Hermaphroditism among the Apodidie. 307 



probable, therefore, that the j^enital glands of developing 

 young, instead of producing exclusively eggs, might produce 

 exclusively sperm, and that the needed males might be more 

 quickly obtained in this way. As far as I can see, there is 

 only one difficulty in believing that this takes place. The male 

 and female glands have, according to Kozubowski, the same 

 shape, and we know from the hermaphrodite specimens above 

 described that the epithelium may produce sperm or eggs. 

 The difficulty, however, lies in the early modification o£ the 

 eleventh pair of legs into egg-pouches. This occurs apparently 

 much earlier in Ajjus cancriformis than in Lepldurus pro- 

 ductus. If this modification has already set in, when the 

 direction of development is changed from parthenogenetic 

 female into male — if, indeed, this change ever takes place — ■ 

 then we should have to assume that these legs, during subse- 

 quent ecdyses, revert to the normal type. 



There is one feature in my hermaphrodite specimens of 

 Lepidurus productus and Apus cancriformis which appears to 

 me of significance in this connexion. The unripe eggs in both 

 specimens were, as stated, in process of degeneration. Those 

 which were quite ripe, i. e. were filled with well-developed 

 yolk-granules, beyond being somewhat small, seemed to be 

 unaffected and to be escaping through the gland towards the 

 egg-pouch. Nearly all the rest, i. e. the unripe eggs, showed 

 signs of shrinking, as if they were being resorbed, the material 

 being probably needed for nourishment. In one or two cases 

 it appears as if there had been a struggle between the yolk- 

 forming tendency of the egg-cells and the rest of the animal 

 organization seeking to resorb the egg-protoplasm. The 

 nuclei were embedded in small compact masses of yolk-disks, 

 while the rest of the staining protoplasm of the egg-cells was 

 full of vacuoles and irregular spaces from which the material 

 had been drawn. It seems obvious that in these two herma- 

 phrodites an egg-producing stage was coming to an end. 



On the other hand, indications are not wanting that a 

 sperm-producing stage is in process of development. The 

 following can be gathered from the figures. The sperm- 

 masses have been in most cases but very slightly drawn upon ; 

 fresh pockets completely filled with sperm-cells appear to be 

 forming (figs. 5 a and 8) ; and, lastly, at the tips of certaiu 

 branches where eggs are degenerating, the cells just below 

 such eggs appear to be actively producing sperm-cells (figs. 9 

 and 10) . The appearances therefore all tend to suggest that 

 a sperm-forming phase was about to succeed an egg-producing 

 phase. It looks, therefore, as if the hermaphrodite specimens 

 were parthenogenetic females in process of becoming males ! 



