88 ORNITHOLOGICAL EXPLORATIONS. 



examination of a very large series of both species has convinced me 

 that all transitional stages occur, as birds are found having the two, 

 three, four, live, or six first primaries with white shafts, and that these 

 characters, consequently, are of little value. The coloration may be 

 depended upon as a safe guide for the identification in the following 

 cases : 



If the bird has wholly blacli: legs, tarsi included, then it belongs to 

 parasiticus {crepidatus), even if light colored underneath. This char- 

 acter, however, is only safe when applied to rather fresh specimens, as 

 the light tarsi often darkens so much in museum specimens that it is 

 difficult to decide upon the original color. 



If the bird is sooty all over, it is parasiticus. 



If it has the tarsi light blue (Saunders, P. Z. S., 1876, p. 331, gives the 

 color as yellowish olive ; that is only the case in dried specimens, not 

 in the fresh bird), simultaneously with uniformly soot-colored under wing 

 coverts, tlien it certainly belongs to longicaudus. 



In all other cases the proportions of the different parts of the bill are 

 the only reliable characters. S. parasiticus is recognizable by having 

 the gonys shorter and the nasal shield longer, while in longicaudus these 

 proportions are reversed. The length of the gonys, as compared with 

 the breadth of the bill across the points of the loral antise is a good 

 character of the bird when still in the flesh, and so is the position of the 

 angle of the gonys in its relative position to the nostrils, the former 

 being placed much in front of the anterior border of the nostrils in 

 parasiticus, while in longicaudus it is placed right below that same point. 

 But museum specimens may be identified by measuring the distance 

 from the anterior border of the nostrils to the tip of the bill and com- 

 pare it with the length of the nasal shield along the culmen. In longi- 

 caudus the two distances are of equal length, while in parasiticus the 

 nasal shield is much longer. 



These characters hold good at least in birds one year old just before 

 the second moult, and in the adults. Whether young ones in the first 

 plumage are distinguishable in the same manner I cannot say with cer- 

 tainty, as 1 have no access to a specemen of longicaudus of that age, but 

 I have little doubt that they are, as the essential characters of parasiticus 

 are very w ell borne out in several young specimens before me. 



