590 ROBERT H. BOWEN 



sufficiently clear to permit comparisons with other insects and 

 it did not seem worth while to pursue the matter further at 

 present. This paper will deal, therefore, only with the mito- 

 chondria and (lolgi apparatus (' acroblasts '), and with them 

 particularly in the older spermatids. 



I must confess that when I first saw Gatenby's figures they 

 aroused considerable scepticism. However, having now 

 examined the material for myself, I find that we are actually 

 in close agreement as to general appearances (so far as Gatenby's 

 account extends), but that my interpretation differs materially 

 from his in regard to several important points. 



Material and Methods. 



For material I have made use particularly of moths belonging 

 to the family Saturniidae, the cocoons of which were 

 readily collected during the winter and early spring in various 

 localities adjacent to New York City. Of these moths this 

 paper deals almost exclusively with Callosamia pro- 

 methea. The cells in this form are rather small, but this 

 disadvantage is offset to some extent by the large size of the 

 testes. Callosamia (and other satm-nids) has been studied 

 by Cook (1910), to whose account reference may be made for 

 the structural features of the testes, &c.^ For purposes of 

 comparison a study was also made of Pygaera bucephala, 

 the form upon which the original work of Platner (1889) and 

 the later classical studies of Meves (1900 and 1903) were based. 

 Material was obtained from ]\Tr. Ij. A\'. Newman of Bexlev, 

 Kent, Pygaera not being native to the United States. 



Various methods of fixation were tried, but I found, in agree- 

 ment with Gatenby, that the best results were obtained AAith 

 Champy or Flemming without acetic acid (diluted in both 

 cases with water). Gatenbv is undoubtedlv correct in his 

 insistence on the necessity of eliminating acetic acid in the 

 case of lepidopteran material, for even in its absence the 

 mitochondria in the spermatocj^tes are very difficult to fix 



^ See also Dederer (1907) on the closely related foriu, Philosamia 

 c y n t h i a . 



