THE ATHALIA GROUP OF THE GENUS MELITa?A. 161 



remarked to me, by an euthusiastic supporter of the claims of 

 this means of differentiation, that the important question was 

 not that of mere pairing, but oi fertile pairing. lu answer to this 

 there are two things to be considered. First, we do not know 

 whether, or how far, these pairings are infertile. We do know 

 that no insect has ever been taken which could possibly be a 

 hybrid between Enchloe cardamines and Bapta temerata, for in- 

 stance, or even between Melitcea athalia and Polygonia c-albiim, 

 but, on the other hand, we know that hybridity interferes greatly 

 with vitality, and even should some of the less violently abnormal 

 pairings prove fertile, the chances of the progeny reaching ma- 

 turity would be exceedingly small, and of those that reached the 

 perfect state— if auy did so — it is improbable that any would 

 happen to be the specimens seen by lepidopterists, as they would 

 probably be weak and fall an easy prey to marauders within the 

 first few hours of their imaginal existence. But giving the fullest 

 possible weight to this aspect of the argument, and granting 

 that none of the above pairings (with the probable exception of 

 the two Taeniocampids) could prove fertile, the very important 

 fact still remains that it is not the differences of the structures in 

 question which per se are the cause of the infertility, as they 

 could only produce this effect directly by making pairing itself 

 impossible, and that at the most they can only be a secondary 

 and correlated cause, not to be compared in importance with 

 such facts as species appearing at different times, inhabiting 

 different altitudes or latitudes, feeding on different (or at any rate 

 on unrelated) food plants, hybernating at a different stage of 

 existence, flying at a different time of day, or possessing any 

 other habit which must either effectually keep them apart, or be 

 practically certain to cause the early death of any hybrid pro- 

 geny which might manage to struggle into existence. 



These considerations appear to me to reduce the importance of 

 these organs, which yrimafacie seem to be on such a vastly higher 

 plane, to the same level as any other structural details, and as 

 the differences are usually very slight, often actually and almost 

 always relatively so, I should be inclined to assign to them, in 

 general, an importance distinctly below neuration, still more 

 markedly below that of the number of fully-developed legs, but 

 possibly above that of scale-forms, even of the androconia, 

 and probably above most other imaginal characteristics which 

 have been used for purposes of classification, such as hairy or 

 glabrous eyes, details of wing-shape and markings (including 

 the presence or absence of " tails "—a very unimportant matter, 

 I believe, especially when the "tail" is filamentous)— markings 

 of the body, number of rings in the antennae, &c. ; on the other 

 hand, I do not believe their phylogenetic and consequent classifi- 

 catory importance to be nearly so great as that of many points 

 in the earlier, especially the egg and newly-hatched larval, stages; 



