STUDY OF VARIATION IN THE RACES OF ZYGAENA FILIPK.NDULAE, L. 109 



similar to the hindwings of sub-species ftlipendulae, by two series 

 diverging from it, one exhibiting a progressive increase in the extent 

 of the nervural pattern, the other exhibiting that of the transverse 

 pattern, and by a number of series between these two, in each of 

 which both nervural and transverse pattern would exist combined 

 together in different proportions. But this scheme is not a creation 

 •of my fancy ; it exists in nature ; it is the Zygaena genus. The two 

 extreme series are embodied by that which leads from Z. rubicundus, 

 Hb., up to the darkest purpuraMs, Br., and by that which leads from 

 Z. laeta, Hb., up to fausta, L., camiolica, Scop., etc. In a paper on 

 this genus in general I will show how its variations can be traced to a 

 beautifully definite and ordinate plan, sufficient to overthrow any 

 theory on natural selection of fortuitous variations. 



Races intermediate between subspecies filipendulae, 

 L., and subspecies stoechadis, Brk. : — 



Race anceps, Obth., Et. Up. Comp., IV., p. 551 (1910). 

 " Bouches-du-Rhone et Var (Hyeres, Sainte-Baume, Saint-Zacharie, 

 Montrieux, Saint-Pons)." Such are the localities given by Oberthur 

 for this very interesting race, which I have already talked of. I need 

 ■only add here that by its build, by the aspect of many individuals 

 very similar to subspecies ftlipendulae, by the tone of red and more 

 ■especially by the marginal band of hindwings being always narrow it 

 belongs on the whole more to ftlipendulae than to stoechadis. On the 

 other hand, however, the reduced red suffusion of the underside of 

 forewings and the numerous five-spotted individuals bring it very near 

 stoechadis. 



Race alpina, Boisd., lcones Hist. Lep., p. 65, pi. 53, fig. 9 

 (1832-41). Oberthur discusses Boisduval'^ description and concludes 

 that his insect must have been the one which is found at Digne ; 

 Boisduval only gives the " Alpes framjaises " as locality ; he thought 

 it was very probably referable to amjelicae, 0. Oberthur keeps it 

 under this heading, but expresses doubts as to its correctness. I do 

 not know it, but from the lengthy descriptions of these authors I 

 should say it is a small mountain race transitional between ftlipendulae 

 and stoechadis and similar to its neighbour in localities less elevated, 

 anceps, by its great variability. Like it, it seems to approach 

 filipendulae by the marginal band of hindwings, more usually narrow 

 and by the very extensive red suffusion of underside of forewing in 

 many individuals, whereas it approaches stoechadis by being often five- 

 spotted and by a few males having the hindwing broadly darkened. 

 The angelicaeformis, mihi, form noticed by Boisduval only occurs 

 occasionally. 



Race pulcherrima-stoechadis, mihi. This interesting race 

 from the hills above Reggio Emilia (Borzeno) and Modena (Rio 

 Spigone in the colli di Scandiano) on the extreme southern limit of 

 subspecies ftlipendulae in Italy, I have already mentioned in dealing 

 with transitions between it and stoechadis. A large percentage of 

 individuals belong distinctly either to the one or to the other, and 

 now and then, in the second case, even to forms medicayinis and 

 stoechadis, much darker than are ever found in race ochsenheimeri from 

 Southern Italy and to the five-spotted one. A. Costantini of Reggio 

 has sent me a large series to examine and he has kindly left me part 



