130 



THE ENTOMOLOGIST S RECORD. 



within the next one. It is worthy of notice, however, that in Alpine 

 races of this sort, consisting in a mixture of very different individual 

 forms, the vast majority of female individuals do not exhibit the 

 average extent of pattern, as might have been expected, but divide 

 into two groups approaching respectively the two extremes, whilst 

 intermediate forms are comparatively scarce. If a curve of frequency 

 were drawn from statistical data it would consist in two cusps, with a 

 strong depression between them. This evidently points to a tendency 

 to dimorphism, somewhat similar to that of female ('alias cnnriis, 

 Fourc, with its distinctly either orange or white forms and extremely 

 rare intermediate ones, or similar to Dri/as paphia, L., with either 

 fulvous or grey females. Grade II. is the one which exhibits most 

 distmctly the secondary variation I have dealt with above. The main 

 line may be mentally pictured as broadening out into parallel lines ; 

 the extreme one on one side consists in those forms in which the 

 nervural and the true, or transverse pattern are equally extensive (the 

 true Alpine bnjoniae, as figured on pi. xxxii. of Uhop. Pal., tig. 27, .'ind 

 the Arctic equivalent ailalainda, Frhst., fig. 37) ; the median liiie 

 consists in forms in which the transverse pattern is very much reduced 

 as compared with the nervural (pseinln-bri/oniae, Vrty., as represented 

 by fig. 86) ; the line on the other side consists in forms with only the 

 nervural pattern, and Eober's radiata, as figured by him in Seitz's 

 CTross-Schiiu'tt., pi. 21, corresponds precisely to the level of grade II., 

 as regards the extent of this pattern. My nymotypical raiicasica, fig. 

 22, also belongs to this grade, but whether the entire race does, still 

 remains to be established with more material at hand than I possess. 

 Grade III. can be roughly described as including those forms which 

 are intermediate between those known in a broad way as bnjoniae and 

 as napi. High Alpine and Arctic races, of most localities, have an 

 average extent of pattern corresponding to this level, and so does the 

 wonderful race of Modling, near Vienna, of which the first generation 

 should, I think, be called interjcrta, Rober., because in it there exists 

 a large predominance of the individual form so named by this author 

 and most characteristic of grade III. At this grade one can detect 

 more clearly than at others the two parallel lines of variation in 

 connection with the reduction in the extent of the nervural 

 pattern noted above. On one of these lines we can imagine the 

 gradual transformation of the Alpine bnjnniae through einibr;/o)tiae, 

 . Vrty., at the level of grade III., on to the following grades ; my fig. 26 

 on pi. xxxii. of Tihop. Pal. might bear this name, but it is not well 

 characterised, because the streaks on basal half of wing are too 

 pronounced ; my fig. 8 on pi. xxxiii. of female oclisenhciweii, Stdgr., 

 shows instead the features of eniihnioniae well ; this Asiatic mountain 

 sub-species of najii (I scarcely think it is a distinct species) is, in fact, 

 the culminating degree of the emibrijoniae line of variation and fixes it 

 as a constant characteristic. On the other line we have interjecta and 

 the Arctic forms similar to my figure 35 of pi. xxxii., with thin streaks 

 of uniform breadth on the entire neuration ; it is the predominating 

 Arctic individual form, and by the extent of the pattern it falls, as a 

 rule, in grade III., so that race arctica, Vrty., belongs to this grade, on 

 an average. Here again the Asiatic races fix this line of variation as 

 a constant characteristic in Eastern Siberia, in Northern China, 

 and in Japan {paeudomelete, Vrty.), being much less variable indi- 



