92 THE KNTOJIOI-OGISTS KKCOIiW. 



form, and my " types " are, in fact, July specimens of the New Forest. 

 Series of the second generation of Central Europe, as far north as 

 Chantonny in Vendee, I possess, are chiefly composed of biviitata, with 

 an occasional diiticnsis. Tlie third brood of the damper localities of 

 Central Italy, such as the one near Florence mentioned above, which 

 produces my typical birittato in the second brood, or such as Poggio, 

 m. 400, in the Isle of Elba, exhibit in the majority of individuals 

 transicns perfectly similar to the English second generation. An 

 interesting series of third brood was collected by the Querela from the 

 end of August to the middle of September, 1921, in the Mainarde 

 Mts., at Atina, m. 500 ; it exhibits at first a majority of biiittata, then, 

 as the season goes on, tran^ieiu supersedes it entirely and becomes 

 more and more like latin/) i, till at the end of the emergence two latln/ri 

 were found, quite similar to the spring form. 



Grade V. : lathyri, Hiibner, is the first generation of all the 

 localities of the species, except those very special ones where the 

 following extreme one replaces it, and it is so well known that I need 

 not describe it here in detail. Its individual variation is far greater 

 than that of the other broods, but all the series I possess from several 

 latitudes and altitudes seem alike on an average, except for size, 

 and the lighter or deeper tone of gray of apical patch ; the one from 

 Waidbruck in S. Tyrol is, on the whole, the largest ; the Florence 

 series is the smallest, although, curiously enough, it is the one which 

 produces the remarkably large (irandis in the second generation ; the 

 third generation from this locality is, however, again a particularly 

 small (liniensis. These marked seasonal differences in size can be 

 explained by a knowledge of the surroundings : there are no springs of 

 water in the Plan di Mugnone, but during the spring months the 

 ground becomes very swampy, when it is on a level, on account of the 

 winter and spring rains ; the vegetation then becomes very luxuriant 

 in the hot days of the end of spring. There then follows the summer 

 drought, when the ground becomes baked and vegetation extremely 

 scanty, and this produces the small dinicnsis. Autumn rains bring out a 

 fresh crop of grasses, but evidently not sufficient for the larva> of 

 lathi/ri, which are then feeding, to grow large. 



Grade VI. : iiii/rencens, Vrty., Ent. Bee, 1919, p. 87, can be 

 described as the highest expression of the latliyri features, more 

 accentuated than they are in the vast majority of cases. The basal 

 suffusion is distinctly blackish instead of gray on both surfaces ; it fills 

 the cell and extends far beyond it on forevving above ; it also exists at 

 the back of cell, along the dorsal margin ; the apical patch is very 

 broad and stretches backwardly by a tapering point as far as the first 

 cubital nervure ; both the cubital nervures have a black streak at their 

 ends ; the aforesaid patch is always black, as it often is in northern 

 races, but only exceptionally in the usual races of Central Italy, such 

 as yrandis, which might well be called cana in the first brood by 

 its very pale gray tinge, evidently due to the same causes as the 

 crescent of that tinge so often produced in the I'^'eris of this region ; 

 also the underside of nijpcficeus is more extensive!}^ 'larkened and of a 

 colder tone. This race I discovered at the mouth oi'the Arno on such 

 swampy grounds that they are under one or two feet of water after the 

 winter and spring rains ; in the last days of May the water had just 

 retired and male siiiajtis was beginning to emerge, whereas in the 



