Reply by the authors

John Coulson criticizes us for one of our assumptions in our population model, of Black-legged Kittiwakes on Fair Isle, namely that this is a closed unit. We share his concern about this and were careful to state this assumption explicitly in the methods. We also pointed out that although there is evidence of emigration/immigration in this species, there are currently no measures for this colony that could be used to parameterise a more complex, and hopefully, more realistic model. A recent study by Suryan & Irons (2001, Auk 118:636-649) on Blacklegged Kittiwakes in Prince William Sound, Alaska found that, although there was evidence of non-natal recruitment, changes in the numbers of birds at a particular colony were significantly influenced by reproductive success at that colony 5 years previously, indicating that natal recruitment (the mean age of first breeding was 5 years) was an important component of colony dynamics. The demography of the Kittiwake in the Pacific and Atlantic may differ (Coulson 2002, Journal of Avian Biology 33: 111-112) so we need to be cautious about extrapolating between the two areas but this finding gives some empirical support for our approach.