KIRUMBO. 



393 



it was this species. Whilst one bird was thus playing, another would frequently 

 answer its cry from a tree hard by." 



Only one family constitutes the super-family COLIOIDE^E, a rank which may 

 fairly be defended by Dr. Murie's words : " If we take one set of regional charac- 

 ters, the feet, the head, the breastbones, the pelvis, and so on, we c;m place it in 

 as many different groups; we can even trace raptorial kin; so that it is hard to say 

 where Colitis could not be wedged in, and plausibly too. Not only is it entitled to 

 be considered aberrant, but to afford the strongest proof of the interlinking of type, 

 not in the chain series so often advocated, but, like the Isle of Man tripodal coat-of- 

 arms, kicking its legs about, and whichever alighting upon, there it stands. But if in 



'. , " ,-,^Y."' 



FIG. 193. Leptosomus discolor, kirumbo. 



true spirit of ornithology we take the bird in its completeness, it will be allowed it 

 does not so closely resemble any acknowledged individual group as to come under 

 its definition." 



As demonstrated by Professor Garrod, the palate is desmognathous (cf. the accom- 

 panying cut, Fig. 194), and the vomer is not ossified, and consequently lacking as a 

 bone, therein agreeing with Alcedo. There are no basipterygoid processes. As to 

 the breastbone, he asserts that it resembles that of the Megalaimidaa more than any 

 other bird. The rostrum is indented, but not deeply cleft. In regard to internal 

 characters, his researches confirm the result of Dr. Murie, based upon ostcological 

 grounds, that the Coliidoe are not to be referred to the Cuculoideae, as has usually 

 been done. They lack the ambiens muscle, and " in the arrangement of its plantar 



