R.G. White, J.E. Rowell and B. Hauer

Supporting Agency: The Center for Field Research and Earthwatch

In this study we examined the effect and predictive value of fall body weight (BW) and fall body condition score (BCS) on pregnancy success in muskoxen. Ten captive female muskoxen were placed on either a high (n=5) or low (n=5) nutritional plane in 1987. The high plane (HP) animals were allowed to graze pasture, had year round free access to brome-grass hay and 550 g/d (12 g/kg BW ^0.75) of a pelleted ration (Fisher Mills, Seattle and Alaska Mill, Anchorage). Low plane (LP) cows were given grazing during the summer with access to hay, but in winter were fed ad lib. brome hay only. Body weight (BW) and body condition scores (BCS) were followed in all 10 cows, between 1988 and 1992. Calving dates, calf birth weights, lactation status in the fall and age were also recorded.

Relationships between pregnancy and BW, BCS, lactation, age, calving date, and calf weight were described by univariate logistic regression (BMDP), based on the raw binary data:

where p is the percent probability of each response variable occurring at variable x; logistic models were checked for goodness of fit (Lemeshow and Hosmer 1982). In all cases, p-values < 0.05 were considered statistically significant.

While both BW and BCS differed significantly (T-test p<0.05) with pregnancy and dietary treatment, only BCS differed with lactational status. In the logistic regression model, Oct BW was the most significant predictor of pregnancy when only weights were considered. When BCS was entered with Oct BW, Sep BCS was the most significant variable and the only one to enter the model, indicating that BCS is a more powerful predictor of breeding success than BW. Neither lactational status or year in combination with other variables or alone, entered the model. This model identified BCS as a very good predictor of pregnancy (100%) but less accurate in predicting non-pregnancy (66.67%), a reflection of the small sample size in the non-pregnant category(n=7).

Both body weight and BCS, measured August through October, can be used to predict fall breeding success in muskoxen. The BCS curve generated by the model (Fig. 4a) gives a 50% probability of pregnancy at BCS 9 (from a possible range of 3-15). Under field conditions, where body weights are hard to obtain, BCS is a rapid, easy means of gaining predictive advantage. From the Oct BW curve (Fig. 4b) a BW of approximately 180 kg is the threshold at which 50% of the females will become pregnant. For individuals involved in intensive muskox husbandry this threshold weight is a critical consideration in harem assembly.