328 GEOL. AND NAT. HIST. SURVEY OF MINNESOTA. 



ond pair of legs. In stage B the post abdomen has developed the per- 

 manent number of setoe, but the ratio of the length of the terminal 

 seta to the length of the body of the member is much greater than in 

 the adult. In passing through stages C and D, the parts of the post- 

 abdomen gradually assume the proportions of the adult. 



Every ]3recaution has been taken to be sure that all the stages 

 above described were stages of one and the same animal. A definite 

 number of each stage was isolated in saucers of water and a record 

 kept of the number placed in each saucer, and also of the stage of 

 growth exhibited by each set at time of isolation. The water used 

 was collected from a pool which did not contain any Ostracodes which 

 in the least resembled those under consideration. As a further pre- 

 caution the water was allowed to stand in the laboratory a couple of 

 weeks. The loss in volume was replaced by river water, taken from 

 the city hydrant. I never have found any Ostracoda in said river. 

 Thus there was no possibility of the water containing the eggs of 

 CypHs herricM. The larval stages which were placed in this water 

 were examined from time to time. In every case the specimens devel 

 oped into the adult form. 



The morphological differences between the shell of stage B and the 

 shell of stage A, and between the shell of stage B and the adult, are 

 certainly as great as the morphological differences between the shells 

 of closely allied species. The morphological differences between 

 stage A and the adult are differences, not only in the shell structure, 

 but also in the structure of the post abdomen. And these differences 

 are as great as those between the genus Ci/pris and the genus Gijpri- 

 dopsis. These facts show that shell structure of Ostracodes, when taken 

 alone, is of almost no taxonomic value. 



These facts also have phylogenetic significance. These various 

 larval stages are resting stages in the development of Ci/prls herricki. 

 Since it is true that the ontogenetic development of an individual is a 

 rapid and compact repetition of its phylogenetic history, these larval 

 forms must represent past stages in the evolution of Cypris herricM. 

 As has been stated above, stage A corresponds very closely to the 

 genus Cypridopsis. The main distinction between the genus Cypris and. 

 the genus Cypridopsis is the difference in the form of the post-abdo- 

 men. In the genus Cypridopsis the body of the post-abdomen is but 

 slightly developed, while the setse are quite long. In stage A, in the 

 larval history of Cypris herricki the post-abdomen is in this rudiment- 

 ary condition; thus it corresponds very closely to the genus Cypridop- 

 sis. Xot only that, but the unbanded forms of this stage correspond 

 very closely to Cypridopsis hystrix Herrick.* Indeed, when this stage 



*C. L. Herrick. Contribution to the Fauna of the Gulf of Mexico and the South. 18S7. P. 30, PI. 

 V, Fig.6. 



