394 
Birge—The Crustacea of the PUmkton. 
very large percentage of the population is found in the upper 
level of the lake and even in the upper meter. This was the 
case during the summer of 1895, when this species was the 
dominant member of the limnetic Crustacea throughout the en¬ 
tire summer. Under these circumstances its vertical distribu¬ 
tion approximates very closely to that of Diaptomus. On the 
other hand, if the species is declining and the young appear in 
small numbers, there is a much larger proportion of the species 
in the lower levels of the lake. This was the case in 1896. In 
August of that year the numbers of Daphnia rapidly declined, 
so that in the latter part of the month there were present less 
than half as many as in the latter part of July, and in connec¬ 
tion with this decline the population of the three upper levels 
was nearly equal. In this year the vertical distribution of 
Daphnia hyalina approximated very closely to that of Cyclops. 
The vertical distribution of D. hyalina illustrates very strik¬ 
ingly the dependence of distribution on specific habit rather 
than on number. 
The illustration given in my former paper (Birge, ’95, plate 
VIII) fairly illustrates the characteristic differences in the sum¬ 
mer distribution of the different genera, and the percentage 
diagram, Fig. 29, given herewith indicates the difference in dis¬ 
tribution during the summer of 1896. 
Diaptomus Oregonensis Lillj. 
Figure 29—Table D, Appendix. 
In general Diaptomus is more abundant in the upper strata 
of the lake than in the lower at all seasons of the year. There 
is rarely less than 70 per cent, of the species in the upper half 
of the lake even in the winter, and the only times when the 
average distribution approaches equality are in late fall and at 
the period of the minimum numbers of the species in the latter 
part of April, or early in May. The other extreme of distribu¬ 
tion is reached when the new broods appear and as their appear¬ 
ance is somewhat irregular the distribution is correspondingly 
variable. The maximum average number in the 0-3 m. level 
was reached in the latter half of May, 1895, where the average 
