SEASONAL POLYMORPHISM. 139 



Mts. (Piceno). The mountain series all bear dates varying from 

 July 20th to August 10th, except a few of the rest of August and 

 beginning of September ; the Porte dei Marmi ones are more abundant 

 in September. Querci and I have not been able to establish if this is 

 the only summer generation of these localities, as it is very likely in dry 

 surroundings, or whether it is followed by a third at Forte dei Marmi 

 and possibly even in the mountains, very late in the season. Querci 

 remembers having seen large numbers of napi at Bolognola late in 

 September. In Florence and on the hills aroiind it, up to 600 m., 

 there is distinctly a second emergence m June and a third one from 

 middle of September to October. The former is of a much larger size 

 and more robust build than the race just described, whereas the third 

 is smaller and frailer, as we shall see in the next grade. To that large 

 second generation no doubt belongs nteridionalin, Riihl, Pal. Gross- 

 schiHftt., p. 714 (1895), described as follows: " Large, lightly coloured 

 specimens ; underside of hindwings nearly unicolorous ; nervures 

 scarcely darkened. Flies in Central Italy." In Hhop. Pal. I wrongly 

 used the name for the individual form of all sizes with no streaks on 

 the underside. In the dry neighbourhood of Florence incridinnalis has 

 markings very limited in extent, as a rule, and usually also greyish 

 rather than black (see FUiop. Pal., pi. XXXII. , fig. 11), as they are, 

 instead, in the third generation. A series collected in the Mainarde 

 Mts. (prov. of Caserta) has much more extensive and perfectly black 

 markings. The second generation of Istria and Dalmatia, described 

 and figured by Stauder in his Weitere Beitrd(/e seems to be quite 

 similar to this last, whilst the third generation has a still more intensely 

 black pattern, but is otherwise quite large and like the second; I name 

 it stauderi, mihi, from his fig. 7 and 8 of PI. I. 



Grade V. has the nervural streaks of the underside about similar to 

 grade IV., but on upperside the true pattern is very much reduced 

 in extent, as compared with this grade. The apical crescent tends to 

 be more broken up by white in each internervular space ; the spots are 

 quite small in both sexes. My fig. 15 on pi. xxxii. of Plho/). Pal., gives 

 an excellent idea of it. I take as typical of this grade and name 

 tenuemaculosa, mihi, the third generation of race meridionalis from 

 Florence ; a perfectly similar one was found in the Mainarde Mts. 

 These consist invariably of small and frail individuals, which have 

 evidently suffered from the summer drought and scanty food ; the 

 butterflies emerge considerably later than the third generation of the 

 other Pieridae, after a very long period (end of June to middle of 

 September), during which no napi are on the wing. We have seen 

 this is not the case in race micro}}ieridionalis, which emerges in July 

 and August. Amongst the latter I have often found tmuemacnlosa as 

 an individual form. Rostagno, in his R/wp. Faunae Ennianae, p. 66, 

 has named barraudi a few dwarf specimens collected very late in the 

 season, which he took to be a fourth generation, but this certainly 

 does not exist. The gray and very limited spots above were due to his 

 specimens being weaklings, and the nervular streaks on underside to a 

 phenomenon I have observed in several localities : the early individuals 

 of the third brood often have no streaks, whereas the latest ones often 

 have very pronounced ones, and females may even exhibit extensive 

 nervural streaks on upperside, recalling the spring brood (form tarda, 

 mihi). 



