Uranotaenia in Solomon Islands — -Belkin 
2L, Tenam, Oct. 18^ 1943 (J.G.Franclemont); 
lM(0-24) Tenam, Sept. 10, 1943 (P. W. 
Oman); IF, 7L(0-32) Tenam, Sept. 9, 1943 
(P. W. Oman); 2M, 1F(K^910) June 21, 1943 
(K. L. Knight); 2M(G^40) Dec. 5, 1943 (A. 
B. Gurney). 
Uranotaenia harnesi is named in honor of 
Arthur W. Barnes, Jr., to whom I am greatly 
indebted for the painstaking rearing of much 
of the material collected on Guadalcanal. 
Variation 
The adults of U. harnesi show a great deal 
of variation in the amount of orbital light 
scaling on the head. Usually the light-scaling 
is extensive and very conspicuous. The tho- 
racic light-scaling shows less variation. The 
modified foreleg of the male is extremely vari- 
able in the degree of development of the 
specialized scales and hairs, but they conform 
to the arrangement as illustrated. It is prob- 
able that some of the differences noted are 
due to the teneral condition of reared speci- 
mens. One female has the apical two seg- 
ments of the left fore tatsus white-scaled. 
The variation in the chaetotaxy of the im- 
mature stages is summarized in Tables 1 and 
2. There is a rather narrow range of variation 
in the branching of the hairs despite the fact 
that many collections were made at all seasons 
of the year. No unusual variations were ob- 
served. 
The material from other islands falls within 
the range of variation exhibited by specimens 
from Guadalcanal. 
Specimens examined: 90M; 92F; 5 IP; 
402L. Individual rearings: 19 larval. 
Taxonomic Discussion 
Uranotaenia harnesi is closely related to three 
New Guinea species of the tihialis-gtowp de- 
scribed by King and Hoogstraal (1947: 585- 
596) : U. setosa, U. neotihialis, and U. fimbriata. 
It is unfortunate that the two earliest nominal 
species in this group, U. tibialis Taylor, 1919, 
and U. antennalis Taylor, 1919, are not defi- 
nitely recognizable at present. Since they were 
339 
both proposed for specimens from Cairns, 
Queensland, it is unlikely that they would 
represent species from the Solomon Islands, 
particularly when it appears from the work 
of King and Hoogstraal that the tibialis -gtou.^ 
is composed of a number of closely related 
species in New Guinea, none of which can 
be definitely associated with Taylor’s species, 
and all of which are distinct from U. harnesi. 
King and Hoogstraal (1947: 593) examined 
specimens of U. harnesi and noted their simi- 
larity to U.fimbriata K. and H., 1947. The two 
species are very closely related but are distinct 
in the following characters of the male of U. 
harnesi: light-scaling of head very wide in- 
stead of narrow; wing with conspicuous light 
scales at base of R, Cu, and lA; fore tibia 
longer, 0.9 instead of 0.7 length of femur; 
fore tarsus 3 longer, 1.1 instead of 1.0; hind 
tarsus white from extreme apex of segment 
2 instead of basal third of segment 3; white- 
scaling of hind tarsus 2 produced ventrally 
from apex, demarcation rather indistinct; 
light-scaling of head, scutum, pleura, and 
wing a rather intense azure-blue instead of 
white. The female of U, fimbriata is unknown 
but it appears likely that it would be separable 
from that of U, harnesi by the extent of the 
light-scaling of the hind tarsus, since this 
character does not show marked sexual dif- 
ferences in this genus. The same character 
will separate U. harnesi females from those of 
U. setosa, which has the hind tarsus white 
from the apex of the third segment only. The 
female of U. tibioclada is unknown, but again 
since the male does not show any white- 
scaling on the apex of the second segment 
it is likely that the female will differ from 
U. harnesi in this character also. 
The pupal stage of U. fimbriata is not 
known, and only the trumpets have been de- 
scribed and figured for U, setosa and U. tibio- 
clada. The trumpets of U. harnesi agree in 
coloration with these two species but appear 
to be intermediate in length. The pupa de- 
scribed and figured by Penn (1949: 32-33) 
as U. albescens is undoubtedly that of a mem- 
