LIFE HIHTOKIES OF NORTH AjMKRICAX WllJt lOWJ.. 41 



should disappear entirely and be completely replaced with an entirely 

 different pliimaj^e, which only an export can tell from that of the 

 somber, mottled female: but such is the case; the wings and the 

 larger scapulars, which are molted only once, are all that remain to 

 distinguish the male. T have seen males molting into the e(rlipse 

 plumage as early as May 10, but usually the molt does not begin 

 until the latter j^art of that month. I have seen drakes in full eclipse 

 plumage as early as July 20, but usually it is not comj)lete until 

 August. It is worn for about a month, the earliest birds beginning 

 to raolt out of it in August. Some birds regain their full plumage 

 in October, but some not until November or even later. Mr. John 

 G. Millais (1902), one of the greatest living authorities on ducks, has 

 made a vei'v thorough and exhaustive study of this subject and has 

 written a jiarticularly full and detailed account of the plumage changes 

 of the mallard. Altliough we ma}^ not vrholl}' agree with all of his 

 interesting conclusions, regarding color changes without molt and 

 control of the molt, we must accept f]\or,\ as probably correct until 

 they are proven erroneous. 



The tendency of several s])ecies of ducks to hybridize is v.'ell known 

 and many interesting hybrids have been described. The mallard 

 seems to be more inclined to hybridism than any othei' species, par- 

 ticularly with its near relative, the black duck. Numerous s])ccimens 

 of hybrids betv/een these two species have been collected, showing 

 various grades of mixed blood: they freely interbreed in captivity 

 and their offsi)ring are ])erfectly fei'tile. Specimens have been de- 

 scribed showing first crosses of mallard ])lood with the muscovy duck, 

 the green-winged teal, the baldpate, and the pintail. In connection 

 with plumages it may be worth mentioning that many sportsmen 

 throughout the West recognize two varieties of mallards, the yellow- 

 legged variety, which is the earlier migrant in the sj)ring and the 

 later in the fall, and the red-legged variety, which is more of a warm- 

 weather bird ; the former is supposed to breed farther north and to 

 frequent the prairies exclusively whereas the latter is more often found 

 in the timbered swamps and streams. Pro])ably the differences in the 

 two varieties are due to age rather than geographical variation. 



Food. — Mallards are essentially frcsh-^ater ducks and find their 

 principal feeding grounds in the sloughs, ponds, lakes, streams, and 

 swamps of the interior, where their food is picked u|) on or above the 

 surface or obtained by partial immersion in shallow water. In Alaska 

 and on the Pacific coast they^ feed largely on dead salmon and salmon 

 eggs, which they obtain in the pools in the rivers. On or near their 

 breeding grounds in the prairie regions they feed largely on v/heat, 

 barley, and corn which they glean from the stubble fields. On their 

 migrations in the central valleys the}' frequent the timbered ponds, 

 everglades, and wooded swamps, alighting among the trees to feed 



