THE ATHALIA GROUP OF THE GENUS MEr,IT^.A. 269 



though both may be regarded as single- brooded, is very great. 

 In deione the difference between the two broods is often con- 

 siderable, and in britomartis—at least as represented at Reazzino 

 — it is most striking. In imrthenie (apart from varia, which, so 

 far as I see at present, there is no reason for connecting with it) 

 the difference in size between the two broods is not, in my 

 experience, considerable, but I have seen no specimens that are 

 undoubtedly North German of the second brood (the mere label 

 "Germania" being almost worse than useless) ; so that, in the 

 face of Borkhausen's description and Godart's illustration, it 

 would be unreasonable to make a general statement to this 

 effect. Speaking somewhat loosely, it may be said that asteria 

 is the smallest species, then varia, and that aurclia, hritomartis, 

 parthenie, athalia, deione, and dictyniia follow in this order ; but 

 this must in no sense be taken as a rule, except in so far as that 

 if hundreds of examples of all the species were measured, the 

 average would probably come out in that order ; but in individual 

 cases the exceptions would be so multitudinous that size alone is 

 most untrustworthy as a guide. Speaking generally, it may be 

 said that in any given species the longer the feeding-time of the 

 larva the larger the resulting imago (though even this must be 

 qualified by taking into consideration the nutritive qualities of 

 the food attainable), and hence it follows that in double-brooded 

 species of this genus, the spring brood, whose larvae feed both in 

 autumn and spring, is generally larger than the autumn brood, 

 which has to get through all its phases in two or three months, 

 or even less. This is strikingly illustrated by the cases of 

 parthenie, berisalensis, and hritomartis in Switzerland. There is 

 little difference between the two broods of the first-named 

 species, the second brood of which begins to feed when the 

 plantains are still fairly young and juicy, and which has about 

 ten weeks of larval life ; the difference is greater in berisalensis, 

 the first brood of which does not generally appear till two or 

 sometimes even three weeks later than parthenie, the second 

 broods being nearly contemporary; whilst in britoinartis, the 

 larval life of whose second brood cannot extend beyond five 

 weeks at most, the difference is very great, the first brood being 

 generally as large as the average parthenie, and the second 

 sometimes as small as the smallest asteria. When aurelia is 

 partially double- brooded, south of the Alps, as at Roveredo, the 

 few second-brood specimens that I have found have been no 

 larger than asteria, and the small size of September athalia, 

 when that species ventures on a partial second brood, is a 

 matter of common knowledge. It follows also from this that 

 the higher the altitude to which a single-brooded species mounts, 

 the smaller will the specimens become, whereas, if a double- 

 brooded species mounts high enough to become single-brooded, 

 the tendency of the specimens will be towards increase in size, 



