THE VAKIOUS MODES OF EMERGENCE. 71 



winter and summer pause may divide them distinctly into a " pre- 

 cocious f?roup " and into a "tardy group," as Ave have ah'oady indi- 

 cated ; besides, also in this case, there exists in every brood generally 

 a period in which a greater number of families emerge ("nucleus " of 

 the brood). Pi/raineis cardtii, L., which emerges from the iniddle of May 

 to the middle of July and again in August and September, and Mdi- 

 taea didi/nia, Esp., from the beginning of May to the summer pause, and 

 during all August till the 5th of September, are examples of species 

 with two graduated broods. 



In other cases, such as occur in Florence, in Coenonympha pam- 

 f kilns, L., in Pohjonimatus icanis, Eott., and in Agriades thersites 

 (Cant.), Chap., there is no interval between the end of the first and 

 the beginning of the second brood, indeed, during some days, these 

 may even overlap ; the individuals of the one, and those of the other, 

 can be distinguished in this case only by the morphological charac- 

 teristics, generally very marked, which they present. Let us examine 

 the mode of behaviour of the three species above mentioned during 

 the entire year. 



The first generation of C. pmnpliilus emerges in a very graduated 

 way from about the 10th of April to the end of July ; the earliest 

 spring individuals often have a characteristic underside to the hind- 

 wings, which are blackish with a bluish gloss (form miirina, Vrty.). ; 

 the rest of that brood has a grey colour with scarcely visible patterns 

 and very limited white space. The second generation emerges from 

 the beginning of July to the middle of Se'ptember ; in July and the 

 first two decades of August a very characteristic form is produced, 

 with a light tawny underside, prominent patterns and large numerous 

 ocelli (form emilyllus, Vrty.) ; these consequently contrast sharply 

 with the tardy individuals of the first generation, which emerge at the 

 same time during July. At the end of August and in September a 

 reversion to very nearly the same form as that of the first generation 

 takes place, although traces of tawny and a slightly more prominent 

 pattern distinguish most specimens from those of the latter (form 

 aesti rails, Ev.). Finally, in October, a very limited emergence of a 

 few individuals often occurs in favourable years and these resemble the 

 early spring ones mentioned above by their blackish underside. 



Now, it must be noted that in Florence by far the greatest number 

 of C. panqjftikis emerge in April and May ; they then dwindle to very 

 small numbers in June and July, until the tawny form makes its 

 appearance, when they again increase considerably ; the end of August 

 and beginning of September emergence is very poor. On the other hand, 

 on the Tuscan coast, at Forte dei Marmi, the chief emergence takes place 

 later, in May and June ; the tawny form of July and the beginning of 

 August is quite scarce and the emergence of the end of August and 

 beginning of September is decidedly abundant. I conclude that in 

 both localities only two real generations occur, as in C. arcania, and 

 that the other emergences are due to precocious or tardy families, as 

 the case may be. The particular extreme forms vturina and emilyllus 

 are the result of particular climatic conditions during the develop- 

 ment of the chrysalis. A phenomenon which still calls for ex- 

 planation if climate can effect the chrysalis stage so deeply is why 

 do July individuals with the characters of the first brood emerge 

 at the same time as others with the very distinct characters of 



