W. E. Agar 313 



In the above table, an " unsuccessful mating " is one in which the 

 eggs were not laid, or if laid, disintegrated in the brood pouch, showing 

 that they had not been fertilized. If the eggs became rounded off and 

 took up their proper position in the ephippium without disintegrating 

 in the few days which elapses between oviposition and the casting of 

 the ephippium, it was assumed that they had been fertilized, and the 

 mating is recorded as successful though only in a few cases was this 

 confirmed by observation of the hatching of the egg. Cases where only 

 one of the two eggs was fertilized and the other disintegrated are counted 

 among the successful matings. 



In compiling the above table, all those failures have been omitted 

 in which there was good reason to believe that the male was either too 

 old or too young, or in which the female had been placed with the male 

 less than 24 hours before the ephippium was cast, since it was found 

 that if copulation were delayed till that time the eggs were rarely laid, 

 even in straight matings. The optimum time for copulation to take 

 place, and therefore for the eggs to be laid, appears to be 3 — 4 days 

 before the ephippium is ready to be cast. At this time it is still very 

 young and soft ; later it becomes rigid and possibly this may oppose a 

 mechanical obstacle to effective copulation or to oviposition. 



A glance at the above table will show that the percentage of suc- 

 cessful interspecific matings is significantly less than in the case of 

 matings between members of the same species. The difference between 

 the percentages for the inter- and intra-clonal matings in D. obtusa is 

 probably within the limits of experimental error, and therefore no 

 significance should be attached to it. 



The first series of crosses were carried out in the spring and early 

 summer of 1914, and resulted in 29 egg-containing ephippia from crosses 

 between the two species. None of these, however, had hatched by the 

 autumn of that year, when circumstances prevented the continuation of 

 the experiment. Some of these ephippia were left in a dry condition, 

 and some in water, and they were all re-examined on Dec. 31, 1918. 

 All the eggs in the dried ephippia had disappeared, but several of the 

 ephippia which had been in the water still contained apparently good 

 eggs. On June 25, 1919, one egg hatched from a mating between two 

 clones of D. pulex made on June 13,. 1914". This egg had therefore been 

 lying dormant in water for five years. So far as I know, there was no 

 change in the conditions to which one could assign the stimulus which 

 caused it suddenly to resume development after such a long dormant 



Journ. of Gen. x 22 



