J. W. H. Harrison and L. Doncaster -243 



form of a reticulum, as in the converse cross (fig. 29). There are 

 usually at this stage two chromatin-nucleoli, each of which is double, 

 with halves which are slightly unequal ; the parts of one or both may 

 become separate, so as to produce three or four distinct masses. 



The testis includes good equatorial plates of both first and second 

 spermatocyte divisions. Counts of first division figures have given 

 numbers ranging from .52 to 60 chromosomes (figs. 30, 31); in those 

 with the higher numbers it is possible that one or two dividing pairs 

 have been counted as two singles in each case. There seems no doubt 

 that the total number is not more than 60, and usually several less, and 

 as in .some figures as many as 18 have been reckoned as large, it is 

 practically certain that a considerable amount of pairing has taken 

 place, since only about eleven large ones are introduced from the 

 hirtarai parent, and the remaining large ones must be formed of pairs 

 of small ones. This is confirmed by the second spermatocyte divisions, 

 in which only about 10 large ones are counted; the small chromosomes 

 in this division are so small as to be counted with difficulty, but the 

 total of second equatorial plates is clearly not far from 50, made up of 

 about 10 large and 40 small (fig. 32). It is probable, therefore, that 

 in the cross hirtaria $ x zonaria ^ more chromosomes find mates in 

 the maturation divisions than in the converse cross, in which over 

 60 are usually present in the first spermatocyte divisions, but as a 

 lower number was found in some equatorial plates of this cross, it is 

 probable that the number which pair varies somewhat in different cells. 



Condumjns and Diaciinsioii. 



The general conclusions which may be drawn from the observations 

 described are then as follows: (1) there are 28 somatic chromosomes in 

 hirtaria, giving 13, of which one consists of a large united to a small, in 

 the spermatocytes. (2) There are 112 somatic and 56 spermatocyte 

 chromosomes in zonaria which are much smaller than most of those of 

 hirtaria. (3) The spermatogonia! number of the hybrids, whichever 

 way the cross is made, is as nearly as can be counted 70, of which 

 fourteen are conspicuously larger, and of these twelve are derived from 

 the hirtaria parent, the other two from zonaria. (4) The early synapsis 

 stage of the spireme in the young spermatocytes and oocytes of the 

 hybrids does not differ greatly from that of the pure species, but it is 

 apparently not followed bj' a normal pachytene ("bouquet") stage with 

 thick coiled thread. (5) Comparatively few of the chromosomes are 



