418 BULLETIN 15 3, UNITED STATES NATIONAL MUSEUM 



ones. Van Someren '^^ writes that it seems to him that somaliensis 

 (which he erroneously refers to as sonialicus) is founded on a female 

 of typical ■fl-avirostris. " Seven adult males from Somaliland show 

 no red on the lower mandible, two females have the basal half and 

 the tip reddish, but this is also found in east African birds. Two 

 birds, differing only in the color of the lower mandible, otherwise 

 alike, can hardh^ occupy the same locality." I have examined an 

 adult female from British Somaliland (Lort Philips collection) and 

 can find no difference either in size or in color between it and typical 

 favirostns. 



The typical race of this hornbill may easily be identified by reason 

 of the fact that the breast feathers are white with black shaft stripes, 

 while in the other two forms these feathers are laterally edged with 

 black and have no shaft stripes, 



2. L. f. leucomdas. — From the Orange River and Natal northward 

 through Damaraland, Bechuanaland, and Southern Rhodesia to the 

 Zambesi Valley. I know of no definite records from southern Mo- 

 zambique or from Nyasaland or Northern Rhodesia. 



This form and the next are very similar but the white spots on 

 primaries 3 to 7 (counting from the outside) are restricted to the 

 outer webs in leucomelas while in elegans there are spots on both 

 webs in these feathers. L. f. leucomelas is the only one of the three 

 races of the yellow-beaked hornbill in which the basal black area <Jn 

 the third (from the outside) pair of rectrices is not at all interrupted 

 by white. It is most widely interrupted in the typical form, less 

 so in elegans^ never in leucomelas. 



3. L. f. elegans. — Angola, Benguella, and Loanda. I have seen but 

 one specimen of this form, marked " South Africa." It was origi- 

 nally in the collection of E. L. Layard and is otherwise without data, 

 and I assume that the locality is wrong as South African specimens 

 are all true leucomelas. 



It is interesting to note in passing that the variations in the wing 

 spots and the distribution of black and white in the rectrices have a 

 geographical significance in this hornbill whereas in Lophoceros 

 crythrorhynchus the highly diversified variations in these charac- 

 ters are purely individual. Have we here one species {erythrorhyn- 

 chus) that is in an early stage of racial differentiation and another 

 in which the process is complete ? If so, it is perfectly obvious that 

 natural selection has nothing to do with the formation of races or, 

 at least, the elimination of certain variations, as these characters can 

 not possibly be of any benefit or detriment to the birds. 



The size variations of the present series may be ajDpreciated from 

 the following table : 



« Nov. Zool., vol. 29, 1922, p. 75. 



