ALTHOUGH Lack1,2 proposed that birds should lay clutches of a size that produces the most recruited offspring, in reality the most frequent clutch size tends to be smaller than the most productive one3. Of the six different but not mutually exclusive hypotheses explaining why Lack's hypothesis1,2 is not correct, four are based on adaptive explanations3-8 whereas two are nonadaptive9-11. One of the latter argues that because of gene flow between habitats of different quality and hence different optimal clutch sizes, clutches may be larger than optimum in poor habitats and smaller than optimum in good habitats 9,10. This is the only hypothesis for which no evidence has been obtained up to now. We now present evidence that great and blue tits (Pants major and P. caeruleus) do not adjust their clutch size in an adaptive way in habitats of different quality. The results support the 'nonadaptive' hypothesis that because of gene flow between habitats of different quality as a result of birds moving from one habitat to another, birds can be unable to produce a clutch size adapted to the habitat in which they breed9,10. © 1990 Nature Publishing Group.