160 Thirty-fourth Annual Meeting. 
water in the bags was thenceforth stirred with an oar continu- 
ously for six days. The results was ample proof that the con- 
clusion was correct. There were two lots of lobsters which re- 
ceived this treatment. Neither of these lots was considered to 
be as promising as the average. However, from one there was 
obtained 748 lobsters in the fourth stage. This is a larger 
number than were obtained in any other experiment, either at 
Wickford, Woods Holl or in any other station where lobster 
culture has been tried, so far is I am aware. From the other 
319 were obtained in the fourth stage; but as one of the bags 
was old, and had a hole in it, the figures fail to give a correct 
idea of the results of the experiment. One of the most encour- 
aging results of this method was the clean and healthy appear- 
ance of the fry in all stages. The continual stirring prevented 
the accumulation of parasites found on the body of nearly all 
of the specimens in the other lots. 
During the following seasons this experiment was follow- 
ed up with others, working upon the same theory, namely, that 
the water should be constantly stirred. To do this it was nec- 
essary to invent a mechanical device which would take the place 
of the oar and designs for such an apparatus were immediately 
made. 
The mechanical device was put into successful operation 
in the season of 1901 at Wickford. The apparatus consisted of 
a number of rotating horizontal paddles, similar to those in 
use over restaurant tables for circulating air, one in each bag, 
run by a gasoline engine. The movement of the paddle blades 
created a constant upward current of water in the bags, which 
kept the fry off the bottom and kept the food suspended in the 
water. Through its use 9,000 lobsters were raised to the fourth 
stage, and in some experiments 50 per cent of the newly hatched 
fry were carried through to this stage. That this was a decided 
step in advance of the old methods will readily be admitted by 
those who have followed the course of previous experiments. 
The apparatus now used (1905) comprises a house-boat 
between the pontoons of which are three small hatching bags 
6 x 6 x 4 feet, made of canvas; tWo side floats constructed of 
6 x 6 in. spruce beams bolted together and buoyed by barrels, each 
supporting five large canvas rearing bags about ten or eleven 
