106 the entomologist's RECORD.' 



the two others ; these in the vast majority of cases constitute that 

 lighter colouring which is wrongly described as " the ground colour," 

 when the mistake is not carried so far as to call " ground colour " even 

 the primary pattern, simply because it has more extent than the 

 secondary pattern. In the Zygaena the dark' indigo or greenish 

 scaling is the primary pattern, the red or yellow is the secondary 

 pattern, and the real ground colour, which is white or yellowish, is 

 only left uncovered in Z. ephialtes, L., or in rings round the spots of 

 the forewing of other species. In this genus the patterns are of the 

 very simplest description, and, in fact, they are not much more 

 complex than in the lowest of the Lepidoptera, the Micropterygidqe,, 

 but the original bands are so fused together that they are not discern- 

 ible at first sight. Variation is thus carried on in its very broadest 

 and simplest lines, making it well suited to a study of its fundamental 

 laws, as I have already mentioned. We must note first of all that 

 there are species of Zygaena which never produce the true primary 

 pattern, the nervural pattern existing alone : this is the purpuralis, 

 Br., group ; the consequence is that the red scaling forms longitudinal 

 bands and never splits transversally into spots. There are, on the 

 other hand, species in which the true pattern exists alone and the 

 nervural pattern is quite absent or very rudimentary : this is 

 the fausta, L., fraxini, Men., and carniolica, Sc, groups; the conse- 

 quence is that the red spots can blend transversally across tbe 

 nervures, but never form longitudinal bands as in the group mentioned 

 above ; in aberrent individuals, they may be united by one central 

 band along the cubital nervure, when the transverse bands are 

 interrupted at this point, bat this, too, is due precisely to the absence 

 of nervural pattern. Between these two extreme groups extend all 

 the other species, in which the nervural and the true primary pattern 

 exist together and in which variation consists in the various combina- 

 tions of their different degrees of development. 



Beturning to lilipendulae, we find it stands in the series of species 

 last mentioned and that it combines the nervural and the true 

 primary pattern. This explains a phenomenon in its variation that 

 would otherwise have been puzzling. I think the order in which I am 

 describing the races is the natural one, because it corresponds to the 

 successive transitions from one to the other as one finds them in 

 nature. Now, in this series it will be noticed that from manni or 

 paulula to pulcherrima the dark markings become less and less 

 extensive and the red ones proportionately more and more so, whereas 

 in the races of subspecies stoechadis, which I am about to deal with, 

 the dark markings increase from the races closely connected to 

 pulcherrima to the most distinct stoechadis ones. This sudden inversion 

 of the process of variation would be difficult to explain. The remarks 

 I have made on the wing-pattern of the Zygaenac give us the clue by 

 showing that the inversion is only apparent and that we have, on the 

 contrary, only one progressive series of variation or, perhaps, rather, 

 two series diverging from pulcherrima and leading up to arctica or 

 paulula on the one hand and to stoechadis proper on the other. In the 

 former it is the nervural pattern which decreases in extent horn paulula 

 to pulcherrima, in the latter it is the true or transverse primary pattern 

 which increases from pulcherrima to the darkest stoechadis. We thus 



