O. A. Merritt Hawkes 151 



heritable ; it does not behave Mendelianly and cannot be fixed." 

 Packard(9), p. 152, writes: "The single median tubercle on the eighth 

 abdominal segment of the specialised Saturnian larvae represents the 

 ' caudal horn ' of Bombyx mori and is evidently the f'usian before the end 

 of embryonic life of what was originally two separate tubercles." The 

 tubercles of this cynthia x ricini cross and the caudal horn being thus 

 equivalent structures, it will be interesting to compare their inheritance 

 behaviour when it is fully worked out, although the indication so far is 

 that the smooth form in this hybrid does behave as a Mendelian reces- 

 sive. I do not know if the structures are anatomically the same, but 

 even if they were it would not necessarily follow that they would behave 

 in the same way in a crossing. Thus one may instance the variability 

 of the dominance or recessiveness of white cocoon colour, and again, 

 whilst I find that the black spots of the cynthia larva are dominant over 

 the plain condition of the ricini larva, Soule(lO) found the black spots 

 of the nearly related P. cynthia advena were recessive when crossed with 

 the plain Gallosamia promethea. 



Tanaka(ll), jd. 197, mentions a knobbed race of Bombyx mori. The 

 knobs, he writes, " are large paired evaginations of the skin formed in 

 the sub-dorsal line."..." No permanent peculiarity is seen in the larvae 

 until they have passed through the third moult, when the knobs come 

 forth for the first time, and their full development is attained only in the 

 last stage of larval life." He found segi'egation of knobbed and smooth 

 from the hybrid, but does not say whether the extracted dominants and 

 recessives remain pure. 



This character appears to be very different from the tubercles of 

 Philosumia or the caudal horns of Bombyx mori, as it is an addition 

 which originates late in larval life, and Tanaka continues, p. 197, " the 

 remnants of certain knobs are still visible in the pupal and imaginal 

 stages." The tubercle belongs to an early stage of Saturnian evolution, 

 but this knob is a mutation added so late that it is even ceasing to be a 

 purely larval structure and is cariied over into the imago. There may 

 also be a fundamental difference between the two, for whilst Tanaka 

 writes of the knob as an " evagination of the skin," the tubercle is an 

 evagination of the body wall related to a definite grouj) of setae. 



Packard(9), p. 151, writes: "The larval characters of this group 

 tend to show that the family has originated from a spiny group and most 

 probably, when we take into account the transformations of Aglia tau, 

 from the Cei-atocampidae....In the European Saturni...we have the 

 perhaps most generalised and primitive members of the group."..." It is 



