CHROMOSOMES IN THE SPERMATOGENESIS OP THE HEMIPTERA. HETEROPTERA. 151 



chromosome. That is to say, there is marked affinity or attraction only between the 

 elements of such a i)air, an attraction exhibited by tlie conjugation process. There is 

 then something correspondent between the elements of a pair, not shared by them 

 with the elements of any other pair, and this can be only a functional peculiarity, one 

 based perhaps upon different metabolic energies. Therefore, as Sutton (1903) has 

 reasoned, a chromosome must be the seat of particular qualities of the individual, not 

 the center of the sum total of the individual's activities. Different chromosomes, that 

 is to say, must have different physiological energies, and the sum of them, that is the 

 whole nuclear element, present the energies of the individual. 



Thus the experimental studies and the morphological ones are in accord in this 

 matter, as Boveri (1904) has shown, and more recentl}' Heider (1906). And these 

 constant size and formal differences enable us to analyze the cell constituents much 

 more fully than we could do a few years ago. 



Another result I would mention here. When I first discovered the constancy of 

 such chromosome pairs, I concluded that the two components of each pair were exactly 

 equal in form and volume, and so have the others who followed me. In the pres- 

 ent paper I have given especial attention to this point, and now find good evidence 

 that the components of each pair are probably constantly slightly different from each 

 other in volume. Tliis is a difficult point to make sure of because it is hard to estimate 

 voluminal mass in such small objects where there is much chance of optical illusion. 

 But in most of those cases of pairs of small diplosomes of approximately equal volume, 

 as those of the coreids, I find that they are alwa3^s slightly different in volume in the 

 first maturation mitosis, then alwaj'S different in this respect in the spermatogonium ; 

 and here one can be fairly certain of his conclusion, because these bodies are nearly 

 spherical and so relatively easy to compare. Again, in Corizus alternatus of the five 

 pairs of autosomes of the spermatogonium, the largest pair {A, a, Fig. 107) is regularly 

 composed of two relatively enormous elements, one slightly more voluminous and 

 nearly straight, the other slightly smaller and horse-shoe shaped. And in Harmostes, 

 where I have studied many spermatogonic divisions, all the autosome pairs are unusu- 

 ally distinct, and in each the two components appear constantly very slightly different 

 in volume. This is clearly the case in Ascaris also. Now in this connection let us 

 recall the discovery of IMiss Stevens (19056) and Wilson (1905a) that when there is a 

 pair of diplosomes of markedly dissimilar volume, as in Tenebrio or Euschistus, the 

 smaller must be the paternal element and the larger the maternal. If this is so for 

 these diplosomes, is it not also probable that in any chromosome pair the slightly 

 smaller element may be paternal and the larger one maternal ? There would certainly 

 seem to be a probability of this, and if it can be shown to be a constant relation it will 



