466 E. ELEANOE CAROTHERS 



2. Description of chromosomal complexes involved in specific 



matings 



An analysis of the male progeny only has been made because 

 the evidence which they furnish is believed to be sufficient and 

 it is free from any uncertainty in identifying homologues, since 

 the figures represent first spermatocytes with the pairs synapsed. 

 Two matings, each with a single male offspring, are shown 

 on plate 1 because both of these matings involved the rather 

 rare condition of atelomitic homologues in pair number 1 . Text 

 figure B gives the form of each of the three pairs of chromosomes 

 under consideration in both parents and progeny of all of the 

 matings. 



a. Mating number 2 {pi. 1 , rows 1 and 2 & parent, row 3 9 

 parent, row Jf. cf offspring). The chromosomal complexes involved 

 ni this mating are shown in horizontal rows 1 to 4. The first two 

 rows represent, respectively, first spermatocyte and spermato- 

 gonial complexes. The latter is given chiefly to enable the 

 reader to niake a more direct comparison with the diploid 

 complex of the female (row 3). The most striking difference 

 between these two complexes appears in column 6, the presence 

 of two accessories in the female complex and one in the male. 

 Row 4 represents a first spermatocyte complex of one of their 

 offspring. 



Taking up the three critical pairs numbers 1, 7, 8, we see: 1) 

 that nmnber 1 is heteromorphic in the male and telomitic in 

 the female. If, then, the point of fiber attachment remains 

 constant from parent to offspring, the chances are equal for 

 this chromosome to be either heteromorphic or telomitic in the 

 progeny, while an atelomitic pair would be impossible. In the 

 one male offspring from this cross both homologues were telomitic. 

 2) Chromosome number 7 was atelomitic in the male and 

 heteromorphic in the female. The chances were, therefore, one 

 to one of its being either atelomitic or heteromorphic in a given 

 offspring. In the one male obtained it was heteromorphic. A 

 further point which may be mentioned here is that the telo- 

 mitic homologue, which on the assumption above mentioned is 



