xl 
also the green is more tinted with yellow or sienna, especially 
in the later emerged examples. These green prolongations 
vary much in width and in the amount of yellow or sienna 
in them. 
2 slightly the larger as a rule: the orange patch at apex 
of fore-wing usually smaller, in a few examples yellowish, 
and in a few others altogether absent. Otherwise this form 
varies but little. 
Locality : Left bank of the Dyala from 350 to 600 ft., in 
March and early April. Females seen ovipositing on young 
flower-buds of a mustard and a mauve flowered Crucifer. 
Z. eupheme tigris Riley. 33 4, 99 2, from the right 
bank of the Tigris, taken March—April 1920, also exhibited 
for comparison. 
(3) Melitaea trivia, subsp. persea Koll. A small series of 
each of three seasonal forms :— 
(a) Spring form, from Mesopotamia and N.W.F., India, 
March and April; comparatively large and with black spots 
well-marked above and on underside, two females being 
especially large examples. 
(6) Summer form, from Mesopotamia, June-July; averaging 
smaller, and with much less black above and beneath. 
(c) Autumn form, from N.W.F., India, Sept.—Oct.; very 
similar to the Spring form. 
THE SUBFAMILIES OF FORMICIDAE. 
Mr. H. DontstHorrE read the following communication :— 
In all the recent works and catalogues on ants up to 1920, 
five subfamilies have been recognised—namely, Ponerinae, 
Dorylinae, Myrmicinae, Dolichoderinae, and Camponotinae, 
and this arrangement is the same as that used by Dalla Torre 
in 1890. 
In 1920 Wheeler, after studying a great number of ant 
larvae of many genera and subgenera in all five subfamilies, 
proposed to raise two more subfamilies—the Pseudomyrminae, 
and the Cerapachyinae. 
Let us see how far Wheeler is justified in this proceeding. 
; f 
— —— 
