SEASONAL POLYMORPHISM. 89 



ni(jrici'lla, fuacedinelLa, and Intijiennella occur in Bi'itain as well as 

 heiiierubiella, which though differing so much in appearance from the 

 others is, I believe, connected with this group. They all have only 

 three pairs of ventrals and all agree in making a somewhat peculiar 

 first case and a second more simple case. 



The group of white species whose larvae spin their cases out of 

 their own silk without admixture of vegetable matter is from the pro- 

 leg point of view an especially interesting one. The largest species 

 which spins a more complicated case than the others has three pairs of 

 ventrals and there is no trace at all of the fourth pair. This is 

 palUatella. In anatipennella and Stainton's ibipennella there are also 

 only three pairs, but the ventral surface of the sixth abdominal 

 segment, instead of being quite smooth as in palUatella, is marked by 

 two scars lying in the situations which would have been occupied by 

 the fourth pair of prolegs had they been present. From this it may be 

 gathered that these two species have lost the fourth pair at a later 

 period. In another group we have vibicella, this also has but three 

 pairs of active ventrals, but it still carries the remnants of the fourth 

 pair in the shape of two minute points on the sixth abdominal 

 segment. It is one step behind aiiafipenncdla in the process of reduc- 

 tion. Among the three-paired species I think we may include 

 conApicuella. The first observer to notice the absence of prolegs on the 

 sixth abdominal was the faithful artist who drew" the figures of the 

 larvHB of limosi pennella and conspicuella in Stainton's Natural Historn 

 of the Tineina (vol. iv., pi. II., fig. 2((, and vol. v., pi. IX., fig. 2rt), as 

 both these \q,\:\-x are there shown with only three pairs. It is now many 

 years ago since I first noticed these figures, but I believe the sight of 

 them first induced me to examine the prolegs of Coleophorid larvae. 



In considering the loss of the fourth pair of prolegs, as we find four- 

 paired larvfe within the genus, we need not go outside this but from the 

 fact that the allied families, Gracilariides and Lithocolletides, have also 

 lost the same pair, we may argue that there exists a tendency to lose 

 this pair in the group of families forming this particular branch of the 

 stirps. I have already described the arched position assumed by the 

 larva when taken from its case, and I believe it retains this positi'^n as 

 far as it is able to do so when carrying its case, that is to say, that the 

 anal claspers are bent slightly under the body in contact with the fioor 

 of the case, while the dor.sal parts of the abdominal segments are pressed 

 against the roof of the case. The first two or three pairs of ventrals 

 would be, owing to the confined space afforded by the case, also in con- 

 tact with the floor, but owing to the incurved anal claspers, the fourth 

 pair may be held above the floor. If this be so then it follows that the 

 fourth pair will be less used than the other pairs, and it may follow that 

 from disuse it gradually dwindles away till it is finally lost. This last 

 condition has only been reached by certain species. 



(To be continued.) 



Seasonal Polymorphism and Races of some European Grypocera 

 and Rhopalocera. 



By ROGER VERITY, M.D. 



{Continued from page 73.) 

 Leptosia shwpis, L.— This species produces on the eastern coast of 

 the Adriatic a remarkable race, which by its gigantic size and other 



