Adaptive precocial reproduction in voles: reproductive costs and multivoltine life-history strategies in seasonal environments

1. We determined whether reproductive costs associated with precocious or later breeding by female Townsend's voles (Microtus townsendii) differ and how this relates to the multivoltine life history of microtines. We related the probability of mothers surviving the lactation period to maternal age and the number of pups per litter using logistic models allowing for a threshold value in either explanatory variable below (for litter size) and above (for age) which survival is constant. 2. Females born at the beginning of the breeding period were precocious, giving birth for the first time by the time they were 85 days old (min 38 days). Females born at other times delayed sexual maturation by up to 6 months. 3. The survival of females who gave birth before a fitted threshold of 85 days old declined precipitously if they gave birth to more than five pups per litter. Survival of older females did not vary in relation to the size of their litters. 4. Precocious females had not reached adult size when they conceived their first litter and substantially increased their structural mass during pregnancy. We estimated that young primiparous females gained on average 127 +/- 9% of their conception weight during their 21-day first pregnancy compared to an average increase of 47 +/- 2% per pregnancy for all other females. It appears that the low survival during lactation experienced by this cohort alone may be a consequence of the cumulative costs of gestation, lactation and somatic growth. 5. We build simple life-table models to estimate the contribution of very precocious (conception at 25 days), precocious (conception at 50 days) and non-precocious females (conception > 50 days) to the population as the probability of surviving the spring and summer, plus the total number of females produced and surviving until the end of the summer. We parameterized such models using the empirical survival estimates. Females who conceived five young very precociously made the greatest contribution to the autumn population. Delaying reproduction had a large cost: a very precocious female starting with five young would contribute 7.21 females, while a precocious female with the same initial litter size would contribute only 2.83 females. This advantage stems from both the longer breeding-life span and from the fact that offspring of very precocious females may themselves reproduce in the year of their birth, such that they compound the precocious investment of their mothers. 6. Thus iterated reproduction and the multivoltine nature of microtine life history select for cohort-specific trade-off between survival and reproduction.