Feeding frequencies of nestling blue tits (Parus caeruleus): costs, benefits and a model of optimal feeding frequency.

1. To examine the ultimate factors determining the frequency with which male and female blue tits (Parus caeruleus) feed their nestlings, I manipulated the brood sizes of 34 pairs breeding in Wytham Woods, England, and recorded automatically the number of visits when the young were 11-13 days of age. At this time adults were caught and weighed and the nestlings weighed. 2. The number of visits/brood increased linearly with brood size, with no evidence of an upper limit to brood feeding frequency. The number of visits·nestlings-1·day-1 decreased from 138 to 108 as brood size increased from 3 to 6, but levelled off at 68-87 visits/nestling as brood size increased from 8 to 15. 3. Female weight-loss during the nestling period increased linearly as the feeding frequency of the brood increased: those females were lightest that fed their young most often; this is of some significance since I previously showed that the lightest females are least likely to survive to the following year. The correlations of feeding frequency with brood size and feeding frequency with female weight are sufficient to account for the commonly observed correlation of brood size and female weight. No relationship between feeding frequency and male weight dynamics was apparent. Feeding frequency of the brood increased as mean nestling weight decreased. Apparently, parents compensate, in part, for low nestling weight by increasing feeding frequency. Low nestling weight, in turn, has proven detrimental to post-fledging survival. 4. Lack's (1947, 1954) hypothesis, that the decline in visits/nestling as brood size increases reflects the inability of parents to sustain high brood feeding rates, is refuted by the observations made in this and a large number of other studies. Royama's (1966) hypothesis that the observed decline in visits/nestling reflects a decrease in energy needed to maintain homeothermy, receives some support. Behavioral and physiological studies, however, suggest that the "Royama effect" is limited to small broods (c. four or fewer young). 5. I propose, instead, a model of optimal feeding frequency, which accounts for observations made by myself and others, and which incorporates Royama's hypothesis as well. The model assumes that parents optimize their investment in their young. On the one hand, an increase in feeding frequency means more food for the young, which enhances offspring survival ("benefit"); on the other hand, increased feeding frequency is "costly" to the parents (reduces parental survivorship). Natural selection favors breeding individuals that maximize the difference between benefits (to the offpsring) and costs (to the parents). The model accounts for many aspects of blue tit reproductive biology, including the observed relationship between nestling weight and brood size and the observation that parent reproductive costs increase with brood size.