BIRDS OF UPPER PEGU. 99 



to illustrate most fully the supposed characteristics of this new 

 species ; but, on the other hand, I have another specimen from 

 the same locality undistinguishable from Etawah, Simla, and 

 Southern Indian specimens, and other specimens again inter- 

 mediate between these ; but in size of wing-, in shape and size of 

 bill, there is not one iota of difference between the grey bird and 

 one of the darkest Darjeeling birds, or between it and others 

 of different shades from Dacca, as well as from other localities. 

 I hardly understand making a species dependent on a slight 

 difference in tint in a case like this, when difference of tint is not 

 even constant in all the individuals from the same locality. These 

 particular Thayetmyo specimens, to which I have referred, 

 are exactly similar in every respect to the grey Dacca birds, and 

 illustrate, I think, clearly Mr. Blyth's remark, that Dicrurus 

 longicaudaUis in passing eastwards and southwards begins to 

 assimilate somewhat or approximate to Dicrurus cineraceus. 



280 bis. — Dicrurus leucophaeus, Vieil. {Vide ante, 

 Stray Feathers, 1874, p. 210.) 



But besides the intermediate forms above referred to, which I 

 have identified with longicaaidatus, and which any one who 

 pleases may sub -divide into two species, — vyrrhops, Hodg., and 

 intermedins, Blyth, both of which, as well as the true longicaudaUis, 

 Hay, occur together at Dacca, and may be there shot together off 

 the same tree, — there are some specimens from Thayetmyo greyer 

 still than those referred to, and absolutely identical with speci- 

 mens from Singapore and Malacca, except in having a somewhat 

 narrower bill. 



I confess that I do not know how to deal with these Dicruri. 

 Nature has drawn no hard-and-fast line between all the innumer- 

 able varieties which bind together with an absolutely perfect 

 chain, no single link apparently wanting, the perfectly grey, 

 comparatively short-tailed, and broad-billed birds from Sumatra 

 and the Straits, which I take to be leiicop/iaus, and the dark 

 comparatively narrow-billed and long-tailed typical longicaudatus. 

 If we compare birds from the opposite ends of the scale, nothing 

 can appear more distinct ; but if we carefully collate hundreds of 

 specimens from very numerous localities, the impossibility of 

 drawing any but entirely arbitrary lines of separation becomes 

 more and more palpable. 



If two nearly allied races are to be distinguished as distinct 

 species, it appears -to me that it is not merely sufficient to define 

 the types of each species, but to lay down such a definition of each 

 species as shall enable observers to refer any specimen they ob- 

 tain certainly and definitely to one or other species ; and this is 

 what it appears to me cannot be done in the case of the four or 

 possibly more races of leucop/ucus, longicaudatus, 8fc. As far as I 



