INDIAN DUCKS AND THEIR ALLIES. 17 



In anothei' letter he remarks on the curious foot that though the normal 

 number of eggs laid is about 10, yet one never sees a family party con- 

 taining more than six or seven young ones, so that the percentage of 

 addled eggs or of accidents to the young after birth must be very great. 



Mr. Doig found on one occasion that otters had been responsible for 

 the destruction of a nest of eggs. He found a nest at Narra in Sind 

 on the 1st of May which had contained 10 incubated eggs, but these, 

 with the exception of one, were all scattered about and broken. Before 

 reaching the island on which the nest was placed he had noticed a family 

 of otters playing about, which all bolted at his approach, and which 

 were doubtless the culprits concerned in the pillage of the nesi. 



The greater number of nests are placed on the ground, well concealed 

 in rushes and grass, often at the edge of some piece of water or stream, 

 frequently on islands, and not seldom in patches of grass well away from 

 water. The ridges between rice-fields seem to be favourite places for 

 them to make their nests upon, the proximity of the food-sup})ly 

 doubtless being an incentive to the birds to make use of such spots. 



Hume thus describes the first nest taken by him. It *' was [)laoed on 

 a drooping braiKdi of a tree which hung down from the canal bank into 

 a thick clump of rushes growing in a jheol that, near the bridge, fringes 

 the canal. The nest was about 9 inches aljove the surface of the 

 water, was entirely concealed in the rushes, and was firmly based on a 

 liorizonttd bifurcation of the bough. It was composed of dry rush, 

 had a good deep hollow, in which down, feathers, and fine grass were 

 intermingled. The nest was at least a foot in diameter, perhaps more, and 

 I suppose 2 inches thick in the centre and 4 inches at the sides. It 

 contained three fresh egos." 



The number of eggs laid seeins to vary considerably ; but from about 

 8 to 10 may be taken as the normal number lai(l, often less, but not 

 often more, though they may occasionally number as many as 14. 



They are much like the eggs of the Mallard in appearance, though 

 rather broader on an average as well as a little shorter. Hume's dimen- 

 sions for the eggs of this duck are, length from 2'08'' to 2*3", breadth 

 l-eS" to 1-18", and the average of fifteen 2-15" X I'TO". 



The eggs in my collection are of two rather distinct iy\)Q9,, the one a 

 broad regular oval, the other a narrower egg with one end very much 

 smaller than the other and distinctly pointed. The texture is the 



