354 
BULLETIN OF THE BUREAU OF FISHERIES 
the eggs were placed on the second tray under the ice hopper and the last tray near the bottom 
part of the stack. Yesterday they were found in excellent condition. There was no noticeable dif- 
ference in the eggs on the two trays, except that on each tray the eggs that were separated showed 
a larger loss than those which were bunched, and the losses apparently larger on the bottom tray. 
We estimated that the total loss is less than 5% . Did not count the dead eggs, as it seemed best 
to close the case at once, and we don’t know the exact number in the case, as the number in bunches 
could not be counted or measured, but approximately the losses thus far have been on the entire 
lot about five per cent of the number packed. 
In another letter, dated April 26, 1909, Race further described the developments 
in the experiment. He said that the 15,000 smelt spawn packed in one of the regu- 
lar transportation egg cases, with exactly the same care with which other classes of 
eggs are packed for long shipments, plainly showed that by packing them as soon 
as taken and without trying to separate the bunches they can be transported safely 
a long distance, provided they are kept well iced in transit. The last eggs in the 
bunches died 3 days previous, having become somewhat foul or sour from being 
closely packed, and this would seem to indicate that by repacking or giving them a 
little fresh air daily, 45 or 50 days’ shipment would show good results on arrival. 
The separated eggs all died, while the bunched eggs were still in good condition. 
In April, 1912, also, several million smelt eggs were planted in Torch Lake, also 
in Michigan. On May 18, 1916, 10,000,000 were placed in the waters of St. Mary’s 
River, and on May 30, 1921, 200,000 were deposited in Sturgeon River of the same 
State, all from Green Lake, Me. 
It is quite probable that at least the earlier shipments to Michigan were made 
under the conditions mentioned above, but the time in transit could not have been 
more than a few days. The smelt eggs of the first two lots — that is, Crystal Lake 
and Torch Lake — were all of the same shipment and doubtless were eggs of the large 
smelt. The others probably were the eggs of the small smelt. 
The incubation period of the large smelt is about 30 days with the temperature 
of the water around 34° F. The incubation period of the small smelt in the same 
season is about 24 days with water temperature around 48° F. 
In Race’s experiment, evidently with eggs of the large smelt that had been 
packed on March 17, the last eggs were dead on April 23, a period of about 37 days, 
or a week over the incubation period. This would seem to suggest that by keeping 
the temperature down by means of ice, the limit endurance had been reached, and it 
is hard to see how it could be expected that under any circumstances eggs could be 
retained without dying or at least hatching for a period of five or six weeks beyond 
the incubation period. 
It would appear that long-distance transportation of smelt eggs, the incubation 
period of which is so short at a temperature so near the freezing point, could not be 
greatly prolonged by reduction of temperature. In the case of eggs of the small 
smelt it would seem that a reduction of temperature to near freezing, if it did not 
kill the eggs, would simply serve to prolong the incubation period to the same length 
of time as was effected in the case of the large smelt. 
If it seems desirable and a good policy to stock distant waters, as those of the 
far West for example, about the only practicable way to transport the eggs that 
occurs to the writer at this time would be by relays, now “taking departure” from 
