416 BASHFORD DEAN. 
PAGE 
I, The breeding habits of Amia . ‘ . 416 
II. The early development 3 ; : . 421 
A. Egg and egg membranes : : : . 421 
B. Rate of development. : : . 423 
C. Segmentation (first to eighth dheseeny. : . 424 
D. Blastula . : 429 
EK. Gastrula: its relations 46 WiSpidostens hubeeane said 
Teleost . ; : ; . 429 
F. Mode of formation of the amber : 436 
G. Origin of the germ layers and early embryonic Gees 437 
III, Conclusions . : : ‘ : : . 440 
I. Tar Breepinc Hapsits oF AmIa. 
The account of the spawning habits of Amia, as recently 
given by Fiilleborn (op. cit.), has been entirely confirmed by 
the present writer. A number of additional notes! are pre- 
sented in the following pages. 
In the beginning of the spring Amia makes its way from 
the deeper water, where it has remained sluggish during the 
winter season, to the shallows in the vicinity of its spawning 
ground. This is usually in the swampy end of a lake, where 
the water is well filled with reeds, stumps, Chara, Potamogeton 
rootlets, here and there broken by little clear channels or 
inlets several feet in depth. In this neighbourhood the fishes 
are early seen, often in numbers, sunning themselves near the 
surface. They are at this period active, and may not be closely 
approached. Like the gar-pike, they are then usually in com- 
panies, the fish well separated. 
The spawning season is an early and somewhat extended 
one, and is apparently induced by the first warm days of spring. 
The time of spawning in 1894, as recorded by Fiulleborn, ex- 
tended from the beginning of May to the first days of June. 
In the spring of 1895, however, although the previous winter 
had been unusually severe, a few warm days in April appear to 
have been the cause of earlier spawning. As with the gar-pike 
1 These include observations made by the present writer at Pewaukee, 
Black Lake, and in the South Carolina rivers. 
