400 BOWEN. 



times that of corresponding cells in the smallest generation. This 

 enormous difference is of special value in determining the size relations 

 of various cellular components and I have constantly taken advantage 

 of this unusual opportunity for critical comparisons. I should further 

 state that there is in Arvcliiis very good evidence of a fourth genera- 

 tion of cells occupying lobe 2, which are slightly larger than those in 

 lobes 1 and 6. My insufficient material does not, however, allow a 

 definite statement on this point. It thus seems not improbable that 

 a pentatomid genus may be found in which each of the lobes would 

 differ from all the others in respect to the relative size of its cells. In 

 line with these comments which refer particularly to the large genera- 

 tions, it may be added that rarely the smallest generations of cells are 

 distinguished by their unusual size relations. Thus in Loxa florida 

 the smallest cells are strikingly smaller than those of the normal and 

 large size. 



(a) The effects of ■polymegaly on various cellular constituents. — It 

 will be convenient here, before going on to an examination of the 

 size differences in the sperms, to consider in detail the possible differ- 

 ences in the spermatocytes and their maturation divisions which might 

 be dependent upon the size relationships of the cells in different lobes. 



As Montgomery ('98) first showed, all the cells in all the lobes seem 

 to be in every way identical and normal up to the period of synapsis 

 which directly precedes the so-called growth period of the primary 

 spermatocytes. The one marked difference is in the number of cells 

 which are in any one of these early stages. In the large-celled lobes 

 the cells are much less numerous than in the lobes with smaller cells 

 due apparently to the lack of room, the available space being much 

 curtailed by the larger size of the cells in later stages and by the added 

 fact that the large-celled lobes are often of markedly smaller volume, 

 as noted above. In the growth period the size differences cj^uickly 

 become apparent and the various generations of sizes become more and 

 more clearly marked, every ^ cell being equally involved at any partic- 

 ular stage. The differential factor of growth seems to be one of quan- 

 tity only, rather than quality. It will be interesting to inquire into the 

 behavior of the various cell elements in large and small sperm cells. 



6 I have not met with a single case of failure to develop the polymegalous 

 inequalities customary for a given genus, and only one case (a specimen, No. 

 94, of Euschistus) has been found in which all the cells were not affected to the 

 normal degree. In this particular case (which was also abnormal in some other 

 respects) scattered cysts occurred among the large generations (in both lobes 

 4 and 6) in which the cells were all normal in size. Cysts of this kind were 

 found in primary spermatocyte, and spermatid stages. There was no clue to 

 their method of origin. 



