278 BRITISH BUTTERFLIES. 



a theory (Rom. Mem. Lep., iv., p. 393) that the species originated 

 in the Pamirs, on the ground that it is there found in all its 

 different forms (except the exclusively British artaxerxes and 

 salmacis), from which he concludes that these forms must have arisen 

 there at the different altitudes, up to 14,000ft., at which it occurs, helped 

 by the different aspects of the slopes on which it flies. This, however, 

 ignores the fact that these differences of conditions are the very cause 

 of the variation, and that these various forms, in spreading over the 

 ground at present occupied by the species, must all have encountered 

 practically identical conditions, often for long periods, so that the 

 differences would tend strongly to become lost, and would have to be 

 re-acquired. As he specially remarks on the fact that the specimens 

 from the highest elevations do not differ from the mountain forms of 

 Europe, his facts really suggest that similar climatic conditions tend 

 to produce similar forms. This certainly holds good, both in altitude 

 and latitude, with regard to the number of broods, but so far as our 

 records and observations go, we have never found any tendency in the 

 single-brooded areas to the production of a second brood, nor in the 

 double-brooded areas to a suppression of the second brood, except 

 under abnormal treatment of the larvae such as described above by 

 Chapman. Where more than two broods occur annually it is quite 

 likely that the actual number may be dependent on the peculiarities of 

 different seasons, the tendency then being to become as nearly con- 

 tinuous-brooded as circumstances will permit ; the localities, however, 

 where more than three broods occur are very few. 



In Europe, speaking generally, the isothermal line of 49° holds 

 good as the division north of which this species is single brooded, the 

 most notable exception being Denmark, for which both Bang-Haas 

 and Aurivillius give two broods, appearing in May-June and August- 

 September respectively ; in the rest of Scandinavia, as far north as it 

 occurs, there is but one brood, appearing in June and July. In 

 Finland also Federley describes it as being not rare tip to 66° N., in 

 June and July. In Holland Snellen gives its time of flight as May, 

 and again from July to September, which seems to imply a somewhat 

 earlier date than is usual with us, and also a probable variation of the 

 time of appearance of the second brood in different seasons ; 

 Lambillion also mentions May as the time of appearance of the first 

 brood throughout Belgium, but only speaks of July and August as the 

 time of the second brood. Throughout France, of course excepting 

 the higher Alps, there are two broods, with indications of a third, as 

 might be expected, in the extreme south, since we have taken it at 

 Hyeres as early as March 28th, and Reverdin records it from Bandol 

 (Var) for March 30th, while Rowland-Brown found it at Beaulieu 

 (also on the Riviera) as late as October 9th. Two broods, again, are 

 the rule in Spain, in the Sierra Nevada as well as in the Central 

 Table-land, but both north and south of this lowest Mountain-chain 

 there are at least three, e.<j., Granada and Algeciras, while at 

 Gibraltar Walker reports it as being on the wing most of the year, from 

 February to November, which seems to imply four or even five broods. 

 Throughout Germany medon is double-brooded where it occurs, the 

 two broods usually appearing at the end of May or the beginning of 

 June, and the end of July or early in August; there are, however, many 



