THE HONEY-GUIDES 9 



Prodotiscus insignis, and finally, on emerging from the forest into 

 the bushveld, Prodotiscus regulus. The genus Prodotiscus is the most 

 divergent from typical Indicator of all the honey-guides, and it is 

 less closely alied to Melignomon than the above brief statement may 

 suggest. It seems, however, that Melignomon zenkeri does represent 

 a first stage on the path that led from Indicator to Prodotiscus, but 

 it is very much nearer to the former than to the latter genus; in fact, 

 Melignomon could readily be called only a subgenus of Indicator. 

 That Prodotiscus regulus is a further development than P. insignis is 

 suggested by the fact that in its immature plumage it has the outer 

 rectrices similar in pattern to those of P. insignis (adult and young), 

 and also by the fact that it is a denizen of more open country, not of 

 the true forests. 



The apparent phylogenetic relationships of the species of honey- 

 guides are expressed in the accompanying diagram (fig. 1). Their 

 classification into genera, species, and races follows that given in my 

 recent systematic revision of the group (Friedmann, 1954b). 



No one character has had a more varied history in the phylogeny 

 of the honey-guides and their relatives than has the form of the tail. 

 Starting from a presumably 12-rectriced group, these birds retained 

 this number while the barbets became 10-rectriced, and it is only 

 in the most divergent of their genera, Prodotiscus, that the honey- 

 guides have lost one pair of rectrices. Also, the genus Indicator 

 gave rise, through the remarkable alteration of an originally plain, 

 unspecialized tail, to the highly peculiar form of the lyre-tailed 

 species, Melichneutes robustus. Furthermore, in at least two species 

 of Indicator {indicator and minor) a rustling flight, possibly involving 

 the tail feathers, is known to take place. In the not distantly related 

 woodpeckers, the tail feathers went through very striking functional 

 modifications as well. 



While not contributing significantly to our understanding of the 

 relationships of the various species of honey-guides, the following 

 anatomical notes, based on an examination of 8 of the 11 species 

 and all additional to the previously published data, seem worthy of 

 record here. Fiirbringer's (1888) extensive observations on the 

 myology of Indicator indicator are not discussed because I have no 

 similar data on the other species with which to compare them. There 

 would seem to be little likelihood of any essential differences in 

 their musculature, however. 



The sternum shows remarkable variation in the development of 

 the keel. In some species of the genus Indicator (subgenera Indicator 

 and Melignothes) the keel is fairly well developed, about as in a number 

 of barbet genera examined; in Indicator xanthonotus (subgenus 

 Pseudofringilla) and in Melichneutes robustus it is noticeably shallower 



309265 — 55 2 



