J. W. H. Harrison 75 



TABLE XIII. 



Hofmozygmis Type bistortata ^ x Homozygous Melanic crepuscularia $ . 



Melanics Types 



Family Parentage Males Females Totals Males Females 



II I^xB'S 56 11 67 — — 



III 19 xB'^ 45 45 — — 



IV I?xB'(J 61 61 — — 



Actual result 1 173 



Expectation if inheritance occurred ) ,„„ 

 on a Mendelian basis ... j 



TABLE XIV A. 



Homozygous Type bistortata $ x Heterozygous Melanic crepuscularia ^ . 



Melanics Types 



^ Percentage 



Females Totals Melanic 



— 30 — 



— 37 — 



Actual result 58 67 46-4 



Expectation if inheritance occurred ) /.„.,. „„_ __ 



on a Mendelian basis ... j '^ 



TABLE XIV B. 



Heterozygous Melanic crepuscularia 9 x Homozygous Type bistortata ^ . 



Melanics Types 



_^ — ^ ^ ^ -^^ ^ Percentage 



Family Parentage Males Females Totals Males Females Totals Melanic 



VII a^^ xl^ 18 12 30 9 36 45 — 



VIII aOxI^ 26 17 43 28 37 65 — 



Actual result 73 110 39-7 



Expectation if inheritance took t qi.k qi.k ka 



place on a Mendelian basis [ 



of typical insects in the last two broods^ is somewhat high. Curiously 

 enough, too, the percentage of types in both cases is exactly 60 7o- 

 However, this excess is in all probability correlated with the slightly 

 greater constitutional vigour possessed by the type forms tending to 

 their preferential survival. 



The matter of the inverse relationship between sex and colour may 

 be connected to some extent with the same phenomenon, but, in face 

 of the corroborative evidence supplied by the two broods, not merely 



1 I have had precisely the same result with two large broods of this hybrid reared 

 during March and April 1920 for another purpose. 



2 The sex ratio (102 ? ? : 81 <f <? ) in the case of these broods is not, I think, significant. 

 In a parallel cross between type crepuscularia ? and type bistortata i I obtained 96 ? ? 



and 88 <f (J . 



