REMARKS ON THE GENUS PER1CROCOTUS. 



183 



Males. 



Locality. 



Number of Pri. % 



mariestvantingtJie * .;„„ 



coloured patch. °J wlH 9- 



Anjango 



Larkhana ('Sindh) 

 Eta wall 

 Atteran R. 

 Amherst 



Andamans (moult- 

 ing uncertain)... 

 Port Blair 

 Sirsa 



5 

 5 



5 



4 

 6* 



4 

 5 

 5 



28 

 2 6 

 26 

 27 

 27 



2-82 

 26 



Locality, 



Number of Pri- _ , 

 maries wanting the -*'*".?'* 

 coloured patch. °J m "? 



2'73 



26 



2-6 



28 



29 



27 



273 



2-8 



I copy the above in detail, because the curious result appears 

 that setting aside the Andaman birds, in which the number of 

 plain primaries is indifferently 4 or 5, all the Indian birds but 

 5 (marked with an asterisk) have five plain primaries, and all 

 the Burmese birds but 2 (also similarly marked) have four. 

 And out of the five Indian exceptions, one is doubtful, and in 

 another I find I had noted that there was a trace ouly of a patch 

 on the fifth. I much regret that at the time I did not go 

 through the whole series; but I have only discovered this curious 

 coincidence in writing out my notes now. 



As to distribution I have the species from almost every part 

 of India, Burmah (including in this Pegu, Arracan, and Tenas- 

 serim) Ceylon and the Andamans, but it is not known to occur 

 in the Nicobars, and is not found, to the best of my knowledge, 

 anywhere in the North- West Punjab, Trans-Jhilum in fact, and 

 it neither ascends the Neilgherries nor the Himalayas. Mr. 

 Sharpe and others quote it from Nepal, but it does not I believe 

 occur there. As far as I can make out the Nepal habitat rests 

 upon Mr. Gray's erroneous identification of Hodgson's plates 



