Factors Determining Vertical Distribution . 421 
regulating factor, since the amount of food in that level in au¬ 
tumn is more than sufficient to support the total crustacean 
population. 
Temperature. 
Temperature may be considered under three heads: (1) the 
rise and fall of the average temperature of the water from 
spring to late autumn, (2) the diurnal variation of temperature, 
(3) the vertical distribution of temperature. 
I have not been able to discover that the warming or cooling 
of the water in spring or fall affects directly the vertical dis¬ 
tribution of any species except Daphnia pulicaria. The move¬ 
ments of this species are undoubtedly determined by the rise or 
fall of the general temperature of the water. It is a sub-ther- 
moclinal species in plankton-poor lakes and in summer it keeps 
as near as possible to the cool water in lake Mendota. 
The diurnal variation of temperature has no noticeable direct 
effect on vertical distribution. 
The most striking fact in the vertical distribution of temper¬ 
ature is the formation in the lake during summer of the thermo- 
cline which forms the lower limit of the Crustacea from July on. 
The Crustacea follow accurately the position of the thermo- 
cline. This layer has a vertical oscillation of two or even three 
meters, being affected by the direction of the wind. In every 
case the lower limit of the Crustacea oscillates with the posi¬ 
tion of the thermocline and follows it downward as it gradually 
descends during the summer. 
The statement made in my former paper (Birge, ’95, p. 481) 
that “during July, only the upper twelve meters are tenanted by 
Crustacea, and over ninety per cent, are in the upper nine 
meters” should be modified so as to read, that ninety-five per cent, 
or more of the Crustacea are found above the thermocline, which 
in July is situated from nine to twelve meters below the sur¬ 
face. Yet, close as is this correspondence between Crustacea 
and thermocline, the temperature is not the fact which limits 
their downward extension. This will be shown under the next 
head. 
I have no doubt, however, that the thermocline is always an 
