233 



page 228) exist also in the case of Ceriodaphnia and in much the 

 same form. 



The relation of temperature to Ceriodaphnia is evident in its 

 seasonal distribution. It does not advance rapidly in its vernal 

 increase until after the water warms to 70, and drops suddenly in 

 numbers when the autumnal decline passes this point. Moreover, 

 seasonal variations in temperature are accompanied by correspond- 

 ing shiftings of the pulses of Ceriodaphnia. Thus in 1898 the water 

 did not reach 70 until about May 20, reaching 73 on May 24, and 

 the vernal pulse of Ceriodaphnia began at once its rise to the maxi- 

 mum of June 7. In 1896 spring was early, 72 being recorded in 

 surface waters on April 24, and we find a vernal pulse rising to a 

 maximum on May 8. So also in 1897, when high temperatures 

 continued into the autumn, the decline passing 71 on October 5, 

 instead of in the first half of September as in other years, we find 

 the pulses of Ceriodaphnia extending into October with unusual 

 amplitude, reaching 5,200 per m. 3 October 5, while the highest 

 record in this month, or later, in other years was 280 per m. 3 Tem- 

 perature rather than season is thus the dominant factor in the 

 seasonal curve of occurrence of Ceriodaphnia. 



The form of this seasonal curve is typically that of a series of 

 recurrent pulses of varying magnitude tending to reach the maxi- 

 mum height in the vernal pulse of May- June, attaining often lower 

 levels in July and rising again in August-September, and falling to 

 a minimum, or even to disappearance, in October. These later 

 pulses do not appear in the disturbed hydrographic conditions of 

 1898 (Table I.), but are clearly delineated in the summer records 

 of other years, especially in the stable conditions of 1897, where 

 well-defined pulses appear in July, August, September, and October, 

 at intervals of approximately four weeks, culminating July 14, 

 August 10, September 14, and October 5. Their maxima attain 

 respectively 5,600, 2,720, 6,000, and 5,200 perm. 3 , and the pulses are 

 delimited in each case by minima of less than 500 per m. 3 They 

 tend to coincide with those of other Entomostraca and to approach 

 those of the Rotifer a. 



The Ceriodaphnia population in channel waters is almost ex- 

 clusively made up of parthenogenetic females. Males were not 

 recorded at any time, though females with ephippial eggs appeared 

 after the October pulse of 1897 and the vernal one of 1898. 



