Manx Shearwater Peter Storey Total Fledglings (No. ringed in burrows) x (No. caught on surface) (No. ringed in burrows and recaptured at surface) In using such an estimate there are two basic assumptions: 1) that the ratio of unringed to ringed birds in the number caught on the surface provides a reliable estimate of the proportion of the birds that were ringed in the burrows, and 2) that by multiplying the ratio by the total caught on the surface, one obtains a reliable estimate of the total number of fledglings (for further details see e.g. Southwood, 1978). Although there are a number of possible biases in such an estimate, it is not believed that any of these are likely to have caused serious errors in this particular case. In order to estimate the total number of breeding pairs, Corkhill multiplied his estimate of the number of fledglings by the reciprocal of the breeding succes (the number of eggs laid which produced fledged young). In the present case such a calculation is not necessary since our aim was to compare the 1978 findings with those of Corkhill's for 1971; a direct comparison of the estimate of the number of fledglings can be made. A total of 452 birds were taken out of their burrows, ringed and replaced. The work was done from 10 to 21 August, mainly by Mr S. Roddis. The approximate position of the nest-site of each bird was recorded and, following Corkhill, the totals subdivided into twc regions, that of South Castle and the remainder of the Neck. During the period 22 August to 11 September 1978, birds were caught on the surface. Again following Corkhill the area was divided into two sections, South Castle and the rest of the Neck