DUCKS. 145 



made occasional excursions from inside the frontiers. Their weapon was the ' bolea- 

 dores,' or bulls, of the same nature as tliose used for catching cattle and horses, and 

 whicli are now sufficiently well known for me to dispense with a description of them. 

 Tiiese 'swan-balls ' differed only in being made of wood, so that they should float on 

 the water if tlie Gauclu) missed his aim. The swans were tamer and easier to aj)- 

 proacli then, and the rider took care always to come down the wind, getting within 

 forty or fifty yards before they took the alarm. Then a desperate ])ush, if the water 

 was not too deej), would gain another ten yards, as the swans are taken at the disad- 

 vantage of being comi)elled to rise dotcn the wind. The balls are whirled, tliroun, 

 and, twisting round the wings and neck of the bird selected, render it quite helpless. 

 Nowadays it is difficult to get witiiin gunshot-range without regular stalkin;'. It 

 nests very early, July and Se])tember, liowcver, being the favorite months. The posi- 

 tion chosen is always in one of the largest and deepest swamps, the nest being placed 

 among the thickest rushes, at some distance from one of the lagoons, but connected 

 with it by a lane of clear water; for the birds always leave the nest by swimming. 

 It is built from the bottom of the swamp, sometimes through four or five feet of 

 water, above the surface of which it rises a foot or a foot and a half. The diameter 

 at the top is about two feet. The general clutch of eggs is either three or four. Thev 

 are of a smooth, glossy cream-color." 



The AnatinoB comprise the group of sub-family rank, which, with a general term, we 

 call 'ducks,' including within it tree-ducks, river-ducks, sea-ducks, and a few minor sec- 

 tions, which at present we cannot satisfactorily place elsewhere. The common char- 

 acter is the shape of the bill which is constructed upon the plan of th;it of the tame 

 duck, rather broad, more or less depressed, with thin and flat lamelhe and mostly nar- 

 row nail, but modified in many ways to conform to the requirements of the different 

 habits and the different food of the members. The sub-family is rather numerous in 

 species, and somewhat polymori)hic, for some of the forms show strong affinities 

 towards the swans, others to the spur-winged geese, others again to the mergansers. 



It will here be necessary to go a little into details in describing the peculiar bulbous 

 enlargement of the windpipe so characteristic of most ducks, since in most works of 

 a general character this feature is usually dismissed by simple mention that such an 

 enlargement occurs. In the females the windjjipe descends regularly to the lower 

 larynx, where it becomes more or less contracted. The rings coalesce into a small 

 pyramid with bony walls, from which the two bronchi depart. In no species known 

 has the female an enlargement like that of the male, with the exception of the Aus- 

 tralian Vintf/o castanea, the female of which has an arrangement similar to that of 

 the male, but smaller, as shown by Prof. Newton. The peculiar structure of the male 

 windpijie consists in a round, bony, bladdery ajipcndage, situated on the left side, just 

 above the bronchiiil tubes, forming the so-called labyrinth, or hicl/a ossea. This aj)- 

 pendage is only absent in a few sea-ducks. In the fresh-water ducks it is of a pretty 

 uniform structure, as ty]>ified by the labyrinth of the mallard. Nevertheless every spe- 

 cies presents minor differences which are constant and ]ieculiar to it. The sheldrake 

 (Tadonui) has a double labyrinth, with the enlargement on the right-hand side. In 

 most of the sea-ducks, the labyrinth is of a somewhat different .structure, it being 

 not uniformly osseous all round, but more or less angular, pierced through by numer- 

 ous openings, the so-calIe<l fenestra*, which are covered by membrane. This difference 

 lias been regarded as of systematic importance in separating river-ducks ami sea-ilucks ; 

 but the fact that the presence or absence of a lobe to the hind toe is not co-cxtensive 



VOL. IV. — 10 



