230 
BULLETIN OF THE BUREAU OF FISHERIES 
Goode apparently had no definite information concerning the actual spawning 
season, for he further remarks: 
It is to be regretted that no one has made careful observations upon the beginning and close 
of the breeding season of this species at different points along the coast, but the spawn appears 
to be deposited, generally, late in the winter and early in the spring. 
In general it may be said that for breeding purposes our smelt ascends fresh- 
water streams variably from late in March until early in May, and farther north the 
season may extend into June. The streams ascended may be of any size, from the 
smallest rivulet to a great river, although in the latter it is probable that they 
divert themselves into smaller tributaries for spawning. This is known to be the 
case in a few instances at least. 
Samuels (1904) says: 
Along the Massachusetts coast and thence north and east the fish enter the brooks and small 
streams in immense numbers. The spawning season varies with the degree of latitude, beginning 
almost with the breaking up of the ice in very early spring in Massachusetts, and becoming later 
in Maine and Nova Scotia, the fish entering the streams that empty into Margarets Bay, N. S., 
early in May, while in the rivers which flow into the Bay Chaleur quite late in that month, incred- 
ible numbers ascend the Jacquet and other rivers as late as May 20 or 25. 
It is probable that all along the coast the first runs of smelts follow the clearing 
of the streams of ice and abatement of freshets — in other words, as soon as the 
streams become suitable — the successively late appearance of smelts being correlated 
with the successively later breaking up of the ice northward. The streams are not 
ascended until clear of ice and turbid or snow water, but the fish appear to congre- 
gate near the outlet of the streams in salt water or tidal creeks. 
The local seasonal variations that affect the time of their ascent may differ in 
different years. Meek (1916, p. 140) said of the European smelt: “It has been 
observed that the migration takes place at night at the surface, the shoal retiring 
from the surface during the day.” 
In the case of the American smelt there appear to be no published observations 
on this fact. It is known that the smelt runs take place at night and always during 
ebb tide, according to the present writer’s observations. Ehrenbaum (1894) says 
that the smelt of the Elbe presents itself in tne spring in very large numbers in the 
brackish and salt water to spawn in fresh water. If the British smelt spawns in 
salt or brackish water, it is contary to all known habits of the fish on this side of 
the ocean; and it would seem that the smelt of northern Europe spawns in fresh 
water. 
Atkins stated in his notes that on May 25, 1877, he collected 53 eggs that he found 
attached to stones, sticks, and weeds, of which 22 were imuregnated and 12 were white. 
He said that most of these were attached to green seaweed, and none of them were 
found above high-water mark, where the water must have been pretty salt when the 
tide was in, yet the eggs developed. He also observed that those low down among 
the seaweed were twice as large as those taken near high-water mark. He stated 
that the eggs were in various stages of growth, from early appearance of trunk of the 
embryo to black eyes. Where the eggs were found were sundry marine organisms, 
such as amphipods, isopods, etc. 
