6 U.S. NATIONAL MUSEUM BULLETIN 2 65 



syringeal muscles), in most respects lucidus is like cwpreus. Therefore, 

 caprius is most unlike the other species of this genus." 



Of the 12 species of the group, the breeding habits of 9 are known, 

 and these are all parasitic. The three whose eggs and young have not 

 yet been discovered are ruficollis, meyerii, smdjlavigularis; it seems safe 

 to assume that these are parasitic as well. In other words, there is no 

 reason to think that any chronological stages in the development of 

 parasitism are to be seen in the members of this genus. Since Chrys- 

 ococcyx is obviously most closely related to other parasitic genera such 

 as Cuculus and Cacomantis, it is reasonable to conclude that the glossy 

 cuckoos were descended from Cuculinae that were already parasitic. 

 As we shall see, some special refinements of brood parasitism are to be 

 found in certain members of the group and not in others; refinements 

 such as egg similarity to host eggs, restriction of parasitism to a small 

 rather than a large number of host species, and atavistic behavior 

 patterns such as nestling and fledgling feeding. However, since until 

 now all the species of the genus were studied, not as an entity, but 

 almost solely according to geographic occurrence, it was impossible to 

 appraise the significance of differences in refinement, or atavism of 

 portions, of the behavior patterns associated with the annual repro- 

 ductive cycles of these birds. 



The chief intention of my comparative survey of adequate series 

 of specimens of all of the 12 species was to enable me to suggest their 

 most probable phylogenetic pattern, according to which I might then 

 arrange and evaluate the data on various aspects of theu" biology. 

 The many hundreds of study specimens in the British Museum and 

 in the American Museum of Natural History were carefidly examined 

 and compared, but even this ample material failed to suggest with 

 incontrovertible definiteness which particular pattern was the only 

 correct interpretation of the relationships and of the past history of 

 the group. It was impossible to conclude that the incomplete, cir- 

 cumstantial evidence of the present forms of the genus and of their 

 distribution pointed to only one interpretation. Too many — in fact 

 almost all — intermediate stages have long since disappeared in this 

 ancient assemblage. Even the "intuitive" grasp of a complex picture — 

 which actually is usually a result of time-requiring "mental digestion" 

 arrived at without any rigidly logical series of steps, and which results 

 in the generally clarified, or at least correlated, arrangement on which 

 taxonomists so often have come to rely and on which they lay such 

 store (often with sound reason) — has been less neatly precise in the 

 present instance than I could have wished it might be. This factor of 

 uncertainty is an almost universal characteristic of evolutionary or 

 phylogenetic reconstructions, and it is by no means peculiar to the 

 present one. I mention it only because so many times authors either 



