3l8 N. M. Stevens. 



III. THE WINTER EGGS. 



/. Early Development and Growth. 



The material for study of the winter eggs was obtained by bring- 

 ing into the laboratory rose twigs with broods of sexual females 

 on the leaves. Adult males were usually found on the same 

 leaves. These young females develop more rapidly in a warm 

 room under glass, and soon begin to lay the fertilized eggs. The 

 winged mothers of the sexual females were sectioned for the study 

 of the ovaries of the sexual female embryos, and young sexual 

 females for early stages in the development of the winter egg. 

 Fig. 19 shows an ovary from such an embryo. It is considerably 

 larger than the ovary of the ordinary parthenogenetic female 

 embryo (Fig. i), and in every case these ovaries show a large num- 

 ber of degenerating oocytes at the posterior end (Fig, 19, a). 

 These cells are more or less shriveled and stain deeply and irregu- 

 larly. Whether this is simply degeneration of a large number — 

 at least half — of the oocytes of an ordinary parthenogenetic ovary 

 in order that the remainder may have room for growth, or 

 whether the parthenogenetic ovary may contain originally both 

 eggs capable of parthenogenetic development and others capable 

 of development only after fertilization, and the former degenerate 

 in order that the latter may develop, I am unable to say. The 

 number of oocytes in the twelve parthenogenetic ovaries is cer- 

 tainly more than double the number of young ever produced by 

 an individual, so that the latter supposition might be possible. 

 Two cases observed in dissection, but never duplicated in fixed 

 material, might support either view, and certainly tend to show 

 that the ovary which produces parthenogenetic embryos and the 

 ovary which produces winter eggs are originally identical. In 

 two individuals egg-strings and winter ovaries with developing 

 eggs were found associated in both groups of ovaries. The num- 

 ber in each varied — in one individual one group contained 5 egg- 

 strings and one winter ovary, the other group 3 egg-strings and 

 3 winter ovaries; in the second individual each group contained 5 

 winter ovaries and one egg-string; on one side there was one 

 parthenogenetic ovary with eggs and embryos (Fig. 20, c and e), 

 on the other side was an egg-string consisting of two partheno- 

 genetic embryos, a winter ovary and a young winter egg (Fig. 21, 

 a, b and ey These individuals with mixed ovaries were found in 



