106 U.S. NATIONAL MUSEUM BULLETIN 2 65 



has retained a fixation on Qerygone in one of its races (C. I. layardi), 

 but has gone far afield in its use of many other birds in New Zealand 

 (C I. lucidus) and especially so in southern Australia {C. I. plagosus), 

 where its hosts total some 75 species of many genera, although it 

 shows very marked preference for species of Acanthiza and to a some- 

 what lesser extent for Sericornis and Malurus. 



C. basalis has the most extensive host catalog of any of the glossy 

 cuckoos, totaling some 100 species but showing a very strong inclina- 

 tion toward the use of wrens of the genus Malurus and, to a lesser 

 degree, species of Acanthiza, Rhipidura, Petroica, Hylacola, and other 

 warblers and honey-eaters, A great reduction in the number of species 

 of victims is characteristic of C. osculans, which is very largely para- 

 sitic on Chthonicola, with the eggs of which its own have a remarkable 

 similarity, and also on Pyrrholaemus. Of the host choice of C. rujicollis 

 we know nothing, but it may be expected to be somewhat similar to 

 that of C. malayanus and C. lucidus. Also, of the Papuan, C. meyerii, 

 there are as yet no data as to what birds it depends on to raise its 

 young. 



The Asiatic C maculatus and C. xanthorhynchus are parasitic on 

 sunbirds and warblers and a few other hosts, but, insofar as present 

 knowledge permits a generalization, the sunbirds, Arachnothera and 

 Aethopyga are their mainstay. 



Of the African species, no hosts are yet known for C. jlavigularis: 

 for the others a large number are now recorded, and there is consider- 

 able overlapping, or what I have here termed homoxenia, among them. 

 However, caprius is far more prone to use weavers, Ploceidae, while 

 klaas tends to use warblers, Sylviidae, and sunbirds, Nectariniidae, to 

 a great extent; cupreus is a less selective parasite, using a common 

 bulbul, Pycnonotus, more often than any other host, but also using 

 weavers, warblers, and sunbirds. 



On the whole, Chrysococcyx presents an uneven performance, in an 

 evolutionary sense, in the development of host-adaptive, individually 

 divergent egg morphism, with the related development of individual, 

 and even of specific, host specificity. The species which show these 

 traits best developed are osculans, with a single egg type and one 

 most regular, favorite host choice, and caprius, with several ovo- 

 morphs, three of which are more or less related each to a regular, 

 frequent host species. Inasmuch as these two cuckoos are quite far 

 apart in the phylogeny within the genus and inasmuch as none of the 

 other species show any comparable degree of adaptive egg morphism, 

 one can only conclude that the genus Chrysococcyx has not revealed a 

 continuity of adaptive evolution lq this important feature of brood 

 parasitism. What has been achieved in this direction has transpired 

 within individual species. 



