154 ip T. A. Chapman on the 
other Blues; if so, it should be possible to find some 
characters in common to optilete and orbitulus which make 
them less specialised than the others, 2.¢. some further 
character than the want of the honey-gland. The only 
trace of such a character I know of is that the segments 
of the larva 7, 8,9 and 10 abdominal are less fused into 
one than in various other Blues, corydon and bellargus for 
example. On the other hand, if they once had the honey- 
gland and have since lost it, it is difficult to explain how 
they have regained the normal armament of dorsal hairs 
that disappear at the site of the honey-gland in those 
species that possess it. 
There is one point worth noting, viz. the structure of 
the Adoeagus. 
In a short but important note on this structure in the 
Plebeiid Blues in the Ent. Record, Vol. xxii, p. 101, 
I pointed out that this structure has a peculiar character 
in semiargus, optilete, pheretes and orbitulus, placing these 
together and separating them from the other sections. 
When I made that note, I had no suspicion that it had 
any bearing on the present question. How or why 
semiargus falls into the group is for the present purpose 
a somewhat obscure puzzle, but leaving it on one side, it 
places the other three species together, all of which have 
a high-level distribution, two of them are without honey- 
gland, and it would not perhaps be rash to expect that 
the third will be found to be so also, but at present this point 
is not known. 
These four species have one other character in common, 
viz. they are without the basal spot between veins 1 
and 2 of the hindwings beneath, a spot rarely absent in 
Plebeiids, except in Tutt’s genus Hirsutina. 
These facts suggest that these three species are closely 
related, and that they are derived from a common ancestor, 
who never had, or who lost the honey-gland. Notwith- 
standing that they differ in facies, perhaps more than any 
other Plebeiids do. 
The hypothesis that these species are at the base of 
the myrmecophilous species seems much more than 
doubtful in at least two respects. The myrmecophilous 
Lycaenids belong toa number of different sections, and the 
idea that they havea common ancestor in species of one of 
these sections, that except in this one respect are fairly 
typical numbers of that section, is contrary to our idea of 
