14-8 Mr. C. F. M. Swynnerton on Rejections [Ibis, 



Museum, and pi'esumably produced in this way. The black 

 deposit was superficial, but in this, of course, it did not differ 

 from the clialk-layer of certain eggs and the heavier blotches 

 of, e. g., certain eggs of Pycrionotus and Phyllastrephus. 



3. The host's and the Cuckoo's defences. — The host's de- 

 fences include, first, means of preventing the Cuckoo from 

 depositing its egg. Attacking or mobbing the Cuckoo is one 

 such means, and here it would seem that the habit of nesting 

 in colonies must be exceedingly useful. I have seen the 

 members of a Weaver colony drive off a Didric Cuckoo, and 

 in a previous instance 1 saw a similar mobbing, at a large 

 colony, of a dark bird, probably a Cuckoo, that I failed to 

 obtain a sufficiently good view of for identification (' Ibis/ 

 1908, p. 11). It must be difficult for a Cuckoo to lay 

 undetected in such a colony, and the mobbing is, of course, 

 more formidable than the attack of a single pair of birds. 

 Even so, the Cuckoo hangs about such colonies and is 

 sometimes successful, and, if detection should be avoided — 

 which is unlikely, — the close collection of nests would, of 

 course, improve its opportunity of matching its egg, if it 

 should have becomeits habit to attempt to do so. Mr. Austin 

 Roberts's observation (Journ. S. A. O. U. ix. 1913, p. 33) that 

 " Chrysococcyx cupreus sometimes deposits its eggs in the 

 nest of Ploceus auricapillus, but apparently only when there 

 are one or two nests in a tree/' has a bearing on this point. 

 As he had spoken of "dozens of nests" in some of the trees, I 

 take his meaning to be that isolated nests are mostly selected. 

 This would seem to testify to the usefulness of the colony. 



Prevention failing, the defence afforded by the colouring 

 of the host's own eggs comes into play. I have already 

 ('Ibis/ 1916, pp. 570 & 573; 1917, p. 271) expressed my 

 opinion that, whatever be the correct explanation of varia- 

 bility in such eggs as the Common Guillemot's, poly- 

 morphism in ti)e eggs of many small Passerines is probably 

 to be explained as having been selected in relation to the 

 baffiing ol: Cuckoos that might otherwise more often match 

 their eggs. The results of my experiments quoted under Con- 

 clusion 7 (p. 132), above, show clearly that there is nothing 



