Color profile: Generic CMYK printer profile

Composite Default screen

845

Nutritional influences on breeding dynamics in elk

Rachel C. Cook, Dennis L. Murray, John G. Cook, Peter Zager, and
Steven L. Monfort

Abstract: Nutrition in summer and early autumn may influence pregnancy rates and recruitment in ungulate herds in
temperate regions, yet this influence on the reproductive endocrinology of wild ungulates is poorly understood. We
examined the effects of 3 levels of summer–autumn nutrition on timing of breeding, pregnancy rates, and concentration
of fecal progestagens (P4) in 30 captive female elk (Cervus elaphus nelsoni). Digestible energy (DE) content for females
on the high-nutrition treatment was maintained at 12.35 kJ/g, while DE content for females on the medium- and low-
nutrition treatments was gradually reduced to 10.89 and 9.42 kJ/g, respectively, by the end of the breeding season (all
feed was offered ad libitum) to mimic the natural decline in forage quality during summer and autumn. In addition, we
used three females as a control; they were maintained on a high-nutrition diet and kept separate from a bull. Based on
fecal P4 concentrations and visual observations, 90, 90, and 10% of the elk on the high-, medium-, and low-nutrition
treatments bred, respectively. Nutritional restriction delayed breeding by an average of 8 and 28 days in the medium-
and low-nutrition animals. Females in the low- and medium-nutrition treatments excreted significantly higher P4
concentrations prior to breeding than did animals in the high-nutrition treatment. Diet failed to affect P4 concentration
during pregnancy, but we found evidence that P4 concentrations were positively correlated with body condition (ingesta-
free body fat content) during early gestation. Females that did not breed failed to exhibit estrous behavior, and hormone-
excretion profiles confirmed that these animals failed to ovulate. These results suggest that inadequate nutrition in
summer and autumn reduce pregnancy rates by preventing estrus and ovulation rather than by inducing early-embryo
mortality.

Résumé : Bien que la nutrition durant l’été et au début de l’automne puisse influencer la fréquence des grossesses et
le recrutement au sein des troupeaux d’ongulés des régions tempérées, cette influence sur l’endocrinologie de la repro-
duction est mal connue dans le cas des ongulés sauvages. Nous avons examiné les effets de 3 niveaux de nutrition
d’été–automne sur le moment de la reproduction, les taux de grossesse et la concentration fécale de progestagène (P4)
chez 30 femelles du Wapiti des Rocheuses (Cervus elaphus nelsoni). L’énergie digestible (DE) de la nourriture offerte
aux femelles dans le régime riche a été maintenue à 12,35 kJ/g et les valeurs de DE chez les femelles soumises aux
régimes moyen et faible en énergie ont été réduites jusqu’à 10,89 et 9,42 kJ/g, respectivement, à la fin de la période de
reproduction (il s’agissait dans tous les cas de nourriture offerte en quantité ad libitum) de façon à imiter les diminu-
tions naturelles de la qualité du brout en été et à l’automne. De plus, nous avons utilisé trois femelles comme témoins;
elles ont été gardées à un régime riche et privées de la présence d’un mâle. D’après les concentrations fécales de P4 et
des observations visuelles, 90 % des femelles soumises au régime riche, 90 % des femelles gardées au régime moyen
et 10 % des femelles gardées au régime faible se sont reproduites. Les restrictions alimentaires ont retardé la reproduc-
tion respectivement de 8 et 28 jours en moyenne chez les femelles gardées au régime moyen et au régime faible. Les
femelles soumises aux régimes faible et moyen ont excrété des concentrations significativement plus élevées de P4
avant la reproduction que les femelles gardées au régime riche. Le régime n’a pas affecté les concentrations de P4 pen-
dant la grossesse, mais certains indices nous portent à croire que les concentrations de P4 étaient en corrélation positive
avec la condition physique (graisses corporelles sans les ingesta) au début de la gestation. Les femelles qui ne se sont
pas reproduites n’ont pas manifesté de comportement oestral et les profils d’excrétion hormonale ont confirmé qu’une
alimentation insuffisante au cours de l’été et de l’automne réduit la fréquence des grossesses en inhibant l’oestrus et
l’ovulation plutôt qu’en déclenchant une mortalité embryonnaire précoce.

[Traduit par la Rédaction] Cook et al. 853

Received June 12, 2000. Accepted March 13, 2001. Published on the NRC Research Press Web site on May 8, 2001.
R.C. Cook1,2 and D.L. Murray. Department of Fish and Wildlife Resources, University of Idaho, Moscow, ID 83844, U.S.A.
J.G. Cook. National Council of the Paper Industry for Air and Stream Improvement (NCASI), 1401 Gekeler Lane, La Grande,
OR 97850, U.S.A.
P. Zager. Idaho Department of Fish and Game, 1540 Warner Avenue, Lewiston, ID 83501, U.S.A.
S.L. Monfort. Conservation and Research Center, 1500 Remount Road, Front Royal, VA 22630, U.S.A.
1
Corresponding author (e-mail: rcook@eou.edu).
2
Present address: NCASI, Forestry and Range Sciences Laboratory, 1401 Gekeler Lane, La Grande, OR 97850, U.S.A.

Can. J. Zool. 79: 845–853 (2001) DOI: 10.1139/cjz-79-5-845 © 2001 NRC Canada

J:\cjz\cjz79\cjz-05\Z01-050.vp
Friday, May 04, 2001 8:44:35 AM
Color profile: Generic CMYK printer profile
Composite Default screen

846 Can. J. Zool. Vol. 79, 2001

Introduction parturition, and an abrupt drop in P4 concentration during

pregnancy is characteristic of fetal loss or abortion. Non-
Productivity and recruitment in many elk (Cervus elaphus breeding (i.e., male absent) or infertile females generally ex-
nelsoni) populations are declining across the northwestern hibit multiple ovulatory “cycles” during the breeding season
United States (Irwin et al. 1994; Gratson and Zager 1999). that can be tracked by monitoring longitudinal P4 excretion
This decline may be attributable, at least in part, to de- patterns.
pressed pregnancy rates in adult females. For example, late- For the current study, fecal P4 concentration was used in
winter adult pregnancy rates are as low as 67–70% in some concert with behavioral assessments to examine the impact of
herds in Idaho (Gratson and Zager 1999) and 45–50% in diet on patterns of estrus, timing of breeding, and pregnancy
Washington and Oregon (Trainer 1971; Stussy 1993; L.C. outcome for captive female elk. Thirty lactating female elk
Bender, unpublished data; B.K. Johnson, unpublished data). were maintained on three nutritional treatments and three
Depressed late-winter pregnancy rates in elk can be related nonlactating females were used as controls to evaluate estrous
to poor winter-forage conditions in some cases (Thorne et al. cyclicity in animals maintained in the absence of a bull.
1976; Nelson and Leege 1982), but early-pregnancy rates are Body-condition and body-mass data were collected to exam-
probably more closely linked to nutrition in summer and ine whether condition and (or) nutritional stress affected fecal
early autumn. Forage conditions in summer and early au- P4 concentration and (or) reproductive performance.
tumn are generally considered to be more than adequate to
meet the requirements of wild ungulates (Nelson and Leege Methods and materials
1982; Christensen et al. 1993). However, in many areas, for-
age quality and (or) quantity may not be as adequate as was Elk and facilities
once perceived (e.g., Verme and Ullrey 1984; Merrill and Thirty-three captive female elk (4 and 6 years old) were housed
Boyce 1991; Parker et al. 1999; Cook 2001). Given the en- in six 1-ha pens near Kamela, Oregon. Elk were captured as calves
ergy demands of lactation, inadequate summer and autumn from a wild herd at the Starkey Experimental Forest and Range
forage quality could adversely impact reproduction, particu- near La Grande, Oregon, in 1991 and 1993 and subsequently
larly of lactating elk. bottle-raised (Cook et al. 1994). Each pen was devoid of vegetation
and contained a barn with 9–12 stalls designed for individual feed-
Nutrition affects fertility and reproduction rates of live- ing and collection of physiological samples (blood, urine, and fe-
stock (Ferrell 1991) and wildlife (Verme 1969). The physio- ces). Facility design permitted researchers to track reproductive
logical causes of poor reproduction following undernutrition activity, evaluate body condition, and manipulate food rations.
have not been studied intensively in wildlife, but extensive
research on domestic livestock suggests that poor nutrition Handling and feeding
can have a direct and adverse affect on hypothalamic– All animals were handled in accordance with an approved
pituitary function (Day et al. 1986; Armstrong and Britt animal-welfare protocol (Wisdom et al. 1993). Thirty female–calf
1987). Although it follows that such changes should affect pairs were randomly assigned to one of three nutritional treatments
hormone production, the mechanisms by which nutritional (high, medium, and low) lasting from 21 June 1997 through 4 No-
stress disrupts reproductive-endocrine function are poorly vember 1997. Manipulation of dietary quality focused on varying
understood. For instance, serum progesterone concentrations digestible energy (DE) content using three hay types and two pellet
during the luteal phase of the estrous cycle in domestic live- types (Table 1). The high-nutrition diet was formulated according
to the livestock literature (National Research Council 1984, 1985)
stock have been reported to increase (Dunn et al. 1974; Wil-
such that DE content would not limit the performance of calves or
liams and Cumming 1982), decrease (Gombe and Hansel females. The low-nutrition diet was designed to mimic DE levels
1973; Imakawa et al. 1983), or remain unchanged (Spitzer et that commonly occur in low-elevation xeric forests (Pinus ponder-
al. 1978) in response to nutritional limitation. Similarly, pro- osa and Pseudotsuga menziesii) on elk range in the Blue Moun-
gesterone concentrations during pregnancy have been shown tains Ecoregion in northeast Oregon (Holechek et al. 1981; Cook et
to be higher in malnourished animals than in those that were al. 1994; J.G. Cook, unpublished data). We assumed that this diet
adequately fed (Vogelsang 1977; Parr 1992), higher in well-fed represented the poorest quality forage encountered by wild elk in
than in malnourished animals (Boone et al. 1975; Vogelsang this region during summer and autumn. The medium-nutrition diet
1977), or similar across nutritional regimes (Vogelsang 1977). was intermediate between the high- and low-nutrition treatments.
Such inconsistencies may result from variations in dietary Prior to 24 July, all elk were fed high-quality pellets and high-
quality hay (30:70 ratio) offered ad libitum. This provided an
treatment, species/breed, age, ambient temperature, or other
overall DE content of 12.35 kJ/g. This regime was used for the
external factors specific to the study. What is clear, however, high-nutrition group throughout the rest of the study except that
is that steroid-hormone production and secretion are influenced the total amount of pellets fed to each female was held constant,
by nutritional status (Ferrell 1991). whereas the hay was fed ad libitum. Beginning on 1 July, the ratio
Fecal and urinary steroid metabolites are now routinely of pellets and hay offered to the medium-nutrition group was
used to assess reproductive status in a diversity of captive switched to a 20:80 ratio. The amount of pellets was again held
(Hodges 1996; Schwarzenberger et al. 1996; Brown et al. constant for the rest of the study, whereas the hay was fed ad libi-
1997) and free-ranging wildlife (Whitten et al. 1998; Berger tum. The strategy was the same for females in the low-nutrition
group, except that the ratio of pellets to hay was set to 10:90. The
et al. 1999) species. In particular, fecal progestagens (P4)
type of hay offered at any given time over summer and autumn was
can be used to monitor reproductive patterns during the peri- the one that most closely matched the target DE levels for the
estrus interval and pregnancy. The level of corpus luteum- nutrition-treatment group. The two pellet types (Table 1) were mixed
derived progesterone is elevated after ovulation and corpus in ratios which ensured that the overall DE consumed matched the
luteum- and (or) placenta-derived P4 remains elevated through- target level for the particular nutrition group. DE for the medium-
out pregnancy. Fecal P4 concentrations remain elevated until nutrition group was reduced from 12.35 to 11.51 kJ/g by late
© 2001 NRC Canada

J:\cjz\cjz79\cjz-05\Z01-050.vp
Friday, May 04, 2001 8:44:35 AM
Color profile: Generic CMYK printer profile
Composite Default screen

Cook et al. 847

Table 1. Crude protein (CP) content, digestible energy (DE) content, and primary ingredients of
rations fed to female elk (Cervus elaphus nelsoni) during a reproductive study in northeastern
Oregon.
Ration CPa DEb Primary ingredients
High-quality pellets 15.9 15.32 Oats, wheat, alfalfa hay
Low-quality pellets 14.2 9.42 Ryegrass screenings, straw, alfalfa hay
High-quality hay 17.4 11.14 Alfalfa
Medium-quality hay 8.3 10.55 Orchard grass, alfalfa
Low-quality hay 7.8 9.55 Fescue and mixed meadow grasses
Note: Crude protein content, gross energy content, and in vitro digestibility of solid feed were determined
for each food type by the Habitat Analysis Laboratory at Washington State University. CP content was
determined by macro-Kjeldahl analysis, gross energy content by bomb calorimetry, and digestibility by two-
stage in vitro trials (Horwitz 1980). DE content of individual feeds was calculated as the product of
digestibility and gross energy content (Robbins 1993). The pelleted rations were manufactured by Pendleton
Grain Growers (Hermiston, Oregon, U.S.A.).
a
Expressed as percent dry matter.
b
Expressed in kilojoules of DE per gram of food (dry-matter basis).

August, and reduced more gradually from 11.51 to 10.89 kJ/g by tions were not observed, estrous behaviors were used to estimate
late October. DE for the low-nutrition group was reduced from breeding date.
12.35 to 10.47 kJ/g by late August, then reduced more gradually
to 9.42 kJ/g by late October. At the end of the rutting period
(4 November), all animals were provided with identical diets of Fecal P4 analysis
high-quality hay and pellets. However, the amount fed was set to Fecal-sample collections were initiated during the anestrous
provide a maintenance DE content such that the body condition of interval (18 August), when fecal P4 concentrations in all females
the elk would remain constant through the rest of autumn. were basal and noncyclic and continued throughout the presump-
All elk were fed twice daily throughout the study. Each morn- tive estrous (mid-September to early November) and early-pregnancy
ing, females were placed in individual stalls and offered pelleted (mid-December) intervals. To reduce laboratory costs, P4 concentra-
food. Hay mangers were used for communal feeding of hay each tions were assessed in a subset of these collected fecal samples for
afternoon. Orts of both pellets and hay were recorded daily. Fe- each female as follows: twice weekly during the prebreeding interval,
males were held in the barns for approximately 2 h a day, enough daily during the peri-estrus interval (5 days before and after the day
time for each to consume the designated pelleted ration. Fecal pel- of observed behavioral estrus), twice weekly through 4 November,
lets were collected from every female twice a week in the morning and then once weekly through 18 December. This strategy was suffi-
during the last 2 weeks of August, daily from 1 September to 17 cient for evaluating cyclic ovarian activity and pregnancy status
November, and weekly from 18 November to 18 December. Pellets throughout the sampling interval, including (i) the endocrine
were stored frozen (–20°C) until analysis. dynamics of the transition from the nonbreeding to the breeding
Each female was weighed twice per week, the two mass esti- season, (ii) the correspondence between reproductive behaviors and
mates (kg) were averaged, and the percentage of mass lost was cal- P4 excretion, and (iii) distinguishing between conceptive and
culated from this average. Body condition was also determined for nonconceptive ovarian cycles within and among treatment groups.
each animal in mid-October by measuring maximum rump fat For each female not exhibiting estrus, 2 or 3 samples per week
thickness via ultrasonography and a body condition score (Cook were analyzed from the beginning of the study to 2 November to
2000). These measurements were used to estimate percent ingesta- determine if cycling had occurred, despite the lack of behavioral
free body fat (IFBF) by the equation estrous observations. We analyzed samples daily from 2 through 16
November to determine if cycling would resume once animals
IFBF = –9.8863584 + 9.1871285x were placed on a higher nutritional plane. For each nonlactating fe-
male, 2–4 samples per week were analyzed from early September
– 1.3831754x2 + 0.083951218x3 until 4 November to obtain clear profiles of cycling animals.
Fecal-sample processing and extraction procedures were similar
where x is LIVINDEX, an arithmetic combination of subcutaneous to those previously described (Monfort et al. 1993; Wasser et al.
rump fat thickness and body condition score (Cook 2000). 1994). Briefly, fecal samples were dried using a rotary savant, and
Fecal samples were also collected from a subset of three non- approximately 0.2 g of crushed fecal material was combined with
lactating elk maintained separate from but immediately adjacent to 10 mL of absolute ethanol, vortexed, and then boiled for 20 min in
a mature bull. These animals served as a control to compare 16 × 125 mm glass tubes. 14C-progesterone (~1400 cpm) was
progestagen-excretion levels throughout a normal estrous cycle added before extraction for monitoring procedural losses. After
with those of animals that were maintained on a nutritional treat- boiling, tubes were centrifuged at room temperature (15 min at
ment and that did not breed. These animals were fed a diet of high- 2500 rpm) and the ethanol supernate was decanted into clean
quality hay and pellets. 16 × 125 mm tubes and evaporated to dryness under a stream of
air. Tubes were rinsed twice with 2–4 mL of absolute ethanol,
Detection of estrus sonicated to remove particulates adhering to vessel walls, and
One adult wild bull was placed in each pen from 5 September evaporated, and finally the extractant was reconstituted in 1 mL of
through 4 November and females were observed daily from dawn methanol. After reconstitution, extractants were vortexed (20 s)
until dusk for estrous behavior and copulations. Behavioral pat- and sonicated (10 min). A 300-µL portion of each extractant was
terns characteristic of estrus were identical with those described stored frozen and an additional 25 µL was counted for 14C-
previously (Morrison 1960; Bubenik 1982). Observed copulations progesterone from a randomly chosen subset of 35 samples to
were recorded for 75% of the bred animals and no female dis- estimate extraction efficiency (percent recovery), which was
played more than one period of estrus. For females whose copula- 74.6 ± 7.14% (mean ± coefficient of variation (CV)).
© 2001 NRC Canada

J:\cjz\cjz79\cjz-05\Z01-050.vp
Friday, May 04, 2001 8:44:36 AM
Color profile: Generic CMYK printer profile
Composite Default screen

848 Can. J. Zool. Vol. 79, 2001

Fecal progestagens were estimated using a radioimmunoassay Huynh–Feldt adjustment was applied to the degrees of freedom
procedure (Wasser et al. 1994) recently validated for elk (Garrott et (SAS Institute Inc. 1988, p. 605).
al. 1998). Inter-assay CVs for two separate internal controls were
10.7% (n = 15, 36–47% binding) and 15.4% (n = 15, 8–13%
binding). Intra-assay CVs were <10% and assay sensitivity was Results
3.75 pg/100 µL. All hormone concentrations are expressed as mass
units of hormone excreted per gram of dry feces. Productivity
More of the females in the high- and medium-nutrition
treatments bred than in the low-nutrition treatment (χ 22 Yates =
Data analysis
18.37, p < 0.001). Only 1 of the 10 females fed a low-
The first portion of the analysis was based on three time periods energy-content diet bred compared with 9 of the 10 females
that were defined with respect to when cows bred. The first time
period (T1) preceded any observed breeding (18 August – 15 Sep-
subjected to each of the other higher energy diets. Non-
tember 1997). We treated values obtained during T1 as a baseline breeding females apparently failed to enter a reproductive
for subsequent analyses. The second time period (T2) was the cycle and thus did not ovulate (Fig. 1). Date of breeding var-
overall breeding period (16 September – 27 October), and the third ied among cows subjected to different treatments (F[2,9] =
time period (T3) followed all observed breeding (28 October – 31.19, p < 0.001). Females fed a moderate-quality diet bred
18 December). An average P4 concentration was calculated for later, on average (7 October ± 1.46 days; mean ± SE), than
each female per time period. We first tested whether average P4 females fed a high-quality diet (29 September ± 1.54 days).
concentrations varied among nutritional treatments (n = 3) and The single pregnant female fed a low-quality diet bred on 27
across time periods (TI, T2, T3) using two-way fixed-effects re- October.
peated-measures ANOVA (PROC GLM; SAS Institute Inc. 1988).
The observed pattern of fecal P4 excretion in one high-
We also used one-way fixed-effects ANOVA (PROC GLM) to test
whether progesterone levels varied among treatments within each nutrition female (Fig. 2) suggested that she experienced
time period. We used the least significant difference multiple- early embryonic mortality approximately 60 days after
comparison procedure for all pairwise comparisons. breeding. Because the short-lived drop in fecal P4 concentra-
We used regression analysis to test for a correlation between tion (25 days post breeding) was not characteristic of corpus
baseline P4 concentration and nutritional stress (measured as per- luteum regression (see Fig. 1), and persistent corpus luteum
cent mass loss) using measures obtained during T1. Percent mass function is unlikely to be sustained for 60 days in non-
loss was calculated from mass at the implementation of the nutri- pregnant individuals, this female apparently aborted coin-
tional treatments (late June) until 15 September. Finally, we used cidently with the sharp decline in fecal P4 concentration.
logistic regression (PROC CATMOD; SAS Institute Inc. 1988) to
test whether eventual pregnancy status differed with respect to av- Progestagen concentration
erage P4 concentration during T1 (baseline levels), nutritional treat-
ment, and their interaction. P4 concentrations (without respect to pregnancy status)
differed according to time period (F[2,54] = 55.03, p < 0.001),
To evaluate the effect of nutritional treatment on pregnant fe-
males, we used one-way fixed-effects ANOVA (PROC GLM) to type of diet (F[2,27] = 18.54, p < 0.001), and the interaction
test whether date of breeding varied among treatments, and we between the two (F[4,54] = 13.68, p < 0.001). During T1, P4
used χ 2 statistics to test whether pregnancy rates of females were levels for the medium- and low-nutrition groups were signif-
homogeneous among treatments (Ott 1993). To test for differences icantly higher than for the high-nutrition females (F[2,27] =
in P4 concentration among treatments, we first synchronized preg- 3.85, p = 0.034). The levels for the low- and medium-
nant females in the high- and medium-nutrition treatments with re- nutrition treatments were 0.99 ± 0.09 (mean ± SE; n = 10)
spect to breeding date, and weekly mean P4 concentrations were and 1.00 ± 0.09 µg/g dry feces (n = 10) compared with
calculated (because only one cow became pregnant in the low- 0.72 ± 0.06 µg/g dry feces (n = 10) for the high-nutrition
nutrition treatment, we could not include her in the analysis). For treatment. Fecal P4 levels differed significantly by dietary
these pregnant females, weekly mean P4 concentrations in the pre- treatment during T2 and T3 (F[2,27] = 5.27, p = 0.012, and
estrus (for 4 weeks) and post-estrus (for 9 weeks) periods were
aligned with the day of estrus (day 0). Because of inter-animal
F[2,27] = 16.00, p = 0.001, respectively). P4 levels were simi-
variation in breeding dates, 9 weeks represented the latest time for lar in females in the high- and medium-nutrition groups but
which data were available for all animals. We used weekly re- were significantly higher than in those in the low-nutrition
peated measures of P4 concentration (n = 13 per animal) by treat- group because of the effect of pregnancy (Fig. 3).
ment (n = 2) to test for differences in P4 concentration between the During T1, mass loss for animals on the low- and
high- and medium-nutrition treatments. medium-nutrition treatments was 5.15 ± 0.8 and 1.16 ±
Within treatments, there was considerable variation among fe- 0.7%, respectively. Females on high-quality food gained
males with respect to body condition during the breeding season, 4.9 ± 0.8%. Fecal P4 concentrations were positively related
and a trend that suggested an important influence of animal condi- to percent mass loss during this time (r 2 = 0.10, p = 0.08).
tion on P4 concentrations during the first 4 weeks post conception Eventual pregnancy status was unrelated to P4 levels during
(a critical time for the health of the embryo). To test for the effect T1 (χ12 = 0.18, p = 0.49), nutritional treatment (χ 22 = 2.04,
of body condition on post-conception P4 levels, we ran regressions
p = 0.36), or the interaction between P4 level and nutritional
for each of the first 4 weeks post conception using total body fat as
the dependent variable. The α value was adjusted from 0.05 for treatment (χ 22 = 1.31, p = 0.52).
sequential tests that used common body-condition data (Johnson Considering only those pregnant females fed diets with a
1998, p. 442). high or moderate DE content, P4 levels varied over time
We considered α ≤ 0.05 to be statistically significant for all (F[12,180] = 41.43, p < 0.001) because of increasing P4 levels
analyses. For all repeated-measures analyses, we used the multi- due to pregnancy, but did not vary among treatments
variate mode of PROC GLM. In all cases, the assumption of (F[1,15] = 0.36, p = 0.56) or among interactions between
sphericity (SAS Institute Inc. 1988, p. 605) was violated and the treatment and time (F[12,180] = 1.39, p = 0.24) (Fig. 3).
© 2001 NRC Canada

J:\cjz\cjz79\cjz-05\Z01-050.vp
Friday, May 04, 2001 8:44:37 AM
Color profile: Generic CMYK printer profile
Composite Default screen

Cook et al. 849

Fig. 1. Fecal progestagen (P4) profiles for 12 captive female elk Fig. 2. Hormonal profile for an adult female elk maintained on a
(Cervus elaphus nelsoni) during autumn 1997. Three animals high-nutrition diet that experienced early-gestation mortality. Day
were provided with adequate food but were not allowed access 0 corresponds to the day of observed copulation.
to a bull (cycling), whereas the remainder were subjected to a
restricted diet but failed to become pregnant (low-nutrition treat-
ment). Cycling animals were aligned to the first day of the first
full estrous cycle (= day 0), whereas for nonpregnant animals
day 0 corresponds to the average breeding date for all bred
animals in the study. Weekly averages from the low-nutrition
treatment include 95% confidence intervals.
5
Progestagens (µg/g feces)

3 Cycling

Low
2

1
that females in poor condition during early gestation exhib-
ited P4 concentrations that were low enough to potentially
0 interrupt pregnancy.
-30 -20 -10 0 10 20 30 40
Moderate undernutrition may delay the onset of estrus,
No. of Days from Estrus and more severe deficiencies may cause complete cessation
of reproductive activity (Verme 1969; Hamilton and Baxter
1980; Cameron 1994). In the present study, estrous cyclicity
Body condition and breeding were delayed in the medium-nutrition group,
Among nutritional treatments, body condition varied whereas only one pregnancy occurred in females maintained
markedly. Total body fat content was 16.6 ± 1.1% (range on a low nutritional plane. For the latter animals, three po-
13.6–23.0%) for the high-nutrition treatment, 9.5 ± 0.6% tential processes could have occurred to prevent pregnancy:
(range 4.8–11.3%) for the medium-nutrition treatment, and (1) failure to ovulate; (2) ovulation, but early embryonic
3.5 ± 0.8% (range 1.1–8.3%) for the low-nutrition treatment. death, with the female immediately resuming a cyclic pat-
During the first 4 weeks post conception, the P4 concentra- tern; or (3) fertilization, but resorption of the embryo 4–
tion was positively related to total body fat content during 6 weeks into pregnancy. Our data indicate that low-nutrition
the second (r 2 = 0.36, p = 0.034) and fourth weeks (r 2 = animals failed to ovulate. Patterns of P4 concentration con-
0.43, p = 0.003). firm that ovulation, ovarian cyclicity, and pregnancy did not
occur. Although it is possible that we failed to detect preg-
Cycling animals nancy if very low levels of luteal progesterone were pro-
All three female elk that were not allowed to breed expe- duced, the absence of estrous-related behavior in 9 of 10
rienced at least one estrous cycle with a full-cycle length of females is consistent with our conclusion.
20.0 ± 1.08 days (mean ± SE) (Fig. 3). Peak luteal phase P4 The actual mechanisms that delay or inhibit reproductive
concentrations ranged from 2.35 to 4.44 µg/g, which ex- activity are not fully understood even in domestic livestock,
ceeded those for noncycling elk, which were 0.22–1.90 µg/g but several mechanisms have been proposed. With lacta-
(Fig. 1). Each female also experienced at least one short tional infertility (Loudon et al. 1983; Gerhart et al. 1997),
cycle before the first complete estrous cycle (Fig. 4); this prolactin produced during lactation or suckling presumably
lasted ~11 days, with peak P4 concentrations that ranged prevents the hypothalamus–pituitary complex from initiating
from 1.50 to 2.20 µg/g. reproductive activity (McNeilly 1979; Howie and McNeilly
1982). Prolactin concentrations fluctuate with the frequency
Discussion and intensity of the suckling stimulus, and high suckling fre-
quencies induce greater prolactin secretion (McNeilly 1979).
Our results demonstrate that nutritional plane impacted re- The nutritional plane of a dam may amplify or prolong lacta-
productive function in adult female elk. Longitudinal evalua- tional infertility by altering the suckling pattern of calves.
tions of fecal P4 excretion before, during, and after the Because the mammary glands fill at a lower rate in dams on
breeding season confirmed the existence of a strong link be- a poor nutritional plane (than in dams on a higher nutritional
tween nutrition and reproductive physiology in elk. Lac- plane), calves suckle more frequently to maintain an ade-
tating cow elk subjected to a low level of nutrition either quate milk intake (Loudon et al. 1983). Thus, until the
failed to ovulate or experienced a delay in the onset of calves are weaned, this increased suckling frequency may in-
estrus. Although more work is needed, our data also suggest hibit reproductive activity via prolactin. Although suckling
© 2001 NRC Canada

J:\cjz\cjz79\cjz-05\Z01-050.vp
Friday, May 04, 2001 8:44:37 AM
Color profile: Generic CMYK printer profile
Composite Default screen

850 Can. J. Zool. Vol. 79, 2001

Fig. 3. Weekly average (± 95% CI) P4 concentrations (µg/g Fig. 4. Hormonal profiles for three captive female elk that were
feces) for high-nutrition pregnant elk (n = 8), medium-nutrition not allowed to breed during autumn 1997. Arrows indicate the
pregnant elk (n = 9), and low-nutrition nonpregnant elk (n = 9) approximate first day of each cycle. The approximate length of
during autumn 1997. Pregnant animals were aligned to the first the cycle is reported for each interval.
day of estrus (= day 0). For nonpregnant animals, day 0 corre-
sponds to the average breeding date for all bred animals in the
study. One high-nutrition animal that became pregnant was
excluded because pregnancy was not maintained.

frequency was not recorded in the present study, J.G. Cook

(unpublished data) documented increased suckling frequen-
cies in elk calves born to undernourished females.
Inadequate secretion of leutinizing hormone (LH) and fol-
licle stimulating hormone (FSH) may inhibit follicular matu-
ration, thereby delaying or preventing ovulation. Prior to the
breeding season, if nutrition is adequate and stress is not a
factor, gonadotropin-releasing hormone initiates pulsatile pi-
tuitary LH and FSH secretion. These gonadotropins, in turn,
promote follicular growth, increase estrogen secretion, and
promote behavioral estrus, which coincides with the pre-
ovulatory LH surge, and finally ovulation (Ferrell 1991).
Moderate undernutrition has been shown to suppress gonad-
otropin pulsatility, which precludes normal maturation and
ovulation of the dominant follicle (Jolly et al. 1995). This
scenario can delay ovulation, but severe undernutrition can white-tailed deer, Odocoileus virginianus (Teer et al. 1965;
completely impair folliculogenesis, leading to anovulation. Roseberry and Klimstra 1970), moose, Alces alces (Testa
Although the mechanisms that lead to suppression or al- and Adams 1998), caribou, Rangifer tarandus (Russell et al.
terations of gonadatropin secretion are complex, elevated or 1998), and domestic livestock (Ayalon 1978), our data did
sporadic progesterone secretion may prevent or suppress LH not reveal increased embryonic mortality in undernourished
and FSH pulsatility. In the present study, female elk that lost elk (e.g., we saw only one incidence of early-gestational
body mass before the breeding season also excreted higher mortality and that occurred in an obese animal from the
P4 concentrations. Nutritional “stress” may have stimulated high-nutrition treatment).
the adrenal cortex to increase production of steroids, includ- We did, however, find a positive relationship between ani-
ing progesterone (Plotka et al. 1983; Jobson et al. 1990). mal condition and postconception P4 excretion. Fecal P4 ex-
Additionally, reduced energy intake may have reduced the cretion in the high-nutrition group was consistently elevated
binding capacity of sex-steroid-binding proteins (Lermite compared with that in the medium-nutrition group during
and Terqui 1991) and limited metabolic clearance of plasma early pregnancy. The positive relation between condition and
P4 (Parr et al. 1993). In concert, these factors may have in- P4 concentration during 2 of 4 weeks post conception is con-
creased unbound plasma progesterone levels in the blood, sistent with the effects of dietary energy concentration.
which, in turn, may have dampened gonadotrophin secretion, While it was not observed in this study, reduced P4 produc-
thereby suppressing follicular growth and maturation (Stock tion in undernourished animals during early pregnancy may
and Fortune 1993; Jolly et al. 1995). lead to increased embryo mortality (Parr et al. 1993; Inskeep
Early embryonic mortality has also been associated with 1995). This observed relationship between body condition
inadequate nutrition. Although it has been documented in and P4 concentration may also be related to increased dietary

© 2001 NRC Canada

J:\cjz\cjz79\cjz-05\Z01-050.vp
Friday, May 04, 2001 8:44:38 AM
Color profile: Generic CMYK printer profile
Composite Default screen

Cook et al. 851

fiber consumption in the reduced-nutrition females (Wasser rates, owing to marginally deficient nutritional conditions.
et al. 1993). In our study, although dietary fiber content var- These marginally deficient conditions also may affect herd
ied as a function of diet, body condition, not dietary treat- demographics, because late-born calves have a greater prob-
ment, was significantly related to fecal P4 concentration. ability of dying (e.g., Clutton-Brock et al. 1982; Keech et al.
Reproductive function can also be altered by influences 2000) and because females in poorer condition have a re-
unrelated to nutrition. Seasonal breeders such as sheep duced probability of over-winter survival (Hobbs 1989).
(Ferrell 1991) and Cervidae (Kelly and Challies 1978;
Harder and Moorhead 1980; Haigh and Hudson 1993) may Acknowledgements
exhibit a short irregular “priming” cycle at the start of the
breeding season. Breeding may or may not occur, but con- Financial support for this study was provided by the Idaho
ception generally does not (Thomas and Femalean 1975). Department of Fish and Game (Federal Aid to Wildlife Res-
Although these priming cycles could be a mechanism for toration Project W-160-R) and Rocky Mountain Elk Founda-
estrous synchronization (Bubenik 1982), we hypothesize that tion. Support of the elk herd was provided by the Northwest
they may enhance flexibility of breeding date, ultimately en- Forestry Association, Boise Cascade Corporation, National
hancing opportunities for females to “choose” mates. For ex- Council for Air and Stream Improvement, U.S. Forest Ser-
ample, several studies showed delayed breeding in females vice Pacific Northwest Forest and Range Experiment Sta-
exposed to immature males (Hines et al. 1985; Noyes et al. tion, Oregon Department of Fish and Wildlife, and Rocky
1996; Komers et al. 1999). Noyes et al. (1996) reported that Mountain Elk Foundation. This support was facilitated
free-ranging female elk with access to 1.5-year-old bulls by L.L. Irwin, L.D. Bryant, R.A. Riggs, and J.G. Kie. We
bred 10–12 days later than females with access to 5-year-old acknowledge the foresight and work of J.W. Thomas that
bulls. This delay is equal in length to the short cycles we established the infrastructure which supported this study.
found in females that were not allowed to breed but were in C.T. Robbins provided guidance on raising and training elk
a pen adjacent to a mature bull. We found no indication of calves. We thank the University of Idaho and the Conserva-
irregular cycling prior to breeding in the 30 lactating elk tion and Research Center, Smithsonian Institution, for logis-
with access to a mature bull (see also Guinness et al. 1971; tical support and the U.S. Forest Service for supplying the
Adam et al. 1985). ultrasound machine. We are also grateful to S.E. George, D.
The results of our study and those of Komers et al. (1999) Lay, and S. Shuckle for their assistance with feeding, fecal
using fallow deer (Dama dama) suggest that stimulation collection, and weighing of the animals and K. Mashburn for
(sight, smell, full access) from a mature bull may be re- her guidance in laboratory procedures.
quired for adequate hormone production necessary for a nor-
mal estrous cycle to occur early in the breeding season. If References
such stimulation does not occur, the female may experience Adam, C.L., Moir, C.E., and Atkinson, T. 1985. Plasma concentra-
a shortened cycle that delays breeding. Normal cycling may tions of progesterone in female red deer (Cervus elaphus) dur-
then occur when females approach a balance between the ing the breeding season, pregnancy, and anoestrus. J. Reprod.
benefits of a preferred, mature male and the costs of delayed Fertil. 74: 631–636.
birth (Komers et al. 1999). Armstrong, J.P., and Britt, J.H. 1987. Nutritionally induced anestrous
Although the mechanisms involved in normal reproductive in gilts: metabolic and endocrine changes associated with cessation
processes are complex, our results reinforce the concept that and resumption of estrous cycles. J. Anim. Sci. 65: 508–523.
inadequate nutrition alters or inhibits breeding. This study Ayalon, N. 1978. A review of embryonic mortality in cattle. J.
shows that inadequate summer–autumn nutrition reduced Reprod. Fertil. 54: 483–893.
pregnancy rates by preventing estrus and ovulation rather Berger, J., Testa, J.W., Roffe, T., and Monfort, S.L. 1999. Conser-
than by preventing embryo implantation or inducing early vation endocrinology: a noninvasive tool to understand relation-
embryonic mortality. Our data indicate that relatively severe ships between carnivore colonization and ecological carrying
DE deficiencies are required to prevent pregnancy. Females capacity. Conserv. Biol. 13: 980–989.
that failed to breed entered the rut in poor condition (<6% Boone, W.R., Hill, J.R., Kennedy, S.W., and Henricks, D.M. 1975. In-
body fat) (Cook 2000); their withers, ribs, and hips were fluence of nutrition on reproduction in the ewe with special empha-
protruding, they were clearly lethargic, and their calves sis on progesterone concentration. Theriogenology, 3: 140–151.
ceased growth by early September. Females with delayed Brown, J.L., Wasser, S.K., Wildt, D.E., Graham, L.H., and Monfort.
S.L. 1997. Faecal steroid analysis for monitoring ovarian and
breeding due to DE deficiency, on the other hand, appeared
testicular function in diverse wild carnivore, primate and ungulate
to be in good condition. Thus, our data have two implica-
species. Int. J. Mamm. Biol. Proceedings Suppl. II. pp. 27–31.
tions for elk biologists. First, low pregnancy rates (e.g., Bubenik, A.B. 1982. Physiology. In Elk of North America: ecology
<80%) of prime-aged females with calves at heel are proba- and management. Edited by J.W. Thomas and D.E. Toweill.
bly symptomatic of summer–autumn ranges with markedly Stackpole Books, Harrisonburg, Pa. pp. 125–180.
deficient nutritional conditions. Forage conditions on these Cameron, R.D. 1994. Reproductive pauses by female caribou. J.
ranges may have an important limiting influence on herd de- Mammal. 75: 10–13.
mographics. Second, high pregnancy rates are not proof of Christensen, A.G., Lyon, L.J., and Unsworth, J.W. 1993. Elk man-
adequate nutritional conditions on summer–autumn ranges agement in the Northern Region: considerations in forest plan
(see also Verme and Ullrey 1984), nor are casual observa- updates or revisions. U.S. For. Serv. Gen. Tech. Rep. INT-303.
tions of females that appear to be in good condition. Our Clutton-Brock, T.H., Guinness, F.E., and Albon, S.D. 1982. Red
data illustrate that delayed breeding, particularly of prime- deer: behavior and ecology of two sexes. University of Chicago
aged, lactating females, can occur despite high pregnancy Press, Chicago.

© 2001 NRC Canada

J:\cjz\cjz79\cjz-05\Z01-050.vp
Friday, May 04, 2001 8:44:39 AM
Color profile: Generic CMYK printer profile
Composite Default screen

852 Can. J. Zool. Vol. 79, 2001

Cook, J.G. 2001. Nutrition and food habits. In Elk of North Amer- Imakawa, K., Kittok, R.J., and Kinder, J.E. 1983. Influence of di-
ica: ecology and management. Edited by D.E. Toweill and J.W. etary energy intake on progesterone concentrations in the beef
Thomas. Stackpole Books, Harrisburg, Pa. In press. heifer. J. Anim. Sci. 56: 454–459.
Cook, J.G., Irwin, L.L., Bryant, L.D., Riggs, R.A., Hengel, D.A., Inskeep, E.K. 1995. Factors that affect fertility during oestrous
and Thomas, J.W. 1994. Studies of elk biology in northeast cycles with short or normal luteal phases in postpartum females.
Oregon: 1993 progress report. National Council of the Paper J. Reprod. Fertil. Suppl. 49: 493–503.
Industry for Air and Stream Improvement, Corvallis, Oreg. Irwin, L.L., Cook, J.G., Riggs, R.A., and Skovlin, J.M. 1994. Ef-
Cook, R.C. 2000. Studies of body condition and reproductive physi- fects of long-term grazing by big game and livestock in the Blue
ology in Rocky Mountain elk. M.Sc. thesis, University of Idaho, Mountains forest ecosystems. U.S. For. Serv. Gen. Tech. Rep.
Moscow. PNW-GTR-325.
Day, M.C., Imakawa, K., Zalesky, D.D., Kittok, R.J., and Kinder, Jobson, N.B., Fisher, M.W., and Suttie, J.M. 1990. Plasma proges-
J.E. 1986. Effects of restriction of dietary energy intake during terone concentrations in cycling and in ovariectomised red deer
the prepubertal period on secretion of luteinizing hormone and hinds: the effect of progesterone supplementation and adrenal
responsiveness of the pituitary to luteinizing hormone-releasing stimulation. Anim. Reprod. Sci. 23: 61–73.
hormone in heifers. J. Anim. Sci. 62: 1641–1648. Johnson, D.E. 1998. Applied multivariate methods for data ana-
Dunn, T.G., Rone, J., Kaltenbach, C.C., van der Walt, L.A., Riley, lysts. Brooks/Cole Publishing Co., Pacific Grove, Calif.
M.L., and Akbar, A.M. 1974. Hormone changes during under- Jolly, P.D., McDougall, S., Fitzpatrick, L.A., Macmillan, K.L., and
feeding of beef females. J. Anim. Sci. 39: 206. [Abstr.] Entwistle, K.W. 1995. Physiological effects of undernutrition on
Ferrell, C.L. 1991. Nutritional influences on reproduction. In Re- postpartum anoestrus in females. J. Reprod. Fertil. Suppl. 49:
production in domestic animals. 4th ed. Edited by P.T. Cupps. 477–492.
Academic Press, San Diego. pp. 577–603. Keech, M.A., Bowyer, R.T., Ver Hoef, J.M., Boertje, R.D., Dale, B.W.,
Garrott, R.A., Monfort, S.L., White, P.J., Mashburn, K.L., and and Stephenson, T.R. 2000. Life-history consequences of maternal
Cook, J.G. 1998. One-sample pregnancy diagnosis in elk using condition in Alaskan Moose. J. Wildl. Manag. 64: 450–462.
fecal steroid metabolites. J. Wildl. Dis. 34: 126–131. Kelly, R.W., and Challies, C.N. 1978. Incidence of ovulation be-
Gerhart, K.L., Russell, D.E., Van DeWetering, D., White, R.G., fore the onset of the rut and during pregnancy in red deer hinds.
and Cameron, R.D. 1997. Pregnancy of adult caribou (Rangifer N.Z. J. Zool. 5: 817–819.
tarandus): evidence for lactational infertility. J. Zool. (Lond.), Komers, P.E., Birgersson, B., and Ekvall, K. 1999. Timing of estrus
242: 17–30. in fallow deer is adjusted to the age of available mates. Am. Nat.
Gombe, S., and Hansel, W. 1973. Plasma LH and progesterone levels 153: 431–436.
in heifers on restricted energy intakes. J. Anim. Sci. 37: 728–733. Lermite, V., and Terqui, M. 1991. Plasma sex steroid-binding pro-
tein in mature heifers: effects of reproductive status, nutritional
Gratson, M.W., and Zager, P. 1999. Study IV: factors influencing
level, and porcine growth hormone, and estradiol-17β treatments.
elk calf recruitment, Project W-160-R-25, Subproject 31: 1998
Biol. Reprod. 44: 864–870.
progress report. Idaho Department of Fish and Game, Lewiston.
Loudon, A.S.I., McNeilly, A.S., and Milne, J.A. 1983. Nutrition
Guinness, F., Lincoln, G.A., and Short, R.V. 1971. The reproduc-
and lactational control of fertility in red deer. Nature (Lond.),
tive cycle of the female red deer, Cervus elaphus L. J. Reprod.
302: 145–147.
Fertil. 27: 427–438.
McNeilly, A.S. 1979. Effect of lactation on infertility. Br. Med.
Haigh, J.C., and Hudson, R.J. 1993. Farming wapiti and red deer. Bull. 35: 151–154.
Mosby Year Book, Inc., St. Louis, Mo. Merrill, E.H., and Boyce, M.S. 1991. Summer range and elk popu-
Hamilton, W.J., and Baxter, K.L. 1980. Reproduction in farmed red lation dynamics in Yellowstone National Park. In The Greater
deer. J. Agric. Sci. (Camb.), 95: 261–273. Yellowstone ecosystem: redefining America’s wilderness heri-
Harder, J.D., and Moorhead, D.L. 1980. Development of corpora tage. Edited by R.B. Keiter and M.S. Boyce. Yale University
lutea and plasma progesterone levels associated with the onset of Press, New Haven, Conn. pp. 263–273.
the breeding season in white-tailed deer (Odocoileus virginianus). Monfort, S.L., Schwartz, C.C., and Wasser, S.K. 1993. Monitoring
Biol. Reprod. 22: 185–191. reproduction in captive moose using urinary and fecal steroid
Hines, W.W., Lemos, J.C., and Hartman, N.A. 1985. Male breeding metabolites. J. Wildl. Manag. 57: 400–47.
efficiency in Roosevelt elk of southwestern Oregon. Wildlife Morrison, J.A. 1960. Characteristics of estrus in captive elk.
Research Rep. No. 15, Oregon Department of Fish and Wildlife, Behaviour, 16: 84–92.
Corvallis. National Research Council. 1984. Nutrient requirements of sheep.
Hobbs, N.T. 1989. Linking energy balance to survival in mule deer: 6th ed. National Academy Press, Washington D.C.
development and test of a simulation model. Wildl. Monogr. National Research Council. 1985. Nutrient requirements of cattle.
No. 143. 6th ed. National Academy Press, Washington D.C.
Hodges, J.K. 1996. Determining and manipulating female repro- Nelson, J.R., and Leege, T.A. 1982. Nutritional requirements and
ductive parameters. In Wild mammals in captivity. Edited by food habits. In Elk of North America: ecology and management.
D.G. Kleiman, M.E. Allen, K.V. Thompson, and S. Lumpkin. Edited by J.W. Thomas and D.E. Toweill. Stackpole Books,
University of Chicago Press, Chicago. pp. 418–419. Harrisburg, Pa. pp. 323–367.
Holechek, J.L., Vavra, M., and Skovlin, J. 1981. Diet quality and Noyes, J.H., Johnson, B.K., Bryant, L.D., Findholt, S.L., and Thomas,
performance of cattle on forest and grassland range. J. Anim. J.W. 1996. Effects of bull age on conception dates and pregnancy
Sci. 53: 291–298. rates of female elk. J. Wildl. Manag. 60: 508–517.
Horwitz, R.J. 1980. Official methods of analysis of the Association Ott, R.L. 1993. An introduction to statistical methods and data
of Official Analytical Chemists. 13th ed. Association of Official analysis. Wadsworth, Inc., Belmont, Calif.
Analytical Chemists, Washington, D.C. Parker, K.L., Gillingham, M.P., Hanley, T.A., and Robbins, C.T.
Howie, P.W., and McNeilly, A.S. 1982. Effects of breast feeding pat- 1999. Energy and protein balance of free-ranging black-tailed
terns on human birth intervals. J. Reprod. Fertil. 65: 545–557. deer in a natural forest environment. Wildl. Monogr. No. 143.

© 2001 NRC Canada

J:\cjz\cjz79\cjz-05\Z01-050.vp
Friday, May 04, 2001 8:44:40 AM
Color profile: Generic CMYK printer profile
Composite Default screen

Cook et al. 853

Parr, R.A. 1992. Nutrition–progesterone interactions during early Thomas, D.C., and Femalean, I.M. 1975. The pattern of reproduc-
pregnancy in sheep. Reprod. Fertil. Dev. 4: 297–300. tion in female Colombian black-tailed deer, Odocoileus hemionus
Parr, R.A., Davis, E.F., Miles, M.A., and Squires, T.J. 1993. Liver columbianus. J. Reprod. Fertil. 44: 261–272.
blood flow and metabolic clearance rate of progesterone in sheep. Thorne, T., Dean, R.E., and Hepworth, W.J. 1976. Nutrition during
Res. Vet. Sci. 55: 311–316. gestation in relation to successful reproduction in elk. J. Wild.
Plotka, E.D., Seal, U.S., Verme, L.J., and Ozoga, J.J. 1983. The Manag. 40: 330–335.
adrenal gland in white-tailed deer: a significant source of Trainer, C.E. 1971. The relationship of physical condition and
progesterone. J. Wildl. Manag. 47: 38–44. fertility of Roosevelt elk (Cervus canadensis roosevelti) in
Robbins, C.T. 1993. Wildlife feeding and nutrition. 2nd ed. Oregon. M.Sc. thesis, Oregon State University, Corvallis.
Academic Press, San Diego, Calif. Verme, L.J. 1969. Reproductive patterns of white-tailed deer related
Roseberry, J.L., and Klimstra, W.D. 1970. Productivity of white- to nutritional plane. J. Wildl. Manag. 33: 881–887.
tailed deer on Crab Orchard National Wildlife Refuge. J. Wildl. Verme, L.J., and Ullrey, D.E. 1984. Physiology and nutrition. In
Manag. 34: 23–28. Whitetailed deer: ecology and management. Edited by L.K. Halls.
Russell, D.E., Gerhart, K.L., White, R.G., and Ven De Wetering, Stackpole Books, Harrisburg, Pa. pp. 91–118.
D.V. 1998. Detection of early pregnancy in caribou: evidence
Vogelsang, R.W. 1977. Blood urea nitrogen, serum cholesterol and
for embryonic mortality. J. Wildl. Manag. 62: 1066–1075.
progestins as affected by nutritional intake. M.Sc. thesis, Vir-
SAS Institute Inc. 1988. SAS/STAT user’s guide, 6.03 ed. SAS
ginia Polytechnic Institute and State University, Blacksburg.
Institute Inc., Cary, N.C.
Wasser, S.K., Monfort, S.L., Souther, J., and Wildt, D.E. 1994. Ex-
Schwarzenberger, F., Møstl, E., Palme, R., and Bamberg, E. 1996.
cretion rates and metabolites of oestradiol and progesterone in
Faecal steroid analysis for non-invasive monitoring of reproduc-
baboons (Papio cynocephalus cynoecphalus) faeces. J. Reprod.
tive status in farm, wild and zoo animals. Anim. Reprod. Sci.
Fertil. 101: 213–220.
42: 515–526.
Spitzer, J.C., Niswender, G.D., Seidler G.E., Jr., and Wiltbank, J.N. Wasser, S.K., Thomas, R., Nair, P.P., Guidry, C., Southers, J., Lucas,
1978. Fertilization and blood levels of progesterone and LH in beef J., Wildt, D.E., and Monfort, S.L. 1993. Effects of dietary fibre
heifers on a restricted energy diet. J. Anim. Sci. 46: 1071–1077. on faecal steroid measurements in baboons (Papio cynocephalus
Stock, A.E., and Fortune, J.E. 1993. Ovarian follicular dominance in cynocephalus). J. Reprod. Fertil. 97: 569–574.
cattle—relationship between prolonged growth of the ovulatory Whitten, P.L., Brockman, D.K., and Stavisky, R.C. 1998. Recent
follicle and endocrine parameters. Endocrinology, 132: 1108–1114. advances in noninvasive techniques to monitor hormone–behavior
Stussy, R.J. 1993. The effects of forage improvement practices on interactions. Am. J. Phys. Anthropol. (Suppl.), 27: 1–23.
Roosevelt elk in the Oregon coast range. M.Sc. thesis, Oregon Williams, A.H., and Cumming, I.A. 1982. Inverse relationship be-
State University, Corvallis. tween progesterone concentration and nutrition in ewes. J. Anim.
Teer, J.G., Thomas, J.W., and Walker, E.A. 1965. Ecology and man- Sci. 98: 517–522.
agement of white-tailed deer in Llano Basin of Texas. Wildl. Wisdom, M.J., Cook, J.G., Rowland, M.W., and Noyes, J.H. 1993.
Monogr. No. 15. Protocols for care and handling of deer and elk at the Starkey
Testa, J.W., and Adams, G.P. 1998. Body condition and adjust- Experimental Forest and Range. U.S. For. Serv. Gen. Tech. Rep.
ments to reproductive effort in female moose (Alces alces). J. PNW-GTR-311.
Mammal. 79: 1345–1354.

© 2001 NRC Canada

J:\cjz\cjz79\cjz-05\Z01-050.vp
Friday, May 04, 2001 8:44:40 AM