7 
COLIAS EDUSA, FAB. (CROCEUS, FOURC.). 137 
form, the third is only partly so. From mid-October onwards, 
and after the tardiest females of the third generation have put 
in an appearance, males emerge with all the characters of the 
gen. vern., followed in due course, and when days are mild, by 
corresponding females. Here then the gen. vern. lasts from 
October to May. In June comes a new and typical generation, 
followed in August and to mid-October only by another typical 
brood. 
Working northward from Tuscany, Dr. Ubaldo Roceci devotes 
a long paper in his interesting and instructive ‘‘Osservazioni sul 
Lepidotteri di Liguria ’* to the phenology and racial characters 
of Colias edusa in Liguria, and gives a formal name—autumnalis 
—to the autumnal form, noting that the three forms as they 
occur in the region of Genoa are distinguishable, though the first 
autumn emergence is differentiated only in minor degree, as 
compared with the vernal, from the gen. @st. 
With regard to the vernal form, I need say no more than that 
he agrees with Dr. Verity’s diagnosis, adding that, whereas in 
some seasons vernalis is common, and the characters of the form 
strongly accentuated, in others it is rare, and the distinguishing 
characters little in evidence. He considers that the summer 
(typical) form has two generations—one from May to about the 
middle of August, the other from the end of August into Sep- 
tember—and the Ligurian examples are larger than those of 
Piedmont. 
It becomes increasingly interesting, therefore, to trace and 
determine the northern limits of the form vernalis, and further 
to ascertain the continuity of the form within such limits. The 
earlier French collectors and authors do not appear to have 
recorded the first flights of Hdusa on the littoral, probably because 
their knowledge was confined to the second (gen. est.) emergences. 
De Graslin, it is true, states (1862) that a form closely resembling 
Chrysotheme, Esp., occurs at Collioure, Pyr. Or., and this should 
correspond with the normal form vernalis. M. Rondou, whose 
own, observations are rather of the Central Pyrenees, also makes 
no mention of a spring race, nor does he cite var. Pyrenaica, Gr. 
Gr., which seems likely to be the ab. minor of Verity, and of the 
same early spring form. But though the records are indefinite 
—Mabille in his ‘Apercu des Lépids . . . del’Aude’ (1885) 
specifies no date of emergence for Hdusa—it may be assumed 
from the known meteorological conditions prevailing in the 
western Mediterranean that the species maintains itself along 
the Spanish seaboard, and along most of the coast from the 
Pyrenees to the Alpes-Maritimes, in a succession of broods 
throughout the year at sea-level and in the plains as upon the 
French Riviera, with the possible exception of that part of the 
Bouches-du-Rhone where the conditions are less favourable. At 
* ‘Atti Soc. Ligustica di Sci. Nat. e Geog.,’ Anno xxx, N. 4, 1920. 
