408 ' BOWEN. 



theoretical calculations. The discrepancy was never satisfactorily 

 explained, so far as I know. The facts here given concerning the 

 sperm heads of Euschistus suggest the possibility^ that Zelen,y and 

 Senay discarded the large sperms and that the two smaller classes 

 were mixed to some extent producing the result which could not be 

 explained. 



II. The Differentiation of the large Generations of Sperm. ^ 



As I have stated in the preceding section, the nuclear and cyto- 

 plasmic components of the large and small cells behave in a very 

 similar manner throughout the early stages of spermatogenesis and 

 up to the formation of the definitive spermatids. Even during the 

 differentiation of the sperm the cytoplasmic elements show little if 

 any distincti^■e differences in the large and small cells (see Bowen 

 '22a). The spermatid nuclei, on the other hand, in addition to 

 differences in size, exhibit others of a rather unique character. The 

 discovery of these differences early in my study of hemipteran testes, 

 led me to think that the differentiation of the large sperm heads 

 differed fundamentally from tlaat of the smaller ones. However, 

 intensive study of the "normal" spermiogenesis brought to light 

 certain features which had been previously overlooked by workers 

 in this field, and finally provided the data for harmonizing the appar- 

 ent differences. The formation of the normal sperms in Miirgantia 

 has been described in detail in the second of these "Studies," and it is 

 my purpose here to point out especially those features wherein the 

 development of the large sperms is different. The period of spermio- 

 genesis will be divided into a series of stages the exact limits of which 

 are given in another place (Bowen '22a). 



The description is taken from Miirgantia histrionica , because in this 

 form tlie relati\'ely small size of the large sperms makes possilile the 

 obtaining of complete (or nearly so) heads in sections, — a difficult 

 matter when the heads become very long, as in Euschistus for example. 

 I have not examined the situation in other genera very thoroughly, 

 but so far as my observations go, the essential features seem every- 

 where to accompany the formation of large sperms. In one respect 

 the large spermatids are less satisfactory for study because a complete 



8 In comparing the plates from my study on the normal sperms of Murgantia 

 (Bowen '22a) with those accompanying this paper, it should be noted that the 

 figure magnifications are somewhat different. In the first named paper the 

 magnification of the plates is 3000, in this paper, 2700. 



