478 Birge—Vertical Distribution oj Pelagic Crustacea. 
pected to confirm this result, at least partially, by finding a 
migration within a limit of several meters. At the same time 
the senior author of this paper had so often found Crustacea at 
the surface by day that he doubted whether the law 
was as simple for lake Mendota as that found in lake Balaton. 
Until our observations forced the conclusion upon us, we had no 
idea that the Crustacea would remain at the same level by day 
and by night. Our work was incited by this paper of France’s, 
and had at first two aims: (1) Assuming France’s law to hold 
for lake Mendota, we wished to determine how deep the Crus¬ 
tacea go by day. While France found that the Crustacea go to 
the bottom of lake Balaton, it was quite possible that in lake 
Mendota, more than twice as deep, they would go only part 
way. (2) We wished to determine the rate of ascent and de¬ 
scent. These original purposes were, of course, interfered with 
by the result of our observations, which were negative as re¬ 
gards any diurnal movement of the Crustacea. 
It does not seem to us at all possible that any vertical move¬ 
ment of numbers of the Crustacea can have taken place to an ex¬ 
tent of 3 m. It is plain that there is no general movement of 
any species to the surface at any time. There was never an 
aggregation of all the Crustacea in the 0-3 m. level. There was 
very rarely such a distribution as to make the number in the 
3-6 m. level larger than that in the 0-3 m. level. In Diaptomus 
this occurred six times, four by day and two by night. In the 
greatest difference the 3-6 m. level contained 59 per cent, of the 
catch, and the 0-3 m. level 34 per cent. In the same species 
the average percentage of the sum of the 0-3 and 3-6 m. levels 
was about 82 per cent. There were 17 cases where this sum 
passed 90 per cent., of which fifteen were by day and two by 
night. There, were eleven cases where the sum fell below 70 
per cent., of which eight were at night and three by day. The 
3-6 m. level was by far the most constant in percentage. While 
naturally varying a good deal, the averages were remarkably 
constant, as the following table will show, which gives for 
each of the four periods the average percentage which this level 
contained, of the whole number of the species of Crustacea 
named. 
