J. W. H. Harrison and L. Doncaster 247 



coupling has frequently been described in connexion with hetero- 

 chromosonies or " sex-chroniosouies " ; the size of the chromatin- 

 nucleolus also indicates that it would give rise to one of the larger 

 chromosomes. Now all the larger chromosomes of hirtana are of 

 greater size than any of zonnria, so that if the size of the sex-chromo- 

 some is any indication of its intensity of action, it is probable that in 

 the cross zonariu % x hirtana ^f , in which all the spermatozoa have 

 the large male-determining chromosome of hirtaria, all the zygotes 

 might receive a sufficient quantity of the male-determining substance 

 to cause them to become male, whether they received any from the 

 female parent or not. In the converse cross, only half the eggs would 

 receive a male sex-determiner from the hirtaria J, and therefore both 

 males and females would be produced. 



This suggestion does not differ essentially from that used by 

 Goldschmidt to explain the results obtained with Li/mantria dispar 

 and L. japonica, the essence of which is that the "potency" of the sex- 

 determiner may be different in different species. It is founded, however, 

 on observati<ms on the size of the chromosomes rather than on the 

 simple results of breeding experiments. It is admittedly very hypo- 

 thetical, and does not explain the excess of females which seems 

 constantly to occur in the cross hirtaria J x zonaria ^ . I hope to be 

 able next season not only to amplify my work on the species used this 

 year, but also to examine other crosses which give comparable results, 

 and to confirm or disprove the hypothesis suggested by observations on 

 ampler and more varied material. 



EXPLANATION OF FIGURES. 



PLATE XVIII. 



All the figures were drawn free-hand, with a Zeiss 3 mm. 1-40 ap. immersion objective 

 and Compens. oc. 12. 



[N.B. In Figs. 19, 22, 23, 26, 27, 30, 31, 32 the reproduction has made some small 

 chromosomes appear in contact which in the original were near together but quite 

 distinctly separate.] 

 Figs. 1 — 14. 7j. hirtaria. 



Figs. 1, 2, B. hirtaria oogonial equatorial plates. 28 chromosomes in each, if the 

 small body at the top of Fig. 1 is not a chromosome. There are in each four 

 very small and two rather small chromosomes. 

 Fig. 3. Oocyte nucleus, cut in two sections. Three chromosomes are cut in both 

 sections (lettered a, b, c in each). Thirteen elongated chromosomes showing 

 traces of doubleness, and compound chromatin-nucleolus (chji.). 

 Fig. 4. Spermatogonia! equatorial plate. Four very small and two rather small 

 chromosomes. Total 28. 



17—2 



