SARCIDIORNIS MELANOTA 39 



stomachs of the birds he examined, others, besides Tickell, have 

 found that cultivated rice forms one of the articles of their diet. 

 They eat all sorts of shoots, roots, seeds, etc., of water-plants, 

 varying this vegetarian food with a little animal stuff now and then, 

 such as worms, spawn, larvse, and perhaps an occasional fish. 



The voice of the Nukhta is, according to Legge, " a low, guttural, 

 quack-like sound, between the voice of a duck and a goose." The 

 few I have heard uttered loud cries, which seemed to me far more 

 like the notes of a goose than of a duck. A pair, whose nest I after- 

 wards found, used to herald my approach to their particular piece of 

 water with loud trumpet-calls, uttered by them, when they first saw 

 me, from their perches high up in the tree. They roost, I believe, 

 always in trees, and not in the water or on the ground, and they are 

 not nocturnal, or even crepuscular, birds in their habits, as are most 

 of their order. 



The African form alluded to by Hume as S. africanus is not 

 specifically distinct from our Indian S. melanota, though it averages 

 a little smaller — the wing being about thirteen or fourteen inches 

 in the male. 



Hume also refers to Sclater's plate of Sarcidiornis, and, referring 

 to the under tail-coverts therein depicted, says that in all the Indian 

 specimens he has seen the tail-coverts are always white. As a matter 

 of fact, although the under tail-coverts in the plate should have been 

 white and not yellow, the bird shown in the plate is not our Nukhta 

 at all, but S. cariuwulata, a much smaller species, found in Brazil, 

 Paraguay, and North Argentina. 



This and other ducks belonging to this subfamily are amongst 

 those requiring a close-time, as all of them are residents or mere 

 local migrants. This close-time might extend from the 1st June to 

 the 1st December. Tickell says that by October most of the young 

 are on the wing, but in some parts of India this is at least a month 

 too early ; and I do not think that the 1st December is too late a 

 date for commencing their slaughter. 



