144 
Psyche 
[August 
(Collection Nos. A-733 and A-761), and Dr. Wheeler’s specimens 
have been compared directly with these. The specimens from 
lots No. 76 and No. 164 are entirely comparable with the topo- 
type specimens except for their slightly larger size. The speci- 
mens from lot No. 87 average smaller in size and tend more 
towards a circular shape. The morphological characters, with the 
exception of some minor differences in the shape and extent of 
the various chitinized areas, as the collar surrounding the anal 
plates, appear to be identical within the limits set by a slight 
allowance for individual variation, in all the specimens examined. 
In addition to the British Guiana records, the writer has 
collected this species in the grounds of the Department of Agri- 
culture, St. Clair, Port-of-Spain, Trinidad, Nov. 23, 1918 beneath 
the bark of Ficus (ulmifolia?) (A-1046), inside the hollowecl-out 
twigs of a large tree of Pithecolobium saman (A-1047), and 
finally on an unknown tree (A-1059), in each case attended by 
ants. 
Some additions to Prof. Newstead’s original description 
may be noted to advantage as aids to the recognition of the 
species: The claw digitules are present, thread-like, curved, not 
quite reaching the tip of the claw: the heav}^ collar at the mar- 
ginal end of the spiracular complex normally bears two stout, 
tapering, blunt-tipped, widely separated spines of small size 
on the outer margin, but these are frequently broken off and 
the spine base is so placed as to make its recognition very difficult 
when the spine is missing; the marginal spines are fairly stout, 
tapering, faintly lanceolate; each anal plate bears from 10 to 
15 setse dorsally, one apical, two subapical, and the remainder 
scattered over the surface; there are four ventral ridge and four 
fringe setse, the latter in two pairs on the edge of the membranous 
fold beneath the plates; ail setse of the anal plates are slender 
and of moderate length; anal ring small, with a few large pores 
and ten setse; the ventral multilocular disk pores are quite 
numerous beneath and behind the anal plates and over the 
posterior portion of the ventral surface, especially near, but not 
quite at, the body margin, up to the posterior spiracles, but 
any transverse rows across the middle of the abdominal segments 
