22 Bulletin Wisconsin Natural History Society. [Vol. 9, Nos. 1-2. 



From this table it will be seen that the time required for de- 

 velopment to hatching is greater in the case of A. punctatum, and 

 that even in the early stages the development is slower. Hence 

 the longer time in the case of A. punctatum is not due entirely, if 

 at all, to the greater thickness of the gelatinous envelopes. 



Embryos of A. tigrinum reared at Lake Forest required for 

 hatching a period of from 18 to 20 days. 



LARVAL HISTORY. 



Larvae of A. tigrinum were reared at Ann Arbor in 1906 and 

 at Lake Forest in 1907 ; A. punctatum was reared at Ann Arbor 

 in 1906 and at Syracuse in 1909; A. jeffersonianum was reared 

 at Syracuse in 1909. In each case the larva were kept in 

 good condition for a little over a month after hatching, cir- 

 cumstances preventing my caring for them for a longer period. 

 The larvae were studied in the living conditon and series pre- 

 served for future reference. The following account deals more 

 particularly with the larvae studied at Syracuse. 



An effort was made to keep the conditions for the two species, 

 jeffersonianum and punctatum, practically alike, in order to avoid 

 possible errors of comparison from the occurrence of modifica- 

 tions due to environment, such as are recorded by Powers ('07). 

 The larvae of both species were jpd at first on entomostraca — 

 chiefly Cyclops and Daphnia.. But soon after A. punctatum 

 hatched the supply of entomostracans failed almost entirely. This 

 diet was then supplemented by scraped raw beef fed every two or 

 three clays to those larvae that would take it. In all cases the 

 larvae took no notice of the meat unless it was moved. A. jeffer- 

 sonianum readily ate the meat, and not satisfied with this, about 

 eight days after hatching commenced to eat each other ; the un- 

 equal rate of growth facilitated this cannibalistic practice. Ordi- 

 narily, one larva partly swallows another nearly as large as him- 

 self, digests a portion of the body, then regurgitates the rest. One 

 larva was found swallowed, tail first, nearly to the eyes, by 

 another less than twice his size. The result of this practice is 

 that in time usually only one larva is left in each aquarium. 



For a long time A. punctatum refused meat and apparently 

 ate nothing, except perhaps the scanty supply of entomostracans. 

 Cannibalism was not noted in A. punctatum during the five 



