Muttkowski—The Fauna of Lake Mendota. 441 
Long before the ice goes 'out completely, some early insect species 
become active. In places where the entry of homoiothermal 
spring water keeps an open area in the lake, insect larvae will 
begin to transform. As a matter of fact, such transformation 
(pupation) may take place even when the lake is completely 
covered with ice. For instance, Pearse (1916) captured perch 
from Lake Wingra in early March, which contained largely 
Tanytarsus dives, half of them pupae; this was at a time when 
the lake was completely covered with ice, which did not go out 
until a week or more afterward. A similar condition exists in 
ponds and other waters, and some species appear to be able to 
measure the time so closely that the period of their emergence 
and the going out of the ice seem to be synchronized to the hour. 
To other species the disappearance of the ice means the in¬ 
ception of larval life. There are not many such species in the 
lake, the Hydroptilid caddisworms being the most prominent. 
But especially among surface-breathers of the other waters 
spring means the beginning of active life; while in the lake it 
means primarily a continuation of larval life for at least the 
insect biota. 
III. FOOD RELATIONS. 
A. Reciprocal .—With the varying influence of molar and 
other agents it is not to be wondered at that the averages of 
food supply fluctuate from month to month and from year to 
year. The greatest fluctuation, however, is the seasonal one, 
and is best expressed in a table of comparative percentages. 
These percentages can refer to bulk, not to numerical repre¬ 
sentation. In table 9 I have attempted a bulk calculation for 
two seasonal periods, and an annual estimate for comparison. 
The seasons represent the two great floral periods of the lake,— 
namely the algal period of May-June and the spermatophyte 
period August-September; or, from an animal standpoint, the 
elimination period ( = transformation) and the multiplication 
period referred to in the discussion of the seasonal cycle. I have 
not included the fish in the summary of the macrofauna; too 
little is known of their numerical representation. (See p. 482). 
A comparison of the bulk percentage and the food relations 
of the various groups shows two important results: (1) The 
scavengers are the dominant forms of the lake; (2) Trichop- 
