180 PROCEEDINGS OF THE ACADEMY OF [Feb., 



The second maturation mitosis (figs. 64, 65) is accordingly an equa- 

 tional division. I have been unable to count the chromosomes in 

 the resultant daughter cells, spermatids, because of their massed ar- 

 rangement there, and therefore have not demonstrated that all the 

 elements become halved in this second mitosis. But all these thirteen 

 chromosomes are constricted or cleft, showing that each is therefore 

 probably longitudinally split; for this reason it is probable that each 

 spermatid receives thirteen chromosomes. 



As was the case in Syrhula so also in Lycosa the mode of division of 

 the bivalent heterochromosomes was not positively determined. In 

 the spermatocytes it can always be distinguished by its smooth outline 

 and compact structure only up to the tune when the other chromo- 

 somes acquire their final shape. Yet among the chromosomes of the 

 first mitosis there seems to be no particular one markedly different from 

 the others. But on lateral view of the spindle (fig. 56) there is some- 

 times one (0 quite different from the others, in the form of two elon- 

 gated rods; and its division is shown in fig. 58 (0- This may be the 

 heterochromosome, but there is no satisfactory evidence for this con- 

 jecture. We found that in the synapsis this was formed, like the ordi- 

 nary chromosome, by a conjugation of two univalent ones, and that 

 each univalent one underwent a longitudinal splitting. This similarity 

 in formation is some evidence that the heterochromosome may behave 

 like the others during the maturation mitoses, namely, that it may 

 undergo a reductional division in the first and an equational in the 

 second mitosis. And wc can say positively that the whole bivalent 

 heterochromosome does not pass undivided into one of the second 

 spermatocytes. 



In both maturation divisions the centrosomes of both spindle poles 

 touch the cell membrane (figs. 55-59, 64). As in the other objects 

 studied by me there is no intermediate cell plate formed after the reduc- 

 tion division, but after all other divisions. 



3. — Occurrence and Time of the Reduction Mitosis. 



Korschelt (1903), in his excellent review upon the maturation phe- 

 nomena, distinguishes two types of maturation: the ''eumitotic," 

 where both mitoses are equational, and the "pseudomitotic," where 

 one of them is reductional. But these should be considered collective 

 terms for groups of divergent opinions, rather than a classification of 

 actually occurring natural phenomena. The general consensus of 

 opinion at the present time, the greater part of all the more recent work 

 on most diverse animals, is conclusive for the decision that a rediic- 



