196 
SOUTH- AFRICAN BUTTERFLIES. 
slender, — first pair, at about middle of their inner edge, with a stout 
(often acuminate) process ; tarsi very long (especially first joint), con- 
siderably longer than tibiae, densely spinulose beneath, — the claws 
large, rather straight, simple. 
Abdomen rather stout, of moderate length or rather long, thickened 
posteriorly ; anal plates in $ usually large and conspicuous. 
Laeva. — Eather thick, often more or less swollen on third thoracic 
segment, and thence abruptly attenuated to head ; usually smooth, but 
in some groups armed with numerous curved fleshy tubercular pro- 
cesses ; head small, smooth ; penultimate segment often more or less 
bifid dorsally into small acute prominences ; exsertile Y-shaped ten- 
tacle on back of first thoracic segment usually rather long ; thoracic 
segments sometimes each with a pair of short filamentous processes 
superiorly and laterally. 
Pupa. — Very variable in form ; rarely almost straight, usually 
with abdomen more or less curved, and thorax and head more or less 
bent backward. Head usually bifid (often very deeply), sometimes 
truncate, round, and blunted. Dorso-thoracic prominence sometimes 
produced into a forward-pointing process. Abdomen often dorsally 
armed with tubercles, which are sometimes developed into conspicuous 
processes. 
For number, diversity, and beauty of its species, the great genus 
Papilio, even if we withdraw from it the magnificent and barely 
separable group Ornithoptera, stands unrivalled. So conspicuous and 
prominent a feature are its members in the butterfly life of the tropics 
and adjacent latitudes, that they have been for considerably more than 
a century more extensively collected and better known generally than 
any others of their tribe. Though modern investigations have shown 
that structurally their affinity to the Heterocera prevents their any 
longer being regarded as the highest or most specialised butterflies, 
they stood for so long at the head of the Order, that they have 
received more study and examination than any other Lepidoptera. 
After comparing the catalogues published by Boisduval (1836), 
Doubleday (1847), G. E. Gray (1852 and 1856), C. and E. Felder 
(1864), Kirby (1^7^ 1877), and G Oberthiir (1879), I think 
that the species of Papilio may fairly be held to exceed four hundred 
in number. Notwithstanding their very great diversity in outline, 
colouring, and pattern, the generic characters throughout this large 
assemblage offer but little modification ; and I agree with the great 
majority of lepidopterists that it is impossible to break up Papilio 
into satisfactory genera. At the same time, as remarked by Mr. 
Distant {Eliop. Malay., p. 323), the genus is readily divisible 
into well-marked groups; and Boisduval, the Folders,'^ and Wal- 
1 Act. C. R. Soc. Zool.-Bot. Vindob., xiv. (1864). This is the most elaborate of the 
published investigations of the structural characters ; but it is, in my opinion, carried into 
a minute analysis more refined than natural. There are no fewer than seventy-five sections 
