mondontidae, cetacean age and growth, testosterone, progesterone, captive
cetaceans
INTRODUCTION
Historically, beluga (Delphinapterus leucas) has been one of the most important
food sources for Inuit?s living in the arctic [Heide-J
rgensen, 1990]. Despite the
continued reliance on this species, little information has been collected concerning
their basic reproductive physiology. For the most part, existing information has been
derived from observation of wild stocks or post-mortem analysis of animals collected
during native harvests [Brodie, 1971; Braham, 1984; Burns and Seaman, 1986;
Doidge, 1990; Heide-J
rgensen, 1990; Heide-J
rgensen and Teilmann, 1994; H
ier
and Heide-J
rgensen, 1994]. Assuming accurate methods exist for aging the animals
harvested [Clark et al., 2000], and despite access restricted by season, this type of
data has been used to document certain biological functions including reproductive
seasonality, growth, and development. Free-ranging male and female beluga have
been estimated to be reproductively mature between 4?7 and 6?9 years of age,
respectively [Brodie, 1971; Braham, 1984; Heide-J
rgensen and Teilmann, 1994]. The
species is believed to exhibit reproductive seasonality with breeding observed from
April?May, although slight geographical variation exists [Brodie, 1971; Heide-
J
rgensen and Teilmann, 1994]. Estimates of total gestation length range widely
from 330 days (Western Greenland) [Heide-J
rgensen and Teilmann, 1994] toB435
days (Canadian) [Brodie, 1971], and lactation is believed to last approximately 24
months [Brodie, 1971].
H
ier and Heide-J
rgensen [1994] evaluated serum collected post mortem
during May and September to describe mean serum testosterone (T) levels of 4.14
nmol/l (1.19 ng/ml) and 0.96 nmol/l (0.27 ng/ml) in mature and immature males,
respectively. In addition, serum progesterone (P) in pregnant females was 27.9 nmol/l
(9.15 ng/ml).
Thirty-two beluga currently reside in nine North American zoological
institutions [L. Garibaldi, personal communication]. Since 1988, these facilities have
been involved in a coordinated captive breeding management effort that has
30 Robeck et al.
resulted in the birth of 21 calves. The ready animal access inherent in zoological
settings, combined with the high tractability of beluga, provides a unique
opportunity to conduct systematic research to study the reproductive biology
of this species. Nevertheless, the only beluga reproductive biology publications
to date have been limited to descriptions of suckling behavior, nursing patterns,
and preliminary data on gestational steroid hormones [Drinnan and Sadler,
1981; Calle et al., 1993, 1996; Russell et al., 1997]. Peak 24-hr nursing activity
was described to occur at B7?10 days of age, and suggested that abnormal
nursing patterns could provide an early indicator of calf distress [Cook et al.,
1992; Russell et al., 1997]. Peak gestational P (60?66 ng/ml) and estradiol
(30?31 pg/ml) concentrations occurred by 4 months of pregnancy [Calle et al.,
1993, 1996].
With known ages and the ability to collect accurate serial morphometric
measurements and serum samples (for reproductive hormone assays), captive
populations provide a unique opportunity to improve our understanding of
reproductive biology in this species [Clark et al., 2000]. The objectives of our
research with captive beluga were to quantify seasonal endocrine changes in females
(serum P) and males (serum T); to establish the age of sexual maturity; to document
behaviors during parturition and the peri-parturient interval; to compare and
contrast nursing patterns in normal and clinically distressed calves, and; to
characterize growth rates in captive born calves and compare these data to estimates
of growth rates in wild beluga.
MATERIALS AND METHODS
Animals and Sample Collection
Blood samples were obtained by voluntary presentation or during routine
venipuncture as part of medical or management procedures from 23 female and
12 male beluga for various intervals from 1983?1998. The animals were located
at nine different facilities: John G. Shedd Aquarium, Chicago, IL; Mystic
Aquarium, Mystic, CT; Point De?ance Zoo and Aquarium, Point De?ance,
WA; SeaWorld of Texas, San Antonio, TX; SeaWorld of California, San Diego,
CA; SeaWorld of Florida, Orlando, FL; US Navy Marine Mammal Program,
San Diego, CA; Vancouver Aquarium Marine Science Centre, Vancouver, B.C.;
and New York Aquarium, Wildlife Conservation Society, Brooklyn, NY. The
US Navy Marine Mammal facility and the New York Aquarium are natural salt
water systems ranging in temperature from 13?211C and 1.7?281C, respectively. All
other facilities housed animals in manufactured salt water at temperatures ranging
from 17?201C.
The blood sampling periods for individual animals ranged from 2?15 years.
Many of the samples or observational data were based on the availability of stored
samples or animal records. During the main sampling period (1996?1998), 17
females ranging from 8?30 years were bled weekly to bi-weekly for 2 years (Table 1).
In addition, 304 serum samples were collected from 10 males that were bled biweekly
to quarterly over a maximum of 15 years (1983?1998). The males? ages ranged from
3?21 years at sampling onset (Table 2).
Reproduction in Captive Beluga 31
T Radioimmunoassay
Male beluga serum or heparinized plasma was analyzed in duplicate using a
double-antibody [125I] RIA (ICN, Costa Mesa, CA) for T according to the
instructions provided except all reagent volumes were halved. The antiserum cross-
reacts 100% with T, 3.4% with 5a-dihydrotestosterone, 2.2% with 5a-androstane-
3b-17b-diol, 2.0% with 11-oxotestosterone, and o1% with all other steroids tested.
P Radioimmunoassay
Female beluga serum or heparinized plasma was analyzed in duplicate using a
double-antibody [125I] RIA that cross reacts with a wide variety of P metabolites. All
female serum extracts were analyzed using [125I] RIA described previously for P
[Brown et al., 1994; Wasser et al., 1994]. The monoclonal antiserum cross-reacts
100% with P, 96% with 5a-pregnane-3b-ol-20-one, 36% with 5a-pregnane-a-ol-20-
one, 15%, with 17a-hydroxyprogesterone, 13% with pregnenolone, 7% with
TABLE 1. Details of captive female beluga
Animal ID EDOBa
Facility
housed
Date range
of serum
samples Samples (n)
Conceptions
during
sampling (n)
Calves with
observational
data (n)
Female 1 1983 Mysb 04/95?09/98 61 0 0
Female 2 1986 VCc 06/95?03/98 21 0 0
Female 3 1985 SWTd No Samples 0 0 2
Female 4 1985 SWT No Samples 0 0 0
Female 5 1986 SWT No Samples 0 0 1
Female 6 1986 JGSe 02/95?06/98 25 1 1
Female 7 1970 WCSf 03/90?09/98 141 1 1
Female 8 1983 Mys 06/95?06/98 29 0 0
Female 9 1986 SWT 09/92?06/97 43 1 1
Female 10 1982 WCS 02/92?08/98 110 1 1
Female 11 1986 SWT 01/92?06/97 47 2 3
Female 12 1986 SWT No Samples 0 0 4
Female 13 1967 Navyg 02/88?08/96 49 0 0
Female 14 1981 JGS 02/96?06/98 77 1 4
Female 15 1983 Mys 02/95?09/98 73 0 0
Female 16 1980 WCS 02/91?08/98 113 1 2
Female 17 1989 PtDh 03/97?12/98 48 0 0
Female 18 1985 SWT No Samples 0 0 0
Female 19 1986 JGS 07/92?06/98 37 1 1
Female 20 1978 Navy 04/88?08/96 44 0 0
Female 21 1984 PtD 02/96?11/98 106 0 0
Female 22 1986 SWT 05/93?05/97 40 0 1
Female 23 1977 SWT No Samples 0 0 2
aEstimated date of birth.
bMystic Aquarium
cVancouver Aquarium.
dSeaWorld of Texas.
eJohn G. Shedd Aquarium.
fNew York Aquarium, Wildlife Conservation Society.
gUS Navy Marine Mammal Program.
hPoint De?ance Zoo and Aquarium.
32 Robeck et al.
5b-pregnane-3a-ol-20-one, 5% with 5b-pregane-3a,17a-diol, 20a-one, and o1%
with pregnanediol-3a-glucuronide and all other steroids tested.
Endocrine Data Analysis
Estimates of luteal phase and estrous cycle durations were limited to females
that were blood sampled a minimum of every 2 weeks. For each individual female, P
concentrations that exceeded 3 ng/ml and were at least 2.0 the mean non-pregnant
P concentration for that particular individual were considered presumptive evidence
of luteal activity. When a sample below this threshold was serially adjacent to a
sample above the threshold, the beginning or end of a luteal phase was de?ned as
median point between these two samples. The value with the highest concentration
during a period of luteal activity was considered the peak. An estrous cycle was
de?ned as the number of days between the beginning of two successive luteal phases.
All of the cycle characteristics were calculated for each individual animal and then
the data were pooled to determine mean values for the population. To determine
seasonal estrous activity, any month(s) in which peak P of a luteal phase occurred
was given a value of one. For analysis, all monthly data were combined across years
to develop a composite 12-month period.
Peri-parturient Observations
Data were not available for all females for each set of observations. For peri-
parturient food intake, values (kg/d) obtained from each animal?s record (recorded
as a standard husbandry practice) were utilized. For trend analysis, mean food
intake was determined for all animals 15 days before and after parturition (Day 0).
The date and time of the following events were recorded for each birth: ?rst vaginal
discharge, appearance of ?uke or rostrum, delivery, and placental passage. For live
calves, initial postpartum nursing time and total daily nursing were recorded.
TABLE 2. Details of captive male beluga
Animal ID EDOBa
Facility
housed
Date range
of serum
samples Samples (n)
Calves
sired (n)
Age at ?rst
conception
Male 1 1985 SWTb 02/88?01/94 43 1 12
Male 2 1987c WCSd 02/91?09/94 45 0 NA
Male 3 1983 VCe 07/91?02/94 7 1 16
Male 4 1982 JGSf 01/94?06/94 15 3 9
Male 5 1985 SWT 05/86?04/81 12 1 13
Male 6 1986 JGS 09/90?03/11 6 0 NA
Male 7 1969 WCS 02/79?07/94 9 2 17
Male 8 1975 Navyg 03/84?08/92 48 0 NA
Male 9 1979 SWT 02/88?04/92 60 4 12
Male 10 1980 WCS 04/85?12/94 59 2 12
aEstimated date of birth.
bSeaWorld of Texas.
cCaptive born.
dNew York Aquarium, Wildlife Conservation Society.
eVancouver Aquarium.
fJohn G. Shedd Aquarium.
gUS Navy Marine Mammal Program.
Reproduction in Captive Beluga 33
Nursing data were collected with 24-hr observations beginning immediately
postpartum and continuing for 30 days.
Age, Growth, and Morphometry
Total body length and weight were collected from 10 captive-born beluga (4
females, 6 males, 2 neonates) maintained at SeaWorld parks from August 1992 until
January 2003. Total body length was assessed as per Norris [1961], whereas body
weights were obtained using a hydraulic scale (Model 747-915-40; Emery Winslow
Scale Company; Seymour, CT). Total body length (cm) and weight (kg) parameters
were used to construct age-based growth models.
A number of growth models have been used in the examination of size at age.
For the present study, both the von Bertalanffy and Gompertz models were
examined for ?t. The Gompertz model provided an adequate ?t as evidenced by an
examination of the residuals. Additionally, due to the prevalence of this particular
model in cetacean age and growth studies [Perrin and Henderson, 1984; Doidge,
1990; Read and Gaskin, 1990; Read et al., 1993; Ferrero and Walker, 1995;
Fernandez and Hohn, 1998], we concluded its use for this dataset would be most
appropriate because it would allow intraspeci?c comparison of growth parameters to
those currently in the literature [Doide, 1990; Heide-J
rgensen and Teilmann, 1994].
The speci?c form of the Gompertz model used was taken from Fitzhugh [1975]
WTorTL ? Winf or Linf x?exp?  b x exp?  k x t??
where WT or TL is the weight or total length of the animal at age, Winf or Linf is the
asymptotic weight or length growth value for this particular dataset, b is the
integration constant, k is the growth rate constant, and t is age (in years).
Expectedly, each animal?s growth data were auto-correlated (runs test for trend
data categorized by animal, P-valueso0.01). Therefore, the nonparametric statistical
technique of bootstrapping, described in Clark and Odell [1999], was necessary to
eliminate the problems associated with autocorrelation [Sokal and Rohlf, 1995]. The
bootstrapping technique allowed calculation of parameter estimates, standard errors
(SE), and 95% con?dence intervals (CI). These calculations then provided the
opportunity to analyze model parameter estimated gender and study-based
differences.
Statistical Analysis
Descriptive statistics were applied to the data using Microsoft Excel
(Microsoft, Redmond, WA). Kruskal-Wallis one-way ANOVA on ranks and
Duncan?s multiple comparison test were used to compare mean monthly serum T
values for mature males and to compare mean serum T levels between adult and
immature males during the breeding season. Student?s t-tests were used to compare
the time interval from ?ukes or rostrum presentation to parturition, and delivery to
initial nursing in primiparous vs. multiparous animals. These data analyses were
carried out using the SigmaStat for Windows (SPSS Statistical Software, San Rafael,
CA). Morphometric statistical analyses were accomplished using the SYSTAT
statistical package (version 8.0 for Windows, SPSS Statistical Software, 1998). Data
are presented as mean7SD; Po0.05 was considered signi?cant.
34 Robeck et al.
RESULTS
Serum T
Parallel displacement curves were obtained by comparing serial dilutions
(range? undiluted  1:32) of pooled beluga serum and T standard preparations.
Recovery of known amounts of unlabeled T (range? 0.05?5.0 ng/ml) added to a
pool of diluted beluga serum was 123.4711.9% (y? 0.06? 1.0x, r2? 0.99). RIA of
elutes after HPLC [Monfort et al., 1991] showed that all immunoreactivity was
associated with a single peak that co-eluted with T. Beluga serum was diluted (1:4?
1:8) and assayed (25 ml) in duplicate. Inter- and intra-assay variation were 8.5% and
o10%, whereas assay sensitivity was 0.05 ng/ml.
Serum P
Parallel displacement curves were obtained by comparing serial dilutions
(range? undiluted  1:32) of pooled beluga serum and P standard preparations.
Recovery of known amounts of unlabeled P (range? 3.75?240 pg/ml) added to a
pool of diluted beluga serum was 106.8713.3% (y?1.09? 1.1x, r2? 0.99). RIA of
elutes after HPLC [Monfort et al., 1991] showed that 480% of immunoreactivity
was associated with a single peak that co-eluted with P. Beluga serum was diluted
(neat?1:2) and assayed (100 ml) in duplicate. Inter- and intra-assay variation were
12.0% and o10%, whereas assay sensitivity was 3.75 pg/ml.
Samples that had been collected before the study-sampling period were assayed
for P by a commercial lab using RIA. Because different methods were used for
sample analysis, only samples assayed by the same method were used to determine
mean concentrations. Despite concentration differences, correlation between
commercial lab results and the monoclonal P RIA during the luteal phase and
early pregnancy was 0.73 (Po0.01). Twenty of the 22 females (91%) sampled
exhibited a total of 54 P peaks (34.4744.5 ng/ml), which were presumed to represent
luteal phase activity or pregnancy. Two of the females did not secrete elevated P
during the study. Thirteen conceptions and 10 births occurred during the serum P
sampling period.
Female Reproductive Maturity
The mean age, length, and weight when luteal concentrations of serum P were
?rst detected was 6.971.5 year (n? 9), 31879 cm (n? 6), and 519783 kg (n? 6),
respectively. The mean age, length, and weight at ?rst conception were 9.172.8 year
(n? 16), 328717 cm (n? 7) and 561772 kg (n? 7), respectively. The mean age at
?rst conception for animals (n? 6) housed with proven sires during the season when
the ?rst conception occurred was 8.073.4 years. Of these females, 4 of 6 conceptions
occurred during their ?rst ovarian cycle at the age of 6 years. The oldest age for
conception in a multiparous female was 20 years.
Seasonality
During periods when data were collected throughout the breeding season, all
but two of the reproductively mature females exhibited at least one elevated serum P
value indicative of luteal activity. The mean number of luteal events for each animal
per season was 1.370.4 (n? 35). All but two luteal events occurred between January
and July, and 70.4% occurred in March, April, and May (Fig. 1). Conceptions
Reproduction in Captive Beluga 35
(n? 13) occurred from February to June and 80.6% occurred in March, April, and
May (Fig. 1).
Estrous Cycle and Pregnancy Characteristics
Mean luteal phase and estrous cycle durations were 30.076.5 days (n? 20) and
47.874.6 days (n? 4), respectively. The mean number of cycles that occurred before
a conception was 0.670.5 (n? 10). The mean gestation length was 475.0720.4 days
(n? 9).
Month
Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec
Te
st
os
te
ro
ne
 n
g/
m
l
0
2
4
6
8
10
12
Adult
Juvenile 
Lu
te
al
 P
ha
se
 a
nd
 C
on
ce
pt
io
ns
0
2
4
6
8
10
12
14
16
18
Luteal Peaks 
Conceptions 
Fig. 1. Demonstrates seasonal variations in female (top graph) and male (lower graph)
reproductive parameters. For the female, the bar charts illustrate frequency of luteal activity
and conceptions. For the male the graph compares monthly mean values of T between adult
(Z8 years) and juvenile (o8 years) beluga. T-values were signi?cantly different (Po0.05)
between adult and juvenile males during Jan?May, July, and August.
36 Robeck et al.
Male Sexual Maturity and Seasonality
The youngest male to sire a calf was 9 years old, and the mean age at which
males ?rst sired a calf was 13.372.6 years (n? 8). Thus, all animalso8 years of age
were considered sexually immature. Based on this division, peak serum T production
in adult males occurred from January?April (Fig. 1). The months of January,
February, and March (mean? 6.274.9 ng/ml) had signi?cantly higher median (5 ng/
ml) values (Po0.01) than the nadir months of August and September (median? 0.9
ng/ml; mean? 1.171.0 ng/ml). Comparisons between median serum T from mature
and immature animals on a month-to-month basis show signi?cant differences
(Po0.05) during January?May, July, and August (Fig 1).
Peri-parturient Observations
From 1990?1999, a total of 25 pregnancies resulted in 21 births. Fourteen of
the pregnancies produced male calves (56%), nine were female (36%) and the sexes
of two were undetermined. One female aborted twin male calves at approximately 11
months of gestation. Three cows experienced dystocia and calves were delivered with
manual assistance. Of the 21 unassisted births, 18 were live births, two died shortly
after birth, and one was stillborn. Two calves were born in a head?rst presentation
(9.5%), and one of these calves died shortly after birth. The mean time from ?rst
observed vaginal discharge to appearance of ?ukes and the subsequent interval until
birth was 3.272.6 hr (range? 0.2?8 hr, n? 17) and 4.472.9 hr (range? 0.4?12.4 hr,
n? 18), respectively. The mean time from delivery to placental passage was 7.671.8
hr (range? 4.6?11.8 hr, n? 18). The mean and median time from parturition until
the onset of nursing was 43.0745.4 hr (range? 5.3?144.0 hr, n? 15) and 22.7 hr,
respectively. The time period from ?rst ?uke appearance to birth and the subsequent
interval until the ?rst postpartum nursing tended to be longer in primiparous cows
(4.471.9 hr, n? 5; 52.3762.1 hr, n? 4, respectively) compared to cows delivering
their second calf (2.770.7 hr, n? 5; 19.1715.3 hr, n? 4, respectively).
For all animals with available data (n? 16), food intake decreased close to and
on the day of parturition (Fig. 2). Seven of 16 (44%) beluga had zero food intake on
the day of parturition. After parturition, ?rst consumption of food by cows occurred
at 1.871.0 days (range? 1?4 days) and food intake returned to normal pre-
parturient levels by 6.273.4 days (range? 1?14 days, n? 14). Two of the animals
were not included in this data set, however, because their food consumption had not
returned to normal baseline levels by 15 days postpartum.
Age, Growth, and Morphometry
Of 18 live births, one died within 30 days from a congenital heart defect, and a
second died within 60 days from an infection; another two animals died before the
age of 2 years from infectious disease. Thus, 14 of 22 births (63.6 %) resulted in
calves that reached 2 years of age. Mean 24-hr nursing time for normal calves (i.e.,
those surviving for 42 years) is illustrated in Figure 3. In addition, nursing patterns
of the four calves that died before the age of 1 year were compared to nursing pro?les
of normal calves (Fig. 3). Animal 1 had a period of increased nursing activity around
Day 18; this animal was diagnosed with pulmonary nocardiosis and despite
treatment died at 60 days [Robeck et al., 1994]. Animal 2 had prolonged elevations
of nursing time from Day 6 to Day 12, but this animal died at Day 29 from a
Reproduction in Captive Beluga 37
congenital heart defect [L. Dalton, unpublished data]. Animal 3 consistently
exhibited a below-average nursing pattern and died by 6 weeks of age from a
bacterial infection. Animal 4 did not nurse initially, was administered supportive
care and antibiotics on Day 5, began nursing by Day 10, but eventually died from a
bacterial infection. For normal calves, the period from parturition to peak nursing
time was 3.972.7 d (range? 1?10 days, n? 13), but for calves that died before the
age of 1 year, peak nursing occurred 5.072.4 days post-partum (range? 3?8 days,
n? 4). The mean body weight and length of beluga calves up to 5 days of age was
61.874.5 kg (n? 5) and 148.076.2 cm (n? 3), respectively.
Days Post Partum
-15 -10 -5 0 5 10 15
K
ilo
gr
am
s/d
ay
-2
0
2
4
6
8
10
12
14
16
18
20
22
24
Fig. 2. Pre- and post-partum food consumption in captive beluga (n? 16).
38 Robeck et al.
Total body length and weight
The relationship between body weight and total length was exponential
(r2? 0.94; Fig. 4). The overlap of the 95% CI for male and female parameter
M
in
u
te
s 
N
ur
si
n
g
0
20
40
60
80
100
Normal Animal Mean
Animal 1
M
in
u
te
s 
N
ur
si
n
g
0
20
40
60
80
100
Normal Animal Mean
Animal 3
M
in
u
te
s 
N
u
rs
in
g
0
20
40
60
80
100
Normal Animal Mean
Animal 2
Days Post Partum
0 2 4 6 8 10 12 14 16 18 20 22 24 26 28 30
M
in
u
te
s 
N
u
rs
in
g
0
20
40
60
80
100
Normal Animal Mean
Animal 4
Fig. 3. Nursing pro?les (minutes of nursing/day) for calves surviving after (pooled mean
data? normal animal mean) or dying before (individual data?Animals 1?4) 1 year of age.
Reproduction in Captive Beluga 39
estimates (Table 3) indicated a lack of sexual dimorphism at this age, which
permitted analyses to be conducted on combined sex data. Present study parameter
estimates differed from those reported in the literature (Table 4).
Age and body weight
Weight at age was described by a Gompertz model (r2? 0.91 for the sexes
combined dataset) and was sexually dimorphic (Fig. 5). Sexual dimorphism was
more apparent than in the total length model. The predicted growth curves for males
and females were clearly different in shape and there was an absence of overlap
between the parameter estimates 95% CI for both maximum predicted length
(WTN) and growth rate constant (k) (Table 3). Model-estimated weights at birth for
the combined sexes data was 88.9 kg, whereas gender-speci?c models returned
estimated birth weights of 70.7 kg and 107.8 kg for males and females, respectively.
Present study parameter estimates were signi?cantly different to those reported by
Heide-Jorgensen and Teilmann [1994] (Table 4).
Age and total body length
A Gompertz function adequately described the relationship between age and
total body length (r2? 0.93 for the sexes combined dataset) and suggested growth
dynamics in total length were sexually speci?c (Fig. 6). Body length at birth for both
0
50
100
150
200
250
300
350
400
450
100 125 150 175 200 225 250 275 300 325
Total length (cm)
W
ei
gh
t (k
g)
wt = 0.00279 *  tl 2.08 
r2 = 0.94; n = 9 
 
Fig. 4. Exponential relationship between body weight (kg) and total length (cm) in captive
beluga. Data combined for sex (males: n? 5; females: n? 4).
40 Robeck et al.
sexes combined was predicted by the model to be 158.7 cm. The predicted body
length of females at birth was slightly greater than that for males (160.7 and 154.3,
respectively). Theoretical maximum body length was greater for males (333.2 cm)
than females (284.1 cm). The steeper incline of the male growth curve derived from
the model-predicted growth curve (Fig. 6) provided visual evidence of support for
gender differences in asymptotic length. The male growth rate constant (k) was less
than the female, whereas growth rate constants of either sex were substantially
greater than those reported in the literature (Table 5).
DISCUSSION
Similar to ?ndings for age at ?rst pregnancy for wild beluga [Heide-J
rgensen
and Teilmann, 1994; Braham, 1984], we found the mean age of ?rst ovulation was
6.9 years for captive females. The mean age for ?rst pregnancies in captive beluga
was 9.1 years, however, which is later than their wild counterparts that apparently
conceive during their ?rst estrous cycles. In captive populations, the inconsistent
availability of breeding males may arti?cially lengthen the period between the ?rst
estrous cycle and ?rst pregnancy. This is supported by the observation that 67% of
captive females (n? 6) maintained in the presence of a proven breeding male became
pregnant at 6 years of age. These data support previous speculation [Heide-
J
rgensen and Teilmann, 1994] that given the right social environment, beluga can
conceive during their ?rst season of cyclic ovarian activity. Similar to the maximum
age of reproduction for wild female beluga (21 years) [Brodie, 1971] the maximum
age of reproduction for captive females was 20 years. Although this may be due to
reproductive senescence [Brodie, 1971], the captive population of female beluga is
too young to establish if or when senescence occurs.
The ability to serially collect hormonal data enabled us to begin to describe the
reproductive biology of captive beluga. We estimated a luteal phase of 30 days and a
TABLE 3. Parameter estimates with associated 95% CI of the weight and total length
morphometric relationship for captive belugas at SeaWorld parks with comparisons to data from
previous studiesa
A B
This study
Heide-
Jorgensen
and
Teilmann
[1994]
Doidge
[1990] This study
Heide-
Jorgensen
and
Teilmann
[1994]
Doidge
[1990]
Both genders 0.002794 0.000514 0.000031 2.080 2.37 2.85
95% CI 0.001339?0.004249 1.983?2.177
Males 0.002224 0.000295 0.000002 2.132 2.47 3.37
95% CI 0.000778?0.003670 2.013?2.252
Female 0.004750 0.000137 0.003515 2.0000 2.21 2.03
95% CI 0.000051?0.009448 1.812?2.188
aWeight in kg, length in cm. Males: n? 5; Females: n? 4. Length Morphometric relationship
weight?A n total lengthB; where A was the initial value of weight when total length was 0 and
B was the exponential growth factor.
Reproduction in Captive Beluga 41
T
A
B
L
E
4.
P
ar
am
et
er
es
ti
m
at
es
w
it
h
as
so
ci
at
ed
95
%
C
I
of
th
e
G
om
pe
rt
z
no
n-
lin
ea
r
re
la
ti
on
sh
ip
be
tw
ee
n
to
ta
l
le
ng
th
an
d
ag
e
fo
r
ca
pt
iv
e
be
lu
ga
s
fr
om
S
ea
W
or
ld
pa
rk
s
w
it
h
co
m
pa
ri
so
ns
to
da
ta
fr
om
pr
ev
io
us
st
ud
ie
sa
T
L
N
b
k
T
hi
s
st
ud
y
B
ur
ns
an
d
Se
am
an
[1
98
5]
D
oi
dg
e
[1
99
0]
H
ei
de
-
Jo
rg
en
se
n
an
d
T
ei
lm
an
n
[1
99
4]
T
hi
s
st
ud
y
B
ur
ns
an
d
Se
am
an
[1
98
5]
D
oi
dg
e
[1
99
0]
H
ei
de
-
Jo
rg
en
se
n
an
d
T
ei
lm
an
n
[1
99
4]
T
hi
s
st
ud
y
B
ur
ns
an
d
Se
am
an
[1
98
5]
D
oi
dg
e
[1
99
0]
H
ei
de
-
Jo
rg
en
se
n
an
d
T
ei
lm
an
n
[1
99
4]
B
ot
h
ge
nd
er
s
33
0
0.
73
0.
43
95
%
C
I
32
0?
34
0
0.
66
?0
.8
0
0.
33
?0
.5
3
M
al
es
33
3
42
7
34
9
48
3b
/4
34
c
0.
77
0.
74
0.
71
0.
95
b
/0
.7
5c
0.
47
0.
14
0.
28
0.
22
b
/0
.2
1c
95
%
C
I
32
2?
34
4
0.
71
?0
.8
2
0.
38
?0
.5
6
F
em
al
es
28
4
35
9
33
0
38
5b
/3
75
c
0.
57
0.
68
0.
60
0.
62
b
/0
.6
5c
0.
71
0.
30
0.
27
0.
27
b
/0
.3
1c
95
%
C
I
27
3?
29
5
0.
46
?0
.6
7
0.
50
?0
.9
1
a L
en
gt
h
in
cm
.a
ge
in
ye
ar
s.
M
al
es
:n
?
5;
fe
m
al
es
:n
?
4
G
om
pe
rt
z
no
n-
lin
ea
r
re
la
ti
on
sh
ip
:T
L
?
T
L
in
f

[e
xp
(
b

ex
p
(
k

t)
);
w
he
re
T
L
w
as
th
e
to
ta
ll
en
gt
h
of
th
e
an
im
al
at
ag
e,
T
L
in
f
w
as
th
e
as
ym
pt
ot
ic
to
ta
ll
en
gt
h
gr
ow
th
va
lu
e,
b
w
as
th
e
in
te
gr
at
io
n
co
ns
ta
nt
,k
w
as
th
e
gr
ow
th
ra
te
co
ns
ta
nt
,
an
d
t
w
as
ag
e
(i
n
ye
ar
s)
.
b
W
es
t
G
re
en
la
nd
.
c W
hi
te
an
d
K
ar
a
Se
as
.
total estrous cycle length of 47 days, which is longer than the bottlenose dolphin (30
days) [Brook, 2000] and the killer whale (40 days) [Walker et al., 1988]. Future
efforts should be directed toward conditioning beluga for daily urine sample
collections, which will permit more detailed analysis of the various components of
the estrous cycle.
Similar to most populations of wild beluga, our peak conception period
occurred in March, April, and May [Brodie, 1971; Heide-J
rgensen and Teilmann,
1994]. Using hormone data, however, we were able to demonstrate that female
beluga are seasonally polyestrous, with animals exhibiting up to two ovulations per
season. For animals that conceived during the breeding season, approximately 50%
conceived during the second estrous cycle. This suggests that in the wild there would
be more than one opportunity for female beluga to conceive during the breeding
season.
Gestation estimates of wild beluga have varied from 330 days to a maximum of
441 days [Brodie, 1971; Sergeant, 1973; Heide-J
rgensen and Teilmann, 1994]. This
variation seems to depend on population location and methodology used to
extrapolate conception dates. Relying on hormone data, we were able to estimate
conception within a 2-week period and determined a longer mean gestation period of
475720 days. The reason that captive animals seem to have a longer gestation than
wild animals is unknown, but may re?ect the inaccuracy of methodologies employed
previously.
0
50
100
150
200
250
300
350
400
0.0 0.5 1.0 1.5 2.0 2.5 3. 0 3 .5 4.0 4 .5 5. 0
Age (years)
W
ei
gh
t (k
g)
females: wt = 528.7 * e ?1.589 * e
r2 = 0.92; n = 4 
-028 * age
males: wt = 371.91 * e ?1.65 * e 
r2 = 0.97; n = 5
-0.79 * age
Fig. 5. Gompertz non-linear relationship between weight (kg) and age (years) by sex for
captive beluga (males: n? 5; females: n? 4).
Reproduction in Captive Beluga 43
TABLE 5. Gompertz non-linear relationship between weight and age by sex parameter estimates
for captive belugas from SeaWorld parks with comparisons to data from previous studiesa
WTN b k
This study
Heide-
Jorgensen
and
Teilmann
[1994] This study
Heide-
Jorgensen
and
Teilmann
[1994] This study
Heide-
Jorgensen
and
Teilmann
[1994]
Both genders 383 1.46 0.53
95% CI 375?391 1.39?1.53 0.45?0.60
Males 372 483b/434c 1.66 0.95b/0.75c 0.79 0.22b/0.21c
95% CI 357?387 1.60?1.72 0.71?0.87
Female 529 385b/375c 1.59 0.62b/0.65c 0.28 0.27b/0.31c
95% CI 273?295 1.34?1.84 0.20?0.35
aLength in cm. age in years. Males: n? 5; females: n? 4. Gompertz non-linear relationship:
TL?TLinf [exp (b exp (k t)); where TL was the total length of the animal at age,
TLinf was the asymptotic total length growth value, b was the integration constant, k was the
growth rate constant, and t was age (in years).
bWest Greenland.
cWhite and Kara Seas.
0
50
100
150
200
250
300
350
400
0.0 0.5 1.0 1.5 2.0 2 .5 3.0 3 .5 4.0 4 .5 5.0
Age (years)
To
ta
l l
en
gt
h 
(cm
)
males: wt = 333.2 * e ?0.766 * e  
r2 = 0.96; n = 5
-0.47 * age 
females: wt = 284.1 * e ?0.568 * e 
r2 = 0.93; n = 4 
-0.51 * age 
Fig. 6. Gompertz non-linear relationship between total length (cm) and age (years) by sex for
captive beluga (males: n? 5; females: n? 4).
44 Robeck et al.
The present study is the ?rst to demonstrate seasonal differences in serum
T levels in mature beluga males. In contrast to T concentrations observed in
other seasonally reproductive cetaceans spinner dolphin (Stenella longirostris)
[T. Robeck, 1984], Paci?c white-sided dolphin (Lagenorhynchus obliquidens)
[M. Yoshioka, unpublished], and similar to cetaceans that produce spermatozoa
throughout the year (e.g., bottlenose dolphins [Tursiops truncatus] and killer
whales [Orcinus orca]) [Schroeder and Keller, 1989; Robeck et al., 1995], mean
monthly T concentrations never dropped below 1 ng/ml. Although minimum
T levels required to maintain sperm production are unknown for this species, the
observed minimum mean of 1 ng/ml suggests that some degree of spermatogenesis
may occur throughout the year. This would agree with the observation of Heide-
J
rgenson and Teilmann [1994] that although epididymal spermatozoa concentra-
tions are higher in May, spermatozoa are still present in October. In bottlenose
dolphins, peak sperm production occurs approximately 60 days after peak serum
T levels [Schroeder and Keller, 1989]. Although further study is required to
determine the timing of spermatogenesis, based on a 60-day interval, the occurrence
of peak T from January?March in captive beluga suggests that peak sperm output
may occur from March?May. This is a logical suggestion because the timing of peak
sperm production in the male would coincide with the occurrence of estrous cycles in
the female.
The earliest age that a captive male sired a calf was 9 years (mean? 13 years) is
in agreement with the estimate for wild beluga of 8?9 years of age [Brodie, 1971], but
is contrary to the 6?7 years of age estimate for Greenland beluga [Heide-J
rgenson
and Teilmann, 1994].
Peri-parturient behaviors exhibited by beluga during Stage 1 and Stage 2 of
labor were similar to those reported for other cetaceans, but there were differences in
the timing and duration of these events. For example, beluga remain in Stage 2 labor
longer (range? 0.4?12.4 hr) than killer whales (range? 1?4 hr) [Robeck et al., 2001],
and bottlenose dolphins (range? 0.8?4.0 hr) [Joseph et al., 2000]. Because it is not
uncommon for beluga to experience dystocia, it is important to consider normal
delivery times when evaluating the progression of labor.
Most peri-parturient reproductive problems of captive cetaceans (e.g.,
dystocia, stillbirth, weak calf, poor maternal care) occur during labor and within a
few hours of birth [Robeck et al., 2001]. Thus, being able to accurately predict the
timing of parturition is an important management tool. Although some species of
cetaceans have been noted to inconsistently exhibit reduced food intake before birth,
food intake in beluga is consistently decreased on the day of parturition. Killer
whales [Katsumata et al., 1998] and bottlenose dolphins [Terasawa et al., 1999]
exhibit a consistent decline in rectal temperature on the day of parturition, but the
predictive value of this approach has not been evaluated in beluga.
Within individual animals, we saw differences in the time from delivery to
nursing for their ?rst (53 hr) compared to their second calf (19 hr). This ?nding
supports the hypothesis proposed by Russell et al. [1997], stating that cows are able
to reduce the time it takes to ?rst nurse their newborn calf as they become more
experienced. This implies that the cow, to a large degree, is responsible for providing
the calf with the best opportunity to learn to nurse.
Nursing patterns among normal calves were similar, with an initial peak
total nursing time around Day 4, after which they gradually decreased reaching
Reproduction in Captive Beluga 45
a plateau around 20 days after parturition. Russell et al. [1997] observed
a slightly longer time to peak postpartum nursing time of 7?10 days. A similar
nursing pattern has been observed in killer whales [Asper et al., 1988; Clark
and Odell, 1999], bottlenose dolphins [Read et al., 1995], and Paci?c white-sided
dolphins [Dalton et al., 1995]. Although some authors attribute this early peak in
nursing to a need for increased caloric intake [Russell et al., 1997], we speculate
that it is related to the learning curve required for the calf to become ef?cient
at obtaining milk. Normal calves are born with fat reserves that provide a caloric
buffer during this learning process. As the calf grows, it is reasonable to assume its
caloric demand will rise accordingly. Reduced nursing time may actually re?ect more
ef?cient nursing by the calf, as well as possible changes in milk quality during
lactation. In addition, a neonatal calf does not have the ability to handle large
volumes of milk during any one feeding. Thus, as the animal?s nursing ef?ciency
improves and the digestive functional capacity increases, total nursing time
decreases.
In support of the hypothesis of Russell et al. [1997] that total nursing time
can serve as an indicator of clinical stress in calves, we saw abnormal nursing
patterns in four calves that were clinically ill (Fig. 3). Nursing patterns have
proven diagnostically useful for evaluating the health of calves from all cetaceans
bred in captivity [Cook et al., 1992; Robeck et al., 1994; Dalton and Robeck,
1995; Robeck and Dalton, 2002]. Once normal nursing patterns have been
determined for a species, they may be useful indicators of the health of the calf or
cow [Amundin, 1986].
Our indications of sexual dimorphism corroborate results reported from
wild beluga [Burns and Seaman, 1986; Heide-Jorgensen and Teilmann, 1994],
although Doidge [1990] found little, if any, gender-based bias in growth in length.
Indeed, sexually dimorphic growth patterns in cetaceans are not lacking from the
literature [Best, 1970; Christensen, 1984; Cockcroft and Ross, 1990; Read et al.,
1993]. Analyses of growth in total body length in the present study were consistent
with those seen in wild beluga [Burns and Seaman, 1986; Doidge, 1990; Heide-
Jorgensen and Teilmann, 1994] in that adult males were larger than females, but
grow at a slower rate. Yet, our data suggested that male beluga reach lighter
asymptotic weights than females while growing faster in this regard, and this
contrasts to what has been reported for wild animals [Heide-Jorgensen and
Teilmann, 1994]. Captive beluga seem to follow similar body length, but not body
weight growth patterns, compared to their wild counterparts. Because adult male
beluga are both larger and heavier than females, this discrepancy warrants further
discussion.
Previous investigators describing growth in other cetacean species have
experienced similar paradoxes, albeit when describing body length and not weight
[Best, 1970; Christensen, 1984; Duf?eld and Miller, 1988; Cockcroft and Ross, 1990;
Read et al., 1993; Read and Tolley, 1997]. In general, two hypotheses are presented
to explain this disparity. One theory is that adolescent male cetaceans may
experience a rapid growth spurt. This would answer the question of how young male
beluga end up heavier than females upon reaching adulthood, whereas our model
suggests otherwise. It is possible that we failed to capture this growth spurt simply
because many of the animals had not yet achieved this milestone within the study
period. The other hypothesis discounts the growth spurt and presents the argument
46 Robeck et al.
that males end up larger (or in this case, heavier) than females simply because they
continue to grow after the females have reached asymptotic weight. If either of these
hypotheses holds true for growth in body weight, as well as length, then it may
explain our observed data. Similarities and differences notwithstanding, the results
from the present study must be viewed cautiously due to the restrictive nature of the
range of the data set and the obvious environmental differences between captive and
wild beluga.
Results from further captive studies may provide useful tools for assessing
baseline life history information pertaining to age and growth dynamics of not only
captive, but also wild beluga. As our animals continue to mature, continued
monitoring of growth and subsequent analyzes should permit more de?nitive
conclusions on attributes of sexual dimorphism and differences between growth in
captive-born and wild beluga.
CONCLUSIONS
In conclusion, we have found that for female beluga, the mean age, length, and
weight at ?rst conception were 9.172.8 years, 318.079.1 cm, 519784 kg. Beluga are
seasonally polyestrous with 80% of conceptions occurring in March?May. The mean
luteal phase and total estrous cycle length were 30.076.5 days and 48.074.6 days,
respectively. The mean gestation length was 475.0720.4 days. For beluga males, the
highest serum T values (6.274.9 ng/ml) were recorded from January?March with
the lowest from August?September (1.171.0 ng/ml). At parturition, times of ?rst
appearance of ?ukes or rostrum to delivery, delivery to placental passage, and
delivery to nursing were 4.472.9 hr, 7.671.8 hr, and 43745 hr, respectively. Food
intake decreased consistently on the day of labor with 44% of parturient females
exhibiting zero intake. Mean 24-hr peak nursing occurred on Day 3.972.7 days, and
nursing patterns, as compared to mean ?normal? consumption, can be used as an
indicator of calf or cow stress or health. Based on the models, captive beluga
exhibited gender-biased differences in growth in body length and weight. Predicted
body weight at birth for either sex and body length at birth for male and female
calves was 88.9 kg, and 154.3 and 160.7 cm, respectively.
ACKNOWLEDGMENTS
The authors acknowledge and thank the veterinarians, curators, keepers,
trainers, and laboratory and veterinary technicians at all the institutions whose
assistance made this project possible. We also thank Dr. L. Dalton, SeaWorld of
Texas, for his assistance. We thank C. Potter, Smithsonian Institution and Kendall
Mashburn, National Zoo?s Conservation and Research Center for technical
assistance and Dr. B. Joseph, The Point De?ance Zoo and Aquarium, for support
of the collaborative whale management program. We thank K. Ramirez and Dr. M.
Greenwell at the John G. Shedd Aquarium for their assistance during this project.
We thank Dr. R. Cook, C. McClave, J. Smith, K. Walsh, M. Hiatt, and G. Skammel
at the Wildlife Conservation Society for their assistance during this project. This
project was funded in part by a grant from the Wildlife Conservation Society?s
Species Survival Fund and by Busch Entertainment Corporation. This is a SeaWorld
Technical contribution number 2004-02-T.
Reproduction in Captive Beluga 47
REFERENCES
Amundin M. 1986. Breeding the bottlenose
dolphin. Intl Zoo Yearb 25:263?71.
Asper ED, Young WG, Walsh MT. 1988. Ob-
servations on the birth and development of a
captive-born killer whale. Intl Zoo Yearb
27:295?304.
Best PB. 1970. The sperm whale (Physeter catodon)
off the west coast of South Africa: 5. Age, growth
and mortality. Investigational Report of the
Division of Sea Fisheries of South Africa 79:1?27
Braham HW. 1984. Review of reproduction in the
white whale, Delphinapterus leucas, narwhale,
Monodon monoceros, and irrawaddy dolphin,
Orcaella brevirostris, with comments on stock
assessment. In: Perrin WF, Brownell RJ Jr,
Demaster DP, editors. Reproduction in whales,
dolphins and porpoises, Special Issue 6. Cam-
bridge: International Whaling Commission.
p 81?90.
Brodie PF. 1971. A reconsideration of aspects of
growth, reproduction and behavior of the white
whale (Delphinapterus leucas) with reference to
the Cumberland Sound, Baf?n Island, popula-
tion. J Fish Res Bd Can 28:1309?18.
Brook FM. 2000. Sonographic testicular and
ovarian assessment in the bottlenose dolphin,
Tursiops truncatus aduncus. In: Duf?eld D,
Robeck TR, editors. The bottlenose dolphin
breeding workshop. Silver Springs, MD:
AZA Marine Mammal Taxon Advisory Group.
p 207?21.
Brown JL, Wasser SK, Wildt DE, Graham LH.
1994. Measurement of fecal estrogen and pro-
gesterone metabolites for assessing ovarian
activity in felids. Biol Reprod 51:776?86.
Burns JJ, Seaman GA. 1986. Investigations of
beluga whales in western and northern Alaska II.
Biology and ecology. Fairbanks, Alaska: Report
by the Alaska Department of Fish and Game.
129 p.
Calle PP, Cook RA, McClave C, Palma S. 1993.
Circulating gestational progesterone and estra-
diol concentrations, parturition, and placental
descriptions of two beluga whales (Delphinap-
terus leucas). Proc IAAAM 24:61.
Calle PP, Cook RA, Robeck TR, Young SJF,
Jones MH. 1996. Circulating gestational proges-
terone and estradiol concentrations in Beluga
whales (Delphinapterus leucas). Proc AAZV
1996:340?2.
Clark ST, Odell DK. 1999. Nursing parameters in
captive killer whales (Orcinus orca). Zoo Biol
18:373?84.
Clark ST, Odell DK, Lacinak CT. 2000. Aspects of
growth in captive killer whales (Orcinus orca).
Mar Mamm Sci 16:110?23.
Christensen I. 1984. Growth and reproduction of
killer whales, Orcinus orca, in Norwegian coastal
waters. In: Perrin WF, Brownell RJ Jr, Demaster
DP, editors. Reproduction in whales, dolphins
and porpoises. Special Issue 6. Cambridge:
International Whaling Commission. p 253?8.
Cockcroft VG, Ross GJB. 1990. Age growth, and
reproduction of bottlenose dolphins Tursiops
truncatus from the east coast of southern Africa.
Fish Bull 88:289?302.
Cook RA, Calle PP, McClave C, Palma S.
1992. Health care and medical problems of
a captive bred and mother reared beluga
whale (Delphinapterus leucas). Proc. IAAAM
23:107?10.
Dalton LM, Robeck TR. 1995. Multifocal pyo-
granulomatous pneumonia in a killer whale calf.
Proc IAAAM 26:84?5.
Dalton LM, Robeck TR, Young WG. 1995.
Growth and development of a paci?c white-
sided dolphin (Lagenorhynchus obliquidens). Proc
IAAAM 26:89?91.
Doidge DW. 1990. Age-length and length-weight
comparisons in beluga, Delphinapterus leucas.
Can Bull Fish Aquatic Sci 224:59?68.
Drinnan RL, Sadler RMFS. 1981. The suckling
behavior of a captive beluga (Delphinapterus
leucas) calf. Appl Anim Ethol 7:179?85.
Duf?eld DA, Miller KW. 1988. Demographic
features of killer whales in oceanaria in the
United Stated and Canada, 1967?1987. Rit
Fiskideildar 11:297?306.
Fernandez S, Hohn AA. 1998. Age, growth, and
calving season of the bottlenose dolphins,
Turiops truncatus, off coastal Texas. Fish Bull
US 96:357?65.
Ferrero RC, Walker WA. 1995. Growth and
reproduction of the common dolphin, Delphinus
delphis Linnaeus, in the offshore waters of the
North Paci?c Ocean. Fish Bull US 93:483?94.
Fitzhugh HA Jr. 1975. Analysis of growth curves
and strategies for altering their shape. J Anim Sci
42:1036?51.
Heide-J
rgensen MP. 1990. Small cetaceans in
Greenland: hunting and biology. N Atl Studies
2:55?8.
Heide-J
rgensen MP, Teilmann J. 1994. Growth,
reproduction, age structure and feeding habits of
white whales (Delphinapterus leucas) in West
Greenland waters. Bioscience 39:195?212.
H
ier R, Heide-J
rgensen MP. 1994. Steroid
hormones and prolactin in white whales (Delphi-
napterus leucas) from West Greenland. Bio-
science 39:227?38.
Joseph BE, Duf?eld DA, Robeck TR. 2000.
Summary data on reproduction of bottlenose
dolphins in controlled environments. In: Duf?eld
D, Robeck TR, editors. The bottlenose dolphin
breeding workshop. Silver Springs, MD: AZA
Marine Mammal Taxon Advisory Group.
p 43?56.
Katsumata E, Katsumata H, Tobayama T, Usuki
S. 1998. Body temperature in reared killer
whales, Orcinus orca. Korea Kwangiu, Republic
of Korea: Recent Progress in Molecular
and Comparative Endocrinology Asia and
Oceania Society of Comparative Endocrinology.
p 448?52.
48 Robeck et al.
Monfort SL, Arthur NP, Wildt, DE. 1991.
Monitoring ovarian function and pregnancy
by evaluating excretion of urinary oestrogen
conjugates in semi-free-ranging Przewalski?s
horses (Equus przewalskii). J. Reprod Fertil
91:155?64.
Norris KS. 1961. Standarized methods for measur-
ing and recording data on the small cetaceans. J
Mamm 42:471?76.
Perrin WF, Henderson JR. 1984. Growth and
reproductive rates in two populations of spinner
dolphins, Stenella longirostris, with different
histories of exploitation. In: Perrin WF, Brownell
RJ Jr, DeMaster DP, editors. Reproduction in
whales, dolphins and porpoises, special Issue 6.
Cambridge, Eng: International Whaling Com-
mission. p 417?30.
Read AJ, Gaskin DE. 1990. Changes in growth
and reproduction of harbour porpoises, Phocoe-
na phocoena, from the Bay of Fundy. Can J Fish
Aquatic Sci 47:2158?63.
Read AJ, Tolley KA. 1997. Postnatal growth and
allometry of harbour porpoises from the Bay of
Fundy. Can J Zool 75:122?30.
Read AJ, Wells RS, Hohn AA, Scott MD.
1993. Patterns of growth in wild bottlenose
dolphins, Tursiops truncatus. J Zool 231:
107?23.
Read K, Mann J, Weiner J, Hecker N. 1995. Infant
development in two aquarium bottlenose dol-
phins. Zoo Biol 14:135?47.
Robeck TR, Atkinson S, Brook F. 2001. Chap 11:
Reproduction In: Dierauf L, Gulland F, editors.
CRC handbook in marine mammal medicine.
Boca Raton: CRC Press. p 193?236.
Robeck TR, Dalton LM. 2002. Zygomycotic
infections in bottlenose dolphins, (Tursiops trun-
catus), a killer whale (Orcinus orca), and paci?c
white-sided dolphins (Lagenorhynchus obliqui-
dens) caused by Saksenaea vasiformis and
Apophysomyces elegans. J Zoo Wild Med 33:
356?66.
Robeck TR, Dalton LM, Young WG. 1994.
Systemic nocardiosis in a 6-week-old beluga calf.
Proc IAAAM 25:109.
Robeck TR, Gross T, McBain J. 1995. Prelimi-
nary results on radioimmunoassay determination
of serum testosterone concentrations in the
killer whale (Orcinus orca). Proc IAAAM 26:
72?3.
Russell JM, Simonoff JS, Nightingale J. 1997.
Nursing behavior of beluga calves (Delphinap-
terus leucas) born in captivity. Zoo Biol 16:
247?62.
Schroeder JP, Keller KV. 1989. Seasonality of
serum testosterone levels and sperm density in
Tursiops truncatus. J Exp Zool 249:316?21.
Sergeant DE. 1973. Biology of the white whales
(Delphinapterus leucas) in western Hudson Bay. J
Fish Res Board Can 1065?90.
Sokal RR, Rohlf FJ. 1995. Biometry, 3rd edition.
New York: WH Freeman and Company.
Terasawa F, Yokoyama Y, Kitamure M. 1999.
Rectal temperature before and after parturition
in bottlenose dolphins. Zoo Biol 18:153?6.
Walker LA, Cornell L, Dahl KD, Czekala NM,
Dargen CM, Joseph BE, Hsueh AJW, Lasley BL.
1988. Urinary concentrations of ovarian steroid
hormone metabolites and bioactive follicle-sti-
mulating hormone in killer whales (Orcinus orca)
during ovarian cycles and pregnancy. Biol
Reprod 39:1013?20.
Wasser SK, Monfort SL, Southers J, Wildt DE.
1994. Excretion rates and metabolites of oestra-
diol and progesterone in baboon (Papio cynoce-
phalus cynocephalus) faeces. J Reprod Fertil
101:213?20.
Wells RS. 1984. Reproductive behavior
and hormonal correlates in Hawaiian spinner
dolphins, Stenella longirostris. In: Perrin WF,
Brownell RJ Jr, Demaster DP, editors. Repro-
duction in whales, dolphins and porpoises,
Special Issue 6. Cambridge: International
Whaling Commission. p 465.
Reproduction in Captive Beluga 49