SEASONAL POLYMORPHISM. 131 



vidually than the European races ; they belong, however, by the 

 extent of their pattern, on an average, to the two following grades and 

 not to this one. 



Grade IV. can best be conveyed by referring the reader to pi. 

 xxxii., fig. 5, and to pi. ixvii., fig. 16, of Ehop. I'aL, which represent 

 about rightly the lesser and the greater extent of the dark pattern to be 

 included herein. It will be seen that it is distinctly more extensive 

 than that of the races of the plains of Central Europe, taken on the 

 whole. The second figure mentioned represents an individual form of 

 race arctica, Vrty., lighter than the average in this race, but still 

 commonly found in it. The first figure represents a female hritannica, 

 Vrty., the race of Ireland and northern Scotland, which I consider 

 typical of this grade, because its average extent of pattern, to my 

 knowledge, falls here ; it is the last race which recalls brijoniae and 

 arctica by producing now and then a female with yellow ground-colour 

 and some with thick nervural streaks, which would belong individually 

 to grade III. As far as I know, also the race of Southern Sweden 

 has its average at this level. It is quite distinct from arctica of the 

 north, and it is the nyraotypical race napi, L., because Linneus gives 

 Fauna Siiecica as first quotation, and the former having been 

 subsequently named, Linneus's name is restricted to the southern race. 



Grade V. includes the widespread races of the lowlands of Central 

 Europe and of some particularly damp localities of Southern Europe. 

 For the present I detect amongst them two perfectly distinct races. 

 By far the commonest is the one I have named septentrionalix in the 

 Ent. Bee, xxviii., p. 79 (April, 1916), taking as typical the English 

 race, as represented by a series from Westcliff-on-Sea and by another 

 from Epping Forest. I have since ascertained that this race spreads 

 as far as Central France and Switzerland in the plains. It must be 

 noted that the grey tinge of the dark pattern predominates over the 

 black one in septentrinnalis. In the race which I have called uinoris in 

 the Ent. Bee. for December, 1921, p. 210, the extent of that pattern is, 

 on the whole, about the same, but it is very predominantly black. I 

 described it from swampy grounds on the coast of Tuscan}-, but I have 

 evidence that it spreads into Northern Italy and probably to Central 

 Europe in particularly damp localities. I notice a perfect resemblance, 

 quite surprising, considering the very different surroundings, between 

 my t^^Dical series and the whitest individual forms collected at the 

 Baths of Valdieri, m. 1375, in the Maritime Alps, in company with 

 much scarcer bryoniae, 0. In Tuscany the second generation of 

 iinwris is micromeridionalis, which I will describe further on ; further 

 north sttbnapaeae, also to be described, and Icovigilda, Frhst., are to be 

 expected. 



Grade VI. stands strikingly apart from the preceding when well 

 characterised series are compared. I take as typical of this grade my 

 first generation viih/aris of race meridionalis, Riihl., as represented by 

 series of specimens from the neighbourhood of Florence, whence were 

 my " types." In the Linnean Society's Journal — Zooloyy, xxxii., 

 p. 177 (May, 1918), I had proposed this name for the more widespread 

 naiii as contrasted with hryuniae and with the Linnean Scandinavian 

 race, which I grouped with the latter. I now fully realise such a dis- 

 tinction of two large groups of races can in no way be made, because 

 they blend into each other and overlap. The name of vulgaris however 



