Bulletin of the British Museum (Natural History)

ind,
pbadon tx barken.
Tieton es

SASS
ERG wy

pee’
aay

bearet:

ene

soe eed s
seater cstersriscett
siybateseprissterepeeterstotee, sees
eatieeeataty SSE

Giihieaststeate

porereeytacrt

rere
pele
aS =+5
whetoss rose

7
Ht

ror)
soetsstevetnizirese
ri cer ee, eyes grarerers
soaepce ve tease!

beet tees

otete cote!

Tote

Trees
sek

+
ieese
beeerhey

3} ss
phetansesetets tes
., bestetsritelesrsrety iat
ss veh ep gnareisety ts teh Satetineteatereesh:
ea Ssht Sos teteebant sett a
Pemcar erates 2
od
pws eh
eeriyaidoe
piseieeas
Sad hat

rit

te titst

Bis

pe
reddierpsiy di bese sere i233;

safeeist

has errs
Wate?

oe
8

Shite
eestoests2oery
setehtied

sisi oon

uae

2

Berson)
bee:
sgeseas} :

ia

te

$,

>

(327

fesse:

Ts

8 9d)

3 ;
2 Brisestetsts)
fits :
. ree
Ast :
otsitt
< >
mt
$ i
a sts. + 9
Heytstand state as HEE mate :
een et be “ be ibang Or" siT tates
sein arti eatin
: af ;
Rae pesstessessrorey. sr reteeee Hittite spaitte:
aise a ie : ; at enits! :
prsraye sae tptise eclese ity. f ; aie : the ethos
Bsteteteses ts : : terete 3 ritttettt att + 4
2: teyih3t st re trical < :
Bisistetits i Hai Mee etiest
ete stest et Bit i iti aa : f
body, v
re! Setatys i af ; at
oo |
1 m3 23% af hes 1
+ . 3 rete +4
eaeihs ts Ha HR i it Hi
i itiatat it fe papitee iS att Hit t
+ ad i
$ 4 . it Hitt 5
i 7 '
= iets: be if ; ii ie . y Hid
: : 3 Biatgititis Hist 32) - ;
a ie iat ths
che + R ! nite ian seatrietantat ths it i
ra His sitet DOERR O sagt
retay i She SESS eee hres ee r Ts steht; it
i iit alts it aint gH RE ST Raat
3 iia in RAE LER Thy ft Raitt!
$e 5 et sapala tet y be pgyists +7 ihe
ait: i i tie Symbian hen En
Hesse sted Ht Here anata as ; a at}
2 : aif : i t atte Ht seats !
ease thes drbigiaigceserts tester iaiit sirereess ii }
t : ag) aig ith biresscd ait pirha pe % +t eae
Hi ‘ EeatGth SSeHENGPeathaas acest iit ih esti
sae Sages toe) } F a8
eet iiiatiat th u Hi
*
: iy
; i t
af {

BULLETIN OF < ey

THE BRITISH’ MUSEUM
(NATURAL HISTORY)

ENTOMOLOGY
VOL. XVIII
1966—1967 |

THE BRITISH MUSEUM (NATURAL HISTORY)
LONDON : 1967

DATES OF PUBLICATION OF THE PARTS

No. 1 29 March 1966
No. 2 . 21 June 1966
No. 3 8 June 1966
No. 4 . Io June 1966
No. 5 7 June 1966
No. 6 . 28 June 1966
No. 7 : . 12 July 1966
No. 8 . 21 September 1966
No. 9 7 February 1967

PRINTED IN GREAT BRITAIN
BY ADLARD AND SON LIMITED
DORKING, SURREY

CONTENTS

ENTOMOLOGY VOLUME XVIII

The genus Ancistrotermes (Isoptera). By W. V. Harris

Revisional notes on the genus Antanartia (Lepidoptera : Nymphalidae).
By T. G. HowArtTH

Revisional notes on African Charaxes (Lepidoptera : Nymphalidae)
Pt. III. By V. G. L. VAN SOMEREN

Indo-oriental Horagini (Lepidoptera: Lycaenidae). By C. F. Cowan
The genus Rhipidocephala (Diptera: Asilidae). By H. OLDROYD

A list of the Odonata types described by F. C. Fraser, now in the
British Museum (Natural History). By D. E. KImMMINns

Revision of the genus Ovosius Distant (Homoptera: Cicadelloidea).
By M. S. K. GHAURI

A revision of the genus Anthene from the Oriental Region (Lepidoptera:
Lycaenidae). By G. E. TITE

Revisional notes on African Charaxes (Lepidoptera : Nymphalidae)
Pt. IV. By V. G. L. VAN SOMEREN

Index to Volume XVIII

PAGE

21

45

103

143

173

229

253

277

317

SU tee ee

Rie - e .. a.” _

7 _ | tec “4 © 1 (i paieuly eopessiteeertis:, sty A! ‘Pa

- ¢
ry Fy

pes (Teds bal Aerhs “Sf hie oe >

i sen &

A Aen iat}

ia <r ah rhea
=

Oe ge

+ Perk ti | a af | ae
‘ts veal? 7 Sah 7

,. a a ae (=
Bi ae as
i ¥ VW M A i
; ii

Ye AL W\
THE GENUS ANCISTROTERMES eu
(ISOPTERA)

W. V. HARRIS

BULLETIN OF
THE BRITISH MUSEUM (NATURAL HISTORY)
ENTOMOLOGY Vol. 18 No. 1
LONDON: 1966

THE GENUS ANCISTROTERMES
(ISOPTERA)

BY

W. V. HARRIS

ri 4y
Termite Research Unit,
Ministry of Overseas Development

Pp. 1-20; 18 Text-figures

BULLETIN OF
THE BRITISH MUSEUM (NATURAL HISTORY)
ENTOMOLOGY Vol. 18 No. 1
LONDON: 1966

THE BULLETIN OF THE BRITISH MUSEUM
(NATURAL HISTORY), instituted im 1949, 1s
issued in five series corresponding to the Departments
of the Museum, and an Historical series.

Parts will appear at irregular intervals as they become
veady. Volumes will contain about three or four
hundred pages, and will not necessarily be completed
within one calendar year.

In 1965 a separate supplementary series of longer
papers was instituted, numbered serially for each
Department.

This paper 1s Vol. 18, No. 1 of the Entomological
series. The abbreviated titles of periodicals cited
follow those of the World List of Scientific Periodicals.

© Trustees of the British Museum (Natural History) 1966

TRUSTEES OF
THE BRITISH MUSEUM (NATURAL HISTORY)

Issued 29 March, 1966 Price Nine Shillings

THE GENUS ANCISTROTERMES (ISOPTERA)
By W. V. HARRIS

CONTENTS
Page
INTRODUCTION . . : ‘ : 3
Genus ANCISTROTERMES SILVESTRI ; : : ; . 6
Key to Imagos . é , , , 2 d ; : ‘ 7
Key to Major Soldiers 7
DESCRIPTIONS OF SPECIES 9
ACKNOWLEDGEMENTS ; : : : : : : : : 20
SEN O PSIsS

The history of the genus is discussed, and eight species are dealt with, of which two are
described as new. Two species are placed in synonymy.

IN-ERO DUCT TON

IsopTERA of the subfamily Macrotermitinae (Termitidae) occur in Africa south of the
Sahara and in tropical Asia. They are characterized by the presence of fungus
combs in chambers forming part of the nest system. The nests are mainly sub-
terranean but some species, under favourable circumstances, build mounds.

In Africa four genera of Macrotermitinae, Macrotermes, Odontotermes, Ancistro-
termes and Microtermes, are numerous and widely distributed. They are dis-
tinguished from the remaining six less-common genera by the comparatively slender
mandibles of the soldier castes, with inner margins not serrated and with no more than
a single marginal tooth. They form a series showing a progressive reduction in size,
and in the number of antennal segments in all castes, and in the complexity of
structure of their fungus combs.

Ancistrotermes is present throughout the extensive areas of savanna woodlands
between latitudes 13° North and 23° South, at elevations below 1,600 metres a.s.l.
It is common in the gallery forests within the savanna, but is uncommon in closed
canopy rain forest on the one hand and in semi-desert thorn bush on the other.
One species, A. Jatinotus, is closely associated with the Brachystegia etc. woodlands
of eastern and southern Africa. The genus as a whole forms a significant part of the
soil macro-fauna of a large part of Africa, and is, in addition, of economic importance
due to attacks on young trees, especially Eucalyptus spp., on plantation crops such as
tea, and on constructional timbers.

Silvestri proposed the name Ancistrotermes as a subgenus in 1912 and as a genus in
1914 for two West African termites, Termes crucifer Sjostedt and T. cavithorax
Sjostedt. Later A. latinotus Silvestri from south-east Congo and Angola, A. guineensis
from Portuguese Guinea and A. periphrasis from the Sudan were added. Two species,
A. lembomboensis Fuller and A. amphidon Sjostedt, are placed in synonomy for reasons
given below. Two new species, equatorius and microdens, are described.

ENTOM. 18, I 1§

4 W. V. HARRIS

In view of the tentative nature of the original references and the absence of subse-
quent records, A. crucifer var. diversana Silvestri and A. ?dubius Sj6stedt are omitted
from this study. It has not been possible to examine the specimens of A. wasmanni
Snyder & Emerson, to which Wasmann gave the name A. cavithorax but included in
his published description significant differences in the soldier mandibles. This
brings the total number of species dealt with here to eight.

A cavithorax oO latinotus

@® crucifer © latinotus -record in literature
© crucifer -record in literature oO microdens

V equatorius ¥ periphrasis

® cuineensis L wasmanni - record in literature

—-+—-—-— boundary of the savanna woodlands

Fic. 1. Distribution of Ancistrotermes in West Africa.

Fie, 2.

THE GENUS ANCISTROTERMES

Distribution of A ncistvotermes in Central and South Africa.

1S

6 W. V. HARRIS

ANCISTROTERMES Silvestri

Termes (Ancistrotermes) Silvestri, 1912, Ann. Mus. Zool. Napoli 3 (22) : 4.
Ancistrotermes Silvestri, 1914, Bol. Lab. Zool. Portici 9:17. [as genus]

Imago. Small (width of head including eyes from 1:58-1:86 mm., length of fore wing from
15-20 mm.); head oval; clypeus inflated, distinctly paler than the rest of the head, with a
dark median line ; eyes circular, large ; ocelli oval, comparatively large and close to the eyes ;
labrum tongue-shaped with broadly rounded tip ; mandibles conforming to the general Macro-
termitinae pattern, with no subsidiary tooth on the first marginal of the right mandible (also
absent in Odontotermes and Microtermes) ; fontanelle a small circular spot ; antennae with 17
segments (19 in Odontotermes and 15 or 16 in Microtermes).

Pronotum approximately semicircular with anterior corners rounded and projecting forwards,
and both anterior and posterior margins slightly sinuate. Wings similar to Microtermes, veins
other than radius and radial sector only feebly chitinized, medius and cubitus not united at the
base, weak branches usually present from medius towards radial sector.

Soldier. Distinctly dimorphic, the major soldiers having the head and mandibles in the range
1:50-2:40 mm., and the minor soldiers 1-14—1-90 mm.

Major soldier with head oval or laterally compressed ; mandibles robust with strongly incurved
tips, left mandible with a small marginal tooth, right mandible with or without marginal tooth
but usually with a small basal spur; labrum not broader than long, pointed ; antennae with
15 segments (Odontotermes has 16 or 17, Microtermes has from 12 to 14) ; pronotum saddle-shaped.

Minor soldier with ovoid head, more or less narrowed towards posterior ; mandibles slender,
only lightly curved, marginal teeth rudimentary or absent ; antennae with 14 or 15 segments.

Worker. There are two worker castes, differing mainly in size; major workers with head
width between 1-4 and 1-6 mm., minor workers between 0-8 and I:o mm.

Head yellow with paler clypeus and two distinct brown spots marking the inner articulations of
the mandibles ; thorax and abdomen cream, with a sprinkling of short, pale setae. Head rectan-
gular, wider than long to base of mandibles; clypeus narrow and inflated ; mandibles as in
imago; antennae with 16 segments (sometimes 15); pronotum saddle-shaped with rounded
anterior lobes separated by a distinct median notch, posterior margin curved or lightly emar-
ginate.

The major workers of Odontotermes are larger, the head is longer than broad and the antennae
have 17 or 18 segments. In Microtermes the worker is smaller, the head is slightly longer than
broad and the antennae have 14 segments.

List of species (* indicates that the imago is known).

I. *cavithorax (Sjost.) West Africa
amphidon Sjost. West Africa

2. *crucifer (Sj6st.) West Africa, Congo, Angola

3. equatorius n. sp. East Africa

4. guineensis (Silv.) West Africa

5. *latinotus (Holmgr.) Central and East Africa
lembomboensis Fuller South Africa

6. microdens n. sp. East Africa

7. *periphrasis Sjost. Sudan and Congo

8. wasmanni Snyder & Emerson Congo

pela ats

THE GENUS ANCISTROTERMES 7

Kry to ANCISTROTERMES Imacos

Width of head including eyes ie aes under 1:77 mm.; hind tibia 1-86 mm. or more;

ocelli smaller A 2
Width of head including eyes HOE more than I- a9 mm. hind tibia ander I* 86 mm.;

ocelli large and broadly oval é ;
Ocelli broadly oval, half the long diameter distant from aves. , : latinotus (p.

ay
Ocelli narrowly oval, under one-third the long diameter from eyes . periphrasis (p. 18)
Upper surface of head with numerous strong bristles. : ; . crucifer (p. 11)
Upper surface of head with a few fine bristles : : : : cavithorax (p. 9)

Kry to ANCISTROTERMES Major SOLDIERS

Left mandible with two microscopic marginal teeth, head and mandibles 1-8—2-0 mm.
wasmanni (p. 20)
Left mandible with a single marginal tooth . : 2
Head laterally compressed, narrowed anteriorly, posterior slightly angular, mandibles
straight with incurved tips, no basal spur on right mandible, head and mandibles
1:82-2:09 mm. : ‘ ‘ ; ; ; d i : guineensis (p. 14)
Head oval or ovoid ‘
Head oval, mandibles robust, each ae marginal tooth, one owith rudimentary basal
spur, head and mandibles 1:82-2:05 mm. ; . crucifer (p. 11)
Head ovoid, right mandible with small marginal tooth and distinct basalspur . : 4

Fics. 3-6. The Imago head in profile. 3. latinotus ; 4. crucifer ; 5. cavithorax ;
6. periphrasis.
Fic. 7. Imago mandibles—cavithorax.
Fic. 8, Fore wing—cavithorax,

W. V. HARRIS

Fics. 9-15. Heads of Major and Minor soldiers. 9. crucifer ; 10. microdens ;

11. cavithorax ; 12. latinotus ; 13. periphrasis ; 14. equatorius ; 15. guineensis.

Fics. 16-18. Right mandible of Major soldier. 16. guineensis ; 17. crucifer ;
18. latinotus,

THE GENUS ANCISTROTERMES 9

4 Head distinctly narrower at posterior, mandibles lightly curved, head and mandibles
1-68—1-82 mm. j 5 ? : ; ; : equatorius (p. 13)

— Head not narrowed towards posterior , :

5 Mandibles curved from base, sides of head lightly carved, head and mandibles

1:50-1:82 mm. ' : : ‘ : cavithorax (p. 9)
— Mandibles straight at base, incurving at tips , 6
6 Pronotum with deep anterior notch, mandibles with oo small marginal teeth, head

and mandibles 1:91-2:18 mm. : ; microdens (p. 17)
- Pronotum with shallow median notch, mandibles with larger marginal ZeCth > 5 , 7
7 Head rectangular-oval, mandibles lightly curved, head and mandibles 1-82—2-27

mm. latinotus (p. 15)
~ Head ovoid, slightly narrower anterior, mandibles strongly curved, head and mandibles

1:82—1-90 mm. ; ‘ F : ‘ : ; ; ; periphrasis (p. 18)

Ancistrotermes cavithorax (Sjéstedt)
(Text-figs. 5, 7, 8 and 11)

Termes cavithorax Sjostedt, 1899, Ent. Nachr. 25 : 37 [soldier from Cameroon].

Termes cavithorax Sjostedt ; Sjostedt, 1900, K. Su. Vetakad. Hand. 34 : 130-132.

Microtermes cavithorax (Sjéstedt) Holmgren, 1913, K. Suv. Vetakad. Hand. 48: 42.

Termes (Microtermes) cavithorax Sjostedt Sjéstedt, 1914, Ark. Zool. 8: 4.

Ancistrotermes cavithorax (Sj6stedt) Sjéstedt, 1926, K. Sv. Vetakad. Hand. 3: 163.
Ancistrotermes cavithorax (Sjéstedt) ; Sjéstedt, 1926, Rev. Zool. Afr. 14: 163-164 [imago from

Ghana].
ee amphidon Sjostedt, 1926, Ent. Tidskr. 47: 240-241 [soldier from Ghana]. syn.n.
Ancistrotermes cavithorax (Sjéstedt) ; Grassé, 1937, Ann. Soc. ent. France 56: 72-78.

The soldiers of Ancistrotermes cavithorax were described briefly by Sjéstedt in 1899
and in greater detail the following year from specimens collected at Victoria,
Cameroon. He described the imago from Ghana in 1926, but later ascribed it to a
new species, amphidon. It is, however, proposed to regard amphidon as a synonym
employed for one series of cavithorax rather larger in size of all castes than the average.
The imago of cavithorax has also been described in detail by Grassé.

Wasmann (rg11) described in detail major and minor soldiers from Sankuru,
Congo as those of Termes cavithorax, and these have been designated types of a new
species A. wasmannt by Emerson and Snyder in Snyder (1949).

Measurements :
Imago from Ghana and Nigeria :
Mean Range (15)
(mm.) (mm.)
Width of head with eyes. 5 : : ; 1-67 1*59-1°77
Diameter of eye : 0°45 0+42—0°52
Diameter of ocellus, long 0-21 0+ 18-0-24
Diameter of ocellus, short 0-16 015-018
Ocellus to eye 0:09 0:08-0:10
Width of pronotum 1°53 I-4I-1-68
Length of pronotum o:9gI 0*QI-0°95
Length of hind tibia 1-80 I+72—-1-82
Length of fore wing 15'7 15*'O-16-0
Width of fore wing 3°9 3°6- 4°5

10 W. V. HARRIS

Soldiers from Ghana and Nigeria :

Major Minor

oom Sg mths te

Mean Range Mean Range

(mm.) (mm.) (mm.) (mm.)
Length of head and mandibles ; 1°62 1:+50-1°82 I*2I I*14—1-36
Length of head é 1°05 O-gI-1-18 0°72 0:68-0:77
Width of head. : I*O0O O:'QI-I*I4 0'73 0:68-0:82
Length of left mandible 0°74 0:+73-0°78 0°58 0*54-0°64
Width of pronotum 0*70 0:64-0:82 0°52 0*50-0°54
Length of hind tibia 0:95 O*QI-1-+04 0:80 0:77-0:86

Illustrations. Imago, soldier and worker castes are figured by Grassé (1937)
together with photographs of the fungus comb present in the nests. Figure of head
and pronotum for major and minor soldiers of amphidon are given by Sjéstedt (1926).

Distribution. Type locality—CAMEROoN: Victoria, in Mus. Stockholm.
Other material. CAMEROON: Mamfe, 1957 (W. Wilkinson).

NIGERIA: Eastern Region—Port Harcourt, Ikot-Ekepene road, Umuahia,
Onitsha, Abakaliki-Ikom road, Enugu, 1957 (W. Wilkinson) ; Northern Region—
Ankpa, Makrudi, Lokoja, Beli, Badeggi, Bida, Jos, Gombe, Kaduna, Zaria, Damaturu,
Argungu, Keffi, 1956-58 (W. A. Sands) ; Western Region—Agege, 1912 (A. D. Pea-
cock) ; Ikeji, 1955 (W. V. Harris) ; Ibadan, Oyo, 1958 (W. A. Sands) ; Benin, 1957
(W. Wilkinson).

GHANA: Accra, 1926 (A. J. Pomeroy) ; Aburi, 1926 (W. H. Patterson) “ amphidon
Sjéstedt, co-types”’. Koforida, 1922 (Baker); Kumasi, 1955 (W. V. Harris) ;
Accra, Ejura, Lawra, Wenchi, Volta Bridge, Tamale, Wa, Bolgatanga, 1959 (W. A.
Sands).

Ivory Coast: Abidjan, 1963 ; Dabou, 1964 (P. Bodot).

SIERRA LEONE: Nijala, 1955 (W. V. H.); Freetown, Bo, 1958 (W. A. S.)
UGANDA: West Nile, Arua, 1965 (K. W. Brown).

Additional published records: Conco: Sankuru, in Sjéstedt (1926).
Ivory Coast: Aboisso, Akakra, Bouaké, Beoumi, Man in Grassé (1937).

The record of A. amphidon from Uganda in Harris (1951) is referred to below under
equatorius N. Sp.

Biology. A. cavithorax is found throughout West Africa in clearings within the
coastal forests, in the mixed woodland savanna, and to the northern limits of the
Guinean woodland.

The subterranean nest is a system of oval chambers, many containing fungus
comb, connected by galleries of varying length. Grassé describes a nest made of
earth situated inside the hollow trunk of a Borassus palm, with chambers measuring
5-6 cm. in length and 3-4 cm. high. Many nests are recorded from mounds of Macro-
termes, Trinervitermes, Cubitermes and Thoracotermes.

THE GENUS ANCISTROTERMES ve

Foraging workers seek out dead wood for food, usually fallen logs and timber in
contact with the ground. Covered runways are built with soil up the trunks of
trees and suitable feeding areas of dead bark are covered with sheets of earth.
Young trees are frequently killed as a result of ring-barking by this termite, es-
pecially in afforestation projects with Eucalyptus. The woodwork of buildings is
also subject to attack.

Swarming is recorded between 18.30 and 18.45 hours at Aboisso in the southern
Ivory Coast during the month of February (Grasse).

Ancistrotermes crucifer (Sjéstedt)
(Text-figs. 4, 9 and 17)

Termes crucifer SjOstedt, 1897, Ent. Tidskr. 18 : 123 [imago from Sierra Leone and Cameroon].
Termes crucifer Sjéstedt ; Sjéstedt, 1900, K. Sv. Vetakad. Handl. 34 : 127-130.

Termes (Ancistrotermes) crucifer Sjéstedt ; Silvestri, 1912, Ann. Mus. Napoli 3: 4-5.
Microtermes crucifer (Sjéstedt) Holmgren, 1912, K. Sv. Vetakad. Handl. 48: 42.

Ancistrotermes crucifer (Sjéstedt) Silvestri, 1914, Boll. Lab. zool. Portici 9: 18.

Ancistrotermes crucifer (SjOstedt) ; Sjdstedt, 1926, K. Sv. Vetakad. Handl. 3 : 162-163.
Ancistrotermes crucifer (Sj6stedt) ; Emerson, 1928, Bull. Amer. Mus. nat. Hist. 57 : 468-469.
Ancistrotermes crucifer (Sjéstedt) ; Grassé, 1937, Ann. Soc. ent. France 106 : 70-72.
Ancistrotermes crucifer (Sjostedt) ; Weidner, 1956, Pub. cult. Comp. Diamantes Angola 29 : 86-87.
Ancistrotermes crucifer (Sjéstedt) ; Harris, 1958, Explor. Parc nat. Upemba 52: 21.
Ancistrotermes crucifey (SjOstedt) ; Harris, 1963, Explor. Parc nat. Garamba 42 : 31.

The preliminary description of Termes crucifer from Sierra Leone and Cameroon
was based by Sjéstedt (1897) on the winged imago. This was followed in 1900 by
detailed descriptions of all castes “‘ nach den Typen ’’, but the list of material studied
indicates soldiers and workers only from Malanga in Angola. This locality is given
by Silvestri (1914) as the source of Ancistrotermes latinotus, with the result that the
first account of crucifer soldiers from West Africa is to be found in that paper. The
two soldier castes are re-described by Emerson (1928) from north-east Congo, while
measurements and figures of Ivory Coast specimens are in Grassé (1937).

Measurements :
Imago : West Africa and Congo
“ Cotypes ’’, Abo — “~ —
(2) Mean Range (9)

(mm.) (mm.) (mm.)

Width of head with eyes . 1 -68-1-°77 rs 7a I -68-1-77
Diameter of eye. , 0*45-0°54 0:48 0*45-0°54
Diameter of ocellus, long. 0°25-0°27 0°22 0+2I-0°23
Diameter of ocellus, short 0:18 0°15 0:14-0°18
Ocellus to eye ‘ : 0+08-—0-09 0:09 0*09—0-I0
Width of pronotum j I*50-1-54 1°55 I +46-1 +68
Length of pronotum g O:'9I 0:92 0:82-1:00
Length of hind tibia : 1+72-1°77 1°82 1+72-1°86
Length of fore wing ; 15-16 15°7 15‘0-16:0

Width of fore wing ; 3°7-4:0 3°9 3°6-4°1

12 W. V. HARRIS

Soldiers : Major Minor
on is —>r Tae
Mean Range (17) Mean Range (4)

(mm.) (mm.) (mm.) (mm.)
Length of head and mandibles : I‘QI 1:82-2:05 1°37 1:°23-1°46
Length of head . : : ; 1°23 1:09-1:28 0:80 0-68—0-86
Width of head. : : ; 1°18 1:09g-1°28 0:80 0:77-0:86
Length of left mandible ‘ ; 0:82 0:77-0:86 0:69 0°64-0°77
Width of pronotum ; : ‘ 0:80 0:77-0:86 0°59 0:+54-0°64
Length of hind tibia. ; : 1:08 0:86-1:09 0:86 0:82-0:91

Illustrations. The imago head is figured by Silvestri (1914). The head and thorax
of the major soldier appears in Silvestri, Emerson (1928), Grassé (1937) and Weidner
(1956). Grassé illustrates the whole minor soldier.

Distribution. Type locality—Sierra Leone and Cameroon (Abo and Mungo)—
imago.

Material examined. CAMEROON : Abo (Buchholz) in Naturhistoriska Riksmuseum,
Stockholm.

GAMBIA: Bunaka Kunda, Busumbula, 1956 (C. R. Wallace).

SIERRA LEONE: Jarama, 1948 (F. A. Squire); Nijala, 1955 (W. V. Harris).

Ivory Coast: Abidjan, 1964 (C. Noirot) ; Dabou, 1964 (P. Bodot).

GHANA: Aburi, Kumasi, 1925 (W. H. Patterson) ; Accra, 1955 (W. V. Harris).

NIGERIA: Northern Region—Sokoto, Zaria, Kaduna, Gombe, Jos, Yola, Bida,
Zinna, Kabba, Gboko ; Western Region—lIlorin-Oyo road, Ibadan, 1956-58 (W. A.
Sands) ; Eastern Region—Onitsha, 1957 (W. Wilkinson).

Conco: Garamba National Park, 1950-52 (H. De Seager); Upemba National
Park, 1947 (G. F. de Witte).

Additional published records. SENEGAL: Dakar, in Sjéstedt (1926).
Ivory Coast: Man, Dimboko, in Grassé (1937).

Toco: Misahohe, Bismarckburg, in Sjéstedt (1900).

Conco: Niangara, Yakulu, in Emerson (1928).

ANGOLA: Dundo, in Weidner (1956).

Biology. Ancistrotermes crucifer occurs throughout the savanna and woodland
areas of West Africa and the Congo down to the northern limits of Brachystegia
woodland in Angola. The subterranean nest system is composed of approximately
spherical chambers, some 2 inches in diameter, connected by galleries. These
chambers contain fungus combs. Nests are also to be found in the walls of mounds
constructed by other termites, for example those of Macrotermes, Cubitermes, Thora-
cotermes and Trinervitermes in Ghana and Nigeria.

The workers build covered runways on dead tree stumps and on the trunks of
living trees in search of dead wood. Damage to economic crops is common, par-
ticularly to groundnuts and young Eucalyptus trees. In Ghana young saplings of
Gmelina and teak have been attacked.

THE GENUS ANCISTROTERMES 13

Winged adults are recorded as swarming at Dundo, Angola at 18.00 hours in the
month of August, and in the Upemba National Park in south-east Congo during
October.

Ancistrotermes equatorius sp. n.
(Text-fig. 14)

Imago. Unknown.

Soldiers. Major soldier; head yellow, not noticeably setose ; mandibles brown; antennae
ivory. Thorax, abdomen and legs ivory.

Head as long as broad, distinctly narrowed towards the posterior, sides curved ; labrum rather
longer than broad, obtusely pointed ; mandibles robust, the distal halves strongly incurved, left
mandible with small tooth midway on inner margin ; gulamentum broad, narrowed towards the
anterior, strongly convex in section ; antennae with 15 segments, III small, IV, V and VI equal.

Pronotum with anterior lobes small and pointed, separated by an acute angled notch ; posterior
margin flatly curved.

Minor soldier ; head pale yellow, mandibles light brown. Thorax and abdomen ivory.

Head ovoid, sides converging towards the posterior; labrum conical; mandibles slender,
tapering, lightly curved at the tips; antennae with 15 segments, III very short ; gulamentum
short and broad, convex in cross-section.

Pronotum with short anterior lobes divided by a shallow, obtuse angled median notch.

Measurements : Major soldier Minor soldier
Sea = = aeeEIER
Mean Range (10) Mean’ Range (8);

(mm.) (mm.) (mm.) (mm.)
Length of head and mandibles ; 1°75 1:68-1:82 I-32 1:°28-1°36
Length of head ‘ j I:IO 1I-+OO-I-I4 0°75 0:73-0:82
Width of head ; 1:04 1:00-1:09 0:77 0:°73-0°82
Length of left mandible 0°79 0:76-0:82 0:65 0:59-0:68
Width of pronotum 0°73 0:68-0:77 0°56 0:+54-0'°59
Length of hind tibia I*02 0:95-1:09 0°85 0:77-0'91

The major soldier is distinguished from other Anctstrotermes by the shape of the
head, which is narrowed towards the posterior. It differs from latinotus in possessing
mandibles more robust and curved, and from amphidon in having a blunter tooth on
the left mandible and a deeper, more angular notch on the anterior margin of the
pronotum.

The minor soldier closely resembles that of cavithorax.

Holotype, major soldier, UGANDA: Karamoja district, 40 miles south-west of
Moroto, 12.x.52 (W. A. Sands), in British Museum (Nat. Hist.)

Further records. UGANDA: Karamoja district, Toro Hills, 8.x.52 (W. A. Sands) ;
Kitgum, attacking cotton, 22.ix.37 (A. M. Gwynn); Busia, inside mound of
Cubitermes, 21.1.50 (W. V. Harris); Soroti, inside Cubitermes mound, 13.x.52
(W. A. Sands).

Described from 20 major soldiers, 14 minor soldiers and a number of workers from
five localities in northern and eastern Uganda,

14 W. V. HARRIS
Ancistrotermes guineensis (Silvestri)

(Text-figs. 15 and 16)

Termes (Ancistrotermes) crucifer subsp. guineensis Silvestri, 1912, Ann. Mus. Civ. Stor. Nat.
Genova (3) 5: 227 [Port. Guinea].

Microtermes guineensis (Silvestri) Holmgren, 1913, Ent. Tidskr. 34 : 333.

Ancistrotermes crucifer var. guineensis (Silvestri) Silvestri, 1914, Bol. Lab. zool. Portici 9 : 20-21.

Ancistrotermes crucifer var. guineensis (Silvestri) ; Sjéstedt, 1926, K. Sv. Vetakad. Handl. (3)
3 : 163.

Ancistrotermes guineensis (Silvestri) ; Grassé, 1937, Ann. Soc. ent. France 106 : 68-70.

Soldiers and workers from Portuguese Guinea were described by Silvestri as a
subspecies of Anczistrotermes crucifer (Sjost.). This was raised to specific rank by
Grassé. Silvestri comments that in relation to crucifer ‘“‘ subspecies haec a forma
typica mandibularum militis majoris brevitate bene distincta est ’’, while Grassé
points out that the major soldier head is more rectangular, and the mandibles more
strongly curved at the tips.

Measurements :
Major soldier Minor soldier
oe ito a | fone ties a |
Mean Range (10) Mean Range (5)
(mm.) (mm.) (mm.) (mm.)
Length of head and madibles ; 1°93 1°82-2-:09 I*52 1:°50-1°54
Length of head : 1°26 1:18-1:36 0-91 0-9!
Width of head : I°I5 I-+14-1°23 0:84 0:82-0:86
Length of left mandible 0°77 0:68-0:82 0°73 0:68-0:77
Width of pronotum 0°87 0-82-0:91 0:67 0:64-0:68
Length of hind tibia I-07 I:00-I'14 0:90 0:86-0:91

These measurements are of soldiers collected in Ghana and Nigeria and are somewhat smaller
than the figures given for Portuguese Guinea and for Ivory Coast (width of head 1-26 mm.).

Illustrations. The heads of major and minor soldiers are figured in Silvestri
(1912) and Grassé (1937).

Distribution. Type locality—-PoRTUGUESE GUINEA: Rio Cassine, in Silvestri
(IgI2).

Material examined. GHANA: Accra, 1926 (A. J. W. Pomeroy); Accra, 1959
(W. A. Sands).

NIGERIA: Northern Region—Ayangba, Beli, Bauchi, Kaduna, Portiskum—Kano
Road, Minna, Riom, 1956-58 (W. A. Sands); Western Region—Olokemeje, 1957
(W. A. Sands) ; Eastern Region—Port Harcourt—Owerri Road, 1957 (W. A. Sands).

CAMEROON : Mamfe, 1857 (W. Wilkinson).
Additional published records. GuINEA: Mamou in Silvestri (1914).

Ivory Coast: Akakro in Grassé (1937).

LAE GENUS ANiChST ROTE RM E'S 15

Biology. A. guineensis occurs in the Guinean woodland formation along the
whole of West Africa, in the more open grass areas and in the dense gallery forest, and
in the coastal savanna-forest mosaic. Soldiers and workers are found mainly on
dead wood, but in some cases nests have been located beneath the mounds of other
termites.

Ancistrotermes latinotus (Holmgren)
(Text-figs. 3, 12 and 18)

Microtermes latinotus Holmgren, 1912, K. Sv. Vetakad. Handl. 48 (4) : 42 [soldier from Luapula,
S.W. Congo].

Microtermes latinotus Holmgren ; Holmgren, 1913, Ent. Tidskr. 34: 332 [imago].

Ancistrotermes latinotus (Holmgren) Silvestri, 1914, Bol. Lab. Zool. Scuol. Sup. Agr. Portici 9:

19.
Ancistrotermes lembomboensis Fuller, 1922, S. Afr. J. nat. Hist. 3: 103. syn. n.
Ancistrotermes latinotus (Holmgren) ; Sjéstedt, 1926, K. Sv. Vetakad. Handl. 3: 161.
Ancistrotermes latinotus (Holmgren) ; Harris, 1948, Proc. R. ent. Soc. Lond. (B) 17: 80.
Ancistrotermes latinotus (Holmgren) ; Grassé & Heim, 1950, Rev. sci. Paris 88 : 3-13.
Ancistrotermes latinotus (Holmgren) ; Weidner, 1956, Pub. cult. Comp. Diamantes Angola 29 : 86.

Holmgren (1912) lists under the genus Microtermes as a new species M. latinotus
from Luapula with the remark “ The figures of a soldier Termes crucifer which
Silvestri gives (1912, Ann. Mus. Zool. Napoli 3 : 4-5) does not agree with the type of
this species at all and belongs most likely to a new species which I designate Micro-
termes latinotus n. sp.’ In the following year this species is included by Holmgren
in his keys to the African Microtermes, the imago being mentioned for the first time,
being separated from crucifer by the larger ocelli and indistinct fontanelle, while the
soldier is said to differ from crucifer, cavithorax and guineensis by a pronotum much
broader than the head. This last is incorrect, being derived from an error in Sil-
vestri’s illustration of his Luapula soldier.

Silvestri (1914) provides the first descriptions and correct figures of A. latinotus.
He makes no reference to Holmgren’s locality and gives as habitat Angola : Malanga.
This refers to specimens of imagos and soldiers from Angola which Sjéstedt had listed
in his Monographie (1900) under Termes cructfer.

Sjdstedt (1926) considers that the correct attribution of latinotus is to Silvestri.
To remove Jatinotus Holmgren as a nomen nudum would render the name inadmissible
for further use in the genus. This would be undesirable from a practical standpoint,
but it is doubtful if the question does, in fact arise.

Fuller’s Ancistrotermes lembomboensis (1922) from Swaziland, Transvaal and
Mozambique is a synonym which appears to have arisen from an incorrect apprecia-
tion of Silvestri’s measurements of the imago ocellus.

16 Wi oVe HARRIS

Measurements :
Imago : Angola, Malanga Zambia and Tanganyika
A Fea | aes Ae our
Male Female Mean __ Range (16)
(mm.) (mm.) (mm.) (mm.)
Width of head with eyes 1-82 1°82 1°82 I+72-1-86
Diameter of eye 0°45 0°45 0°45 0*4I-0'50
Diameter of ocellus, long 0:18 0°23 0-21 0:18-0:27
Diameter of ocellus, short . 0°13 0°13 O-14 0+13-0-16
Ocellus to eye 0°13 0°13 0°13 0:09-0:18
Width of pronotum 1°64 1°68 1°72 1-60-1 +84
Length of pronotum I-02 1-09 1-02 0+82-1-14
Length of hind tibia . ‘ 1°86 1-86 1°95 1 -86-2-00
Length of fore wing . . 18-0 18-0 57°97 16-0-20°0
Width of fore wing. F 4°0 4°0 4°3 3°8-4°8
Soldiers. Rhodesia to Tanganyika :
Major Minor
' io = FER, ts *. can,
Mean Range (25) Mean Range (10)
(mm.) (mm.) (mm.) (mm.)
Length of head and mandibles a 2:06 1:82-2:27 I*79 1:68-1:90
Length of head ; ; 1°24 I:*OQ-I-4I 0°97 0:86-1-94
Width of head , ; : I*I7 1:00-1:36 0°95 0:77-1:06
Length of left mandible ‘ ‘ 0°87 0:82-0:95 0°83 0:77-0:90
Width of pronotum 0°83 0:73-1:00 0:67 0:59-0°73
Length of hind tibia 1:07 1:00-1:14 0°90 0:82-1:00

Illustrations. The imago head and the head and thorax of the major soldier are
figured in Silvestri (1914). Sjdstedt (1926) has a small figure of the minor soldier
head.

Distribution. Type locality—South-east Conco: Luapula (Duchessa d’ Aosta).

Material examined. ANGOLA: Malanga (Mechow) in Naturhistoriska Riks-
museum, Stockholm.

ZAMBIA: N’changa, 1931 (C. T. Macnamara); Samfya, nr. Lake Bangweulu,
Fort Roseberry, Kitwe, Ndola, Choma, Kafue Bridge, 1957 (W. G. H. Coaton) ;
nr. Tunduma, 1953 (Sands & Wilkinson).

RHODESIA: Chipinga District, 1939 (A. Cuthbertson) ; Trelawney, 1949 (G. H.
Bunz) ; Gokwe, 1962 (M. G. Bingham).

MALAwi: Mlange, 1932 (C. Smee); Karonga, Bua River, Kasungu, Fort John-
stone, Zomba, Lower Shire River, Domasi, 1953 (Sands & Wilkinson) ; Lilongwe,
1959 (W. V. Harris).

SWAZILAND: Lebombo Flats, 1916 (C. Fuller—‘ cotypes’”’ of lembomboensis).

SouTH AFRICA: Messina, 1916 (C. Fuller).

TANGANYIKA : Uvinza, 1934; Kigoma, 1934; Shinyanga, 1934; Muheza, 1934;
Tabora, 1935; Songea, 1935; Kasulu, 1935; Tabora, 1935; Morogoro, 1937 ;
Lindi, 1938; Uzinza, 1938; Usangu, 1939 (W. V. Harris); Handeni, Daluni,

THE GENUS ANCISTROTERMES 17

Ngomeni, 1951, Mkomasi, Mwakijembe, Gombelo, 1952 (P. B. Kemp) ; Nachingwea
(M. Bigger).

KENYA: Kinango, 1952 (P. B. Kemp) ; Ngombeni and Lukongo near Mombasa,
1952 (W. A. Sands).

Additional published records. Conco: Libenge, in Sjéstedt (1926).
CENTRAL AFRICAN REPUBLIC: Bossembelé, in Grassé & Heim (1950).

MOZAMBIQUE : Xinavane nr. Lourenco Marques, in Fuller (1922).
The specimens labelled “‘ Sudafrika, Periguey col.’ in the Stockholm collection
have a further label ““ Kap? Rhodesia? ”’.

Biology. A. latinotus occurs mainly to the south of the equator, and in association
with the Brachystegia—Isoberlinia woodlands or “‘ miombo’’. In north-east Tangan-
yika and adjacent Kenya, where the vegetation is in general the drier Acacza-
Commiphora thicket, A. latinotus is found in isolated patches of miombo on higher
ground.

The nest system is a loose agglomeration of interconnected chambers, roughly
spherical and from 2 to 3 inches in diameter. These chambers contain fungus combs
of a typically convoluted form. Nests have been located by trenching, or during
terracing operations on sloping ground. They have also been found in the outer
walls of mounds constructed in woodland by Macrotermes and Pseudacanthotermes,
and in open grasslands within the miombo by Cubitermes, Crenitermes and Trinervi-
termes.

Workers feed on fallen logs and branches, and construct covered runways on the
trunks of trees in search of dead wood. They are attracted to wooden pegs, fence
posts, etc. They have been recorded attacking Eucalyptus seedlings, tea bushes
and cotton.

The soldiers have a vague odour, likened by one observer to that of the grass
Cymbopogon.

Swarming has been observed at various times between 15.00 hours and sunset,
but mainly between 17.00 and 18.00 hours, during the rainy season (October to April
in eastern Tanganyika). The appearance of winged adults is not restricted to periods
of overcast sky.

Ancistrotermes microdens sp. n.
(Text-fig. 10)

Imago. Unknown.

Soldiers. Major soldier; head broadly oval with flatly curved posterior ; labrum large, as
wide as long, with obtuse tip ; mandibles robust, basally straight, then curving strongly towards
the tip, each mandible with a single median marginal tooth, the right mandible with a short
basal spur ; antennae 14-15 segments, depending on the separation of a short III from IV.

Pronotum with small, pointed anterior lobes separated by a deep notch; posterior margin
lightly sinuate.

Minor soldier ; head ovoid, narrower at the posterior ; labrum large, pointed ; mandibles
slender, straight with in-curved tips, marginal teeth inconspicuous ; antennae 15 segments,
III very short.

18 W.LIV AHAR RES

Pronotum with short, rounded anterior lobes separated by a rounded notch ; posterior margin
almost straight.

Measurements :

Major soldiers Minor soldiers
f os cas | f¥ > YY

Mean Range Mean Range

(mm.) (mm.) (mm.) (mm.)
Length of head and mandibles 1°99 I-9I-2-18 1-63 1°54-1°72
Length of head : 1°26 1:°18-1°41 0:92 0:86-1:00
Width of head : : , I-15 I:*O0Q-1°25 0:91 0-86-0'9I
Length of left mandible : : 0:78 0:73-0:82 0°79 0:*77-0:82
Width of pronotum 0:88 0:+82-0°95 0°71 0:68-0'77
Length of hind tibia I-05 I:00-1'14 0°92 O:'9I-0:95

The major soldier agrees with the published description of A. wasmanni apart
from the “ two very small teeth in the middle of the left mandible ’’. It differs from
latinotus in the reduced marginal tooth on the left mandible and the considerably
deeper anterior marginal notch of the pronotum. A. cavithorax has a more rounded
head, and periphrasis has more strongly in-curved mandibles.

The minor soldier is larger than pertphrasis, with the head proportionately longer.
Compared with /atinotus the mandibles are more strongly curved, the head narrower
at the posterior, and the pronotum distinctly wider.

Holotype, major soldier, KENyA: Kisumu, on lake shore in upper levels of
Macrotermes mound, 18.iv.52 (W. A. Sands), in British Museum (Nat. Hist.).

Paratypes, major and minor soldiers from type colony; KENYA: West Suk,
4.x.52. UGANDA: Toror Hills, Karamoja District, 8.x.52; Moroto, river valley,
7.X.52; Moroto, 5.x.52 (W. A. Sands).

Described from major and minor soldiers in six series from western Kenya and
adjacent Karamoja District of Uganda.

Ancistrotermes periphrasis Sjéstedt
(Text-figs. 6 and 13)

Ancistrotermes periphrasis Sjostedt, 1924, Rev. Zool. Afr. 12 : 496 [minor soldier from the Sudan].

Ancistrotermes periphrasis Sjéstedt ; Sjdstedt, 1926, Rev. Zool. Afr. 14: 149-150 [imago from
Congo].

Ancistrotermes periphrasis Sjéstedt ; Sjéstedt, 1926, Ent. Tidskr. 47 : 241-242 [major soldier
from Ghana].

This species was established by Sjéstedt for soldiers collected in Kordofan Province,
Sudan. Later the imago was described from the Upper Uelé area of the Congo, in
association with similar soldiers. In the same year specimens of two soldier castes
from Ghana were considered to represent the minor soldier, as already described, and
the hitherto unknown major soldier. Only the minor soldier has been figured.

Additional material from Kordofan has now become available, consisting of six
series of both major and minor soldiers. The minor soldiers are similar to the holo-
type of Sjéstedt. The opportunity is taken to add to the original descriptions. It

THE GENUS ANCISTROTERMES 19

has not been possible to examine the major soldier from Ghana, but the minor soldier
is sufficiently different from the Sudan specimens as to caste doubt on this record.

Imago. ‘‘ Close to latinotus, from which it differs slightly but significantly in size, is darker
in colour, and possesses cross veins in the sub-costal area of the wing which are absent in Jatinotus.
The ocelli are much closer to the eyes than in /atinotus ’’.

Other characters mentioned by Sjéstedt are: head oval, dark brown with pale yellow, fairly
large clypeus ; eyes quite circular; ocelli oval, the ratio of the long and short diameters being
8 : 5, the distance between ocellus and eye equal to one quarter of the long diameter ; antennae
with 17 segments ; pronotum with a deep median anterior notch, and posterior margin lightly
indented ; wings with oblique cross veins in the subcostal area.

Measurements of three morphotypes from Iri, Upper Uelé :

(mm.)
Width of head with eyes. é : ; ‘ ‘ 1-86
Diameter of eye : , F : ; ; ; 0°54
Diameter of ocellus, long. ; ‘ ; ° ‘ 0°25
Diameter of ocellus, short 0-15-0°18
Ocellus to eye 0:06-0:08
Width of pronotum 1 -64-1 +72
Length of pronotum I *00-I-09
Length of hind tibia . é : ‘ ; ‘ ’ 1 -86—1-96
Length of fore wing . ‘ ; ; , ; F 20
Width of fore wing. . : ; : : ‘ 4°6

Soldiers. Major soldier ; head orange-yellow, mandibles dark brown, antennae pale yellow.
Pronotum, abdomen and legs pale yellow, with numerous pale yellow bristles.

Head broadly rectangular, with sides and posterior curved ; labrum short and broad with
obtuse tip ; mandibles strongly in-curved at the tip, left mandible with a small median tooth,
below which the margin is straight until a small notch separates it from the large basal process ;
right mandible without a median tooth, but basal spur present ; antennae with 15 segments,
III small, IV and V equal and shorter than II; gulamentum broad with lightly curved sides,
strongly convex in cross-section.

Pronotum with small, pointed anterior lobes and emarginate posterior.

Minor soldier ; head yellow, mandibles brown, antennae pale yellow. Pronotum, abdomen
and legs pale yellow with inconspicuous short, pale bristles.

Head oval ; labrum large with obtuse tip ; mandibles long and slender with slightly in-curved
tips and smooth inner margins ; antennae with 14-15 segments.

Anterior lobes of pronotum not so acutely pointed as in major soldier, posterior straight.

Measurements :

Major Minor

range range

(mm.) (mm.)
Length of head and mandibles . 1 -82-1:-90 I+36-1°41
Length of head I +23-1 +32 0°77-0:82
Width of head I-I1-1-18 0°77
Length of left mandible 0:77-0:86 0-68
Width of pronotum 0-73-0°86 0-59
Length of hind tibia I *04-I -09 0°82-0-86

The major soldier differs from guineensis in having a rounded posterior margin to
the head, and in lacking the rounded anterior lobes of that species. The marginal
tooth of the left mandible is smaller.

20 Wa°Vi BARRIS

Sjdstedt describes the minor soldier as being “ nearest to cavithorax, but with
more strongly in-curved mandibles ”’.

Illustrations. The head and pronotum of the minor soldier are figured by Sjéstedt
in his Revision (1926).

Distribution. Type locality—SupAN: Kordofan Prov., El Amira-Bir Joghan
(R. Ebner), minor soldier and worker, Sjéstedt No. 507 in Mus. Stockholm. Morpho-
type imagos—ConcGo: Uelé Prov., Iri, éclosion dans la pluie, 5 hrs. 21.iv.25 (H.
Schouteden), Sjéstedt No. 517a in Mus. Stockholm, and in Mus. Tervuren, associated
with minor soldiers.

Additional material. SuDAN: Kordofan Prov. Kadugli, vii.52 ; Kadugli-Talodi
road, attacked by Pheidole ants, vi.52 ; El Berdab, attacking cotton plants, xi.52 ;
Jebel Debri, ix.52; Lamma, x.52; Lake Keilad, vi.52; Jebel Miri, vi.52 (C.
Sweeney).

Ancistrotermes wasmanni Snyder & Emerson

Termes cavithorvax Sjostedt; Wasmann, I911, Rev. Zool. Afy. 1: 102 [soldiers from Congo].
Ancistrotermes cavithorax (Sjéstedt) Sjéstedt, 1926, K. Suv. Vetakad. Handl. (3) 3 : 163 [locality].
Ancistrotermes wasmanni Snyder & Emerson in Snyder, 1949, Smithson. Misc. Coll. 112 : 247.

A series of soldiers and workers from the Congo was identified by Wasmann as
Termes cavithorax, with the comment that they agreed with Sjéstedt’s description
apart from the mandibles of the major soldier being “ strongly bent and sharp
pointed, the right one without teeth, the left with two very small teeth in the middle,
the hind one only visible under magnification’. Sjdéstedt includes this locality in
his account of Ancistrotermes cavithorax in his Revision.

Snyder & Emerson consider the series to represent a new species, to which they
give the name wasmanm, without further discussion, in Snyder’s Catalog.

It has not been possible to examine Wasmann’s specimens in this study.

Measurements. Wasmann gives the following measurement :

Major Minor

soldiers soldiers

(mm.) (mm.)
Total length . ; : : ‘ . ; 4°0-4°3 2°5-3°6
Length of head and mandibles . ; ‘ : 1-8-2-0 I+2-1°6

Illustrations. None.
Type locality. ConGo: Sankuru, in Wasmann (1911).

ACKNOWLEDGEMENTS

I am indebted to Prof. L. Brundin and Dr. P. I. Persson of the Naturhistoriska
Riksmuseum, Stockholm and Dr. P. Basilewsky of the Musée Royal d’Afrique
Centrale, Tervuren for the loan of type material, and to Dr. W. G. H. Coaton,
Pretoria for specimens from South Africa and Zambia. My thanks are due to the
Trustees of the British Museum (Nat. Hist.) for providing facilities for this study.

A LIST OF SUPPLEMENTS
TO THE ENTOMOLOGICAL SERIES
OF THE BULLETIN OF
THE BRITISH MUSEUM (NATURAL HISTORY)

. Masner, L. The types of Proctotrupoidea (Hymenoptera) in the British
Museum (Natural History) and in the Hope Department of Entomology, Oxford.
Pp. 143. February, 1965. £5.

2. Nixon, G. E. J. A reclassification of the tribe Microgasterini (Hymenoptera :

Braconidae). Pp. 284; 348 Text-figures. August, 1965. 6.

3. Watson, A. A revision of the Ethiopian Drepanidae (Lepidoptera). Pp. 177;

18 plates, 270 Text-figures. August, 1965. £4 4s.

4. SANDS, W.A. A revision of the Termite Subfamily Nasutitermitinae (Isoptera,

Termitidae) from the Ethiopian Region. Pp. 172; 500 Text-figures. October,

1965. £355.

5. AumaD,I. The Leptocorisinae (Heteroptera: Alydidae) of the World. Pp. 156;

475 Text-figures. November, 1965. {£2 15s.

6. Oxapba,T. Diptera from Nepal. Cryptochaetidae, Diastatidae & Drosophilidae.

In press.

7. GILIOMEE, J. H. Morphology of Adult Male form, Coccidae (Homoptera:

Coccoidea). In press.

PRINTED IN GREAT BRITAIN BY ADLARD & SON LIMITED, BARTHOLOMEW PRESS, DORKING

ge >
27 JUN 1966
REVISIONAL NOTES Ly SS

ON THE GENUS ANTANARTIA
(LEPIDOPTERA : NYMPHALIDAE)

T. G. HOWARTH

BULLETIN OF
THE BRITISH MUSEUM (NATURAL HISTORY)
ENTOMOLOGY Vol. 18 No.2
LONDON: 1966

REVISIONAL NOTES
ON THE GENUS ANTANARTIA
(LEPIDOPTERA : NYMPHALIDAE)

BY

T. G. HOWARTH _\_

British Museum (Natural History) )

Pp. 21-43 ; 21 Text-figures, 5 Plates

BULLETIN OF
THE BRITISH MUSEUM (NATURAL HISTORY)
ENTOMOLOGY Vol. 18. No. 2
LONDON: 1966

THE BULLETIN OF THE BRITISH MUSEUM
(NATURAL HISTORY), instituted im 1949, 1s
issued in five series corresponding to the Departments
of the Museum, and an Historical series.

Parts will appear at irregular intervals as they become
veady. Volumes will contain about three or four
hundred pages, and will not necessarily be completed
within one calendar year.

In 1965 a separate supplementary series of longer
papers was instituted, numbered serially for each
Department.

This paper is Vol. 18, No. 2 of the Entomological
series. The abbreviated titles of periodicals cited
follow those of the World List of Scientific Periodicals.

© Trustees of the British Museum (Natural History) 1966

TRUSTEES OF
THE BRITISH MUSEUM (NATURAL HISTORY)

Issued 21 June, 1966 Price {1 6s.

REVISIONAL NOTES
ON THE GENUS ANTANARTIA
(LEPIDOPTERA : NYMPHALIDAE)

By T. G. HOWARTH

CONTENTS
Page
INTRODUCTION . 2 : : : ; : E , : : 23
KEY TO SPECIES, SUBSPECIES AND FoRMS . ‘ é é : ; 25
DESCRIPTIONS . , ‘ ; : “ d ‘ i ; : 26
ABNORMAL SCALING . P , : : : : ; , , 36
PARALLEL MODIFICATION . : 2 : : : : : é 37
ACKNOWLEDGEMENTS é , ‘ F : : ; : 3 38
REFERENCES . : : ; ; ; ; é ; : : 38
SYNOPSIS

In the present paper one new species, consisting of four subspecies, is described together with
four new subspecies and a possible interspecific hybrid. One name has been synonymized and
attention is also drawn to the existence of two clines. All the known species and subspecies are
figured together with figures of the male and female genitalia, and a key to the species, subspecies
and forms is given. Notes on scale defects and parallel modification are also given.

INTRODUCTION

THE genus Antanartia Rothschild & Jordan (1903 : 508) (Type: delius Drury) was
erected to include the five African, Malagasy and Mascarene species which were
originally placed in Hypanartia Hiibner, which now contains only New World species.
The species of Antanartia inhabit Africa from Sierra Leone through the West Coast
to Ethiopia, southwards to Cape Province and then eastwards to the Comoro Islands,
Madagascar and the islands of Réunion and Mauritius. In the tropics most of the
species are montane in habit, preferring altitudes of between 3,000 and 13,000 feet.
The present investigation was begun when some material was sent to the Depart-
ment of Entomology, British Museum (Natural History) for determination and it
was realized that there were several previously unrecognized subspecies present
amongst the museum material. In the course of this investigation all the characters
mentioned by Rothschild & Jordan were examined with a view to placing the species
in a more natural phylogenetic order than that given by Aurivillius (in Seitz, 1913).
When Rothschild & Jordan separated Antanartia from Hypanartia one of the
characters they mentioned was the presence, in the male of Hypanartia, of a mesial
hook on the eighth abdominal tergite. This structure, named the superuncus by
Kusnezov in 1915, is unusual in Rhopalocera but it is extremely well developed in
ENTOM. 18, 2 2

24 T. G. HOWARTH

Hypanartia. In Antanartia there are several characters present that are not in the
American genus. In Antanartia, the uncus is slightly bifurcate, but not deeply chan-
neled asin Hypanartia and the brachia below it are well developed. The valvae are
unusual in that they possess an inner and outer harpe, the latter, arising basad of the
outer margin, is a massive structure strongly armoured with large teeth and spines.
It is held in position by an outer flange of the cucullus and a lobe or lobes along the
costal or dorsal edge of the valva. The flange and lobes have numerous strong setae
which help to retain the harpe in position. In one species, abyssinica (Felder), the
outer harpe is Y-shaped with the apex widely bifurcated. In some cases the aedeagus
has two barbs ventrally towards the tip and in others these are absent. The saccus
is reduced or absent and the vinculum is produced ventrally and posteriorly so that
the whole genitalia lie well within the body cavity when at rest.

The female genitalia are unusual in that they have the sterigma or intersegmental
membranes of the abdominal sternites adjacent to the ostium heavily chitinized and
in the form of pouches. It seems probable that these have been developed to prevent
the accidental rupture of the abdominal wall by the heavily armoured harpes of the
male during copulation.

In both sexes the genitalia can be divided into two groups of three species, those
that have the barbed aedeagus in the male together with those without the signa on
the ostium bursa of the female, namely delius (Drury), schaeneia (Trimen) and
borbonica (Oberthiir), and those that have an unbarbed aedeagus in the male and a
well-developed signa in the female, namely hippomene (Hiibner), the newly described
dimorphica and abyssinica (Felder).

Rothschild & Jordan when dealing with dimorphica sp. n. (as hippomene
Hiibner), schaeneia (Trimen) and abyssinica (Felder) drew attention to the variation
in the small area of modified scales situated on the underside of the fore wing near the
base of space Ia in both sexes. This character, like the genitalic characters men-
tioned above, tends to illustrate the same specific relationships so that the six species
divide into two, the delius group and the hippomene group of species. In the first
the specialized scale area is confined to space 1a by the vein Ia but in the second it
extends beyond the vein into space 1b above it, though abyssinica (Felder), being
more variable, tends to link the two groups.

The critical areas of the venation in the genus are situated in the apical area of the
discoidal cell of the fore wing and in the area of veins 3 and 4 of the hind wing and
these have been figured for each species. Further remarks on the aberrant nature of
the venation of Antanartia borbonica (Oberthiir) may be found under that species
(p. 30).

All the type material is in the British Museum (Natural History) except where
otherwise stated.

In the descriptions the length of the fore wing is measured from the base to the
apex and where two measurements are given, these denote the smallest and largest
specimens in the series. In order to save unnecessary repetition when there is a
long series of paratypes only the more important data i.e., the locality, altitude and
date are given with an overall figure for the country concerned.

NOTES ON THE GENUS ANTANARTIA

KEY TO THE SPECIES, SUBSPECIES AND Forms OF ANTANARTIA, BASED ON THE

La

nn

—
Lan!

~—

WING MARKINGS

Upperside fore wing with no clearly defined yellow/red transverse band from
mid-costa to near tornus

Upperside fore wing with clearly defined transverse band F

Upperside hind wing pale discal area not suffused with dark red or brown

Upperside hind wing pale discal area suffused with dark red or brown

Upperside fore wing cell not darkened beyond central bar . delius delius (p.

Upperside fore wing cell darkened beyond central bar

delius delius f{. nigrescens (p.
Underside fore wing transverse area pale. ‘ . delius guineensis (p.

Underside fore wing transverse area dark, red-brown or brown

delius delius f. amauroptera (p.

Hing wing with no tail at vein 4
Hind wing with tail at vein 4
Upperside fore wing cell bar broken and darker than base

abyssinica vansomereni (p.

Upperside fore wing cell bar not broken and not darker than base .
Upperside hind wing ocellus in space 2 smaller than those in 3 and 4, often

blind, marginal band abruptly broader in space 5 abyssinica abyssinica (p.

Upperside hind wing ocelli in spaces 2-4 more equal, that in 2 with pupil,

marginal band gradually widened in space 5 . . abyssinica jacksoni (p.

Upperside hind wing with well defined post discal black line .
Upperside hind wing without post discal line ;
Upperside hind wing band narrow and divided by dark interneural lines, Q upper-

side fore wing transverse band cream ; : schaeneia diluta (p.

Upperside hind wing band wider and interneural lines inconspicuous, ? upper-
side fore wing transverse band not cream :
Underside fore wing inner edge of transverse band stepped at vein 2 and not

at vein I 3 ; . schaeneia schaeneia (p.

Underside fore wing | inner ‘edge of transverse band not stepped at vein 2 but

stepped at veinI . : : schaeneia dubia (p.

Underside fore wing with broad violet submarginal line from vein 6 to tornus
Underside fore wing without violet submarginal line
Upperside fore wing post discal spots in spaces 5 and 6 white

borbonica borbonica (p.

Upperside fore wing post discal spots in spaces 5 and 6 suffused with orange

borbonica mauritiana (p.

Underside fore wing basal cell markings outlined in bluish white or cream
Underside fore wing basal cell markings outlined in brown or red .
Hind wing tail at vein 4 short (+ 3 mm.) and broad. Upperside bands ochre

hippomene hippomene (p.

Hind wing tail at vein 4 long (+ 4 mm.) and thin. Upperside bands orange

hippomene madegassorum (p.

Upperside bands orange-ochre

Upperside bands deep reddish orange . : ; dimorphica mortoni (p.

Upperside hind wing inner black marginal line not terminating at vein 3 .
Upperside hind wing inner black marginal line terminating at vein 3

dimorphica comoroica (p.

Upperside hind wing with dark submarginal spots or ocelli in spaces 4-6

dimorphica aethiopica (p.

Upperside hind wing with no dark submarginal spots in spaces 4—6

dimorphica dimorphica (p.

25

29)

26 T. G. HOWARTH
The DELIUS Group

Antanartia delius delius (Drury)
(Pl. 1, figs. 1, 2,3,4 and5. Text-figs. 1, 8 and 15)

Papilio Nymphalis Phaleratus delius Drury, 1782 : 18, pl. 14, figs. 5, 6.
Papilio Eurocilia Fabricius, 1793 : 79, No. 247.

Vanessa Demonica Godart, 1819 : 301, No. 14.

Antanartia delius {. khamitugensis Dufrane, 1945 : 99 Syn. n.

This, the type-species of the genus, is also the most variable and ranges from Sierra
Leone in the west, to Uganda, Kenya and Tanzania in the east. When arranging
the long series in the collection of the British Museum (Natural History) according
to distribution, it was noticed that specimens from the eastern part of the range
differed from those from further west and it was at first thought that the former
might well represent an undescribed subspecies. However it was found that Dufrane
(1945) had already separated a male specimen from the rather intermediate area of
Kivu and named it kamitugensis, which necessitated a closer examination to determine
the limits of these two forms or races and it was discovered that a rather ill-defined
cline rather than two distinct races was involved. Prior to this, all the specimens
had been measured and averaged and the results are given in the following Table.

Fore Wing Length from Base to Apex in Millimetres

Males Females
| re “~ cm | f- A rao
Max. Min. Average Max. Min. Average
W. AFRICA
Sierra Leone—Congo eae. * | 24 28°3 36 26 32°8
104 6, 35 2
E. AFRICA
Uganda—Tanzania . c. £90. 23 26°5 30 23 29
72 3; 229

It will be seen from these that there is a marked difference in size from west to east.
The western specimens have, on the upperside of the fore wing, the subapical and
submarginal series of white spots usually well developed. On the underside of the
fore wing the blind pupil of the indistinct ocellus in the discal area of spaces Ia and 1b
is broadly edged with bluish white, the transverse discal area is yellowish and the
basal cell spot often has no red-brown pupil and the red-brown centre of the central
cell bar tapers and does not usually reach the median vein. The specimens from
Uganda and further east are generally smaller and darker and usually lack the white
subapical spots on the fore wing upperside except that in space 4 which is usually
present. On the underside of the fore wing the basal cell spot has a distinct reddish
pupil and the central cell bar is broadly divided with the same colour. The discal
markings towards the hind margin in spaces ra and rb are more distinct and usually
reach vein 2, the outer and inner lines are more yellow and the yellowish transverse
discal area is often suffused with orange at the base of space 2 and the adjoining

NOTES ON THE GENUS ANTANARTIA 27

interspaces. There is a great deal of individual variation throughout this insect’s
range, the males being usually darker than the females. On the upperside of the
fore wing, specimens frequently have the dark basal markings of the first interspace
suffused with brown which extends beyond half-way to the tornus and into the cell and
adjacent interspaces, sometimes even obscuring the black cell markings and coalescing
with the black apical area. An extreme example of this type of variation is form
nigrescens Suffert (1904 : 108) in which the fore wing is suffused with blackish brown
with only a half crescent of orange present in the discal area from the inner margin to
vein 3. On the hind wing upperside the black basal suffusion is almost confluent
with the black discocellular bar. On the underside of both wings the markings are
not noticeably darkened. In Uganda a different type of variation occurs and affects
the paler markings of both wing surfaces. In form amauroptera Sharpe (1904 : 181)
(Pl. 1, fig. 5) the yellowish orange markings of the upperside of both wings and the
underside of the fore wing are darkened to a deep brownish red or burgundy with the
darker markings showing through, except adjacent to the costa of the mid-cell of
the underside of the fore wing, where it is cream. There is another form transitional
to the above in which the deep velvety red-brown of the disc of the upperside of the
fore wing appears darker than the plain brown of the apical and marginal areas.
There is an even more extreme form that has the whole of the upperside of both
wings an almost unicolorous deep brown, with the exception of the marginal markings
of the hind wing, which remain quite distinct. On the underside, the discal area of
the fore wing is a dull brown and both wings, with the exception of the basal and
cell bars and spots, lack the rufous tone present in form amauroptera.

Antanartia delius guineensis ssp. n.
(Pl. 1, figs. 6 and 7)

In general appearance this subspecies is nearer to the Ugandan specimens than to
those from the West Coast mainland.

3g. Upperside. Fore wing length 26-30 mm. The orange-red transverse area suffused
and darkened from the median vein towards the hind margin, which is the reverse of the colora-
tion of the western mainland specimens, in which the costal area is darkened. Both series of
white spots in the apical area distinct ; the red discal area of the hind wing much darkened
basally, so that the black-brown discocellular bar and the costa are hardly discernible. One
specimen has the whole basal area as far as the inner submarginal line a uniform brown. Under-
side. Fore wing very similar to the typical Ugandan specimens having the yellowish transverse
discal area tinged with orange and the black basal markings in space rb and the cell outlined in
brilliant lilac. There is a diffuse patch of lilac beyond the cell and the inner submarginal line in
spaces 3 and 4 is edged with this same colour. Hind wing markings generally more distinct and
more suffused with lilac discally than in specimens from Uganda.

Q. Unknown.

Holotype 3. FERNANDO Po (Hewitson coll.).

Paratypes g. 1, same data as holotype; r, Fernando Po (Godman-Salvin coll.) ;
1, Fernando Po (L. Frazer coll.).

28 T. G. HOWARTH

Distribution. This subspecies is confined to the island of Fernando Po in the
Gulf of Guinea, though there is another male which is indistinguishable from it
labelled “‘Angola’’ (Grose-Smith coll. ex. Joicey Bequest) which may probably be
wrongly labelled.

A probable interspecific hybrid between Antanartia delius (Drury)
and Antanartia schaeneia (Trimen)

(Pl. 2, figs. 8 and g. Text-fig. ro)

A female specimen of Antanartia was received from Mr. T. H. E. Jackson of Kenya
which was taken at Budongo, Bunyoro, Uganda in June, 1938 by C. Cripps. On the
upperside, this specimen resembles a dark example of schaeneia superficially, as it
has the sharply defined inner margin to the transverse orange band of the fore wing
and the hind wing completely brown as far as the orange margin so characteristic
of schaeneia. The orange bands however are slightly deeper in colour and are nearer
to delius in this respect, that of the hind wing has the dark interneural submarginal
bars which both species have, though they are more prominent in schaeneia. In
space Ib of the fore wing there is a prominent dark spot in the orange band which
neither species has. On the underside however this spot is even more noticeable as it
is larger and surrounded with ochre, while the remainder of the band is brownish in
coloration similar to that only found in delius form amauroptera Sharp from Toro.
It is interesting to note that one or two specimens of delius also have a trace of a
spot in the above mentioned position. Apart from the coloration of the fore wing
band, the underside is indistinguishable from that of schaeneia. Antanartia delius
and schaeneia fly together in Uganda and their genitalia appear sufficiently close as
not to preclude the possibility of a successful mating and as the genitalia of this
specimen appear mid-way between the two species I conclude that it may possibly
be the result of a feral cross between them.

Antanartia schaeneia schaeneia (Trimen)

(Pl. 2, figs. 10, 11 and 12. Text-figs. 2, 9 and 16)

Pyrameis Hippomene (Hiibner) Trimen, 1862: 121. [Misidentification from Boisduval,
1833 : I9T].

Eurema schaeneia Trimen, 1879 : 329.

Hypanartia commixta Butler, 1880 : 336.

The nominate subspecies occurs in South Africa in Cape Province, Natal and as
far north as Transvaal. According to Swanepoel (1953) it inhabits rain-forests,
preferring the higher elevations where mists are prevalent. Specimens from still
further north in Rhodesia, Zambia, Tanzania, Kenya and Uganda belong to a distinct
subspecies,

NOTES ON THE GENUS ANTANARTIA 29

Antanartia schaeneia dubia ssp. n.
(Pl. 2, figs. 13, 14 and 15)

3g. Upperside. Fore wing length 21-29 mm., similar to nominate subspecies but the orange
transverse band slightly broader and the inner edge sometimes stepped at vein 1. Hind wing,
the marginal orange band between vein 4 and apex not smudged at apex and with the submarginal
shading not as distinct, particularly in spaces 4 and 5. Underside. Fore wing, the inner edge
of pale transverse band appears slightly smoother centrally, due to the outer dark bar in space 2
barely reaching vein 2. However at vein 1 it is stepped or cut off abruptly, whereas in the
nominate subspecies the dark bar in space 2 is shaded outwardly and diagonally towards the
tornus and vein 1, so that the inner edge of the transverse band is straighter in this area. Hind
wing, similar to ssp. schaeneia but slightly more rufous.

g. As male but markings slightly broader and paler generally on both upper and underside.
Fore wing length 25—30 mm.

Holotype 3. MALAwi: Nyika, Kasungu Mt., 7,200 ft., 4.iii.96 (R. Crawshay).
Allotype 2. MALaAwi: Slopes of Mt. Mlanje, vi.1913 (S. A. Neave).

Paratypes : REPUBLIQUE DU CONGO, 3 g: Kirisimbi Volcano, Kivu District, ix ;
W. slopes of Ruwenzori, xii, 2,500 m.; Forest 90 km. W. of Lake Albert-Edward,
I,000 m. ii. UGANDA, 36 3,62: Ruwenzori District ; E. Ruwenzori, 6—13,000 ft.,
ii; W. side Bwamba Pass, 5,500—7,500 ft., xii-i; N. Ruwenzori, 6,000-8,500 ft., xi ;
Mt. Ruwenzori, 6,000—9,000 ft., ii; Namwamba Valley, 6,500 ft., xiii ; Fort Portal,
ii; Kigezi district, Mafuga Forest, 7,500-8,500 ft., iti; Rutenga Forest, vi;
Kanaba Gap, vi; Mt. Kokanjero, S.W. of Elgon, 6,400 ft., viii ; Subugo Forest, xii ;
Kakamega Forest, Mau, xii. RWANDA, I 4, I 2: Rugege Forest, Lake Kivu,
8,000 ft., xii; Mkoko River, iv. KENyA, 82 3, 13 2: E. Slopes Aberdare Mts.,
7,000-8,500 ft., ii; Yala River, Kakunga Forest, 4,800—-5,300 ft., v; S. Foot and
Bridge ; Meru district, Mt. Kenya, ix; Mt. Kenya, 9,500 ft.,i; Mt. Kenya, Naro
Moro, 8,000 ft., viii ; Subukia, xi; Nguru Hills ; Nyeri; Kikuyu, Roromo ; Nairobi ;
Patsho, Nandi Country, xii; Rau, Nandi Country, ii. TAnzanta, 8 3g, 3 9: W.
Kilimanjaro, 4,500—5,000 ft., xii-ii ; Mt. Kilimanjaro, Marango, ii ; Magazine Hill, v.
MALAWI, 7 3, 122: 3 35, 8 9, same data as allotype ; Mlanje Boma, 2,400 ft., iv—v ;
Mlanje Plateau, 6,500 ft., xi; Nyankowa Mt., W. of Lake Nyasa, 6,500 ft., iv ; Lake
Shirwa, Chikala. RHopEsIA, 7 g, 4 9: 1 g, Umtali, iii; 1 3g, Laurenceville,
Vumba, v; I 4g, 2 9, Mt. Selinda, ii, v, ix; [B.M. (N.H.)]. 1 3, 2 9, Mt. Selinda,
ix ; I §, Vumba Mts, ii; [National Museum, Nairobi]. 2 ¢, Vumba Mts, v; [B. K.
West coll.].

Distribution. This is a montane subspecies, generally inhabiting elevations over
4,000 feet in the Kivu district of the République du Congo, Uganda, Rwanda, Kenya,
Tanzania, Malawi and Rhodesia.

Antanartia schaeneia diluta Rothschild & Jordan
(Pl. 3, figs. 16 and 17)

Vanessa schoeneia [sic] (Trimen) Oberthiir, 1883 : 723.
Antanartia schaeneia diluta Rothschild & Jordan, 1903 : 510.

30 T. G. HOWARTH

This very distinct subspecies is characterized by the normal transverse orange
band of the fore wing and the marginal band of the hind wing upperside being much
narrower and paler in the male and almost white in the female.

Antanartia borbonica (Oberthiir)

This beautiful species, which is confined to the Mascarene Islands, is wrongly
stated by Aurivillius (in Seitz, 1913 : 228) to occur also in Madagascar. Though
faintly resembling the previous species or even hippomene madegassorum (Aurivillius)
in appearance, it is in fact a very distinct species which has evolved a marked
venational difference from the other members of the genus, due no doubt to the
extreme isolation of its habitat. Rothschild & Jordan (1903 : 509) in their definition
of the genus Antanartia state “‘ The third subcostal branch of the fore wing stands,
moreover, much farther from the cell than in Hypanartia, The cell of the hind wing
is closed, the cross-vein standing distally of the point of the origin of M!, while it is
placed opposite M? or proximally of it in Hypanartia dione, kefersteini, lindigi etc.”
The first part of this statement cannot apply to either genus as the third subcostal
branch (vein g) and also vein ro arise near the end of the cell in all species of both
Antanartia and Hypanartia, except borbonica, which is unique in having veins 9 and
10 on a common stalk from the cell-end. The second part, relating to the cross-vein,
is only partly correct in that only Hypanartia arcaei (Salvin) has the vein placed
proximally, with the exception of borbonica, which has the cross-vein standing well
beyond the point of origin of veins 3 and 4. From these facts it would appear that
Rothschild & Jordan based their venational characters for the genus Antanartia on
those of borbonica, the one aberrant member.

Antanartia borbonica borbonica (Oberthiir)
(Pl. 3, fig. 18. Text-figs. 3, rr and 17)

Vanessa hippomene (Hiibner) Boisduval, 1833 : 191, pl. 8, figs. 3, 4.
Vanessa borbonica Oberthiir, 1880 : 164.

This, the nominate subspecies, is confined to the island of Réunion.

Antanartia borbonica mauritiana Manders
(Pl. 3, fig. 19)

Antanartia hippomene mauritiana Manders, 1908 : 437.
Antanartia borbonica mauritiana Manders; Aurivillius (in Seitz), 1913 : 228.

This subspecies, which has only been taken on the island of Mauritius and which
may now be extinct, is easily separable from ssp. borbonica by its smaller size and the
orange coloration of the post-discal spots in spaces 4-6 of the upperside of the fore
wing, which are white in the nominate subspecies, and the inner edge of the orange
transverse band of the same wing surface, which is sinuate in mauritiana and straight
in ssp. borbonica.

NOTES ON THE GENUS ANTANARTIA 31
The HIPPOMENE Group
Antanartia abyssinica (Felder)

This species is the smallest of the genus and is distinguished by its lack of a tail at
vein 40n the hind wing. On examination of a long series from the whole of its known
distributional areas it was noticed that there are three distinct subspecies involved.

Antanartia abyssinica abyssinica (Felder)
(Pl. 3, fig. 20. Text-figs. 4, 12 and 18)

Pyvameis abyssinica Felder, 1867 : 397, 589.
The nominate subspecies is confined to Ethiopia.

Antanartia abyssinica jacksoni ssp. n.
(Pl. 3; fig. 21)

6, 2. Similar to the nominate subspecies but has the submarginal ocelli of the upperside of
the hind wing more equal in size and that in space 2 usually with a pupil, moreover the ochre
submarginal band is not broadened so noticeably in space 5. Fore wing length g, 17-21 mm.,
9, 20-23 mm.

Holotype 3. KeEnya: Mt. Elgon, vii.1937 (T. H. E. Jackson).
Allotype 9, same data as holotype.

Paratypes: KENYA, 60 3g, 24 2: Aberdare Mts, 7,000-9,500 ft., ii; Mt. Elgon,
5,100-5,800 ft., vi, vii, x; Mt. Kenya, 4,500-8,500 ft., ii, ix ; Crater Lake, N.W. of
Meru, 5,000-7,000 ft., 11; Masai Reserve, Mara River, v; Hoey’s Bridge, 6,200 ft. ;
Embi; Patsho, Nandi Country, xii; Nairobi; Escarpment, 6,600-9,000 ft., 1, ix,
x, xl, xu; Kikuyu Escarpment, Kijabe to Limoru, Uganda R., 6,800-7,400 ft., 111 ;
Lake Nakuru, 6,100 ft., ix ; Eldoma Ravine, iii; Yambeni Range, Mt. Kenya Dist.,
3,000 ft., x11; Nyeri,i; Kikuyu, Roromo. TANZANIA, 13 ¢, 4 2: Old Moschi, iv ;
Ngorongoro Crater, Arushu Dist., 5,800—7,800 ft., ii; W. Shore of Lake Manyara,
ii-v ; Meru, 7,000 ft., xii, ii; W. Kilimanjaro 4,500-5,000 ft., xlici; Edge of
Olomoti Crater, Arushu Dist., 10,000 ft., ii.

Distribution. Apparently widely distributed in Kenya and Tanzania in suitable
localities at altitudes from 3,000—10,000 feet.

Antanartia abyssinica vansomereni ssp. n.
(Pl. 3, figs. 22 and 23)

6, 9. Easily distinguished from the other two subspecies by the upperside of the fore wing
having the orange-ochre transverse band much broader and the dark basal area more tawny,
so that the black central cell-bar, which is broken in this subspecies, is more conspicuous. The
marginal band of the hind wing is broader and almost encloses the distinct spots and ocelli at its
inner edge. Fore wing length, 3, 19-22 mm., 2, 20-24 mm.

Holotype 3. REPUBLIQUE DU Conco: Upper Oso River, N.W. Kivu, 4,000 ft.
Forest with some grass, ii.1g24 (T. A. Barns).

32 T. G. HOWARTH

Allotype 2. Same data as holotype.

Paratypes: UGANDA: 7 g, I 9, E. Ruwenzori, 5,000-13,000 ft., ii. 1906 (G. Legge
& A. F. R. Wollaston) ; 1 2, Ruwenzori, 6,000-8,000 ft. (Scott Elliot) ; 2 3, Toro,
xi-xii.1900 (H. B. Rattray). REPUBLIQUE DU CONGO: 4 3, I 9, Mt. Niragongwe,
N. of Lake Kivu, I900-3,000 m., ix.1907 (R. Grauer) ; I 3, Karissimbi Forest, Kivu,
1919 (T. A. Barns) ; 2 3, Mikeno Mt., N. Kivu, x.rg19g (T. A. Barns) ; 1 3, Bukeyei,
6,000 ft., 21.v.1926 (fF. G. Jackson) ; 1 2, Congo, 1926 (F. G. Jackson) ; 3 2, Kitanga;
3 3d, I 9, Marienseen to Issawi, I,000 m., vii.1907 (R. Grauer); 1 3, Nr. Kisenyi,
N.E. Shore of Lake Kivu, 3,600—5,000 ft., vii, vili.1926 (F. G. Jackson). WANDA:
3 6, I 9, Kissenyi, Lake Kivu, x.1907 (R. Grauer).

Distribution. This subspecies seems confined to the Kivu and adjacent areas of
the République du Congo, western Uganda and Rwanda. There is one specimen
labelled ‘‘ Mkoma Mt. South Urindi Dist., E. Tanganyika, T. A. Barns ” the accuracy
of which seems extremely doubtful, as this locality is about one thousand miles away
from the headquarters of vansomerent. Inthe Ruwenzori Range it has been recorded
up to 13,000 feet.

These two newly described subspecies are named after Mr. T. H. E. Jackson and
Dr. V. G. L. van Someren of Kenya, both of whom have done so much for the study of
African Rhopalocera.

Antanartia hippomene (Hiibner)

Up to the present time hippomene has been thought to be a single species inhabiting
the African continent from Ethiopia to South Africa and the West Cameroons area,
with a subspecies in Madagascar. A closer examination of the extensive series in the
British Museum (Natural History) has shown however that only the South African
and Malagasy populations should be attributed to hippomene and that the others
belong to an undescribed species with four races, one in Ethiopia, another in Central
and East Africa as far south as Rhodesia and the Transvaal, another in West
Cameroon, Nigeria and Fernando Po and yet another in the Comoro Islands.

Hiibner’s figure of this species leaves no doubt as to its identity but the brief
description by Aurivillius [in Seitz, 1913 : 228, pl. 52 (d)] is of little diagnostic value
and he figures the newly separated species dimorphica not hippomene. It should
be pointed out that in the second line of his description the word “‘ forewing ’’ should
read “ hind wing ’”’.

Antanartia hippomene hippomene (Hiibner)

(Pl. 3, figs. 24 and 25. Text-figs. 5, 13, 19)
Hypanartia hippomene Hiibner, 1823: 2, pl. 25.
Eurema hippomene (Hiibner) Trimen, 1887 : 204.
Antanartia hippomene (Hiibner) Swanepoel, 1953 : 209, pl. 11, fig. 19 [in colour].

3, . Characterized by the palpi being cream ventrally with sparse black hairs situated out-
wardly in this area. Fore legs cream with a narrow brown stripe on the front inner surface,
the long hairs of the male in this area cream. Antennal shaft brown above, whitish below and
at the base of club, which is blackish brown with a reddish tip. Upperside. Fore wing has

NOTES ON THE GENUS ANTANARTIA 33

outer margin sinuate and bent inwards at spaces 3 and 4, which gives the apical area a slightly
falcate appearance. The yellowish ochre transverse discal band tapers sharply to a blunt point
and only just touches the hind margin. There is a small white spot usually present in space 3.
The coloration of the cell-base not noticeably different from the remainder of the wing base.
Hind wing, black brown basal area extends along the costa to apex and into the yellow ochre
marginal band along the veins, so that the inner edge of band is scalloped. The two ocelli in
spaces 2 and 3 are well developed, that in space 2 tending to have a yellowish orange ring round
it, which is more noticeable outwardly. The two marginal lines in the tornal area broad, the
inner extending to vein 4 and joining that of the rather blunt tail. Underside. Fore wing,
transverse discal band cream slightly tinged with orange basad, the dark basal marks in cell
outlined in cream, with a small cream triangular spot at extreme base of cell. The sexes are
very similar on the underside. Fore wing length g, 22-26 mm., 9, 23-30 mm.

Distribution. The nominate subspecies is confined to suitable forested areas in
South Africa from the Cape to the Transvaal.

Antanartia hippomene madegassorum (Aurivillius)
(Pl. 4, figs. 26 and 27)
Hypanartia hippomene var. madegassorum Aurivillius, 1898 : 129.

3,9. Similar to ssp. hippomene but the antennae are reddish brown throughout, outer margins
of fore and hind wings more wavy and the tails of hind wing longer and more sharply pointed.
Upperside. Fore wing, central cell bar separated from base by small patch of orange, and
transverse discal band this same colour. Hind wing marginal band also orange. Underside.
Fore wing, ground colour darker particularly towards apex but transverse discal band ochre
and not orange, as in nominate subspecies. The sexes are similar in coloration. Fore wing
length g, 23-25 mm., 9, 25-27 mm.

Distribution. Confined to the island of Madagascar and now believed to be

extremely rare. Specimens from Beforo and S. Betseleo in British Museum (Natural
History).

Antanartia dimorphica dimorphica sp. n.
(Pl. 4, figs. 28, 29, 30, 31, 32 and 33. Text-figs. 6, 14 and 20)

This species may be separated from hippomene by its more truncate fore wings,
particularly in the female, and the deeper orange colour of the central part of the
transverse band on the underside of the fore wing. The species exhibits considerable
sexual dimorphism on the underside of the hind wing, the female resembling
hippomene madegassorum in tone and the male being even darker.

6. Palpislightly more hirsute than hippomene and the black hairs longer ; fore legs browner,
with brown and cream hairs inwardly, which give the leg the appearance of having a much
broader stripe. Upperside. Fore wing, as hippomene but with less falcate apex and the orange
transverse band not tapering to a point at hind margin. Usually without a small white spot in
submargin of space 3. The base of the cell rufous-brown in contrast to the distinct black cell bar
and the remaining basal coloration. Hind wing, the yellow orange marginal band extends to
the apex, which is not darkened as in hippomene, and the inner edge generally straighter ; inner
black submarginal line of tornus not often joined to black line of tail at vein 4. Underside.
Fore wing, as hippomene but ground colour darker and the transverse discal band broader and
orange as are the basal striae of the cell. There is no pale spot at the wing base. Hind wing,
darker than hippomene with more of a purple gloss and the broad submarginal area has a smoother
appearance.

34 T. G. HOWARTH

Q. Similar to the male but wings slightly broader and the margin of hind wing slightly more
wavy and the underside of hind wing considerably paler. Fore wing length ¢ 20-27 mm.,
Q, 21-26 mm,

Holotype g. UGANDA: Ruwenzori Range, xii.1934-i1.1935. B.M. E. Afr. Exp.
Namwamba Valley, 8,300 ft. (7. H. E. Jackson).

Allotype 2, same data as holotype but taken at 6,500 ft.

Paratypes, I g, I 2 with same data as allotype. A large series of 122 9, 55 2 from
the following localities and altitudes. REPUBLIQUE DU ConGo: Rutchuru—Kabali ;
Kisenyi, 3,600-5,000 ft.; S.W. Corner of L. Kivu, 6,300 ft., Karissimbi Volcano,
Kivu; N.W. Lake Tanganyika, 1,700-1,900 m.; N.W. Kivu, 4,300 ft. UGANDA:
E. Ruwenzori, 5,000-13,000 ft.; N. Ruwenzori, 6,000-8,500 ft.; Ruwenzori,
Bwamba Pass, 5,500-7,500 ft.; Mt. Niragongwe, N. of L. Kivu, 1,900-3,000 m. ;
Mt. Mikeno, N. of L. Kivu, 1,900-2,400 m. ; Mafugi Forest, Kigesi, 7,500-8,500 ft. ;
Rutenga Forest, Kigesi; Kayonza, Kigesi, 4,000 ft.; Kanaba Gap, Kigesi; Toro ;
Rutschuru plain; Mt. Kokajero, S.W. Elgon, 6,400 ft.; Daro or Durro Forest,
Toro, 4,000-5,000 ft.; Mpanga Forest, Toro, 4,800 ft.; Mbale; Subugo Forest ;
Mau. RWANDA: Rugege Forest, E. of S. end of Lake Kivu. SupAN: Lotti Forest.
KeENnyA: Patsho, Nandi Country ; Nandi Station ; Nandi Plateau, 5,700-6,200 ft. ;
Eldoma Ravine; Escarpment, 6,500-9,000 ft. ; Kipiperi, Aberdare Range, 8,000-
9,000 ft. ; Dabida, Rabai-Mombasa, 6,000 ft. ; E. slopes, Aberdares, 7,000-8,500 ft. ;
Roromo, Kikuyu ; Hoey’s Bridge ; Mt. Kenya, Meru dist. (Mt. Meru, Kenya dist.) ;
Mara River, Masai Reserve ; Nairobi, Kinangoli, 9,000 ft.; Kibwezi ; Nyeri ; Mbolo
Hill, Voi District, 5,500 ft.; Mt. Mbololo, 5,000 ft.; Kilimanjaro, Mt. Elgon ;
Lumbwa, Wandanyi, 5,000 ft.; Mt. Kulal, 5,500 ft.; Mt. Marsabit. TANZANIA:
Magazine Hill; Longido, W. Kilimanjaro, Ngare—Nairobi, 4,000-5,000 ft.; Mt.
Rungwe, nr. N. Langenburg, 5,000-6,000 ft.; Itumba District, Meru, 8,000 ft. ;
Lushoto ; District of Great Craters; Tukuyu. MALaAwi: Mlanje Plateau, 6,000-
7,000 ft.; Kasungu Mt., 7,425 ft. ; Nyika Plateau. RHODESIA: Vumba (2 3, I 9,
National Museum, Bulawayo, Rhodesia) ; Laurenceville, 10 miles S. of Umtali
(2 3, H. Cookson coll.) ; Chitora Hills, 30 miles S. of Umtali (x g, H. Cookson coll.).
TRANSVAAL : Woodbush (1 3, National Museum, Nairobi) ; Mariepskop, Pilgrims
Rest district (r g, Transvaal Museum, Pretoria) ; Graskop (1 g, Transvaal Museum,
Pretoria) ; Haenertsburg (xr 9, National Museum, Bulawayo. 1 g, 1 9, Transvaal
Museum, Pretoria).

Distribution. A montane species occurring from about 4,000 to 13,000 feet in
suitable habitats from the eastern edge of the République du Congo to Mt. Marsabit,
Kenya in the north and as far south as the Transvaal. It was at first thought that
the Rhodesian and Transvaal specimens might belong to a separate race, as the
transverse orange band of the upperside of the fore wing is stepped at vein 1 and the
dark basal area of the hind wing upperside extends into the orange marginal band
along the veins, but there are similar specimens amongst those from Tanzania and
Malawi and it would seem as if these southern specimens may be at the extremity of a
cline.

NOTES ON THE GENUS ANTANARTIA 35

Antanartia dimorphica aethiopica ssp. n.
(Pl. 5, figs. 34 and 35)

3, @. Distinguished from the nominate subspecies by presence on the upperside of hind wing
of two or three extra black spots or ocelli situated just inside the dark basal area in spaces 4-6,
that in space 4 sometimes having a faint blue ocellus. In this respect it resembles Antanartia
abyssinica (Felder), the only other species in the genus that exhibits this character ; this resem-
blance is increased still further by the absence of tails in abyssinica and the very short tails in
this race. Fore wing length g, 20-24 mm., 9, 21-24 mm.

Holotype g. EruiopiA: Djem Djem Forest, 8,000-9,000 ft., 7-9.x.1926 (H.
Scott).

Allotype 9. Eruiop1A: Dangila, 6,700 ft., 40 miles S. of Lake Tana, 3. viii. 1926
(R. E. Cheeseman).

Paratypes: 2 3g, same data as holotype; 2 g, same data as allotype; 5 g, 1 9,
Abera, Djamdjam ; 1 4, 2 9, Abera, S. of L. Abasa ; 12 J, Gimera (Kaffa) to N. end
of L. Rudolf; 1 3, Badattino, Gindeberat ; 1 g, Dereta Mts., Kaffa; 2 ¢, Kollu
(Schoa) ; 3 g¢, Kankati, Djimma, S.W. Abyssinia; 2 3, Aberasch, E. of L. Abai ;
1 9, Arbe, S. of L. Abasa; 2 4, 2 9, Charada Forest, Kaffa, 6,000 ft.; x ¢, I 9,
Mt. Zuquala, over 9,000 ft. ; Gamo Prov., Bonghé Valley; 2 3, Gughé Highlands,
9,500-10,000 ft. ; 2 g, Luguala Crater, 9,000 ft. ; Mt. Chillalo, c. 9,000 ft.; 5 3g, 4 9,
Scioa ; 5 2, Scioatalit (M.S.) ; 1 g, 2 9, Chercher ; 2 9, Jam-Jam; 1 g, Muti, 6,000-
8,000 ft.

Distribution. This race is confined to the highlands of Ethiopia at altitudes up to
10,000 feet.

Rothschild & Jordan (1903 : 509) when dealing with specimens of this subspecies
stated ‘“‘ The Abyssinian specimens of hippomene are all short tailed, but do not
present any constant differences from East and South African ones’ but that they
were found to have been placed over a blank label in the Rothschild collection.

Antanartia dimorphica comoroica ssp. n.
(Pl. 5, figs. 40 and 41)

. Can be separated from the other subspecies by the broad orange marginal band of the
upperside of hind wing, which almost obscures the indistinct and blind ocellus in space 3 and the
inner submarginal black line which terminates abruptly at vein 3 and does not invade the orange
band towards vein 4. There is a small white spot in space 3 of the upperside of fore wing. Fore
wing length 23-25 mm.

6. Unknown.

Holotype 2. GRAND Comoro Is. (Rothschild Bequest).
Paratypes: 3 2, same data as holotype.

Distribution. This subspecies is confined to Grand Comoro Island and it is of
interest that the southern hippomene has reached Madagascar as madegassorum
(Aurivillius), whereas the more northern dimorphica has only succeeded in reaching
the Comoro Islands.

36 T. G. HOWARTH

Antanartia dimorphica mortoni ssp. n.
(Pl. 5, figs. 36, 37, 38 and 39)

3, 2. Separated from the nominate subspecies by the broader and deeper orange coloration
of the transverse discal band of the fore wing of both upper and undersides, and the marginal
band of hind wing upperside which is noticeably scalloped at its inner edge in some specimens.
Two specimens have the band widened considerably in space 5 in the same manner as in abyssinica.
Fore legs of male have the hairs of inner surface cream as in hippomene. Upperside. Fore wing,
no small white spot in space 3 and black central cell bar stands out clearly against the rufous tone
of the cell. Fore wing length g, 21-25 mm., 9, 23-24 mm.

Holotype 3. W. CAMERooNS: Mt. Cameroon. On path from Buea to Mann’s
Springs, 7,500-10,000 ft. In montane grassland. 30.xii.1958 (J. K. Morton).

Allotype 2. NIGERIA: Obudu Plateau, 5,200 ft., 1.ii.1965 (R. G. T. St. Leger).

Paratypes : 9 3, same data as holotype (5 in J. K. Morton coll.) ; 3 g, NIGERIA:
Obudu Plateau, 5,200 ft., 31.1.1965 ; 2 9, same data as allotype; 1 g, Bamenda
Plateau (P. J. L. Roche). 19, W. CAMEROONS: Bamenda, iv.1g1g (D. Cator coll.).
I g, 2 9, FERNANDO Po: 3,000-4,000 ft. vi.1926 (T. A. Barns).

Distribution. West Cameroons, Nigeria and the island of Fernando Po.

This subspecies is named after Prof. J. K. Morton of Fourah Bay College, Freetown,
Sierra Leone, who captured the first series of this insect which, when submitted for
determination, was instrumental in the commencement of the present investigation.

ABNORMAL SCALING IN THE GENUS ANTANARTIA

During the examination of the series of this genus it was noticed that several
specimens had the orange-red of the transverse bands of the upperside of the fore
wings reduced in intensity and it was thought that this reduction might be due to a
scale defect. This proved to be correct and as the specimens varied in degree of
abnormality every specimen was examined microscopically.

Normally the scales are arranged over the wing surface with their bases towards
the wing base in roughly parallel double lines running concentrically from the wing
base and interspaces, each double row consisting of an upper and under layer of
scales arranged alternately (see Bayard, 1932). The scales of the lower or basal
layer are adpressed to the wing membrane and partially overlap each other and do
not appear to be so heavily pigmented as those of the upper. The upper layer
consists of slightly longer scales than the lower and overlap one another as well as
covering and concealing the lower basal layer though in fact they are slightly inclined
to the wing surface. In the specimens exhibiting abnormal scaling, the individual
scales are rolled and twisted longitudinally to a point and in extreme cases they
resemble thick hairs. Associated with this deformity is either a lack of, reduction of,
or alteration to, the pigment. In all the specimens examined only the paler scales
were affected, i.e. white, yellow, orange or red and none was found with the dark
brown or black scales deformed. In some specimens only a few scales of one layer
are affected, sometimes scattered and sometimes in small patches and in others a
more general defect is present, affecting most if not all the paler scales and it is these

NOTES ON THE GENUS ANTANARTIA 37

that are discernible to the unaided eye due to a general alteration in colour. The
white scales of the apical spots of the upperside of the fore wing consist of only the
basal layer and these scales seem to be the first affected when a scale defect is present
in aspecimen. In some examples that have the majority of the pale scales defective
the scales of the white apical spots are often absent, though their sockets are present
on the wing membrane denoting a transition between the two types of defect men-
tioned by Cockayne (1922: 48). In general it seems as if the scales of the fore wings
are affected more than those of the hind wings and the uppersides more than the
undersides. In the following Table which gives the total number of specimens of
each species and subspecies examined and the percentage found to have abnormal
scaling, it will be seen that of the six species, hippomene is the most affected and delius
the least, while borbonica has no defective scaling though this result may well be due
to insufficient material.

Specimens of ANTANARTIA with Abnormal Scaling

Total examined % of abnormalities
' fo re | i is |
3 ? ) ?
delius delius ; ; : 237 63 1°3 —
delius guineensis ; : 5 —_— = —
schaeneia schaeneia . : 8 15 12°5 —
schaeneia dubia . : 3 142 42 2:8 9°5
schaeneia diluta : , 52 16 —_ 6°3
borbonica borbonica . , 3 10 — —
borbonica mauritiana . : 8 5 — —
abyssinica abyssinica . 4 30 12 16°6
abyssinica jacksoni. ; 129 42 15°5 33°3
abyssinica vansomerent ; 25 10 32-0 10°'0
hippomene hippomene . ; 15 20 40:0 30-0
hippomene madegassorum . 4 2 — —
dimorphica dimorphica : 95 59 27:0 13°3
dimorphica aethiopica . , 65 23 21°5 39°1
dimorphica mortoni. : 5 2 -- —
dimorphica comoroica . : — 3 — —

PARALLEL MODIFICATION IN THE GENUS ANTANARTIA

There is a small but significant amount of parallel modification or similarity,
amongst the races of the species that are sympatric. In Ethiopia particularly, the
males of the subspecies of the three species that occur there—schaeneia diluta,
abyssimica abyssinica and dimorphica aethiopica, are generally smaller and have
shorter and blunter tails, if present, than elsewhere. On the upperside of the fore
wings they tend to have the white subapical spots often tinged with ochre and the
orange bands of both wings are slightly narrower. On the upperside of the hind
wings dimorphica aethiopica has a series of well-defined extra spots basad of the
orange marginal band very similar to those of abyssinica abyssinica and also the
apex is clouded in a similar manner. Generally the Ethiopian races are nearer in

ENTOM. 18, 2 3

38 T. G. HOWARTH

appearance to those from South Africa, where they occur, than to those from the
central part of the African continent, i.e. schaeneia diluta is nearer to schaeneia
schaeneia than to schaeneia dubia and dimorphica aethiopica is nearer to hippomene
hippomene than to dimorphica dimorphica, hippomene being the replacement species
for dimorphica in South Africa. In the Malagasy Republic and the Mascarene
island of Réunion the two species hippomene and borbonica occur separately but are
similar in pattern, shape and colour. They both have long pointed tails, very
crenate wing margins and the same broad, deep orange bands on the upperside
though hippomene madegassorum retains the normal ochre colour of the transverse
band on the underside of the fore wing of the nominate race.

ACKNOWLEDGEMENTS

The author wishes to express his sincere thanks and appreciation to those colleagues
in the Department of Entomology who have helped in numerous ways with this
paper and to the many correspondents and specialists who have been so generous with
their advice and information and for the loan of material. Particular mention must
be made to Mr. N. D. Riley for reading through the manuscript and to the following :
Dr. R. Agenjo (Madrid, Spain), Dr. L. A. Berger (Tervuren, Belgium), Dr. G. Bernardi
(Paris, France), Mr. R. H. Carcasson (Nairobi, Kenya), Messrs. H. & D. M. Cookson
(Umtali, Rhodesia), Mr. C. G. Dickson (Cape Town, South Africa), Dr. R. M. Fox
(Pittsburg, U.S.A.), Mr. T. H. E. Jackson (Kitale, Kenya), Dr. P. Malzy (Tananarive,
Malagasy Rep.), Prof. J. K. Morton (Freetown, Sierra Leone), Mr. K. M. Pennington
(Balgowan, Natal), Dr. E. Pinhey (Bulawayo, Rhodesia), Dr. P. J. L. Roche (London,
England), Mr. R. G. T. St. Leger (Enugu, Nigeria), Dr. V. G. L. van Someren (Ngong,
Kenya), Dr. G. van Son (Pretoria, South Africa), Dr. P. Viette (Paris, France),
Mr. B. K. West (Dartford, England).

REFERENCES

AuRIVILLIus, C. 1898. Rhopalocera Aethiopica. Die Tagfalter des Aethiopischen Faunenge-
bietes. K. svenska VetenskAkad. Handl. 31 (5): 1-561, 6 pl.

—— In SeiTz, A., 1913, The Macrolepidoptera of the World. 13, 613 pp., 80 pl. Stuttgart.

Bayarp, A. 1932. Observations élémentaires sur les ecailles des Lépidoptéres. Bull. Soc.
fr. Microsc. 1 (3) : 61-67.

BoispuvaL, J. B. A. D. DE. 1833. Mémoire sur les Lépidoptéres de Madagascar, Bourbon et
Maurice. Nouv. Annls Mus. Hist. nat., Paris 2 (3) : 149-270, 7 pl.

Boorman, J. & RocHE, P. 1959. The Nigerian Butterflies. Part 5. Nymphalidae. (3)
19 pp, 47 pl. Ibadan, Nigeria.

Butter, A. G. 1880. On a collection of Lepidoptera from Madagascar, with descriptions of
new Genera and Species. Ann. Mag. nat. Hist. 5 (5) : 333-343.

CocKAYNE, E. A. 1921. Structural abnormalities in Lepidoptera. Lond. Nat. 10-69, 1 pl.

DiaxonoFF, A. 1954. Considerations on the terminology of the genitalia of Lepidoptera.
Lepid. News 8: 67-74. 2 figs.

Drury, D, 1782. Tilustrations of Natural History of Exotic Insects 3. 18 pl. London.

DUFRANE, A. 1945. Lépidoptéres du Kivu. Bull. Soc. ent. Belg. 81 : 90-143.

Fapricius, J.C. 1793. Entomologia Systematica. 3. 487 pp. Hafniae.

FELDER, C. & R. 1867. Reise Novara. 3: 379-535. 27 pl. Wien.

GoparT, J. B. 1819. Encyclopédie Méthodique, Histoire Naturelle. 9: 828 pp. Paris.

NOTES ON THE GENUS ANTANARTIA 39

HUBNER, J. 1823. Sammlung exotischer Schmetterlinge.2. 225 pl.

ManvDeErRS, N. 1908. The butterflies of Mauritius & Bourbon. Tvans. ent. Soc. Lond. 4: 429-
454. pl. 29.

OBERTHUR, C. 1880. Spedizione Italiana nell’ Africa Equatoriale—Lepidotteri dello Scioa.

Part1. Annali Mus. civ. Stor. nat. Giacomo Doria 15 : 129-186, pl. I.

1883, ditto, Part 2. 18: 709-740, pl. 9.

ROTHSCHILD, W. & JORDAN, K. 1903. Lepidoptera collected by Oscar Neumann in North-
East Africa. Novit. zool. 10 : 491-542.

SHARPE, E.M. 1904. Descriptions of new Lepidoptera from Equatorial Africa. Entomologist.
37 : 181-183.

SUFFERT, E. 1904. Neue Nymphaliden aus Africa. Dt. ent. Z. Iris. 17: 108-123.

SWANEPOEL, D. A. 1953. Butterflies of South Africa. 320 pp. 17 pl. Cape Town.

TRIMEN, R. 1862. Rhopalocera Africae australis 1: 1-190. 1 pl. Cape Town.

1879. On some hitherto undescribed butterflies inhabiting South Africa. Tyvans. ent.

Soc. Lond. 4 : 323-345.

1887. South African Butterflies.1. 355 pp. 6 pl. London.

40

Ty. G. BOWARTIA

Fics. 1-7. Male genitalia of Antanartia Rothschild & Jordan and Hypanartia Hiibner with
left-hand valva removed. 1, Antanartia delius (Drury) ; 2, A. schaeneia (Trimen) ; 3, A.
borbonica (Oberthiir) ; 4, A. abyssinica (Felder) ; 5, A. hippomene (Hiibner) ; 6, A. dimor-
phicasp.n.; 7, Hypanartia kefersteini (Doubleday & Hewitson) with the eighth abdominal

tergite and superuncus,

NOTES ON THE GENUS ANTANARTIA

41

Fics. 8-11. Female genitalia of Antanartia Rothschild & Jordan.

8, Antanartia delius

(Drury) ; 9, A. schaeneia (Trimen) ; 10, A. delius x schaeneia hybrid (?); 11, A. bor-

bonica (Oberthiir).

42 T. G. HOWARTH

tart t

Fics. 12-14. Female genitalia of Antanartia Rothschild & Jordan. 12, Antanartia
abyssinica (Felder) ; 13, A. hippomene (Hiibner) ; 14, A. dimorphica sp. n.

NOTES ON THE GENUS ANTANARTIA 43

oe SN

Fics. 15-21. Venation of Antanartia Rothschild & Jordan and Hypanartia Hiibner :
Upper. Area adjacent to the upper apex of the discoidal cell of fore wing. Lower.
Discocellular area adjacent to veins 3 and 4 of hind wing. 15, Antanartia delius (Drury) ;
16, A. schaeneia (Trimen); 17, A. borbonica (Oberthiir) ; 18, A. abyssinica (Felder) ;
19, A. hippomene (Hiibner); 20, A. dimorphica sp. n.; 21, Hypanartia kefersteini

(Doubleday & Hewitson).

PLATE 1
Upper and undersides of Antanartia Rothschild & Jordan

Fic.1. Antanartia delius delius (Drury), g Cameroons. Photos. Brit. Mus. (N.H.) Nos. 37294,
37295. Fic. 2. A. delius delius (Drury), 2 Sierra Leone. Photos. Brit. Mus. (N.H.) Nos.
37296, 37297. Fic. 3. A. delius delius (Drury), g S. Sudan. Photos. Brit. Mus. (N.H.)
Nos. 37300, 37301. Fic. 4. A. delius delius (Drury), 9 Kenya. Photos. Brit. Mus. (N.H.)
Nos. 37302, 37303. Fic. 5. A. delius delius f. amauroptera Sharpe, g¢ Kenya. Photos. Brit.
Mus. (N.H.) Nos. 37304, 37305. Fics. 6 and 7. A. delius guineensis ssp. n. Holotype 4,
Fernando Po. Photos. Brit. Mus. (N.H.) Nos. 37298, 37299.

Bull. B.M. (N.H.) Entom. 18, 2 NA) Dt, 4 BA Ds

PLATE 3
Upper and undersides of Antanartia Rothschild & Jordan

Fics. 16,17. Antanartia schaeneia diluta Rothschild & Jordan g and 9 respectively, Ethiopia.
Photos. Brit. Mus. (N.H.) Nos. 37314, 37315, 37316, 37317. Fic. 18. <A. borbonica borbonica
(Oberthiir), ¢, Réunion. Photos. Brit. Mus. (N.H.) Nos. 37346, 37347. Fic.19. <A. borbonica
mauritiana Manders, Paratype ¢, Mauritius. Photos. Brit. Mus. (N.H.) Nos. 37318, 37319.
Fic. 20. A. abyssinica abyssinica (Felder), g, Ethiopia. Photos. Brit. Mus. (N.H.) Nos.
37320, 37321. Fic.21. A. abyssinica jacksoni ssp. n., Paratype g, Kenya. Photos. Brit. Mus.
(N.H.) Nos. 37322, 37323. Fics. 22 and 23. <A. abyssinica vansomereni ssp. n. Holotype 4,
N.W. Kivu. Photos. Brit. Mus. (N.H.) Nos. 37324, 37325. Fic. 24. <A. hippomene hippomene
(Hiibner), 3, Natal. Photos. Brit. Mus. (N.H.) Nos. 37326, 37327. Fic. 25. <A. hippomene
hippomene (Hiibner), 9, S. Africa.. Photos. Brit. Mus. (N.H.) Nos. 37328, 37329.

Bull. B.M. (N.H.) Entom. 18, 2

PLATE 4
Upper and undersides of Antanartia Rothschild & Jordan

Fics. 26, 27. Antanartia hippomene madegassorum (Aurivillius), ¢ Madagascar. Photos.
Brit. Mus. (N.H.) Nos. 37330, 37331. Fics. 28, 29, 30, 31. A. dimorphica dimorphica sp. n.,
Holotype g and Allotype ? respectively, Uganda. Photos. Brit. Mus. (N.H.) Nos. 37336, 37337,
37338, 37339. Fics. 32, 33. A. dimorphica dimorphica sp. n. Paratype ¢ and 9 respectively,
Rhodesia. Photos. Brit. Mus. (N.H.) Nos. 37342, 37343, 37344, 37345-

PLATE 4

Bull. B.M. (N.H.) Entom. 18, 2

PLATE 5
Upper and undersides of Antanartia Rothschild & Jordan

Fics. 34, 35. Antanartia dimorphica aethiopica ssp. n., Holotype g, Ethiopia. Photos,
Brit. Mus. (N.H.) Nos. 37340, 37341. Fics. 36, 37. A. dimorphica mortoni ssp. n., Paratype ¢.
Cameroon, Photos. Brit. Mus. (N.H.) Nos. 37334, 37335. Fics. 38, 39. A. dimorphica mortoni
ssp. n., Allotype 9, Nigeria. Photos. Brit. Mus. (N.H.) Nos. 37400a, 37401a. FIGS. 40, 41.
A. dimorphica comoroica ssp. n., Holotype 9, Comoro Is. Photos. Brit. Mus. (N.H.) Nos.
37332, 37333-

Bull. B.M. (N.H.) Entom. 18, 2 PLATE 5

*_) £4

ia

33> ere =
at a
ig Lee
A
i > Se eae Na
va UJ
i,
ge 5 a
a d

™
-~
x a - ’ »
- : zr
« ?
me
.

A LIST OF SUPPLEMENTS
TO THE ENTOMOLOGICAL SERIES
OF THE BULLETIN OF
THE BRITISH MUSEUM (NATURAL HISTORY)

1. MASNER, L. The types of Proctotrupoidea (Hymenoptera) in the British
Museum (Natural History) and in the Hope Department of Entomology, Oxford.
Pp. 143. February, 1965. £5.

2. Nixon, G. E. J. A reclassification of the tribe Microgasterini (Hymenoptera :
Braconidae). Pp. 284; 348 Text-figures. August, 1965. 6.

3. Watson, A. A revision of the Ethiopian Drepanidae (Lepidoptera). Pp. 177;
18 plates, 270 Text-figures. August, 1965. {4 4s.

4. SANDS, W. A. A revision of the Termite Subfamily Nasutitermitinae (Isoptera,
Termitidae) from the Ethiopian Region. Pp. 172; 500 Text-figures. October,
1965. £3 5s.

5. AHMAD, I. The Leptocorisinae (Heteroptera: Alydidae) of the World. Pp. 156;
475 Text figures. November, 1965. {2 I5s.

6. OxaDA, T. Diptera from Nepal. Cryptochaetidae, Diastatidae & Drosophilidae.
Pp. 129; 328 Text-figures. 3.

7. GILIOMEE, J. H. Morphology and Taxonomy of Adult Males of the Family

Coccidae (Homoptera: Coccoidea). In press.
8. FLETCHER, D. S. A revision of the Ethiopian species and a check list of the

world species of Cleora (Lepidoptera: Geometridae). In press.

PRINTED IN GREAT BRITAIN BY ADLARD & SON LIMITED, BARTHOLOMEW PRESS, DORKING

———

f 8 JUN 196
REVISIONAL NOTES ON AFRICAN ™« «<
CHARAXES
(LEPIDOPTERA : NYMPHALIDAE)
PART III

V. G. L. van SOMEREN

BULLETIN OF
THE BRITISH MUSEUM (NATURAL HISTORY)

ENTOMOLOGY Vol. 18 No. 3
LONDON : 1966

ge B®
= 2

| 1966
REVISIONAL NOTES ON AFRICAN oe
ae
(LEPIDOPTERA : NYMPHALIDAE) CUES
PART III

BY

V. G. L. van SOMEREN

Ka/ f
P.O. Box 24947, Karen, Kenya fs

Pp. 45-101 ; 5 Maps, 16 Plates

BULLETIN OF
THE BRITISH MUSEUM (NATURAL HISTORY)
ENTOMOLOGY Vol. 18 No. 3
LONDON : 1966

THE BULLETIN OF THE BRITISH MUSEUM
(NATURAL HISTORY), instituted in 1949, 1s

issued in five series corresponding to the Departments
of the Museum, and an Historical series.

Parts will appear at irregular intervals as they become
veady. Volumes will contain about three or four
hundred pages, and will not necessarily be completed
within one calendar year.

In 1965 a separate supplementary series of longer
papers was instituted, numbered serially for each
Department.

This paper is Vol. 18, No. 3 of the Entomological
serves. The abbreviated titles of periodicals cited
follow those of the World List of Scientific Periodicals.

© Trustees of the British Museum (Natural History) 1966

TRUSTEES OF
THE BRITISH MUSEUM (NATURAL HISTORY)

Issued 8 June, 1966 Price £2 16s.

REVISIONAL NOTES ON AFRICAN CHARAXES
(LEPIDOPTERA : NYMPHALIDAE) PART III

By V. G. L. van SOMEREN

CONTENTS

Page
eh ci F : : : ; : : 47
. Charaxes violetta GRosE- SMITH . AND Its Races ‘ : : : : 47
Descriptions and Notes. . , ‘ ‘ : : : ; 50
Systematic List : : : 2 ‘ : , : 54
2. THE Charaxes etesipe-penricei Complex : : ; : ; : 54
Descriptions and Notes. : ; : : g : : 59
Systematic List : : 69

3. Charaxes pembanus JORDAN AND Charaves isambarie VAN SOMEREN &
JACKSON : ‘ ; ; ; : , 2 : 70
Descriptions and Notes : ; 5 : : ‘ : ‘ 70
Systematic List : ; , : ; , ; : . 74
4. THE Charaxes berkeleyi GRouP ; ; ; ; : : . 74
Descriptions and Notes. : 2 ‘ : , ; ‘ 74
Systematic List E : ‘ ‘ : : : ‘ : 97
ACKNOWLEDGEMENTS . : : ‘ : : , Z ; : 99
REFERENCES , ; : : : : ; F F i : 99
INDEX 7 . : : ; ; , : ; : : < 2800

SYNOPSIS

One complex and twelve species, and their races and forms, of the genus Charaxes are dealt
with in this paper. One new species, six new subspecies and one new form are described, and
two subspecies and one race are raised to specific rank.

I. CHARAXES VIOLETTA GRosE-SMITH AND ITS RACES

It is well known that the species violetta has a wide range, extending from Delagoa
Bay, through southern Mozambique to the eastern border of Rhodesia! (Vumba
Mts. area), thence to Malawi, Tanzania and into Kenya. It has generally been
assumed that only one race existed throughout this wide distribution. The types
of both sexes came from Delagoa Bay.

Rothschild & Jordan (1900 : 374) did however draw attention to the fact that
specimens from tropical (German and British) East Africa had wider bands on fore
and hind wings, both above and below, than individuals from Delagoa Bay. Auri-
villius in “ Seitz ’’, rgr1, lists only one race and gives the range, as Delagoa Bay to
Mombasa. Stichel (1939 : 438) and Peters (1952: 51) mention only one race.

1In this paper, the current names for the following African countries have been adopted. Rhodesia

(Southern Rhodesia), Zambia (Northern Rhodesia), Malawi (Nyasaland) and Tanzania (Tanganyika
with Zanzibar)). The locality labels on the specimens have not been changed.

ENTOM. 18, 3 4

48 V.G.L. VAN SOMEREN

In my series of papers on the Charaxes of Kenya and Uganda (van Someren &
Rogers, 1928. Pt. 8), I described and figured specimens from Mombasa, assuming
that these could be taken as typical violetta. In a further reference to the species,
(van Someren, 1935 : 185) I extended the range of the species northward to Meru,
near Mt. Kenya.

In the course of correspondence with Dr. R. M. Fox of the Carnegie Museum, U.S.A.,
I was informed that he was describing a new race of Ch. violetta from the Nguru Hills,
Tanzania, because specimens from that locality differed from those of Amani (Usam-
bara foothills), Dar-es-Salaam and Mombasa, which he considered to be nominotypical
violetta. I have since seen the published description in which he refers to my previous
references to the species as quoted above. No reference or comparison was made
with topotypical Delagoa Bay material, though I suggested this should be done
before publication of his paper. A preliminary survey which I had carried out
suggested that a detailed study of material from throughout the known range was
essential to a reasonable appreciation of the degree of variation in the species.

First of all, it was necessary to have access to the types, which fortunately are
preserved in the British Museum (Natural History). These specimens, a male and
female, are still sufficiently well preserved to exhibit the salient characters. Other
topotypical material seems difficult to trace and examine, but I have brought together
five authentic females and four males and these exhibit a constancy of upperside
characters, though rather variable below, which are described in detail hereafter,
and depicted on PI. 1, figs. 1-8.

Through the kindness of various contributors I have assembled considerable
series from eastern Rhodesia (Vumba to Selinda) ; from the Beira area of Mozambique
(Dondo and Amatongas) ; from Malawi and the southern areas of Tanzania, north
to Mt. Kilimanjaro, and also from the Usambaras and from the coast belt from
Dar-es-Salaam to Mombasa and the coastal forests of Kenya, inland to the Teita
Range and from the area N.E. of Mt. Kenya at Meru and the Njombeni hills.

This considerable material shows clearly that the observation made by Rothschild
& Jordan in 1900 was correct, viz., that specimens from the northern tropical areas
have considerably wider discal bars, above and below, especially in the females, than
Delagoa Bay insects. Thus specimens from the extremes of the range are markedly
different.

In order to obtain an overall picture of the species it is necessary to arrange the
material in geographical sequence, starting in the south and working northwards
and attempting to correlate the specimens with topography and ecological environ-
ment so far as it is known, using as a basis a standard topographical map and the
Vegetation Map produced by UNESCO in 1958.

The line of distribution of violetta as exemplified by material under examination is
shown on map A.

As a result of this arrangement, it is noted that violetta, apart from the two com-
munities at the extremes of the range, presents a confusing picture, evidenced by
instability of pattern in the female, particularly in regard to the width of the fore wing
white bar and size of the hind wing patch and white bar below.

REVISIONAL NOTES ON AFRICAN CHARAXES

Ls acoceth
F

my R, CONGO ‘<
—— UGANDA
/
~ /
} i conan)
! Pa
is t eens Ss
Conco Meare)
N r
L, KIVU 4) RUANDA , “A
4 ; SERENGET
paras : A
’
Posies
i
URunodl
TANGANYIKA \ ay
L. MANDA
> } ‘ PAG aS
‘ 4 4
L. MWERU ~“RUMGWE
Jp ea ea
: NJOMBE
} Ne
i)
j
KATANGA yy ;
oe NYA
) \
Peas a [i - '
i Tie a BANGWEULU Y : tee
) ~z~-~
\ Se 4
) Re J
See
_ yy t
Norid
= Ni
oa f}
a pr ty a
a
Vor MLANJE MTS.
moacchnes . ¥ ra
~ 5 74
L. KARIBA an
ot an SOUTHERN RHODESIA
\
“4 a \
i
‘
1
‘\
5
N
Nan
~
x
Lo. U
>
) aa
\ t R. SABI
eee ¢
~ , /
3£ CHUANALAND > po A
~
y \
TRANSVAAL \
\ R, LIMPOPO
\
A.
/
-
(
/ A
~ South AFRICA
Lye aS DELAGCA BAY
~-

LOURENCO MARQUES

Map A

NAIROBI

sp

i

x
KILI

rN

BEIRA

@oepvpe>r

SOMALIA

R, TANA
‘
XN
, Ni
KENYA
R, GALANA f
MANJARO ®@
Nt

\
USAMBARA

MTS.

R, RUFIGI

od veuta e &
Oe

MC

Breuca

®
A @ Bi tsicas
oO

DAR-ES-SALAAM

A wsaria

LINDI

A NEWALA

R, RU

VUMA

R. LURIO

VIOZAMB I QUE

CHARAXES

CHARAXES

CHARAXES

CHARAXES

CHARAXES

KEY.
VIOLETTA
VIOLETTA,

VIOLETTA
CLINE.

VIOLETTA
MELLONI,

VIOLETTA
MARITIMA,

VIOLETTA
MERU,

49

50 V.G.L. VAN SOMEREN

Nevertheless, the overall pattern of specimens from Vumba northward to Malawi,
the central area of Tanzania and to the western side of Mt. Kilimanjaro appears to
present features suggesting a cline from nominotypical voletta of Delagoa Bay to the
broad-banded, more stable races found in Kenya. Thus, in a long series from the
area of Mt. Selinda, one or two may be narrow-banded ; similarly, specimens from the
lower zones of Usambara may tend toward the broad-banded coastal Kenya race ;
the majority being transitional.

It is often a matter of personal opinion as to whether such an aggregate is deserving
of a subspecific name. As indicated previously, Dr. Fox has published the name
mellont, applicable to specimens from the Nguru Mts. near Morogoro (Fox, 1963 :
222-224) but the female type is only one variation of this aggregate. It may be
noted here, that Charaxes violetta is a species in which the black colour, especially
in the females, tends to fade, assuming a brownish-black or brownish-purply colour
in the dark areas above.

DESCRIPTIONS AND NOTES
Charaxes violetta violetta Grose-Smith

(Pl. 1, figs. 1-8; Pl. 2, figs. ro-12, 14-16; PI. 3, figs. 17, 18 and 21)

Charaxes violetta Grose-Smith, 1885 : 247 [¢].
Charaxes violetta Grose-Smith & Kirby, vi.1887:1, pl. 1, figs. 1-3. [g, 9].

MALE: Fore wing length 37-38 mm. Upperside. Ground colour of fore wing strongly blue-
black with slightly more greeny-blue sheen at base ; base of costa brown. Discal spots blue ;
streak on costa often whitish ; mark below costa a streak at end of cell followed by larger ovate
spots at bases of 4—5, a larger spot sub-basal in 3, a larger spot in 2, that in 1b more quadrate and
contiguous with post-discal mark in same area; a longer blue streak in 1a often extending
inward proximad. Post-discal spots : two subapical white, upper one more distinct, remainder
blue, those in 5—4 often vestigial, that in 3 more distinct, the spot in 1b larger and touching the
discal spot, the longer mark in 1a variable. Margin of wing slightly incised, with slight indication
of internervular whitish spots sometimes present in areas 2-4. Hind wing basal area and border
blue-black, disc of wing with large blue patch whitish toward centre and at junction with ashy-
grey of inner fold, not extending through to costa but represented by one or two discrete spots
in6and 7. Dark border with row of small bluish spots, double at anal angle ; margin with line
of bluish lunate marks whitish at veins. Tails shortish, tapering rapidly, upper 5 mm., lower
3mm. ; margin of wing slightly serrate. Underside. Fore wing basal area olive-ochre, followed
by a broad darker olive discal band accentuated in narrow white and black ; distad to this a paler
bar similar in colour to the wide distal border with a darker triangular patch between ; double
tornal spot with a smaller one above usually well marked, internally bordered with crescentic
olive-ochre accentuated proximally by black. Black lines at base rather faint or thin. Hind
wing ground colour olive-ochre, with a darker olive discal band outlined narrowly in black and
white, followed by a paler zone which is contiguous with an irregular darker olive post-discal
zigzag line; border olive-ochre carrying submarginal spots, distinct or obscured, whitish and
violet in colour ; marginal lunate marks whitish, distinct in lower half, indistinct towards upper
angle.

Variation. In some specimens the underside pattern is in greater contrast, the distal borders
of the wings being much paler than the basal dark areas ; the whitish discal lines on both wings
broader and more distinct.

FEMALE. Fore wing length 40-45 mm. Upperside. Base of wing blackish-brown with some
purply-blue sheen in side light ; apical half of wing blacker ; discal curved white band extends

REVISIONAL NOTES ON AFRICAN CHARAXES 51

from the costa to hind border, passing through bases of 5—6, sub-basal in 4, slightly off-centre in
3-2, the mark in 1b indented on both ends, shorter than the mark above proximad, followed
by a longer streak in 1a; the last two marks with some violet-blue scaling at ends. The white
mark on the costa extends proximad beyond the next mark. The band is narrow at end of
cell, about 4-5 mm. wide, slightly broader in 5, the longest mark in 2; the outer border of the
bar is thus more curved than the inner. The post-discal spots in subapex usually limited to two,
the upper elongate, the lower smaller and more round. Hind wing basal area brownish-black ;
distal border blacker ; the disc of the wing with a dull violet-blue patch rather dyslegnic on
inner border, but more clear cut on outer edge ; crossed in the discal line by a whitish bar which
runs toward but does not reach the inner fold ; the outer border of the patch irregular in its
upper half. The black border carries a complete row of white elongate spots which are tinged
with violet ; admarginal line of lunate marks lilac and white. Tails thin and tapering, upper
6 mm., lower 3 mm. Underside. Fore wing basal triangle olive-grey, crossed by black lines,
proximally edged in white it becomes darker where it abuts on the white discal band, which is a
replica of the upperside, but marks in 1a—1b are more clear-cut. Beyond this, a dark olive-
brown triangular patch fills the area between the upper distal edge of the bar and the distal
border of the wing ; the two post-discal subapical spots are white, followed by large indistinct
golden-olive marks bordered distally with lilac scaling, culminating in the double black tornal
spot in rb with a smaller spot above. The outer border of the wing is slightly satiny. Hind
wing basal triangle olive-greyish, bordered distally by an almost straight whitish line in the
upper half, then shading darker to the discal white bar, which runs from the costa to above the
anal angle, slightly expanded on the outer border at about mid-point ; this is followed by a dark
satiny zone which narrows above the anal angle and becomes paler in the submarginal area
carrying a series of obscure lunate lilac marks with black at the veins, accentuated toward the
anal angle, where the double black marks are strong ; margin with indistinct lunate olive-ochre
marks.

Four other topotypical females agree in having the fore wing white bar considerably narrower
than in the type specimen, more particularly in the posterior half of it, so that the bar is almost
parallel-sided and there is a decided kink on the proximal side in rb. As in the type, the hind
wing patch is restricted on the proximal side.

Range: My investigations suggest that the nominate form of vzoletta is restricted
to Laurenco Marques and the area of Delagoa Bay, and slightly northward in the
coastal forest zone. It does not extend directly westward into the Transvaal, but
occurs on the east side of Rhodesia in the region of the Vumba Mts. in a slightly
different form, described hereafter.

Charaxes violetta transitional form
(Pl. 2, figs. 9 and 13)

Mate. Fore wing length 40 mm. Upperside. Ground colour blue-black with slight greeny
sheen at base. Discal row of spots blue, the spot in 1b rather quadrate, the streak in 1a slightly
longer ; the post-discal spots in the subapex white, the remainder blue, the spots in 4—5 missing.
(In this respect the fore wing is similar to the nominate form.) Hind wing similar to nominate,
but discal patch may be slightly extended toward base. The majority of males however differ
as follows: the fore wing discal spots are stronger and the post-discal series complete, the blue
marks in 1a—1b larger ; the discal patch in hind wing larger and less strongly blue. (PI. 2, fig. 9.)

FEMALE. Upperside. Fore wing basal area not so black as nominate or may be considerably
paler ; discal band narrow in area beyond cell but expanding gradually and widest toward the
hind margin. Hind wing discal patch larger and extended basad, usually carried up to costa,
paler and not so blue on borders, Underside. Somewhat variable. (Pl. 2, fig. 13.)

52 V.G.L. VAN SOMEREN

The species is plentiful in the area inland from Beira in the Dundo-Amatongas
and extends northward into Malawi, thence to the Southern Highlands of Tanzania.
This aggregate exhibits a fair degree of stability of characters tending towards, but by
no means identical with, nominate violetta. (Mlange to N. Malawi.)

We now consider specimens from Tanzania: Newala, Iringa, Morogoro area, the
higher zones of the Usambara Mts, and the area west of Kilimanjaro (Arusha-Meru).
It is from amongst this aggregate that the name melloni was applied to specimens
from Nguru, Morogoro district, because these differed from examples from the coast
line. Dr. Fox presumed that melloni represented a localized breeding community,
but in actual fact his specimens represented one variation (and was represented by
only one female) of a somewhat variable aggregate. (cf. Pl. 2-4, particularly type
of female melloni and PI. 2, fig. 16, Kimboza forest.)

It will be noted that although the aggregate exhibits features common to the
majority, some tend toward the Malawi and some toward the coastal forms. Thus
the name melloni Fox (1963: 222) is applicable to a form within this unstable

aggregate.

Charaxes violetta maritima ssp. n.
(Pl. 3, figs. rg and 20, 22-24)

Specimens of Charaxes violetta from the Kenya Coast differ markedly from examples
of nominate violetta from Delagoa Bay (cf. Pl. 2, figs. 1 and 2) more particularly in
the case of the female sex. This fact had already been noted by Rothschild & Jordan
in 1900, but Kenya examples have been lumped with nominate violetta even to the
present day, due no doubt to the confusing picture given by representatives of the
species in the intermediate areas of its range.

Mate. Fore wing length 36 mm. Upperside. Ground colour blue-black with slight greeny
sheen at base ; discal spots not extending to costa, strongly blue, slightly paler in 1b and longer
streak in Ia; post-discal spots: two subapical white, remaining spots of increasing size from
5 to 1b where the mark is contiguous with the discal mark; margin of wing with very small
obscure internervular blue dots. Hind wing base black; border black; disc of wing with
largish patch strongly blue scaled on outer border, slightly less so on inner, more whitish in
discal line, represented toward costa by an elongate blue mark and by two spots at subcosta.
Black border with submarginal bluish-white spots double at anal angle; margin with rather
divided bluish-white lunules ; fringe very slightly white between veins. Underside. Ground
colour rather darker than nominate race but pattern similar though white lines narrower but
clear cut.

FEMALE. Fore wing length 44 mm. Uppferside. Base of wing brownish-black shading
blacker distally, the junction of the black with the discal band rather dyslegnic. Discal white
band very broad, commencing at the costa as a long streak extending basad, from then on the
band widens very considerably and is 15 mm. broad at hind margin, the broadening of the band
distad toward the tornus thus encroaches on the distal half of the wing, so that the post-discal
spots, which are white in subapex and then bluish, stop in 3. Hind wing, basal brownish-black
area very restricted, merging rapidly into the very broad discal patch, which is largely whitish
with a slight bluish tinge proximally and very slightly on distal border ; the patch usually runs
right through to the costa as in the type, or it may be more strongly represented by the large

REVISIONAL NOTES ON AFRICAN CHARAXES 53

quadrate white spot in the discal line, as in Pl. 3, fig. 22. The black border carries a row of
submarginal bluish-white spots ; the margin a series of irregular lunate bluish-white marks up
to 6. Tails, upper 5 mm., lower 3 mm. Underside. Ground colour rather paler than in other
races, a lighter olive-brownish ; the black marks show up strongly in the basal area of the fore
wing and attornus. The white bar broad as above ; the post-discal whitish spots other than the
two subapical rather indistinct. Hind wing ground colour paler, the basal area not so strongly
defined from the white discal bar, which is broad and the post-discal zigzag dark line not so
heavy ; the lunate marginal marks well marked.

Holotype male. KENYA: Rabai Hills, vii.1922 (van Someren).
Allotype female. KENYA: Rabai Hills, v.1928 (van Someren).

This is a smallish race and remarkably constant, the chief variation being in the
intensity of the fore wing basal area and the degree of bluish suffusion on the proximal
side of the hind wing patch.

This subspecies has been bred in considerable numbers on Afzelia cuanzensts
Welw. (Caesalpinaceae) and Brachystegia (edulis Burtt Davy & Hutch.) spzciformis
Benth, but mainly on the former.

Range: This race has its chief locus in the forests of the Kenya Coast belt : Rabai
Hills, Shimba Hills, Sekoke-Arabuko forest and it extends inland to the Teita Hills.
It also occurs in the forests of the Shimoni-Vanga area, and in the Tanga-Dar-es-
Salaam (Pugu forest) areas. There is some indication of variation in the lower
forests of the Usambara Range at Amani (PI. 3, fig. 24).

How far the species goes along the Sabaki and Tana Rivers gallery forest is not
known, but there appears to be a considerable break between the race maritima
and that described hereafter.

Charaxes violetta meru ssp. n.
(Pl. 4, figs. 25 and 26, 29 and 30)

This subspecies is considerably larger than the race maritima and because of its
rather isolated position has evolved characters which entitle it to subspecific rank.

There is a certain superficial resemblance between this race and the aggregate from
Tanzania, but whereas these latter are unstable, the race meru is very stable and has
no contact with this cline, to which the name melloni has been applied.

MALE. Fore wing length 40-43 mm. Upperside. Ground colour blue-black with strong
greeny-blue sheen at base of wing in side light. Pattern very similar to maritima but discal and
post-discal blue spots more distinct ; the marginal internervular white spots more distinct.
The rather variable blue marks in 1a—1b conjoined with the post-discal. Hind wing discal patch
larger and less blue, with a more pronounced white discal zone. Underside. As in the other
subspecies, and very similar to maritima but white line on fore and hind wings rather broader ;
tails shorter and more robust, upper 4 mm., lower 2 mm.

FEMALE. Fore wing length 45-47 mm., thus larger than maritima. Upperside. Pattern
generally similar, but base of fore wing rather darker, the white bar broad, but considerably
narrower in subcosta at end of cell but widening appreciably in 3 then more gradually to hind
margin where it expands proximally and distad in 1a-1b. The hind wing patch margin with
distinct internervular white spots is large, extending basad and merging into the basal dark

54 V.G. L. VAN SOMEREN

area, the inner border suffused with violet-blue scaling ; the outer border clear cut and tinged
with violet-blue but crossed by a distinct white discal bar which extends to above anal angle ;
marginal border black with complete row of submarginal bluish-white spots ; the margin with
distinct whitish or bluish-white lunules ; fringe whitish between veins. Tails moderately long,
upper 7 mm., lower 4 mm. Underside. Generally darker than in maritima with white bars
more in contrast. In the fore wing the post-discal and submarginal spots more distinct, though
the tornal black spots tend to be smaller. On the hind wing the zigzag dark line is strong and the
submarginal spots are more distinct.

Holotype male. KerNnyA: Meru, N.E. of Mt. Kenya. British Museum (N.H.).
(M.S. Berkeley).

Allotype female. KENYA: Meru, near Mt. Kenya. xi.1935. British Museum
(NFL).

Range : The only areas from which this race has been recorded are Meru, north-east
of Mt. Kenya in both the upper and lower forests, and further north-east in the
Njombeni Hills.

SYSTEMATIC LIST
Charaxes violetta Grose-Smith

Charaxes violetta violetta Grose-Smith, 1885. Type locality: Delagoa Bay, Mozam-
bique. Range: Restricted to the southern end of Mozambique
in the Delagoa Bay, Laurenco Marques region.

A cline toward the Malawi aggregate ; some tendency toward nominate in
southern zones. Range: Eastern boundary of Rhodesia from Mt.
Selinda northwards along the Vumba Range.

A cline in the Malawi area, Mlange, north up the west side of Lake Nyasa
intergrading into the mixed aggregate of southern and east-central
Tanzania to western foothills of Kilimanjaro.

As melloni Fox, 1963. Type locality: Nguru Hills north of
Morogoro, Tanzania.

maritima ssp.n. Type locality: Rabai Hills, Kenya Coast zone. Range:
The coastal strip of N.E. Tanzania (Dar-es-Salaam, Tanga), Kenya
Coast forests; Shimba Hills, Sekoke-Arabuko forest, Ganda
Forest to Lamu ; also at Taveta.

meru ssp. n. Type locality: Lower Meru Forest, Mt. Kenya. Range:
The lower forests on eastern side Mt. Kenya, Meru, also Njombeni
Hills.

2. THE CHARAXES ETESIPE-PENRICEI ComMpiLex

Rothschild & Jordan reviewed the members of this complex in 1go0o, treating
elestpe etesipe as a nominate race with a distribution from West Africa to Uganda,
etesipe abyssinicus as the Ethiopian race, e. tavetensis as the East African race,
e. cacuthis as an insular race from Madagascar, and Charaxes penricei as a distinct
species from Angola.

REVISIONAL NOTES ON AFRICAN CHARAXES 55

During the past fifty years much additional material and information has accumu-
lated ; Lathy described etestpe paradoxa from the Comoro Islands in 1925, Joicey &
Talbot (1922) published the description of penricei dealbata from Portuguese Congo,
and other races and forms were added subsequently.

The last general review of the complex was by the late Professor Hale Carpenter
(1947). This elaborate paper was concerned mainly with the males of the species
comprising the group. Having dealt first of all with the named species and races
listed up to that time, and indicated their distribution, he provided a map showing
their respective ranges. This map gives the general impression that there is a very
considerable overlap of penricer and tavetensis, but in a footnote Carpenter states
“ the alternating position of the names ¢avetensis and penricei ... in Rhodesia and
south-eastern Congo, must not be taken to indicate details of distribution only the
general occurrence in the same area’”’. This map was based presumably on records
and on actual specimens and I shall return to this point later on.

Having reviewed the original references to members of this complex, Carpenter
selects certain points in the fore wing pattern and the general underside coloration to
demonstrate the relationship or divergence, one to the other. He takes the width
and position of the blue band at vein r and furnishes a table of percentage values,
and a statistical analysis of this data; later on the character and position of the
hind wing band is discussed. The arrangement of Table r (Carpenter, 1947 : 95)
suggests that there are three species in this complex : efesipe with three subspecies,
penricet with two, and cacuthis. It would appear that this division is based on the
following: (A) the evidence of measurements given in the table. (B) the apparent
overlap of penricei and etesipe tavetensis. (C) the present isolation of cacuthis and
the distinctive character of its female. In my view these data are inconclusive.

A new approach, aimed at elucidating the relationship of members of this complex,
has been proposed by Dr. G. van Son (in lit.). Having found that the antenna-wing
ratio was of value in separating very similar species of Satyrines, he suggested that a
series of antenna-wing measurements might throw light on the etesipe-penricer
problem. I accordingly submitted a small representative series of specimens to him.
The resultant ratios proved of considerable interest and were as follows :

TABLE I
Ch. penricet penricet . : ‘ : 3, 0°417, 0-418, 0°420, 0-416, 0-420
(Mumbwa, Zambia) 9, 0°376, 0°376, 0°374
etesipe etesipe ; : : : 3, 0°401, 0°402, 0-400, 0400
(Uganda) 2, 0°337, 0°336, 0°336
e. tavetensis . , : , : 3, 0438, 0°436, 0-440, 0°443, 0°436
(Kenya, Coast) Q, 0355, 0°357, 0°356
é. abyssinicus ; : ; : dh, 0°427, 0°427
(Gimma, Ethiopia)
ssp. n. , ; : ‘ : 3, 0°414, 0°416, 0°413, 0-412, 0-413
(W. Tanzania) dg, 0°412, 0-414 (No females sent.)

These data were submitted to Dr. C. B. Cottrell for analysis with the results shown
in Table IT,

56 V.G.L. VAN SOMEREN

TABLE II

penricet é. etesipe é. tavetensis e. abyssinicus e. ssp. 0.
No. in sample,
males only. ‘ 6 4 6 2 4
Observed range - 0*416—420 0+ 400-402 0*436-445 0°427 0*412—416
Mean . ; : 04185 0+ 4008 0+4388 0°427 00-4134
Standard deviation . 00017 00010 0*0027 — 0*0025
Standard error r 0:0009 0*0005 O:OOIT _ 0:0009
Coefficient of

variation . : 0-41 0:13 0:60 —_ 0-60

Dr. van Son (in lit.) suggests that each member of this complex should be rated as
a species. In my opinion, it would be rash to evaluate the antenna-wing ratio to
the degree suggested by Dr. van Son. It is however possible that these figures may
throw some light on the relationship of two or more entities in the group, on a sub-
specific plane. It will be noted that :

1. The data for the new subspecies come nearer to penricei than any other. The
large spots in fore wing are those of penricei ; the hind wing band is more
proximad than in éavetensts.

2. etesipe etesipe and penricer show a wide divergence in ratios and their general
pattern and appearance are different.

3. etestpe etesipe and e. tavetensis show the greatest difference in antenna-wing
ratios, but in upper side pattern they are not very widely different.

4. é. abyssinicus is intermediate between efesipe etesipe and e. tavetensis and the
upperside pattern inclines toward the former.

I think this aspect of the problem should be looked into further.

I am unable to accept, in their entirety, the views expressed by Carpenter (1947).
In order to check his data, I have brought together a very considerable amount of
material covering the entire known range of the insects involved and I have examined
types where possible or photographs of them and topotypes in all cases. In addition
I have examined the actual specimens “ doubtfully ” allocated by Carpenter, such
as those mentioned on page 96 and 102 of his review.

It is well known that in many groups of nymphalid butterflies, the underside
pattern (and not necessarily the coloration) is diagnostic of the species. On this
basis, it would seem probable that in the distant past before Africa was subjected to
vast changes in vegetational coverage, efesipe and penricer had a common ancestor.
Changes in topography and the elimination and separation of forest areas which we
know took place, caused a splitting up of this ancestral stock, the etesipe stock remain-
ing in the still well-forested areas to the north and the stock which gave rise to
penricei adapted itself to sparsely wooded and savanna types of country. There is
evidence suggesting that the penricet stock extended from west to east, well south of
the Equator, at a period prior to the complete separation of Madagascar and the
Comoros from the mainland of Africa. There is also some evidence suggesting that at
a later date, the efesipe stock and that of penricei spread, converged, but have not
interbred ; that etfesipe, in the form of race tavetensis, is a comparatively recent

57

REVISIONAL NOTES ON AFRICAN CHARAXES

yVOSVOVOVIN)

ad dv

ets
/ONVIOLNS VG
C557 -
~ rs
@ voluay HLNOS ie RY,
4O NOINY 1
{oriva izes eek H
7 t Vt
ie / Sau \
@ | és
(IvWASNWaL } H
- I
\ £7 H volusy

ieee of
eV wares : 1S3M
R/ ae { HLNOS
vVy \ !

< open nyatinos «_-—J

-— ames

3nd1 SWYZOW
Pf yee |
SON,

¥

0 eed ’ en | VIO9N
‘SGNV1S | O¥OWOD un? , ) ~~} vV NV
Vv al Of ’

ae
Cot *~

onace e @

viavWoSs

y "y

eo 4y VINVZNVL

O9NOD 7 e

vIdOIlH1L nvans \

6

SITHLONOVO

“vxoOavVuVd
IZ01YN3d

“JVALANVONVIE

I3901YN3d

“wivelv3ad
139 1YN3d

“139 1YN3d
}Z901YN3d

*VEW3d
adil'Ss3ia

“SISN3L3IAVL
Jd VSaLa

*tNOGYOS
aditsala

“SNOINISSASY

S3IXVYEVHD

S3XVYVH9

S3XVUVHO

SAXVYVHO

SIXVYVHD

S4XVUVHO

SAXVYVHO

3d1S313 S3XVYVHO

*3d1S313

3d1S313 S3XVYVHD
“ASM

ee Ee,

Piha

@o0e 6 © 44 <P

58 V.G. L. VAN SOMEREN

invader of the once penricei territory, following the establishment of coastal forest
belts and gallery forests down the east coast, thus creating suitable environment for
etestpe stock. On the other hand, the penricei stock on the Comoros and Madagascar,
by prior complete isolation, evolved into insular entities which can now be considered
species.

Viewing the complex as we see it today, Rothschild & Jordan were right in con-
sidering penricer to be a species distinct from efesipe, but it is difficult to understand
why they considered cacuthis to be a subspecies of the latter. Carpenter supports
the view that cacuthis is a species.

When the distribution of the various elements of this complex are mapped out in
some detail (vide map B), it would appear at first sight that each represented a
replacement subspecies of a single species, but with one or two exceptions, viz.
dealbata and etesipe in N.W. Angola, and the apparent overlap of penricei and etesipe
tavetensis in S. Tanzania, S. Malawi and eastern Rhodesia.

According to Carpenter’s map, the overlap was considerable in some areas, but
some of this apparent intermingling is due to misplacing of some specimens. There
would appear to be a definite territorial overlap of etesipe etesipe and penricei dealbata
in N.W. Angola. The latter was described from two males taken at Kibikolo do
Zombo in Portuguese Congo (north of the mouth of the Congo River) and one male
from Pungo Andongo in N.W. Angola. It would seem that efesipe cuts across
between these two localities for I have examined specimens of nominate e¢esipe from
Luanda and the Congo Prov. of Angola.

Other areas of overlap were suggested by Carpenter (1947 : 102) ; it is stated that
penricet and tavetensis both occurred at Geita in the Biharamulo district of Tanzania
to the south-west of Lake Victoria. I have examined these specimens and they are
examples of a race of penricer which are described later. Neither nominate etesipe
nor tavetensis seem to occur directly south of Lake Victoria, but the former does
extend to the eastern shore of the lake.

The records of tavetensis apparently overlapping with penrice: in Zambia, Malawi
and the eastern side of Rhodesia, require careful investigation in the light of
ecological preference displayed by these species. In the main, penricez is an insect
of savanna, whereas efesipe is an insect of forest in bulk or gallery forest. These
habitats may be defined and clearly separate or they may merge where savanna
gives way to forest ; or savanna country may be traversed by rivers with fringing
forest. Although there is this apparent overlap of distribution when depicted on a
small scale map, I can obtain no evidence from correspondents that the two insects
fly together in the same spot and might be taken in the same trap! Unfortunately
the majority of data labels fail to give information regarding ecological environment.

Some specimens I have examined were erroneously placed to ¢avetensis or even
dealbata, but they are in my opinion varieties of male penricez with a strong blue
suffusion over the lower fore wing spots and over the hind wing band; the fore
wing spots are large as in nominate penricer, the hind wing band is discal or partly
discal, not post-discal. Such examples are from Songea in S. Tanzania, Chintechi in
Malawi, Mumbwa in Zambia, Lundi in Rhodesia and south-east Katanga.

REVISIONAL NOTES ON AFRICAN CHARAXES 59

Carpenter (1947 : 99) deals very briefly with the female forms, and in his summing-
up states “it is, in my opinion, almost impossible to say which females are tavetensis
and which penrice: when taken in Nyasaland where the two overlap’”’. But do they
really overlap? The antenna-wing ratio should be of assistance (vide Tables I and
II). Females of nominate penricer have the pattern dead white, suffused with some
blue (with the exception of ab. flavus Lathy) ; in tavetensis, the fore wing pattern is
creamy.

It is generally assumed that the male genitalia in this complex exhibit no real
structural differences, except perhaps in the case of cacuthis (Carpenter, 1947 : 98,
text-figures).

The foregoing remarks will have indicated the difficulties encountered in attempting
to unravel this complex. I have quoted at length the diverse views on the problem
which have appeared in print and given my views on them.

The following section will provide a brief description of the species, subspecies and
forms, which come within this complex.

DESCRIPTIONS AND NOTES

Charaxes etesipe etesipe (Godart)
(Pl. 4, figs. 28 and 32; PI. 5, figs. 33-35 and 37-39)

Nymbphalis etesipe Godart, 1824 : 355 [9].
Nymphalis etheta Godart, 1824 : 356 [3].

Nominate examples from Sierra Leone are, on the whole, rather smaller than
specimens from the eastern Congo and Uganda ; this applies more to the female sex,
but there appears to be complete gradation between the extremes.

Mate. Fore wing length 36-40mm. Upperside. Ground colour blue-black (more brownish-black
in old specimens) with a greenish lustre more pronounced toward the base of the wing. Pattern as
follows : a crescentic white spot toward the end of the cell, followed by a large white spot sub-
basal in 6 with a small spot below in 5 often vestigial; large spots sub-basal in 3-2 in discal
line ; complete series of post-discal spots : two spots in line in 7-6, spot in 5 set in slightly, small
or vestigial ; larger spots in 4-3, all upper spots whitish ; spots in 2 to hind margin larger, and
bluish, spot in tb double. These post-discal spots follow the contour of the wing margin,
which carries a complete series of blue marks increasing in size from 7 to 1b, where the mark is
double. Hind wing ground colour blue-black ; subcostal white mark in discal line usually
present or may be vestigial ; a post-discal series of conspicuous blue spots from 6 to anal angle,
beyond a series of submarginal white spots, double at anal angle, but spot in 5 very small ;
margin above upper tail reddish or maroon, the crescentic marks to anal angle greenish. The
outer margin of the wing is characteristic, having a pronounced projection at the base of the
upper tail and a rounded projection at anal angle below lower tail. Underside. The pattern
is characteristic and highly complicated and is well depicted on the accompanying plates.
Ground colour of the fore wing cell whitish to cream crossed by three reddish bars outlined in
black ; reddish bars cross the sub-bases of 1b—3 ; the discal and post-discal spots of above, here
represented by larger greenish-white marks distally and proximally accentuated by black marks,
large and double at tornus; distal portion of wing greyish or greyish-brown incised by two
triangular creamy marks, base to margin, the lower along vein 4, the upper along vein 7; the
submargin with a fine black line; margin black. Hind wing ground colour greyish olive ;
costa conspicuously creamy-white from base to post-discal line; with a quadrate white spot

60 V.G.L. VAN SOMEREN

below in discal line; basal bars reddish-brown, followed by two elongate white marks in the
cell ; the discal line with whitish spots ; post-discal black dots in double row, large on upper part
of inner line, small on outer ; submargin with line of whitish linear marks, large at base of upper
tail and at anal angle, accentuated by a triangular black mark at base of upper tail; margin
reddish above upper tail, greenish toward anal angle.

Variation. Males from Uganda, besides being larger on an average, have the upper fore wing
spots conspicuously white; the margin of the wing with conspicuous internervular glaucous
marks ; the costal spot on the hind wing may be absent, small or double ; the post-discal line
of blue contiguous spots small, and the bar narrow, or the spots large.

FEMALE. Fore wing length 40-42 mm. in topotypical specimens. Upperside. The original descrip-
tion of the type of the species is reasonably accurate but states that the ground colour is blue-
black and the oblique bands crossing fore and hind wings are white, no mention being made of the
shading to the borders of the hind wing band. In specimens I have seen, the ground colour is
brownish-black with a strong greenish-blue sheen when fresh, the distal half of the wing blacker.
The pattern of the fore wing is basically that of the male but modified : the spot in the cell is
small or often absent and the spots beyond may be four or reduced to one subcostal and one in 3.
The post-discal or discocellular bar is well marked and inclined discally in hind portion; the
three subapical spots are small, the others increasing in size on the proximal side so that the
band in fa is in line with the costal mark in the hind wing. The colour of the bar is seldom white
and is usually creamy or even yellowish tinged. The marginal marks are limited to a few in
fringe between the veins. The hind wing bar varies somewhat in width and extent toward the
anal angle; it is well marked at the costa, slightly narrower in the space below, then of more
even width but tapering above anal angle at inner fold. The band is whitish or yellowish white
with a varying amount of bluish or greeny-bluish scaling on the borders. The submarginal
line of spots may be complete, or absent above upper tail, and whitish in colour; the margin
above the upper tail is dark reddish, then strongly black at base of tails, then greenish to anal
angle. Underside. Pattern somewhat as in the male, but the discal band more clearly indicated
and wider with, on the distal side, a complete row of submarginal black marks of increasing size,
largest and double at the tornus ; then follows a thin blackish admarginal line ; marginal border
greenish buff. Hind wing pattern as in the male, but subcostal mark larger and the light discal
bar of above may be slightly indicated by creamy interspaces between the reddish bars in the
base and the black post-discal spots ; border as in the male but marks accentuated.

Variation. I have examined one specimen from Calabar, Nigeria, in which the spots of the
fore wing post-discal line are well separated and blue-tinged except for the subapical spots ; the
hind wing band is strongly suffused with blue. This variety would appear to be analagous to
form caeruleotincta Carpenter which occurs amongst the Uganda aggregate. This specimen bears
a strong likeness to male Ch. penricei dealbata Joicey & Talbot or even to the blue variant of
male penricei and should not be confused with either (Pl. 5, fig. 38). Mr. R. G. T. St. Leger,
of Enugo, Nigeria, is of the opinion that the creamy-yellow and bluish barred female forms
are seasonal.

Charaxes etesipe : Eastern Uganda aggregate

As already mentioned, specimens of efesipe, especially the females, tend to be considerably
larger than nominate Sierra Leone examples. The majority of males have a fore wing length of
42mm. There is already a tendency in many examples toward a widening of the blue band in
the post-discal zone of the hind wing (PI. 5, fig. 33).

Females are more variable and are consistently larger, not infrequently with fore wings of
58-60 mm. Though of very considerable expanse, the oblique bars on the fore wing and hind
wing tend to be generally narrower throughout, because there is not the abrupt difference in
length of the spots in 4 toward the hind margin, and the discal spots in 2-3 are narrow and not
much shorter than those in 1a—1b; the marginal spots are very small and may be represented
only by those of 6-7.

REVISIONAL NOTES ON AFRICAN CHARAXES 61

Variation. (a) The form caeruleotincta Carpenter has already been referred to and so far, has
only been taken in the region around Entebbe ; (b) a form in which the fore wing bar is almost
white, that of the hind wing white with strong blue borders and tapering rapidly toward the
inner margin; (c) a form in which the fore wing bar is creamy yellow, the hind wing band
creamy-yellow with greeny-blue suffusion on the borders; (d) the form castoroides Poulton
(1926 : 549), in which the cell spot in fore wing is orange, the discal spots ochreous and the fore
wing bar strongly orange-ochreous, as is also the hind wing band except for the costal spot and
one below which are white. In all these forms there is a tendency for suppression of the sub-
marginal white marks above the upper tail to upper angle and lack of a green border between
the tails but the anal lobe is margined with greeny-ochre or green.

Range: The nominate race ranges from Sierra Leone to Nigeria and the Cameroons
and northern Angola through the Congo Republic to southern Sudan ; in slightly
larger form in Uganda and in Kavirondo in north-west Kenya. The nominate form
is also recorded from Upembe in Katanga. To the south of Lake Victoria and the
eastern side of Lake Tanganyika efestpe overlaps with, but is replaced by a race of
penricet.

The Uganda form of etestpe has been bred in numbers on Ricinus communis and
Phyllanthus sp. (Euphorbiaceae).

Charaxes etesipe abyssinicus Rothschild
(Pl. 5, figs. 36 and 40)
Charaxes etesipe abyssinicus Rothschild in Rothschild & Jordan, 1900: 458.

In the original description this Abyssinian race was contrasted with e. tavetensis
from which it differs very considerably ; it however resembles nominate etesipe
fairly closely on the upper side.

Mate. Fore wing length 40-42 mm. Upperside. Ground colour blue-black with a strong
blue sheen at bases of wings. Pattern of fore wing as in nominate race, but post-discal spots in
1a—2 slightly larger and more strongly blue or violet-blue. Hind wing pattern as in the nominate
form, but subcostal spot usually blue and the post-discal series of spots larger and more contigu-
ous so that the band is wider and in this respect resembles the wide-banded males of the Uganda
strain, the colour is however more intensely dark blue or violet-blue, cf. Pl. 5, fig. 36. Underside.
The pattern is generally similar to that of the nominate race but the ground colour is darker,
more chocolate-brown, and the bars at the bases of the wings a deeper chestnut red-brown ;
the black lines and spots generally heavier.

FEMALE. This sex was unknown to Rothschild & Jordan and remained so when Aurivillius
(1911) dealt with the species. It is not mentioned by Carpenter (1947). The first full description
appears to be that by Ungemach (1932 : 53), a translation of which follows :

On the fore wing the discal band is pure white and broad, on 1b 8-9 mm. This continues
toward the costal border in a series of round spots, white in spaces 4-8. Two similar spots of
the same size, I-2 mm. are present between the cell and the median line in 3-4. There is a
straight brownish-grey marginal band which contains, toward the apex, some minute white dots.
The median band which continues into the hind wing is at first, in space 7 also pure white but is
tinged with blue in space 6; from there to the anal angle the band is 12-14 mm. broad. The
underside markings show through in a bluish tint, and the whole of the distal half containing
several acute-angled spots between the veins, is blue. The marginal spots in Ic and the marginal
line up to space 4 are also bluish. The underside of both sexes (pattern) is similar to efesipe,
but the shade of the ground colour is very different . . . reddish-brown instead of olive-green.

ENTOM. 18, 3 5

62 V.G. L. VAN SOMEREN

Through the kindness of Dr. G. Bernardi, of the Paris Museum I am able to give a
photograph of a female taken by Ungemach (PI. 5, fig. 40).

Range: This subspecies is found mainly in the high country to the west of the
Abyssinian Rift. Ungemach records it from Youbdo and Tchella and I have
examples from Gimma and Adola. Specimens from the southern Sudan that I have
examined do not appear to differ from Uganda material, though Carpenter thought
them to be “ transitional ’’ towards abyssinicus.

Charaxes etesipe patrizii Storace
Charaxes etesipe patrizii Storace, 1949 : 21.

This race was founded on one male specimen from Somalia and is characterized by
its small size, the length of the fore wing being 36:5 mm. Its validity can only be
established by the examination of more material.

Charaxes etesipe tavetensis Rothschild

(Pl. 6, figs. 41, 42, 45, 46)

Charaxes tavetensis Rothschild 1894 : 535.
Charaxes etesipe tavetensis Rothschild & Jordan, 1900: 458.

The type is a very damaged male captured in the Taveta district, Kenya. Roths-
child & Jordan associated with it three females from Dar-es-Salaam, Tanzania, and
a male from Zomba, Malawi. This race is now known to extend down the east coast
to as far as N. Zululand. Some of the records given by Carpenter (1947) from more
inland areas such as the central block of Zambia, Katanga, and the south shore of
Lake Victoria seem erroneous, for a check on the actual specimens indicate that they
do not belong to this race of efestpe ; his distributional map is thus at fault in these
respects. I shall refer to these specimens in the appropriate place.

Mate. Fore wing length 38-41 mm. Upperside. Ground colour blue-black with blue or
greeny-blue sheen at base of wings. Fore wing pattern as in nominate etesipe, but discal white
spots slightly bolder ; the line of post-discal spots more incurved in area 3 conforming to the
slightly incised outer margin of the wing; the spots in 1a—2 usually bluer. The hind wing pat-
tern basically similar to etesipe but the post-discal blue bar a brighter blue and shading to whitish
at inner fold, the whole band wider, the blue in 1-4 being fused as a block though the spots in
6-7 are elongate and free; the costal spot in discal line is white. Submarginal white marks
elongate, marginal border above upper tail maroon, then from tail to anal angle greeny-blue.
Underside. Pattern as in elesipe etesipe but basal reddish-brown bars in fore wing rather stron-
ger; the pale triangle in 3-4 less distinct ; on the hind wing the discal white to cream inter-
spaces rather larger.

FEMALE. Fore wing length 40-44 mm. Upperside. Pattern conforming to the general
pattern of the nominate race but fore and hind wing light bars broader. The spots in the fore
wing discal line larger and there is a compensating reduction in the length of the post-discal spots
in 3-4, but an increase in the marks in 1a—2, thus the bar tapers more rapidly from the subapical
mark in 2 toward the costa ; the spots in 2-3 may be separate or just touching. The hind wing
bar is of almost equal width from costa to inner fold though tapering slightly toward the anal

REVISIONAL NOTES ON AFRICAN CHARAXES 63

angle. The submarginal linear white marks are usually clear and distinct, but occasionally
smaller and obscured above upper tail to upper angle. The margin above upper tail usually
strongly reddish, then greenish to anal angle ; extreme edge black with white fringe internervu-
larly. The fore wing bar is usually creamy, the hind wing band whiter but flushed with bluish
to greenish on the borders. In other specimens the fore wing bar may be ochreous, the hind
wing band also ochreous except for the whiter mark at costa. Underside. Differs from the
nominate in the manner described for the males. There is an enlargement of the white or creamy
areas due to the increase in width of the pattern above, thus the underside appears lighter.

There is some degree of instability in the width of the hind wing blue band in
southern representatives of this race in the male sex, and an even greater degree of
variation in the size of the females.

Range: This subspecies occurs in the western foothills of Kilimanjaro (Arusha
and Moshi), and east of the Pare Mts., in the Taveta district ; it is plentiful on the
Teita Range and in the coastal forests of Kenya, extending southward to Songea and
the Ruvuma River in Tanzania, where it apparently meets penricer. It is recorded
from the southern end of Malawi; in the forests of Mozambique inland from Beira ;
it occurs on the eastern side of the Vumba Mts., in Rhodesia and reaches as far
south as N. Zululand.

Females have been observed laying on Entada pursaetha DC. (gigas auct.) (Mimo-
saceae) in the Teita and Coast districts of Kenya ; several families have been raised
on this plant.

Charaxes etesipe gordoni van Someren

(Pl. 6, figs. 43 and 47)
Charaxes etesipe gordont van Someren, 1935 : 198.

Described from six very uniform males. Upperside very similar to efesipe tavetensis but fore
wing white spots rather stronger; on the hind wing the submarginal linear white marks more
developed. On the underside the basal chestnut-brown marks stronger in fore wing, those of
hind wing also stronger ; the double row of black submarginal marks strong, the tornal double
spot marked. The dark marks toward the upper angle of the hind wing strong. The ground
colour of fore and hind wings darker.

At the time of describing this race the female was not known ; aspecimen has since been taken.

FEMALE. Fore wing length 45 mm. Upperside. Ground colour black with strong green
sheen toward base ; spot in cell moderately large ; discal spots large and rounded with exception
of minute spot in sub-base 5 ; upper post-discal spots large, that in 5 smaller, others increasing
in size to 3-4, then spots broad to hind margin; all marks creamy white ; marginal spots 6-7
present but obscured. Hind wing with broad bar, white at costa, then white with slight creamy
tinge but strongly bluish on outer border ; the dark pattern of below shows through slightly ;
submarginal linear white marks bold except for that in 5 which is small ; marginal border above
upper tail maroon, then becoming greenish to ochre beyond anal angle. Tails very long and
slender, upper 10 mm., lower 12 mm. Underside. Strong chestnut-brown bars present at base
of fore wing and in subcostal base of hind wing, as in the male; fore wing with pale bar well
marked. Distal half of hind wing, ground colour greyish-olive, with black marks clear in sub-
margin, the black mark at base of upper tail very strong ; margin above upper tail maroon then
ochreous to greenish to anal angle.

64 V.G.L. VAN SOMEREN

Neallotype female. KENyA: At lower Meru forest near Mt. Kenya, ili.1950
(R. E. V. Saunders). British Museum (N.H.).

Range: So far only recorded from the north-east side of Mt. Kenya at Meru and
Njombeni Hills.

Females noted laying on Entada sp. with small leaflets. Larvae found on younger
sprays of leaves at forest edge ; refused to feed on Entada abyssinicus.

Charaxes etesipe pemba ssp. n.
(Pl. 6, figs. 46 and 48)

Carpenter (1947: 99) records a female specimen from Pemba and a male from
Zanzibar and places the latter to ¢avetensis, and assumes that the Pemba female
also belongs here.

Pemba Island is well north-east of Zanzibar and the two islands are not by any
means similar ; the former has a considerable amount of forest coverage and there
are some known endemics on it, such as Charaxes pembanus Rothschild and a local
race of Euphaedra neophron.,

I have recently received a small series of etesipe from Pemba Island taken by Dr.
Arthur Rydon and some from Mr. T. H. E. Jackson. These specimens indicate that
these examples of efestpe are not tavetensis but represent an insular and distinct
subspecies.

Mace. Fore wing length 41-42 mm. Upperside. With rather acuminate wings, the outer
margin being strongly incised in 3-4, more so than in ¢avetensis, and much more than in nominate
etesipe. The line of post-discal spots is thus more curved, conforming to the contour of the
outer margin. These spots and those of the discal line are bold and show up clearly in the strong
blue-black ground. The hind wing blue band is post-discal and is narrower than in efesipe
tavetensis from 1c—4, the spots in 5—6 are free and there is an indication of a spot in 6; the sub-
costal mark is absent or just faintly indicated. The reduction of the band is on the inner side.
The submarginal linear white marks are strongly developed ; the margin above the upper tail is
without red or there is just an indication of red in the otherwise greenish line which extends to
the anal lobe. Underside. Generally darker than tavetensis, the basal brown lines are almost
black ; the hind portion of the fore wing almost black shading to bluish-grey and the double
line of black submarginal spots with bluish connecting ground pronounced in 1b and above at the
tornus ; the upper outer half of the wing more greyish-olive, less buffy.

FEMALE. This is relatively large, fore wing length 48mm. Upperside. Ground colour black
with strong green sheen at bases of wings. Spot in cell distinct, upper subcostal large with two
white streaks above at costa ; two minute dots in 4 and 5, large spots in 3—2 contiguous or fused
with the post-discal spots ; post-discal spots in sub-apex rather long ovoids, the lower spots
increasing in size to hind margin; this bar is creamy, or white, tinged slightly with cream.
The hind wing band, continuous with fore wing bar, is whitish at costa, then white with bluish
or greeny-blue on borders ; submarginal linear white marks distinct and bold ; margin maroon
above upper tail or mixed with greenish, rest of margin green to anal lobe; edge black with
white fringe between veins. Tails long and slender, upper 6 mm., lower 7 mm. Underside.
Basal chestnut bars strong; the upperside pattern strongly represented below and whitish ;
the submarginal row of graduated black spots strong and the mark in 1b at tornus very marked ;
admarginal black line distinct. Hind wing costal white border strong ; basal brown lines strong,
and the discal band indicated by creamy interspaces clear; distal third of wing olive-greyish
traversed by a double row of black post-discal spots ; submarginal white line outlined distally in

——,
2 ati

——

REVISIONAL NOTES ON AFRICAN CHARAXES 65

black distinct, especially the black spot at base of upper tail; margin ochreous-olive; edge
black with white fringe between veins. The underside is thus more heavily marked and more
contrasty than in favetensis.

Holotype male. TANZANIA: Pemba Island, north east of Zanzibar, ix.1963
(A. Rydon). British Museum (N.H.).

Allotype. female, same data. British Museum (N.H.).

Charaxes penricei penricei Rothschild

(Pl. 7, figs. 49-51, 53-56)
Charaxes penricei Rothschild in Rothschild & Jordan, 1900 : 460 [3].
Charaxes peculiaris Lathy, 1906 : 125.

Male. Fore wing length 38-40 mm., more pointed than in efesipe etesipe, outer margin
distinctly incised. Upperside. Ground colour blue-black with strong greeny-blue sheen at base ;
pattern somewhat as in female efesipe thus differing considerably from male efesipe. The spots
in the distal portion of the wing are large and arranged as follows: one large spot at end of
cell, triangular or quadrate; a large spot, elongate-quadrate sub-basal in 6 with a small spot
sub-basal in 5; large spots sub-basal in 3 and 2, round and quadrate; two large subapical
spots in 7-6 followed by larger more ovate spots in 4-3 set in more proximad ; a large double
fused somewhat quadrate mark indented on inner border in 1b followed by a longer narrower
mark in 1a. The upper spots are white, those in 1a—2 with bluish scaling on borders. The
position of the oblique bar is thus set in toward the discal line so that there is a wide distance
between its outer edge and the margin of the wing. Margin with distinct white internervular
spots, double in rb. Hind wing ground colour blue-black, crossed by a broad band commencing
at the costa and extending toward the inner fold ; the costal mark in the discal line is white,
slightly bluish distally, it may be fused with, or slightly separated from, the smaller bluish spot
below which may be fused with the larger contiguous spots which extend to the inner fold above
the anal angle. This bar is white in the central line, bluish on the borders, particularly strong
on the distal edge. The dark pattern of below is slightly apparent in the band. There is a
complete row of marks in the submargin, which may be linear or triangular and edged distally
with blue; double in anal lobe. Margin above upper tail usually maroon then narrowly blue
or bluish-green to anal angle ; edge black with white fringe between veins, strong at anal lobe.
Underside. The coloration and pattern is confusingly like etesipe tavetensis though less variable ;
the chief difference lies in the position of the oblique white bar which is much more proximad in
penricei and being wider shows up more distinctly. On the hind wing the costal and subcostal
white is more prominent because it is whiter and less creamy tinged. In other respects the mark-
ings are similarly placed (Pl. 7, figs. 49 and 50).

Variation. It is appropriate to draw attention to certain variations in the male sex. Two
such variants are depicted on PI. 7, figs. 51 and 56, which are figured alongside penricei dealbata
which they somewhat resemble. This departure from the nominate is mainly in the degree of
bluish scaling suffusing the spots in the fore wing band and over the discal band in the hind wing.
This may be accompanied by a reduction in the width of the hind wing band in 6~7 as in fig. 51
or a division of the mark in 6 as in fig. 56. Such variants seem to occur in greater degree on the
margin of the range of the nominate race, to the east, such as at Lundiin Rhodesia, Zomba in
Malawi, Songea in Tanzania and less frequently in Zambia.

The aberration peculiaris Lathy (1906 : 125) is merely one in which the bluish scaling is replaced
by greenish-blue.

An aberration from Lundi is interesting in that the mark in fore wing in area 1b-2 on the
underside is a long double streak.

FEMALE. The typical female is in most respects an enlarged replica of the male, but with less
strong blue-black ground and less blue scaling in the pattern, Fore wing length 43-45 mm,

66 V.G.L. VAN SOMEREN

Upperside. Ground colour black with bases of wings scaled with olive and some greeny-blue
sheen. Fore wing pattern as in the male but larger and bolder, all marks pure white occasionally
with a slight amount of blue-green on proximal edge of large spot in tb. Hind wing with a
comparatively wide band filling the discal area from costa to the inner fold above the anal angle,
the band is mostly white with some blue on the distal border ; there is an indication of the under-
side pattern showing through. Submargin and margin as in the male. Tails long and slender,
upper 8 mm., lower 9 mm. The typical female is thus somewhat like a pale barred female of
etesipe but the fore wing bar is whiter and the hind wing band bluer, and the shape of the fore
wing is different. Underside. As in the male but white pattern bolder, in keeping with the
white pattern of above (Pl. 7, figs. 53 and 54).

Variation. In very fresh specimens the fore wing band may be very slightly creamy but this
soon becomes white. However, one does find very rarely a specimen with the light pattern above
definitely creamy-yellowish ; this is ab. flavus Lathy (1926 : 95) which is recorded from Chintechi
on the west side of Lake Nyasa (Pl. 7, fig. 55).

Range : The nominate fenricei ranges from about central Angola eastward through
the western block of Zambia and southern Katanga northward to the Southern High-

lands of Tanzania, just north of Lake Nyasa, south of Malawi to Rhodesia mostly in
the Vumba area and just south (vide map B).

Charaxes penricei dealbata Joicey & Talbot
(Pl.9, fie. $2 (PAS, fie.57)
Charaxes penricei f. dealbata Joicey & Talbot, 1922 : 337-338.

Originally placed as a variety of penricez, it seems best to treat it as a subspecies,
since there are no records of nominate penricez within the area of its distribution.
This subspecies was described from two males captured in Portuguese Congo, and
associated with them was a specimen from Pungo Andongo, which is in north-west
Angola. This last was cited as a paratype, and was figured by Carpenter (1947) ;
it has been made available to me for study by the British Museum (N.H.).

Mate. General shape that of penricei (thus unlike etesipe), fore wing length 36 mm. Upper-
side. Ground colour brown-black but probably faded from a blue-black. Fore wing pattern
as in nominate penricei ; upper white spots slightly larger to 3, the outer spot in 2 bluish with
white proximally, the spots in 2—3 well separated ; the double blue spot in rb more quadrate,
that in 1a narrow and slightly longer; the hind portion of the band is thus narrower than in
nominate penricei. The hind wing band starts in 6 with a blue spot confluent with the white
costal mark, the next two spots are free, the remainder contiguous and slightly wider, that at
the side of the inner fold, the longest. There is no indication of a white central area, the band
is thus solidly blue and is just off the discal line at the costal end. Submarginal white spots
small but more pronounced in anal angle; margin with very little reddish above upper tail,
but greenish to anal lobe. Underside. Fore wing basal lines are strong but dark brown with
slight black outline ; the light areas have a bluish tinge, the lower half of the distal portion of
fore wing also bluish, otherwise general pattern as in nominate penricet.

The female seems to be unknown.
Range: Portuguese Congo to the north of the mouth of the Congo River, also

Pungo Andongo in north-west Angola. This range is divided by the presence of

etesipe etestpe, which occurs in the Luanda area of Angola, northward to the Congo
River,

REVISIONAL NOTES ON AFRICAN CHARAXES 67

Charaxes penricei tanganyikae ssp. n.
(Pl. 8, figs. 58-60)

I have before me a series of 12 males and a few females which exhibit characters
suggestive of dealbata, thus belonging to the species penricei, rather than to etesipe
or to any of its races.

The points indicative of close relationship to penricei are: MALE. Upperside. The size of the
spots in the fore wing and the position of the outer row forming the bar, i.e. the spots are set
well in from the outer margin, thus more proximad than in favetensis ; the band is also wider
in its lower half. In the hind wing, the blue band is also more proximad and, as in the case of
dealbata, this band lacks any white except for the costal quadrate spot which may be large and
continuous with the next blue spot or small and separated from it. There is some variation in
the width of this hind wing band ; if wide, it extends proximad toward the mid-discal line and
there may be a few whitish scales in the centre. In this latter instance the specimen bears a
strong resemblance to the blue-banded variations of nominate penricei, already mentioned
under that species. Underside. In general appearance some examples are almost as dark as in
etesipe abyssinicus, thus considerably darker than nominate penricei, but the position of the
light bar in the fore wing and the white interspaces on the hind wing are similar to penricet.

FEMALE. Upperside. Pattern similar to that of nominate penricez in both fore and hind
wing, but with slightly less blue. Underside. Markings are similar to those in the male except
for a strong indication of white bars of above.?

Holotype male. TANZANIA: Kigoma. East side Lake Tanganyika. xii. 1961
(Japanese Expedition).

Allotype female. TANZANIA: Geita, Biharamulo district, x.1937 (7. H. E. Jack-
son). British Museum (N.H.).

Range: This subspecies is found mainly on the eastern side of Lake Tanganyika
from Kigoma-Mpanda and on both sides of the south end of the lake and thus extends
slightly into the Albertville area of Katanga. To the north, it extends into the
Biharamulo district of western Tanzania and the Geita area to the south west of
Lake Victoria. This distribution embraces localities from which came specimens
determined by Carpenter as (a) penricei female. Specimen r. (b) male, not typical
penricet ..? intermediate penricei tavetensis. Specimen 5.

These specimens have been examined by me and carefully checked ; they belong
to penricei tanganytkae.

Charaxes paradoxa Lathy stat. n.
(Pl. 8, figs. 6x and 62)

Charaxes etesipe paradoxa Lathy, 1926: 94 [3].

The relationship of paradoxa from the Comoro Islands to etestpe and penricer has
been debated by Carpenter (1947: 95, Table I) who gave it as his opinion that
paradoxa is related to tavetensis and thus a subspecies of eteszpe.

2 Variation. Since the description of the female was written, other similar examples have been
received from the Kigomaarea. In addition, one female has enlarged fore wing spots, strongly tinged
with blue, especially those of 1a-2; the hind wing patch is broad, strongly suffused with blue but with
white central marks in 5-6, and the costal mark elongate-quadrate. TANzanrta: Kigoma area, Mihumo,
vi.1965 (Japanese Expedition).

68 V.G.L. VAN SOMEREN

Unfortunately, I have been unable to examine actual specimens, and must rely on
my interpretation of the excellent photographs of both sexes kindly supplied by the
British Museum (N.H.).

Mate. It will be noted that the fore wing inner spots are large, considerably larger than in
etesipe or tavetensis ; that the post-discal series forming the main bar are also large, and that the
lower spots in 1a, 1b—2 though more restricted than in penricei are very similar to ssp. tangan-
ytkae ; that this row is incurved at 3, following the contour of the wing ; that on the hind wing
the blue band is set in as it is in fanganyikae and is in line with the costal white spot. The
submarginal spots are large ; the tails relatively short and more robust. On the underside the
pattern comes within the variation of the tanganyikae aggregate for the male ; that of the female
is distinctive and unlike that of tavetensis.

In my opinion paradoxa appears to have been derived from penricei through a
form very like tanganytkae (to which it bears a close resemblance) but because of its
long isolation and stability of characters it should rank as a species.

Range : Grand Comoro Island.

Charaxes cacuthis Hewitson

(Pl. 8, figs. 63 and 64)

Charaxes cacuthis Hewitson, 1863 : 63 [d].
Charaxes antanala Lucas, 1872: 1 [9].

It is generally agreed that cacuthis has been derived from a penricei-like ancestor,
and that by reason of prolonged isolation has evolved characteristic features in both
sexes, divergent from any mainland form, and should now be considered an insular
species.

MALE. Fore wing length 36-38 mm. Upperside. Ground colour blue-black with a strong
blue sheen basad. The pattern is similar to the blue variation of nominate pemnricei and very
similar to penricei tanganyikae ; the forward spots are large and white, the blue marks in 1a—2
extended more basad so that the band is here broader. The marginal spots are indistinct and
glaucous. The hind wing blue band is wide, almost straight on its inner border, slightly angled
at 4 on outer margin; the costal spot is white and directly below are two white marks in disc
of the band and some whitish scales along the inner border in some examples. Margin above
upper tail slightly reddish, then blue to green to anal lobe ; submarginal spots strong but small.
Outer margin of hind wing to upper tail rather dentate. Underside. Basic pattern typical of
the group ; basal marks chestnut with black edges, all black in tb; the marks well developed ;
the tornal black mark strong. Hind wing costal border strong, subcostal discal white mark
distinct but small followed by more buffy white marks in discal line ; margin above upper tail
more reddish than above ; black mark at base of upper tail strong ; margin to anal angle buffy-
olive and green.

FEMALE. This sex exhibits considerable divergence from any mainland representative of the
group. Fore wing length 40-42 mm. Upperside. Ground colour brown-black with some
purply-green sheen at base. Cell with indications of dark marks from below, subterminal spot
orange-tawny conjoined to mark in 4 by more obscure orange spot ; discal spots in 4—6 elongate,
conjoined to post-discal elongate spots by rays; these spots tawny-orange to orange as are the
more elongate spots in 2-3, which are a fusion of the discal and post-discal spots, the line of
fusion indicated by thin blackish lines ; the large quadrate spot in tb and the elongate mark in

REVISIONAL NOTES ON AFRICAN CHARAXES 69

1a white, the former with some orange scaling on outer border and bluish scaling on inner ; there
is some blue on distal end of mark in 1a. Margin of wing rather dentate in mid area and with
tawny spots in 6-7. Hind wing with a broad white discal band with strong blue scaling on
outer border on lower half, only slightly blue on inner ; some indication of dark marks of below
in disc of band. Submarginal white spots strong or more obscured in region of tails ; margin
with strong brick-red above upper tail then bluish-white with greenish to anal lobe. Tails long
and slender 9-10 mm., black with a conspicuous white line in centre. Underside. Basic pattern
as in the male but white areas enhanced by broad bar to fore wing and strong discal band in
hind wing ; the lines and spots distinct on a paler ground.

Range: Madagascar, mainly in the centre and northern half of the island.

SYSTEMATIC LIST
Charaxes etesipe (Godart)

Charaxes etesipe etesipe (Godart, 1824). Type locality : Sierra Leone. Range: Sierra
Leone, Nigeria eastward to S. Sudan, Congo, N.W. Angola, N.
Katanga, Uganda and N.W. Kenya.

f. castoroides Poulton, 1926. Jinja, Uganda.

f. caeruleotincta Carpenter, 1945. Entebbe, Uganda.

abyssinicus Rothschild, tg00. Type locality: Shoa, Ethiopia. Range:
Ethiopia, west of the Abyssinian Rift, mainly in high country.

patrizit Storace, 1949. Type locality: Ola Uagér (Oltregiuba), Somalia.

tavetensis Rothschild, 1894. Type locality: Taveta, Kenya [3], Dar-es-
Salaam, Tanzania [2], Range: Western foot-hills, Kilimanjaro
(Moshi Arusha); east of Pare Mts., Taveta forest, Teita Hills ;
Coast forests, Kenya. Tanzania north of Ruvuma River, inland
to Songea district. Malawi south, Mozambique inland from
Beira ; eastern Rhodesia, south to Zululand.

gordont van Someren, 1935. Type locality: Meru, Mt. Kenya. Range:
North-eastern side of Mt. Kenya and the Njombeni Hills, Kenya.

pemba, ssp. n. Type locality: Tanzania, Pemba Island. Range: Pemba
Island off east coast, N.E. of Zanzibar.

Synonym: etheta (Godart, 1824). Type locality: ‘‘ Cote occidentale
d’ Afrique ”’.

Charaxes penricei Rothschild

Charaxes penricet penricei Rothschild, 1g00. Type locality : Quebe River, Angola.

Range: Central Angola, southern Katanga, Zambia, the southern
Highlands of Tanzania north of Lake Nyasa, South Malawi, and
eastern Rhodesia.

ab. peculiaris Lathy, 1906 (male). Kawama, N.E. Rhodesia.

ab. flavus Lathy, 1925 (female). Chintechi, Malawi.

dealbata Joicey & Talbot, 1922. Type locality: Kibokolo do Zombo,
Portuguese Congo [3]. Paratype [g] Pungo Andongo, N. Angola.
Range: Portuguese Congo and N.W. Angola at Pungo Andongo.

70 V.G.L. VAN SOMEREN

tanganytkae, ssp.n. Type locality: Kigoma, east side Lake Tanganyika.
Range: The eastern and southern shores Lake Tanganyika,
also in Biharamulo district, Tanzania, Geita, south-west Lake
Victoria.

Charaxes paradoxa Lathy
Charaxes paradoxa Lathy, 1926. Type locality: Grand Comoro Island. Range:
Comoro Islands.

Charaxes cacuthis Hewitson

Charaxes cacuthis Hewitson, 1863. Type locality: Madagascar. Range: Limited
to Madagascar.

3. CHARAXES PEMBANUS Jorpan and CHARAXES USAMBARAE van
SOMEREN & JACKSON

Charaxes pembanus Jordan
(Pl. 9, figs. 67 and 68, 71 and 72)
Charaxes pembanus Jordan, 1925 : 289.

This species was described by Jordan from a single rather damaged male captured
on Pemba Island, which lies off the East African coast, north-east of Zanzibar and
about midway between Mombasa and Tanga.

This specimen remained unique until March 1963 when Commdr. Hollebone cap-
tured a pair, unfortunately extremely tattered when handed over, but still recogniz-
able. This exciting capture induced Dr. Arthur Rydon of Mombasa, to fly over to
Pemba for a week. By using a battery of hanging traps he was lucky enough to
secure a perfect pair. On leaving, a couple of traps were left in charge of a friend
and this resulted in two more females being captured. A short time afterwards, the
island was visited by Mr. T. H. E. Jackson, but only a single rather damaged male
was added to the small series from which I now give full descriptions of both sexes.

Mate. Fore wing length 33-35 mm.; margin of wing slightly concave at 3-4 with a slight
projection on vein 2, apex of wing thus rather pointed, outer margin slightly dentate and narrowly
white fringed between veins. Upperside. Ground colour blue-black with slight greeny sheen
in side light ; two small subapical blue spots, and a larger blue spot beyond end of cell. Hind
wing ground colour blue-black, duller toward inner fold ; there is the slightest trace of a greenish
post-discal wavy line opposite the tails ; submarginal spots bluish to whitish, small and puncti-
form ; marginal border with greenish-blue lunate marks, with some dull red scaling in marks
above upper tail, becoming more olive in region of tails and analangle. Tails thin, upper 6 mm.,
lower 6 mm., black with fine bluish-white central line. Underside. Fore wing ground colour
very dark brown with stronger brown in the distal curve of the wing, separated by a paler satiny
triangle toward the apex ; thin black lines strongly accentuated by white edging, conspicuous in
the basal half of the wing; post-discal row of black spots not conspicuous except in 1b; sub-
marginal row obscured except at tornus in tb. Hind wing ground colour as fore wing, rather
darker distally ; fine black lines outlined in white in basal area. A satiny line crosses the disc

REVISIONAL NOTES ON AFRICAN CHARAXES ype

and this joins another satiny post-discal bar just above the anal angle, these bars finely outlined
in black; the outer border of the post-discal bar strongly bordered by a series of crescentic
maroon marks extending from costa to above anal angle ; submarginal line lilac-greyish adjacent
to series of maroon lunules outlined in greenish above upper tail and more olive to anal lobe
carrying admarginal black triangular marks. The underside is thus ornate, but not conspicu-
ously patterned.

FEMALE (Neallotype). Fore wing length 43 mm. Outer margin not very incurved at 3-4,
but slightly dentate with a stronger projection at vein 2. Upperside. Ground colour brownish-
black with a slight purply sheen toward base, darker on distal half of the wing. Wing crossed
by two lines of creamy-white spots as follows: discal line, two large spots beyond end of cell in
6-5, a minute dot based in 4, followed by larger marks increasing in size from 3 to hind margin,
mark in 1a a long streak ; post-discal spots, three subapical in 7~5, followed by slightly larger
spots in 4-2 placed in a slight curve conforming to the contour of wing, double spots in 1b
contiguous with discal mark and set in slightly, the post-discal spots slightly brownish scaled
distally. Margin with very slight white fringe between veins. Hind wing ground colour
blackish-brown, slightly more greyish at inner fold ; discal band creamy white, extending from
costa to above anal angle, almost uniform in width but tapering toward anal lobe ; some longish
white hair-like scales on inner border over cell area softening the inner edge. Submarginal row
of linear lilac-white marks complete, double at anal lobe; marginal row of crescentic maroon
marks strong above upper tail, then olive to anal angle; tails long, but of thin and tapering,
upper 8 mm., lower 7 mm., black with whitish central line. Margin of hind wing rather strongly
dentate-serrate. Underside. Ground colour of fore wing greyish-brown, more strongly reddish-
brown distally toward curve of wing ; basal area with black lines outlined with white as in the
male; discal bar of upper side strongly represented, bordered on inner edge with black lines,
the spot in 1b with black marks distally ; post-discal row of spots distinct, upper ones to 2 buffy
in colour ; submarginal black spots faint except for the large double mark at tornus 1b. Hind
wing basal area brownish with thin basal black lines; discal whitish line of above clear cut,
outlined proximally in black and distally by the double row of olive and maroon crescentic
marks extending from costa to above anal angle, the row strongly angled in 5; submarginal
line of lilac-white marks edged black distally, complete and contiguous with the marginal lunate
marks, maroon above upper tail then olive to anal angle, distally bordered with bluish-grey at
edge.

Neallotype female. Pemba Island, off Kenya—Tanzania coast N.E. of Zanzibar,
ix.1963 (A. Rydon), deposited in the British Museum (N.H.).

A second female taken on Pemba, x.1963 by Mrs. Mike Heincen, agrees with the
type in all essentials. The fore wing series of creamy spots are slightly more united
in area 2, and the hind wing band is slightly more irregular on the outer border.

A third specimen (the first 2 to be captured, Hollebone, iii.1963) presents some
divergent features: the two rows of fore wing spots are larger and completely
separated up to the long streak in ra. The basal area of the fore wing is strongly
blue-black and the hind wing band is suffused with blue scaling on the borders,
except in the upper third. On the underside, the ground colour is more generally
reddish-brown, and the markings accentuated.

Range: So far as is known, this insect is confined to the Island of Pemba which
lies north-east of Zanzibar Island, but separated from it by a deep fissure some 450

72 V.G.L. VAN SOMEREN

fathoms deep ; Zanzibar, on the other hand, is part of a coral shelf and separated
from the mainland by a shallow channel 20 fathoms deep (Map C).

Pemba Island was probably separated from the mainland long before Zanzibar
was ; it now has several indigenous races of birds and insects, indicating long isolation.
The fauna of Zanzibar, on the other hand, is mainland in character. Pemba has some
residual areas of mixed forest toward the north and Albizzia spp. are common.
Dr. Rydon suggests that pembanus may lay on Albizzia. Specimens were taken at
the forest edge and in adjoining clove plantations.

38K
a 4
SA
nx
—|
ta, Alton
USAMBARA MTS, j
Xe PVs
oN Zi
yee ee °
R, RUVU $, ™
aS ane.
| SLAND
ftmen su
TANZANIA ...
6
| SLAND
\ a

REVISIONAL NOTES ON AFRICAN CHARAXES 73
Charaxes usambarae van Someren & Jackson stat. n.
(Pl. 9, figs. 65 and 66, 69 and 70)

Charaxes pembanus usambarae van Someren & Jackson, 1952 : 275.

When we described this insect from the Usambara in 1952, we placed it as a sub-
species of pembanus because of certain similarities, especially on the underside of the
males. The male type of fembanus was a rather damaged specimen and the female
was not then known. The association appeared justifiable at the time ; however,
since the acquisition of more material from Pemba has made a more critical examina-
tion possible, I am of the opinion that pembanus and usambarae should not be
considered conspecific for the following reasons :

Males. The shape of the two are different, wsambarae having the fore wing much
more falcate, the apex thus more pointed; the outer border much more
incised or concave, and the margin crenate-dentate ; the hind wing margin
more serrate ; the tails more slender and longer ; the pattern on the underside
is different in detail.

Females. The shape of the fore wings are different, wsambarae being longer, more
falcate, more incised on outer border, more dentate on margin; the tails
longer and more slender. Apart from the fact that the fore wing pattern of
known usambarae females is like ethalion f. rosae, thus differing entirely from
pembanus, the position of the post-discal spots are different, they are set further
in from the apex and the angle formed between the three subapical and spots
in 3-4 is more acute. The underside pattern is different. (cf. etheocles,
ethalion and viola in regard to shape.)

Mate. Fore wing length 37 mm. Shape falcate, apex pointed, outer border incised and
margin crenate-dentate. Upperside. Ground colour blue-black with slight green-blue sheen
at base ; two small subapical blue spots followed by one subcostal beyond end of cell and one
spot at end of cell; margin with blue internervular streaks and white fringe between veins.
Hind wing blue-black duller in region of inner fold; a very slight indication of a post-discal
greeny line from above anal angle to 6; submargin with triangular blue marks white at base,
strong at anal angle to 4 then as very small dots to 6. Margin with strong greenish contiguous
lunules with slight admixture of red or dull reddish above upper tail. Tails long and slender,
upper 7 mm., lower 6 mm. ; margin of wing serrate to 7. Underside. Ground colour fore wing
very dark mahogany-brown, base of wing with thin black lines; costal area with a quadrate
satiny patch followed by a quadrate dark brown area, upper part of apex satiny ; a post-discal
zigzag black line runs from subapex to Ib, followed by a submarginal row of dark spots becoming
black in 1b-2 ; margin with slight white fringe between veins. Hind wing ground colour dark
mahogany or chocolate-brown, upper part of basal area with fine black lines ; a slight indication
of a satiny line followed by a dark bar in discal line from costa to cell followed by another satiny
line faintly outlined in black proximally, is followed by a line of contiguous olive and maroon
lunules from costa to anal angle, proximally and distally edged with black; marginal line
greenish, and dark maroon above upper tail, then olive with lilac spots to anal angle ; margin
of wing serrate-dentate.

FemaLe. As already noted, all the known females of usambarae have a pattern rather similar
to that of Charaxes ethalion f. rosae, thus resembling also the females of Charaxes cithaeron and
Charaxes violetta in some degree, and quite different from the female of pembanus.

74 Vv. G. L. VAN SOMEREN

Fore wing length 43 mm. Shape falcate, outer margin incised or concave, and dentate at
veins. Upperside. Ground colour brownish-black at base of wing, blacker on distal half;
disc of wing with a curved white band commencing at the costa at apex of cell to base of 3 then
widening to streak on hind margin ; some violet-blue scaling on ends of white marks in 1a—1b
especially on proximal end ; there are two white spots beyond the cell ; post-discal spots white
and small, three in a subapical row, spot in 4 set well in, those in 2—3 more linear, that in 2
touching the large white mark in discal line. Edge of wing with white fringe between veins.
Hind wing ground colour brownish-black at base, slightly darker on distal border ; discal band
extending from costa to above anal angle white, with strong violet-blue suffusion on borders ;
submarginal linear violet-whitish marks complete from upper angle to anal lobe followed by a
black zone adjacent to the marginal border which is reddish above upper tail then olive to anal
angle ; edge black with slight white fringe between veins. Margin of wing serrate; tails long
and slender, upper 10 mm., lower 10 mm. Underside. Fore wing ground colour brownish,
slightly darker on outer border and in space between discal and post-discal rows of spots ; apex
greyish and surround of upper post-discal spots also greyish ; submarginal dark spots diffuse in
upper half becoming bolder and blacker in 3—2, double black mark in tornus at 1b bordered
outwardly with white. Hind wing ground colour brownish, black lines in upper half of base
distinct ; discal pale band whitish in upper half but suffused with brownish scales in lower half ;
post-discal lunate olive and dull crimson zigzag line, black edged internally, very distinct.
Submarginal lilac-whitish linear marks, outlined in black distally, continuous from upper angle
to anal lobe ; margin dull reddish above upper tail then olive-greenish to anal lobe.

The specimen from which this description is made is in good condition and does
not differ in any essential from the rather damaged type and paratype.

Range: So far, only recorded from the Usambara Range in Tanzania, and mostly
from the Amani area. Nothing is known of its breeding habits.

SYSTEMATIC LIST

Charaxes pembanus Jordan

Charaxes pembanus Jordan, 1925. Type locality: Tanzania, Pemba Island.
Range: Confined to Pemba Island.

Charaxes usambarae van Someren & Jackson

Charaxes usambarae van Someren & Jackson, 1952. Type locality: Tanzania,
Usambara. Range: Only recorded from the Usambara Range.

4. THE CHARAXES BERKELEYI Group

Charaxes berkeleyi van Someren & Jackson?
(Pl. rr, fig. 81-83)

Charaxes aubyni Poulton [vecte van Someren & Jackson]; van Someren & Jackson, 1952 : 274
[as form or species, g and 9].
Charaxes berkeleyi van Someren & Jackson, 1957: 52.

3 While this paper was in press, further information has come to hand which may alter the concept of
Ch. berkeleyi. Nomino-typical berkeleyi was described from east of the Rift Valley:—Ngong area,
Aberdares and Nyeri, Mt. Kenya. Specimens have now been received from west of the Rift Valley:—
Visoi, Kaptagat, Endebess, and east Elgon which are a distinct subspecies of Ch. berkeleyi. They fly
together with Charaxes evansi van Someren & Rogers, 1932. Ch. evansi was described from East
Elgon, and was placed as a subspecies of etheocles by van Someren & Jackson, 1957. The matter is still
under investigation.

REVISIONAL NOTES ON AFRICAN CHARAXES 75

The original description reads as follows:

Male, upperside ground colour deep black with a blue reflection especially at base of fore wing.
A blue spot in cell, another beyond with a very minute dot below it ; two subapical dots bluish-
white. Margin greyish, broadly interrupted by black veins.

Hind wing with a narrow post-discal wavy green line from veins 1-5, followed by a sub-
marginal series of small blue dots in each cellule. Marginal lunules green, their centres filled
with red above upper tail. Tails long, thin, green with black margins, equal length, 6 mm.
(Fore wing length 33-35 mm.)

Underside, fore wing dark greyish with a strong silvery satin sheen especially at the apex and
distally to median line. Black basal and discal marks similarly placed as in the “‘ etheocles ”’
group ; post-discal zigzag line reddish, edged proximally with green then black ; submarginal
spots strongly blue.

Female : Upperside ground colour black with blue sheen in reflected light, some rusty scaling
on distal half of fore wing especially at margin. Discal and post-discal spots entirely separate,
the latter small and orange in colour, the lower double spot placed on either side of vein 2, the
former (discal bar) with proximal edge straight and pale yellow in colour, that in 1a whitish.
There is often a small orange spot at end of cell.

Hind wing discal band white margined with blue ; complete series of submarginal blue spots ;
marginal border reddish above upper tail then greenish to anal angle. Tails long, 8-7 mm.
greenish with black edges.

Underside silvery grey with whitish discal band present on both wings ; otherwise markings as
in the male, but bolder.

Holotype male. KENyA: Ngong area, bred, ii. 1955.

Allotype female, same data (van Someren). Types and two paratypes in the
British Museum.

The original descriptions are applicable to a majority in a long series of captured
and bred specimens. The males are very constant above, but in some specimens the
ground colour of the underside is rusty-brownish instead of greyish. Females may
vary slightly as follows :

Variation. (a) The fore wing discal and post-discal spots may be paler, almost creamy with a
slight ochre tinge to upper spots, the underside greyish and the whole pattern bold.

(b) The fore wing spots may be smaller and strongly orange and well separated.

(c) The upperside markings as in nominate race but the underside ground colour may be rusty-
ochreous so that the ochreous discal bars on both wings are less in contrast, and are even suffused.
This variant may appear in one family in the proportion 10 rusty to 15 greyish. (A similar
variation occurs in both sexes of Chavaxes etheocles evanst van Someren and is not seasonal.)

Range: Kenya, mostly east of the Rift Valley, Mt. Kenya, Nyeri, Aberdares,
mostly on the eastern side ; Kikuyu Escarpment, Limuro, Nairobi-Ngong (Karen-
Langata). West of the Rift, it occurs at Visoi-Eldama Ravine, Kaptagat, Mau,
Kiptiget and Mwangoris. This species, in part of its range, overlaps Charaxes
etheocles evansi with which it could be confused.

The status of this species was established by a series of bred families from known
female parents. Females have been noted laying on Albizzia gummifera (Mimo-
saceae) mainly in the Ngong-Karen area ; occasionally on Ochna holstit (Ochnaceae),
also at Karen ; on Scutia myrtina (Rhamnaceae) in the Kinangop area of the Aber-
dares. The eggs are usually laid on the upper surface of young leaves, but may be

V.G. L. VAN SOMEREN

RUDOLF J 1
1
'
L. ALBERT '
% !
Fa ELGON : SOMALIA
Pd * 2 {
r t + j
\ nN R, TANA 1
L, EDWARD P !
+> ABERDARES se '
-
‘ \
Ne st & \
an @ NAIROBI ‘

x ENYA
L. «ive MDRuANDA 4 KEN
aa R, GALANA
r™ J 2 .
‘ nS - Ss

we, = TEITA
eh a5 *
#tvE HILLS

ie ge
iin KIL iar tain?

MH ; ~
URUND Pd

Co Tiga

fone! % 5 MOMBASA

: "a .

: %

3 USAMBARA MTS e :

J .

‘ F cone 2 PEMBA
Oh trann®™ 44

TANGANYIKA

L.

TANZANIA
e ZANZIBAR

UNGURU MTS,

R, RUFIGI

0 Aaria

L, MWERU

Pa tnes
bebe Litt!
ai

-
‘
a

ee
;

LINDI

o
”
:

R, RUVUMA

MOCAMB IQUE
R, LURIO

q MOZAMBIQUE
£;

KARIBA

‘
\
(
CHINDE
5 N KEY
* ‘
RECHUANAL AND C
ECHUANA Rng BEITRA S CHARAXES AUBYNI
AUBYNI
R, SAB
SABI + CHARAXES AUBYNI
ECKETTI
¢ CHARAXES AUBYNI
AUSTRALIS
/ LIMPOPO
y 4
ps oe CHARAXES BERKELEYI

Map D

REVISIONAL NOTES ON AFRICAN CHARAXES 77

deposited on old. The eggs are greenish-cream when first laid but if fertile show a
brownish ring around the upper disc. The larvae at all stages can be readily dis-
tinguished from those of ethalion, viola kirki, aubyni ecketti and dilutus, all of which
may be found on the same food plant.

Charaxes aubyni aubyni van Someren & Jackson

(Pl. 10, figs. 73-75)

Charaxes etheocles etheocles 9 f. aubyni Poulton, 1926: 552.
Charaxes etheocles 2 f. aubyni Poulton ; van Someren & Rogers, 1932 : 168.
Charaxes aubyni [Poulton] van Someren & Jackson, 1952 : 272.

For purposes of this revision it is desirable to give full descriptions of both sexes,
and to indicate variations.

Mave. Fore wing length 35-37 mm. Shape very similar to male Charaxes viola but consider-
ably larger, pointed, falcate, outer border concave at 3-4, hind angle at 1a—1b projecting rather
strongly. Upperside. Ground colour deep black with greenish sheen, strongly greenish at
base of costa and base of cell fore wing. Spotting of fore wing somewhat as in Charaxes viola,
a subcostal blue spot at mid cell (sometimes vestigial) followed by two spots beyond end of cell,
upper one large, and whitish, two subapical spots whitish or ochreous, followed in discal line by
faintly indicated bluish spots to 1b; marginal border with large greyish-green lunules separated
by ends of dark veins, and in between the fringe is white. Hind wing ground colour black with
greenish sheen over disc ; a wavy greenish post-discal line present from 5 to just short of the
anal angle ; submarginal spots usually distinct especially toward upper angle, white or bluish ;
marginal lunules well marked, greenish-maroon to upper tail then green to anal angle. Tails
long, 5-6 mm., black with yellow-green central line. Underside. Ground colour dark brownish-
grey with satiny bars crossing the cell, the disc, and apical triangle. Fore wing basal and discal
black marks thin in upper half but stronger over 1b—2 ; double tornal black spots and one above
strong.

Hind wing ground colour as fore wing, with satiny bars through the base and through the
discal line ; post-discal zigzag olive and maroon line strong and complete; marginal border
strongly maroon above upper tail, then olive to anal angle.

FEMALE. Fore wing length 43-45 mm. Outer border in spaces 3—4 concave, apex falcate,
margin slightly bluntly dentate. Upperside. Ground colour brownish-black with slightly
greenish sheen. Fore wing pattern creamy-white to cream as follows: a spot often present in
the cell and one at base of 4 (often small), discal row, two beyond end of cell, spots from 3—1a
increasing in size and forming a long triangle with its base on hind margin ; post-discal spots,
three subapical in a row, spots in 4—3 set in slightly, spot in 2 more distad, spots in 1b more diffuse
and may merge slightly into discal mark in same area. Marginal lunules greyish, rather diffuse
and indistinct though stronger at hind angle. Hind wing ground colour black crossed in the
discal line by a triangular band widest at the costa and tapering toward the inner fold above
anal angle, creamy in colour but shaded with greenish or greenish-blue especially on inner border.
Submarginal row of linear spots may be distinct or slightly obscured in upper half, white or
bluish toward the anal angle ; marginal border strong, reddish above upper tail then greenish
toanalangle. Tails long and somewhat slender 6—5 mm., black with greeny-yellow median line
Underside. Fore wing ground colour greyish-brown, more strongly brownish in curve of the
wing, disc crossed by a paler satiny bar; black lines strong at base and toward tornus ; the
pattern of the upper side strongly represented, creamy-white. Hind wing ground colour rather
browner than fore wing, upperside pattern strongly indicated, the discal bar bordered distally by

ENToo. 18, 3 6

78 V.G.L. VAN SOMEREN

the zigzag olive and red-brown post-discal line strongly marked in lower half; submarginal
linear whitish-lilac strong ; border above upper tail maroon, then greenish to anal angle and
carrying black dots.

Variation. (a) Colour of fore wing pattern similar to the nominate, but discal and post-discal
spots enlarged and joined together by rays.

(b) The pattern of the upperside fore wing ochreous to orange-ochreous ; the hind wing band
slightly ochreous ; underside pattern yellowish ; ground colour more brownish.

(c) 9 form ochrefascia van Someren & Jackson (1957 : 47) (Pl. 10, fig. 75) somewhat like (b) but
fore wing spots wider at hind margin in 1-3 and the discal and post-discal upper spots orange-
ochreous, the two series joined together by rusty-rufous rays, the rusty colour often extending
to the outer margin especially at apex. Hind wing band ochreous, margined by greenish-blue
underside pattern more strongly ochre tinged.

Range: The nominate race was described from the Teita Hills, Kenya (Dabida,
Wandanyi, Chawia) ; it is now known to occur on the adjoining hills of Mbololo
and Sagala, but so far has not been taken on the Chyulu Range. In Tanzania it is
found on Kilimanjaro and Monduli, the Usambara Range and Unguru Mts.

A specimen in the British Museum (N.H.) said to have been taken at Lindi has
been examined; it belongs to the nominate race. Lindi is at sea-level on the
Tanzania coast, and since the species aubynz is an insect of high elevations it is prob-
able that the locality given is incorrect. (See also comment on “ Lindi”’ as type
locality of Charaxes xiphares maudet Talbot, another montane species.)

The food plants of this race are Albizzia sassa and A. gummifera (Mimosaceae).
For full biological notes see van Someren & Jackson (1952, 1957).

Charaxes aubyni ecketti van Someren & Jackson

(Pl. ro, figs. 77-79)

Chavaxes aubyni Poulton [vecte van Someren & Jackson] form ; van Someren & Jackson, 1952:
273-274 [Nairobi area to Mt. Kenya].
Charaxes aubyni ecketti van Someren & Jackson, 1957: 48.

Mate. Fore wing length 33-35 mm., outer border slightly incurved and falcate. Upperside.
Differs from nominate aubyni and aubyni australis by its generally smaller size and smaller and
fewer fore wing spots. The spots are limited to one large blue subcostal spot beyond end of cell
with a minute dot below, two ochre tinged subapical spots ; marginal greeny-grey lunules smaller.
Hind wing marginal border narrow, mostly greenish with just a trace of reddish above upper tail ;
submarginal row of bluish to whitish spots distinct or rather obscured, and internal to this a
trace of a greenish post-discal wavy line opposite the region of the tails, sometimes absent.
Tails shorter than in nominate. Underside. Deeper greyish-brown than nominate race but
markings less strong except at base and tornal area; satiny bars present through disc and post-
discal zone. Hind wing rather more tinged with purply-brown ; the post-discal line of greenish
and maroon lunate marks strong; the border more reddish above upper tail then olive to anal
angle, narrow but distinct.

FEMALE. Fore wing length 38 mm. slightly falcate. Upperside. Ground colour blacker
than nominate aubyni; discal and post-discal spots stronger yellowish-cream with an ochre
tinge, thus richer than nominate race, but more restricted ; discal spots: two beyond end of cell,
a small spot at base of 4 may be present, spots in 3 to hind margin increasing in size but the whole
bar is narrower than in aubyni aubyni ; post-discal spots, about equal in size, are: three sub-
apical in a row, spot in 4 set in a little with one below in 3, spot in 2 set out slightly ; no spot

REVISIONAL NOTES ON AFRICAN CHARAXES 79

in tb or only a trace touching mark in discal row. Margin with only a trace of diffuse lunate
marks. Hind wing discal band creamy to yellowish ochre, of almost equal width though tapering
to inner fold, is bordered with greenish-blue ; submarginal spots small and distinct or somewhat
obscured in the upper half; marginal border as usual, reddish above upper tail, then greenish
to anal angle. Underside. Fore wing ground colour darker greyish-brown with slight satiny
bar at apex ; discal creamy bar distinct, upper post-discal spots strong culminating in strong
tornal black marks on distal side. Hind wing discal band of above strong but narrower below ;
bordered distally by the zigzag post-discal line of greenish-ochre and red which is strongly
developed ; submarginal linear marks whitish to mauve outlined distally in black and with
black dots toward the anal angle ; marginal border maroon above upper tail greenish to anal
angle.

Variation. (a) Pattern as in typical female but pattern of fore wing almost white, otherwise
similar.

(b) 2 form ochretincta van Someren & Jackson (1957: 49); (Pl. 10, fig. 79) the pattern of fore
wing bolder, and orange-ochre sometimes with a pinkish tinge ; hind wing band though whitish
at costa is strongly orange-ochreous otherwise.

Range: Kenya, mainly in the Nairobi-Ngong area and Kikuyu, extending to the
Aberdares, Mt. Kenya, and Mau, Sotik and upper Mara River.

For full biological notes vide van Someren & Jackson (1957). The female lays on
Albizzia gummifera (Mimosaceae).

Charaxes aubyni australis van Someren & Jackson
(Pl. ro, fig. 80)

Charaxes aubyni Poulton [vecte van Someren & Jackson] ssp.; van Someren & Jackson, 1952 :
273 [Vumba Mts. 9].
Charaxes aubyni australis van Someren & Jackson, 1957: 47 [4].

This race was founded on five males taken at Nyamkowa, Malawi, anda single female
from Vumba, Rhodesia, was associated with them, since it presented features which
seemed to indicate relationship. This was a mistake, but at that time we had not
appreciated the significance of the wide low-lying Zambezi Valley as being an ecologi-
cal barrier between the montane areas of southern Malawi and the eastern mountains
of Rhodesia. I have now had the opportunity of examining a small series of males
and females from various parts of Malawi, and the males agree in all essentials with
the type, but the females indicate clearly that the ‘““ Vumba ”’ female does not belong
here, but is, in fact, a female of Charaxes gallaghert van Son, which was not described
until 1962, based on a male; the female was described in 1963. The females are
remarkably alike, but the males totally different. The authentic female of aubyni
australis will be described hereafter, and a neallotype selected.

Mate. Fore wing length 33-34 mm.; shape rather strongly falcate, thus very similar to
aubyni ecketti. Upperside. Ground colour black with green sheen at base. Spots small,
represented by a small blue spot beyond the end of the cell, subapical spots, reduced to the two
upper ones, are small and white; marginal lunules present but rather obscured. Hind wing
ground colour black, with only the slightest trace of a post-discal greenish line in lower part,
opposite the tails, or this may be entirely wanting ; submarginal line of small whitish dots most
evident in region of tails; marginal border narrow, mostly greenish with slight trace of red
scaling above upper tail ; tails long and slender, black, with white median line, 6—5 mm. Under-

80 V.G.L. VAN SOMEREN

side. Fore wing ground colour dark brownish-grey with a slight golden bloom over disc, satiny
bands not strong except at subcosta in apex; pattern thus not strongly marked except for
tornal black mark. Hind wing ground colour as fore wing, satiny discal band strongest toward
costa ; post-discal zigzag line moderately strong, ochre and maroon ; submarginal line of greyish
linear marks not strong; marginal border narrow and only slightly reddish above upper tail
and mostly greenish to anal angle. The whole underside is thus less strongly marked or varie-
gated than in the nominate race, but is nearer to aubyni ecketti of the Kenya Highlands.

FEMALE. Fore wing length 37-38 mm. (type 37 mm.), thus smaller than aubyni aubyni.
Upperside. Ground colour black with greenish sheen at base ; pattern as in other races, discal
bar creamy-whitish ; the post-discal spots free from subcosta to 2, if carried to 1b, the spot merges
with the discal mark in same area; the main portion of the discal bar is not strongly triangular,
for the increase in width of the spots is gradual from the rather larger rounded spot in 3. Hind
wing ground colour black; discal band, widest in cell area, is creamy bordered with greenish-blue
especially in lower half ; submarginal line of white to bluish spots complete ; marginal border
narrowish, reddish above upper tail then green to anal angle ; extreme edge black ; tails moder-
ately long, 7-6 mm., black with greenish-ochre median line. Underside. Ground colour greyish-
brown, darker in curve of fore wing and in basal area of hind wing; satiny bars fairly strong,
otherwise pattern similar to other races.

Neallotype female. MALAwI: Blantyre, 27.v.1962 (H. McKay).
Other material. MALAw1: Mlosa Stream, Mlanje, 2 2 (Handman & Martin).

Variation. (a) The pattern of the upperside rather restricted in areas 1a—1b of fore wing so
that the lower part of the bar is narrow and all the post-discal spots including that in rb free ;
the colour of the spots a rich yellowish-cream. The underside ground colour more silvery greyish
so that the whole pattern appears bolder.

(b) The fore wing pattern with larger spots than usual, with slight indication of creamy scales
joining the discal and post-discal spots in upper half. Underside light pattern bolder. This
variation is found in all races of aubyni.

Range: Malawi, in all the high forested slopes of Mt. Mlanje, especially on the
Mlosa Stream; also at Blantyre, Limbe, Zomba, Nyamkowa and on the Nyika
Plateau. In Tanzania recorded from the Southern Highlands: Rungwe, Poroto,
Njombe and Iringa. A specimen labelled “‘ Lindi” in the British Museum (N.H.)
belongs to the nominate race and probably came from the Usambara Range.

This race has not been bred; it probably lays on Albizzia.

Charaxes marieps van Someren & Jackson
(Pl. 11, figs. 88 and 89)

Charaxes marieps van Someren & Jackson, 1957 : 50.

Mate. Fore wing length 33-35 mm. Upperside. Ground colour black, without any blue
sheen in side light. No blue spots in subcostal area but the margin of fore wing with obscured
bluish internervular spots. Hind wing black, slightly duller on inner fold. There is a suggestion
of a bluish-green post-discal series of lunules from the anal angle to 6, more clearly seen in some
specimens. The submarginal row of lilac and whitish spots extends from the anal angle to upper
angle ; marginal border greenish-red above upper tail, olive to anal angle. Tails short, fairly
robust, of about equal length, 5 mm. Underside. Ground colour dark ashy-grey-brown with
some satiny reflections in the discal line and upper part of apex of fore wing, and between, a
darker brownish band ; a series of dark somewhat indistinct marks in post-discal line increasing

REVISIONAL NOTES ON AFRICAN CHARAXES 81

in size to tornal black marks. Basal black lines distinct. Hind wing ground colour more
brownish than fore, with a satiny bar in sub-basal area ; a darker brown discal bar outlined in
black followed by the post-discal zigzag olive and maroon line ; submarginal line of linear marks
whitish-lilac, distinct to anal angle ; marginal border maroon above upper tail then greenish to
anal angle which carries two black dots.

FEMALE. Upperside. Ground colour brownish-black with a strong blue iridescence over
base of fore wing and disc of hind wing. The distal portion of the fore wing rather brownish,
or even rusty in some specimens. The pattern of the fore wing, especially in discal line, is
restricted, the spots being well separated and not expanded to any extent toward the hind mar-
gin, consisting of two spots beyond end of cell, the upper large and lower small, a spot at base of
4 set well in, spots in 3-2 larger and arrowhead in shape, spot in upper part rb oval, the lower
spot larger, the spots in 1b may be separate or contiguous, the streak in 1a extended distad.
Post-discal spots, three in a row but well in from the rather blunt apex, the spot in 4 only slightly
set in so that the angle with above is not strong ; spots in 3-2 slightly larger ; a single or double
spot in 1b, the lower free or just touching streak in ra. All marks are rusty-ochre except those
of ta—1b which are whitish with increasing blue scaling to hind margin. Margin of wing with
diffuse indistinct internervular rusty spots. Hind wing ground colour brownish-black at base,
strongly black distally ; discal bar comparatively narrow and restricted just reaching subcosta
as a separated spot, the rest conjoined and of equal length, but stopping short of the inner fold
at 2; the band bluish, with a white median bar. The black border is thus wide, but the sub-
marginal linear marks are strong, whitish or slightly tinged lilac distally ; marginal border
relatively wide, reddish above upper tail, then olive to anal angle ; extreme edge black. Tails
mostly black with median olive streak, robust, the lower often with an upward turn; upper
tail 7 mm. lower 5 mm. Underside. Ground colour a strong reddish-brown with satiny
highlights on discal line and upper part of apex fore wing; basal black spots and lines well
marked ; discal marks of above here buffy accentuated proximally by black lines ; post-discal
spots, though indistinct, have a heavy black line proximally increasing in thickness, and distally
in 1b—2 the tornal black marks are strong. Hind wing ground colour reddish-brown with a
golden bloom ; black lines at base clear ; the discal bar is hardly indicated except as a narrow
angular line accentuated proximally in black. The post-discal zigzag line of maroon and greenish
lunules is however strong, outlined proximally in black and more greenish above anal angle.
Submarginal, pale line strong up to anal angle, whitish to lilac in colour ; the border well marked
and broad, reddish above upper tail olive-green to anal angle. Tails olive with black border.

Variation. (a) In some specimens the distal portion of the fore wing is increasingly rufous
toward the margin, and the post-discal spots are somewhat obscured. The hind wing violet-blue
band is restricted and narrow especially toward the costa.

(b) This aberration has the ground colour of the fore wing even more brownish, the discal and
post-discal spots rusty in colour and there is indication of connecting rays between the discal
and post-discal series in the upper half.

Range: This isolated and distinctive species seems to have no near relative and
is confined, so far as is known, to the high elevations of Mariepskop in eastern
Transvaal.

There is no record of the species having been bred.

Charaxes karkloof karkloof van Someren & Jackson
(Pl. x1, figs. 84 and 85)

Charaxes karkloof van Someren & Jackson, 1957: 5I.

MALE. Fore wing length 33-35 mm.; general facies very similar to ethalion, but apex more
pointed, outer border more concave and extreme margin with white fringe widely broken by

82 V.G.L. VAN SOMEREN

black interspaces at vein ends. Upperside. Ground colour very deep black with only a slight
suggestion of greenish-sheen along base of costa ; blue spots limited to two very small subapical
dots and one hardly visible diffuse subcostal mark beyond end of cell. Hind wing uniformly
black except for a complete row of minute submarginal whitish dots, double at anal angle ;
border just above upper tail slightly reddish then mixed with olive to anal angle, rather narrow
and often discontinuous. Tails relatively short, slender and straight, 4-5 mm. Underside.
Ground colour olive-grey-brown with strong satiny highlights especially through the disc of fore
wing and upper part of apical area and in between a dark quadrate brownish mark, the brownish
colour extending to the outer border in the curve of the wing. Fine black lines toward base of
wing and on either side of the lower part of the satiny bar; submargin often with a series of
darker spots extending from apex to tornus where the spots in rb and 2 are strong. Hind wing
slightly more olive-brownish with a rusty bloom in disc but traversed by a slight satiny band in
sub-basal area and through disc, both outlined in black ; zigzag post-discal olive and maroon line
strong from costa to above anal angle; submarginal linear marks of lilac-grey ill-defined in
upper half but bolder in lower, outlined in black distally, represented on anal angle as two black
dots ; marginal border maroon above upper tail then olive to anal angle ; extreme edge narrowly
black. There is often a pale “ flare’ internal to the submarginal line in region of tails.

FEMALE. Fore wing length 36-37 mm. Upperside. Ground colour deep brownish-black
with strong purply sheen at base in side light ; distal portion of wing blacker. Wing crossed
by two rows of spots, discal and post-discal, completely separated even at hind margin; discal
row, two spots beyond end of cell, upper large, lower a small dot, no spot in 4, that in 3 rather
angular, spot in 2 larger and free followed by a double mark in 1b, a longer streak in 1a. The
upper three spots are orange, the next slightly whitish, the remaining marks forming a long
triangle whitish tinged with violet, often strongly. The post-discal spots are strongly orange,
the three subapical placed in a row at right angles to the costa, spot in 4 set in a little with spot
in 3 in same line that in 2 slightly out, spot in 1b often indistinct or it may be in contact with
discal mark. Margin of wing with indistinct orange internervular marks. Hind wing ground
colour as fore wing, discal band strong, narrow at costa, represented in 6 by a lilac spot often
free from the central whitish bar, which is strongly bordered with lilac especially on the outer
border and at tapered end, which does not cross the inner fold. Submarginal row of linear
lilac-blue marks, double at anal angle, rather small but distinct; marginal border strong,
reddish above and just below upper tail then olive green to anal angle. Tails long and robust,
upper 7 mm., lower 6 mm. and almost entirely black. Underside. Fore wing dull drab brown
at the base, more rusty in post-discal line then rufous to the border ; slightly satiny in discal line.
Black lines at basal area well marked ; discal whitish bar fairly strong and in its lower portion
outlined in black on inner border ; post-discal spots rusty and rather indistinct but bordered
with black lines on inner edge in 1b—2; tornal double black marks edged whitish distally ;
margin of wing with obscured darker rusty-brown contiguous marks. Hind wing generally
rusty-grey, the basal black lines fine but distinct ; the discal band of above, indicated in the
costal half by whitish-buff, but fading out toward middle of disc, but black lined proximally ;
post-discal zigzag lines strong, rusty-brown toward costa then more maroon, proximally edged
in black; submarginal line continuous, widest opposite tails, whitish with slight lilac above
tails, separated from the reddish and olive border by thin black line ending as double dot in
anal angle.

A few years ago Mr. K. Pennington reported to me that he had seen a female
ovipositing on a tree in the Karkloof area, but he was unable to hunt for eggs or larvae
at the time. He unfortunately did not have the food plant identified. The life
history has now been worked out by Mr. Felix van der Riet, assisted by Mr. R. W.
Wells of Pietermaritzburg. Ova and larvae were collected from a plant of Ochna
arborea (Ochnaceae). The larvae were brought through to pupal stage and imagines
of both sexes were subsequently obtained,

REVISIONAL NOTES ON AFRICAN CHARAXES 83

Mr. Wells kindly sent me colour slides of the egg and larvae, but, as he unfortunately
did not make written descriptions at the time, I can only describe the stages from
these slides.

Ovum of usual Charaxes form, pale ochreous in colour, developing a brownish ring,
if fertile, at the junction of the slightly fluted flattened top with the sphere.

Larva. Ist instar, general colour ochre with greenish tinge, anal segment with
two brownish-black spines ; head black surmounted by a pair of outer and inner
horns and central tubercles also black.

2nd instar: body now greenish covered with fine irrorations pale tipped ; tail,
spines divergent, outwardly curved and rufous in colour; head now rufous, the
horns now decidedly rufous, outward and upward curved, and covered in fine
tubercles.

3rd instar: general colour stronger green, the body irrorations more pronounced ;
the head, twice the size of the previous one, still with rufous horns, the second long
pair white tipped; the disc of the “ face’’ tinged green.

4th instar: body now deep leaf-green, the surface markedly irrorated, each seg-
ment with almost luminous blue spots. Ventro-lateral line yellow; tail spines
greenish-ochre. From the slides, there would appear to be no dorsal ornamentation
in the majority, but a few may have a yellowish dorsolateral mark on 6th segment.

Head now mostly green with yellow marginal border and black spot at lower
angle; “‘jaws”’ blackish ; central short horns rufous at tips, next long pair green
at base, rufous at ends, outer pair mostly rufous with green bases.

Pupa: Not figured.

This species has recently been taken in some numbers in the moist sub-tropical
forests of the Port St. Johns area, by Messrs. Teare & McMaster and by the latter at
Cwen Cwe near Stutterheim. I have examined a series of fresh males and females
and cannot detect any difference between these and nominate specimens from Kark-
loof. There is considerable variation in size amongst the females; there is also
variation in regard to the width of the fore wing discal bar in areas Ia—1b, the mark
in rb may extend distad and so meet the post-discal orange spot in the same area, or
it may be well short of it.

The taking of karkloof at Port St. Johns at the beginning of 1964 confirms my
identification of a pair of this species captured by Mr. C. R. Barrett in the same area
in 1961 and sent to me as a variety of Charaxes ethalion. Both species fly together
in this area, the ethalion being a strongly orange-spotted one superficially resembling
karkloof.

Range : From the high mist forests of Karkloof near Howick, Natal and now known
from Port St. Johns and from near Stutterheim.

Charaxes karkloof capensis ssp. n.
(Pl. rz, figs. 86 and 87)

MALE. Fore wing length 35 mm. Shape of wing similar to nominate race, outer border
slightly more concave and bluntly dentate, more especially toward the hind angle at 1b-2.

84 V.G.L. VAN SOMEREN

Upperside. Ground colour both wings resembling typical karkloof ; fore wing with indistinct
blue spots, one beyond end of cell and two subapical; the hind wing with small punctiform
white submarginal dots, the marginal border narrow and broken. Underside. Very similar to
nominate, ground colour slightly more greyish, satiny bands strong through discs; dark lines
similar, but tornal black spot less well marked ; in the hind wing the post-discal zigzag line is
indistinct in upper half but stronger toward the anal angle; the submarginal pale line is less
defined except in region of tails.

FEMALE. Fore wing length 42 mm., thus larger than nominate, but ground colour and pattern
similar, except that the discal fore wing bar is wider in 1a—1b-—2, so that the lower portion forms
a strong triangular patch which is 12 mm. wide at base and is whitish-lilac with a lilac-blue tinge ;
the large subcostal spot is ochre. The post-discal spots are relatively small, of about equal
size and extend from subapex to 2 with an indistinct spot in rb, all spots orange-tawny in colour.
Hind wing discal band is wide extending from costa to margin of inner fold, whitish towards
costa but strongly violet-blue on outer border and lower end ; the submarginal linear lilac marks
with white central dot well marked from anal angle to upper tail, fade out beyond ; marginal
border reddish above upper tail becoming olive to anal angle. Underside. Rather more greyish
than nominate, but with strong brownish on distal border of fore wing and base of hind wing.
The distal fore wing bar well represented but tornal black spot less strong and black line in
lower post-discal line faint. Hind wing band shows up in upper half but suffused brownish
in lower portion ; post-discal zigzag olive and maroon line strong at lower end but fades out
toward costa ; submarginal line whitish-lilac complete, but strongest in region of tails ; marginal
border comparatively narrow, maroon and olive-green as usual.

Holotype male. SoutTH AFRICA: Cape Province, at Van Staaden’s Pass, 22.11.1963
(C. W. Wykeham). In Wykeham collection.

Allotype female. Same data.
Range: Cape Province, Van Staaden’s Pass. Swellendam.

Charaxes alpinus alpinus van Someren & Jackson stat. n.
(Pl. 12, figs. 90 and 91)
Charaxes ethalion alpinus van Someren & Jackson, 1957: 42.

Mae. Fore wing length 32-34 mm.; shape somewhat falcate, the apex pointed and the
outer border concave. Upperside. Ground colour velvety black ; fore wing immaculate but
for two minute subapical blue spots. Hind wing with submarginal white or bluish-white dots,
double at anal angle; marginal border slightly reddish above upper tail, then olive to anal
angle. Tails short and robust. Underside. Ground colour dark greyish-brown with a golden-
rusty bloom on distal half of fore wing; a satiny band crosses the disc and the apex is also
satiny and in between is a darker brownish zone extending from costa to hind margin and this
is bordered by lunate or rounded post-discal dark marks terminating toward the tornus as blackish
linear marks, but the tornal spots are faint; the submarginal dark spots are obscured. The
basal black marks are strong and accentuated by some white scaling around. Hind wing darker
brownish-grey with a slight indication of a discal satiny band; basal black lines faint ; post-
discal zigzag line of lunate maroon and greenish marks, not very strong in most specimens ;
submarginal greyish-mauve line fairly well indicated and outlined distally in black ; anal angle
with two black dots.

FEMALE. Fore wing length 42-45 mm., thus larger than ethalion f. ethalion. Upperside.
Ground colour more strongly black, but pattern very similar to ethalion, but fore wing bar is
narrower, made up as follows: two spots beyond cell end, upper large, lower small; a small
dot toward base of 4, set well in, spots in 2-3 sub-basal, that in 2 set in a little, large spot in 1b

REVISIONAL NOTES ON AFRICAN CHARAXES 85

hardly wider than streak in ta. Post-discal spots all well marked from the subapex to 1b where
the double mark is free or just in contact with the lower portion of discal mark. All fore wing
marks are white. Hind wing ground colour black ; discal band wide at costa to mid area, then
tapering toward inner fold above anal angle, but not extending through the fold, but there
represented by a pale spot ; the band is obliquely white in the disc but shaded blue on the borders.
The width at costa is noticeably wider than the discal bar ta—1b of the fore wing. Submarginal
bluish-white linear marks complete and distinct ; marginal border reddish above upper tail,
then greenish-olive to anal angle. Tails robust, 6 mm. and 5 mm. Underside. Somewhat
variable. Fore wing ground colour greyish-brown with satiny sheen over upper part of apex
and through the discal zone ; the distal border more brownish in the curve. Basal black marks
distinct ; pattern of upperside clearly indicated, strongly marked in some specimens ; tornal
dark marks not very strong. Hind wing ground colour generally darker than fore, and brownish
tinged but with a sub-basal satiny bar; discal line usually well marked, so also the zigzag
greenish and maroon line of lunate marks terminating in parallel black lines above anal angle ;
submarginal pale line, lilac with black outline distally, clearly marked, with double black dot in
anal angle.

Variation. Similar to the foregoing in fore wing pattern above, but marks slightly tinged with
ochre to 2, marks in tb—1a white, with slight blue scaling on distal ends. Underside more rufous,
especially on hind wing where the discal bar is slightly obscured on its lower portion.

Range: The nominate race occurs on the higher elevations of the Vumba Mts.,
Rhodesia.

Charaxes alpinus nyikensis ssp. n.

(Pl. 12, figs. 92-97)

MALE. Fore wing length 36 mm., apex rather pointed, out margin slightly concave but wing
not strongly falcate. Upperside. Deep velvety black on both wings; fore wing almost im-
maculate except for two small subapical blue dots and an obscure spot sometimes present in
subcosta beyond cell. Hind wing with obscured submarginal bluish spots more apparent in
region of the tails, or the spots may be vestigial ; marginal border very narrow, with a slight
indication of reddish above upper tail or greenish-olive throughout, then olive at anal angle ;
tails short but not so robust as in nominate race from Vumba, black with a trace of green at
base and mid-line. Underside. Ground colour very dark brown but with a rusty-golden
bloom over the distal part of the fore wing; the discal area crossed by a satiny band, more
noticeable at the costal end and separated from the satiny apical triangle by a dark brown
quadrate patch. A post-discal dark wavy line runs proximad to a series of darker spots ex-
tending from apex to tornus, culminating in the double tornal spots in 1b which are not very
heavy. The black basal marks are present but rather obscured by the generally dark ground.
Hind wing ground colour darker than fore, and though the pattern is essentially similar to
nominate alpinus, it is obscured by the general dark ground colour.

FEMALE. Fore wing length 41-43 mm. Upperside. Ground colour brownish-black with
strong greeny sheen at base of fore wing in side light. Pattern as in nominate alpinus, but spot
at base of 4 in discal series large and set well in, thus there is a strong angling of the discal line ;
the upper discal marks are whitish, those in 1b-1a shaded with greenish-blue at both ends.
Post-discal spots complete to 2, the two upper subapical whitish, the remainder ochreous to
orange-ochre ; the angle, formed by the three subapical spots and the spot in 4, is obtuse. In
the type specimen there is an indication of a spot in 1b conjoined to the discal mark. Hind wing
ground colour as fore with a darker, blacker border. Discal band extends from costa, here
rather narrower than in nominate race, but widening over cell area then tapering toward inner
fold but not crossing it, above anal angle, though indicated by greyish scaling; the central
area of the band is whitish, the borders strongly shaded with bluish-green or blue scaling.

86 V.G.L. VAN SOMEREN

Submarginal row of linear bluish marks complete ; marginal border reddish above upper tail,
then olive to anal angle. Tails in type missing except for left upper, but all tails present in
another specimen, upper tail 7 mm. slightly spatulate, lower more slender, 6 mm., black with
ochre central line. Underside. Ground colour more rusty than in Vumba specimens with a
slight golden bloom over the disc, darker brownish on the distal border of fore wing; basal
black lines fine, but obscured in the hind wing. In the type, the discal bar is distinct up to 3,
but the post-discal row of spots is obscured though visible and increasingly outlined in black
proximally from 3 to 1b; the tornal black spots large and clear. The hind wing discal band
is hardly or very slightly indicated due to the general suffusion of golden-brown over the disc ;
the post-discal line of contiguous lunate maroon and olive marks distinct in the type, but ob-
scured in other specimens in varying degree; the submarginal pale line is obscured but the
marginal border of reddish and olive lunules distinct.

Variation. (a) The lower marks of discal bar fore wing and the discal band in hind wing at
costa broad and strongly bluish.

(b) The lower marks in fore wing discal bar white, but discal band of hind wing suffused with
violet-blue especially on the borders.

Holotype male. Northern Malawi-Zambian border on Nyika Plateau, 29.x.1962
(C. B. Cottrell).

Allotype female. Nyika Plateau, 20.x.1962 (C. B. Cottrell).

Five additional males and three females were taken in 1963 on the Nyika Plateau.
A pair of these will be donated to the British Museum (N.H.).

Range: So far, only recorded from the Nyika Plateau, Malawi.

Nothing is known of the breeding habits of these two races of Charaxes alpinus.
The genitalia of the two races agree.

Charaxes manica Trimen
(Pl. 13, figs. g8—-106 ; Pl. 14, figs. 107-115)

Charaxes manica Trimen, 1894: 43.
Charaxes etheocles etheocles g f. cytila Rothschild in Rothschild & Jordan, 1900: 484.

The type was described by Trimen (1894: 42) and the type locality is given as
Manica, south-east Africa, i.e. on the border of Rhodesia and Mozambique (P. E.
Africa). It was described as a species and based on a female specimen ; two other
specimens were taken. Males captured in the same area were assigned to Charaxes
ephyra Godart and though it was noted that “ the underside is darker and with more
ferruginous tinge than usual’’ ... compared with west coast examples, they were
not associated with the female manica. This male was subsequently described and
named cytila by Rothschild in Rothschild & Jordan (1900 : 484), the type locality
being Bandawe, Malawi.

The figure of the 9 type manica, depicts the form with the base of the fore and hind
wings greeny-blue. Rothschild (1900, pl. 12, fig. 7) figures the form with these
areas violet-blue. Both forms and intermediates occur in a long series ; the under-
side ground colour in all is identical.

When Rothschild & Jordan dealt with this complex group of Charaxes in 1900,
they were of the opinion that there was insufficient evidence on which to associate
males and females as of the same species ; they therefore dealt first with the males,

REVISIONAL NOTES ON AFRICAN CHARAXES 87

then the females. Asa result we find that a male and a female of what we now know
to be sexes of the same species, were given different names. They however place
manica Trimen and phaeus Hewitson amongst the female forms of the composite
“ etheocles ’’.

Aurivillius in “ Seitz ’’ (1911) more or less follows Rothschild & Jordan’s grouping.

In 1921 N. D. Riley reported on a collection of butterflies taken in Zambia by
H. Dollman ; and under the heading Charaxes etheocles Cramer, he (quoting from the
field notes and breeding results obtained by Dollman) showed that the male form
cytila Rothschild was always associated with the female manica Trimen, and to a
lesser degree 9, f. phaeus (sic) came into the association, for the series of specimens
under review contained 20 male cytila, 19 9 f. manica, and 1 9 f. phaeus. It would
appear that Riley accepted Rothschild & Jordan’s view that phaeus was a form of
“etheocles ’’, and finding a female in the series which appeared to agree with phaeus,
Hewitson associated this insect with manica, as others apparently had done.

In our original attempt to unravel the etheocles complex (van Someren & Jackson,
1952), we followed suit, but noted that phaeus Hewitson had been described three
years earlier than manica Trimen, and if the two females were to be associated as
belonging to the same subspecies, the name phaeus would have to apply, having
priority. Further examination of actual phaeus material showed that the male of
true phaeus belonged to the species viola and not etheocles Cramer and we therefore
listed phaeus and other female forms under the heading viola phaeus (van Someren &
Jackson, 1952 : 269). For some unaccountable reason we failed to check the males
bred by Dollman, which had been identified as cytila by Riley, nor did we examine
the type of cytila Rothschild. We thus amalgamated two species under the one
heading. We discovered our mistake too late to make an alteration to our revision
but in our subsequent paper (van Someren & Jackson, 1957: 42) we rectified our
error. We recognized manica Trimen as a species, naming the “ phaeus ’’-like
female, pseudophaeus and retained the name cytila Rothschild for the male of the
species. These forms of manica are excellently depicted on Plate VI, figs. 1-3 of
Riley’s paper.

Though Riley placed manica to the species etheocles, as accepted at that time, we
believe that Trimen was right in describing it as a full species, for the following
reasons: (I) the general shape of both male and female differs from etheocles, being
more robust and compact, and less falcate in the fore wing ; (2) there is an overlap of
manica with a race of etheocles in the north-east part of Angola, in southern Katanga
and on the west of Tanzania on the eastern side of Lake Tanganyika. There is also
considerable overlap of manica and true phaeus Hewitson.

It is usually assumed that the advanced forms manica and pseudophaeus have
evolved in mimetic relationship to the male and female forms of the larger species
Charaxes bohemani Felder, with which they fly in a large part of their range. If we
accept the view that both manica and pseudophaeus evolved to their present pattern
and coloration, by natural selection, as a form of protective resemblance, it is of
interest to note that although both sexes of the “ model”’ occur in equal numbers,
pseudophaeus resembling the male of bohemani is seldom taken, and indeed appears

88 V.G.L. VAN SOMEREN

to be absent in some localities. We have already noted that in the long series bred
and taken by Dollman in the Solwezi area of Zambia, there was only one pseudophaeus
to 19 f. manica. Ina series of 20 females from Mumbwa, Zambia all are manica ;
from other sources the following are recorded: 48 manica to 12 pseudophaeus, 29
manica to 6 pseudophaeus ; 32 manica to 8 pseudophaeus ; 18 manica to 2 pseudo-
phaeus.

The question arises, if the modification of pattern and colour is an evolutionary
trend toward a mimetic protective resemblance, why should the female bohemani
appear to have exerted a greater influence than the male, though both are equally
common?

In this regard it is interesting to note that in the case of viola phaeus the converse
holds good ; if the pattern and coloration are mimetic, it is the male of bohemani
which has exerted the influence ... the female not at all!

This brings us to the question, from what were manica and pseudophaeus derived?
Dr. K. Jordan was the first to suggest that the primitive female of cytila was
“ etheocles’’-like ; this was conveyed in a letter to Prof. E. B. Poulton, who recorded
it in his paper (1925 : 542). He drew attention to certain transitional forms which
occurred in the Malawi area but they were only briefly referred to and not designated
by name. They are the forms chintecht and protomanica, which are dealt with later.

The original description of cytila is very brief and is mainly contained in the “ key ”’.
Our reference (van Someren & Jackson, 1952 : 269) is equally brief and confusing,
since at that time we failed to distinguish between Charaxes viola phaeus and manica,
but the description given is that of cyt#/a and not viola. However, I think it advisable
to give a full description of the male associated with manica.

MALE: Fore wing length 33-34 mm. Shape slightly falcate but not so strongly as in races of
viola ; outer margin thus concave in 2-4, more so than in ethalion, which it otherwise resembles.
Hind angle or tornus rather strong. Upperside. Ground colour deep black with slight bluish-
green sheen in side light, more greenish toward base of costa. Usually two large bluish subapical
post-discal spots, well marked or very occasionally obscured, usually one large bluish subcostal
mark beyond end of cell and one in cell toward end ; margin with obscured greenish internervular
marks ; extreme edge white, strongly divided by dark ends of veins. Hind wing, ground colour
deep black, duller on inner fold, the lower part of the disc with a slight greenish sheen extending
toward a series of greenish lunules running from anal angle to 5; this followed by a complete
though rather obscured submarginal series of greenish-blue spots, double at anal angle, occasion-
ally with white centres in region above tails; marginal border rather narrow, reddish above
upper tail, more greenish to anal angle, the band sometimes outlined with bluish scales. Tails
comparatively short and fine, 4-5 mm. long, black with central greenish line. Underside.
Fore wing ground colour, a dark mahogany-brown, the black lines in basal half of wing variable,
faint or strong ; the disc of the wing crossed by a lighter satiny band outlined proximally in its
lower half by a wavy black line, distally by a darker band extending from costa to 1b, this
accentuated in the lower quarter by black in 1b—3, followed by a narrower satiny bar expanded
in upper half of apical area; concavity of wing darker brown, accentuated on inner side by
tornal black double spot extending up as a series of diffuse dark submarginal spots ; margin with
darker ends to veins ; edge with white scaling often continuous. Hind wing, mahogany-brown
with a ferruginous bloom with a slight indication of a satiny discal bar, bordered by darker
brown in upper quarter, followed by a series of contiguous lunate greenish-maroon marks from
costa to above the anal angle ; submargin with a series of linear lilac marks outlined distally in

Ee

REVISIONAL NOTES ON AFRICAN CHARAXES 89

black in the form of dots, double at anal angle in region of tails ; border maroon above upper
tail then olive to anal angle, narrowly edged black ; extreme edge whitish.

There is some slight variation in the underside ground colour, some specimens being more
rusty-red than others, but all are dark.

The various forms of the female are now dealt with.

The type female of manica depicted by Trimen (1894, pl. vi., fig. g) is a specimen
with an incomplete fore wing white band, the white area stopping short in upper
part of rb, of the greeny-blue phase, the greeny scaling not going right through to
the base of the fore wing which is brownish; the hind wing patch is greeny-blue,
not going through to the costa but interrupted by a median black patch. The
nominate form is described below.

FEMALE: Fore wing length 36-42 mm. therefore size somewhat variable. Upperside. Basal
area greenish-blue all over or less green toward base of cell. Curved white band broad, 7-8 mm.,
commencing at costa and including the upper part of the cell end, the basal area of 3, passing
obliquely through the middle of 2 into 1b as a quadrate mark, indented on the inner end or
merely as a streak in the upper part ; there may be some white scaling in 1a. Distal portion
of wing black, usually with two white or bluish-white subapical spots of the post-discal series ;
occasionally the post-discal spots are slightly indicated to 3, usually bluish, rarely white. Hind
wing costal and outer border black, the disc with a large bluish-green patch sharply defined on
outer border, the blue scaling may go through to the costa but more often stops short in 6-7
or may be represented by a slight extension along the discal and post-discal lines. In the
nominate form the submarginal row of linear marks are whitish or bluish and well separated,
double at anal angle; the marginal border more reddish above upper tail, then mixed with
olive to anal angle. Tails comparatively long, rather slender, the upper slightly spatulate at
end, upper 8 mm., lower 5-6 mm., black, edged with strong olive central line. Underside.
Base of fore wing dark reddish-brown carrying rather thin black lines; white band strongly
marked to 1b, where the spot is strongly angled ; there is a somewhat triangular satiny patch
between the upper part of the band and the paler upper part of the apex, this is followed by
indication of the two post-discal spots of above, then by a series of darker spots culminating in
the black tornal marks in 2 and 1b; the concave margin of the wing reddish brown. Hind wing
general colour reddish-brown with rather ill-defined black lines ; a slightly darker brown discal
bar present followed by the zigzag post-discal maroon and olive line from costa to above anal
angle ; submarginal line greyish-blue edged black distally ; border olive-reddish above upper tail,
then olive to anal angle; black dots present above tail area.

Variation. Form figured by Rothschild & Jordan (1900, pl. 12, fig. 7). In these varieties
the blue basal area in fore wing and in the hind wing disc may be blue or strongly violet-blue in
side light ; in the fore wing the blue may be uniform or dusky at base of cell. The blue in the
hind wing seldom goes right through to the costa; this is usually black or it may be invaded
by some blue in the discal and post-discal lines. The submarginal line in hind wing almost
obscured and dark blue. The underside is like that of the nominate form, or may be darker,
the black lines more strongly indicated.

2 form pseudophaeus van Someren & Jackson, 1957 : 46

This name was given to the ‘‘ phaeus ’’-like form of manica. It occurs in two phases, one in
which the base of the fore wing and the disc of the hind wing are strongly blue, the other with
these areas greenish blue. It bears a strong resemblance to phaeus Hewitson on the upper side,
but is much larger and of a different shape; the underside is much darker. It more or less
resembles f. manica on the upperside but it lacks entirely the white curved band in the fore wing ;
the fore wing post-discal bluish or white spots may be generally more in evidence. The hind
wing discal patch may go through to the costa or it may stop at 6 or it may be represented by

go V.G.L. VAN SOMEREN

blue marks in the discal and post-discal lines. The blue patch may be restricted on the outer
edge resulting in a wider black border, but it goes through to the base of the wing where it may
be slightly dusky. The submarginal line may be clear or ill defined. Underside. Similar to
manica but without the fore wing white bar thus the satiny area is more extended and fills the
disc of the wing.

This form is comparatively scarce ; it occurs in the ratio of roughly 1 to 25 manica.
I have already touched briefly on the curious anomaly regarding the so-called mimetic
patterns and coloration of forms manica and pseudophaeus in relationship to their
models, the two sexes of Charaxes bohemam. It might be suggested that of the two,
manica is the more advanced and the more stable genetically, but it is also the more
complicated, moreover, primitive forms of both occur in about equal numbers.

I propose to deal now with these primitive forms, to which Jordan drew attention :

2 form chintecht van Someren & Jackson, 1952 : 269

Fore wing length 40-42 mm. Upperside. Base greeny-brown shading to greeny-blue to
about mid-cell and the proximal border of the discal bar where the blue is brightest ; the distal
portion of the wing black, traversed by rows of discal and post-discal spots; the discal row
consisting of two spots beyond the end of the cell, one small spot base of 4 ; spotsin 3,2, 1b and
Ia increasing in size toward the hind border ; spot in 3 free, others contiguous and forming a
triangle, the inner border almost straight, the spots in 1a and Ib extending distad toward the
tornus. The upper spots are ochreous-orange shading to white in Ia and 1b. Post-discal spots
orange, the three subapical ones in line, spot in 4 set in, that in 3 directly below or slightly inset,
remaining spots extending to rb hardly visible ; a double whitish spot submarginal at tornus is
sometimes present. Hind wing basal area brownish with greeny-blue suffusion becoming bluer
on the inner border of discal bar which is whitish and extending right through to the costa but
tapering toward anal angle, strongly blue especially on lower part of outer border. Outer black
border of wing with a distinct series of linear white marks accentuated with mauve distally ;
marginal border dull reddish above upper tail then olive green to anal angle; extreme edge
black ; tails long and fine, 8 and 7 mm. Underside. Strongly vinaceous red-brown at base
and distal border of fore wing, in between a satiny zone from upper part of apex to hind border
between discal and post-discal spots, the former buffy, shading to whitish on hind portion, the
latter more strongly buffy-ochreous, distinct in upper section fading out in lower but with a
curved black line in 1b opposite the double black tornal mark which is outwardly accentuated
with whitish. Hind wing, the patch of upper side here represented by a light buffy-brown
irregular band, widest at costa then fading out toward the anal angle. Post-discal zigzag line
strong, consisting of contiguous olive-ochre lunules strongly bordered in maroon distally, more
greenish above anal angle, outlined in black. Submarginal line well marked, whitish-lilac ;
border above upper tail maroon then olive to anal angle carrying black dots, double at anal
angle.

Were it not for the strong greeny-blue suffusion over the base of the fore and hind wings,
this form would look like a Charaxes etheocles female with orange fore wing spots, and even more
like a Charaxes ethalion female.

Variation. (a) A form in which the fore wing discal and post-discal spots are white, excepting
the three subapical, which are orange. The hind wing band wide and strongly suffused with
blue-green, thus encroaching on the black border which is narrowed, but the submarginal linear
marks are strong.

(b) A form in which the discal and post-discal spots in fore wing are all strongly orange except
for a streak in 1a; the hind wing band narrow and bordered with purply-blue ; the hind wing
submarginal line obscured; the bases of the wings only slightly greeny-blue. This form
closely resembles the orange-marked ‘‘ dry-season ’’ form of female ethalion.

REVISIONAL NOTES ON AFRICAN CHARAXES 91
2 form protomanica forma n.

We mentioned this interesting form in our previous paper (van Someren & Jackson,
1952 : 270) but did not designate it by name. It is a modification of form chintechi
toward form manica.

Fore wing length 40-42 mm. Upperside. Base exhibiting a varying degree of greeny-blue or
violet-blue suffusion distally circumscribed by a curved white band asin manica the costal portion
of which is narrowed leaving the two large spots beyond the end of the cell entirely free. These
two spots and the post-discal row from subapex to 2 are ochreous. The lower part of the
white band in 1a—1b tinged with bluish proximally. Hind wing basal area brownish with a
distinct greeny or violet-blue suffusion; disc of wing with a large bluish patch, slightly paler
in the discal line extends through to the costa and tapers at the anal end where it merges into
the inner fold; black border widest above 5 toward tornus then decreasing toward anal angle.
the outer border of the discal patch angled between 4 and 5; submarginal linear marks rather
obscured; border as usual, reddish above upper tail, greenish to anal angle. Underside. As
in chintechi, fore wing rather darker at the base, the white bar very distinct and here touches
the two spots beyond the end of the cell and in the hind portion there is a dark vertical mark
in lower portion of 1b opposite the well marked black tornal spots above which there is a black
mark in 2. Hind wing as in form chintechi.

Holotype female. Matawt, Chintechi, 1922 (T. H. Lloyd). British Museum (N.H.).

There are six other very similar specimens in the British Museum (N.H.) collection
all from Chintechi and this form has recently been taken at Dondo and Amatongas
by K. M. Pennington.

Variation. In the most primitive, the costal portion of the curved white fore wing band is
very narrow, the two large marks beyond the end of the cell entirely free and strongly orange ;
the post-discal spots large and orange in colour.

Specimens showing an advance to the manica pattern have the costal portion of the white band
of fore wing upperside very wide and incorporating the two spots beyond the end of the cell ;
the hinder portion of the band in ta—1b rather narrowed by a strong suffusion of violet-blue
proximally and over ta, post-discal spots white in subspex. The hind wing patch is strongly
violet-blue and does not extend through to the costa except as dots in the discal and post-discal
lines. The submarginal linear marks are distinct.

Range: These forms are recorded from Chintechi and Limbe in Malawi; also
from Dondo and Amatongas in Beira area of Mozambique. The species has a very
wide range. I have examined material from:

MOZAMBIQUE: Beira, Mzimbita Forest, Dondo, Amatongas.

RHODESIA: Bindura, Umtali, Vumba, Nyamadzi R., Mt. Selinda, Enkeldorn,
Sinoia, Salisbury, Bazely Bdg., Odzi R.

ZAMBIA: Mwinilunga, Kitwe, Chingola, Solwezi, Chisamba, Mkushi, Katete,
Lusaka, Mumbwa.

Marawi: Nkata Bay, Manchowa, Chintechi, Limbe, Bandawa.

Conco ReEpusiic: S. Katanga, Elizabethville, nr. Lake Mweru.

TANZANIA: East shore Lake Tanganyika, Kigoma, Kungwe, Rungwe, Tabora,
Morogoro, Newala Ruvuma River.

ANGOLA: Enclave into Zambia. Record from Loanda requires verification.

92 Vv. G. L. VAN SOMEREN
Charaxes fulgurata Aurivillius
(Pl. 15, figs. 119-123 ; Pl. 16, figs. 124-129)

Charaxes Ephyra Godart var.; Dewitz, 1887 : 371.
Charaxes fulgurata Aurivillius, 1889 : 236.
Charaxes etheocles etheocles 2 f. lunigera Rothschild in Rothschild & Jordan, 1g00: 488.

This species was first referred to by Dewitz in 1887 as a variety of Charaxes Ephyra
and the characters of both male and female were mentioned and both sexes were
excellently figured. The specimens came from Angola and were taken at Pungo
Andongo (9° 45’ S., 15° 40’ E.) by Major von Homeyer.

Aurivillius (1899 : 236) applied the name fulgurata to these specimens thus giving
them specific rank.

Rothschild & Jordan (1900 : 482-483, 488) refer to these descriptions, the name
fulgurata being lumped with the comprehensive group of male “ etheocles’’, and the
female was given the name /umigera and they explain why they think it desirable to
apply names to female forms. Briefly, in this particular case, they assumed, because
of lack of evidence to the contrary, that the male cited by Dewitz was the male of all
three females figured and not of only one of them. However, in this case Aurivillius
(1889) cited Dewitz (1887, pl. 17, figs. 10, rr) as types of his fulgurata male and female.
The name lunigera proposed for fig. 11 is therefore redundant. However, Aurivillius
in “ Seitz’”’ (1911 : 137) accepts the name /unigera Rothschild, and this lead was
followed by van Someren & Jackson (1952 and 1957). This form resembles Charaxes
manica 2 f. manica Trimen.

Riley (1921 : 243) described the form mima from Zambia. This form resembles
Charaxes viola phaeus, 2 f. phaeus Hewitson. Riley records and accepts the opinion
of Dollman (2m lit.) that fulgurata is a good species distinct from manica and phaeus.
In spite of this, we note that Poulton (1925 : 541) confuses some of the female forms
of what he refers to as etheocles f. phaeus and females of fulgurata. Van Someren &
Jackson (1952) kept fulgurata as a distinct species, but still confused manica and
phaeus ; this however was rectified in van Someren & Jackson (1957).

It seems desirable therefore that a full description of this species be given.

Mate. Fore wing length 32 mm. (average). The shape is much more that of Charaxes viola
than ethalion or manica; the fore wing is falcate with the outer border concave, but not so
acuminate as viola. Upperside. Fore wing, black with a strong green sheen at base, along the
costa and on the outer border ; there are large greenish-blue spots in the sub-costal area: one
in the cell, one just beyond the cell end, and two subapical ; in addition there is a series of post-
discal rather ill-defined greenish marks or lunules extending from rb to 5, they may be free or
just reaching the broad greenish glaucous border of the wing which is divided by the dark veins ;
the white fringe is very faint. Hind wing, ground colour black, more brownish on the inner fold.
A post-discal row of greenish lunules extends from 5-6 to anal angle, followed by a submarginal
row of bluish well separated spots, more whitish in area above the tails, double at anal angle;
border reddish above upper tail, then greenish to greeny-red to anal angle; extreme margin
black with slight white fringe ; tails short but robust, 4-5 mm. Underside. Fore wing, dark
shiny grey-brown, more tinged with rusty at base and outer border ; black spots in base, strong ;
the discal satiny area strong in disc and to post-discal line, paler in upper part of apex ; post-
discal line of spots very indistinct but bordered proximally with a thin black line, darker distally

REVISIONAL NOTES ON AFRICAN CHARAXES 93

to black mark in 1b and ending in the double tornal black spots with one in area above. Hind
wing, ground colour grey-brown but flushed with rusty-red ; black lines at base of wing fine but
clear, the line through disc turning up above anal angle where there is a pale quadrate mark at
the margin ; post-discal bar clearly marked proximally in black, extending from costa to anal
angle below the quadrate mark, upper part of bar ochreous to ochre red terminating at anal
angle where the double black line is heavier ; submarginal line of spots lilac with black dots ;
border reddish then ochre-red ; double black spot in anal angle. Upper tail reddish, lower ochre,
both black bordered.

FEMALE. Fore wing length 46 mm. (average). Upperside. Base of wing brownish-grey
heavily scaled with pale blue, more strongly blue toward the discal white bar, the blue scaling
extending into the bar in 1a and partly into 1b. The bar is comparatively wide, 10 mm. at
costa where it includes the cell end and bases of 5—7 extending to 2 then is rather angled distally
at upper part of 1b and goes to the hind border, where it is suffused with blue. The distal half of
the wing is black, slightly paler on the border, carrying two large whitish or bluish subapical
spots, sometimes a third in upper post-discal line and extending down to 3 as more diffuse
indistinct spots. The type female shows the subapical marks as angled, a character more often
noted in the form mima. Hind wing ground colour greyish at extreme base and on inner fold,
disc of wing with a large pale blue patch, somewhat triangular in shape, base toward inner fold,
the upper side not reaching the costa except as two ill-defined spots in discal and post-discal
line; border of wing black, widest at upper angle, carrying submarginal linear marks, well
marked or ill-defined, lilac or bluish in colour slightly more whitish toward anal angle ; border
above upper tail reddish, then more olive to anal angle. Tails rather slender, 6-7 mm. long,
black with central light line. Underside. Wather darker brown than male and with more
rusty suffusion ; fore wing with white bar distinct, and the subapical spots well reproduced ;
the black tornal spots larger and clearer. Hind wing pattern similar to that of the male rather
more suffused with brownish, but the post-discal maroon and olive lunules more distinct.

This form bears a resemblance to the female of Charaxes manica and is slightly
like the female of ‘‘Charaxes vetula’’ Rothschild.

2 form mima Riley, 1921 : 243
(Pl. 16, figs. 126 and 127)

There are three variants of this form; the nominate form being dominant.

Fore wing length 34-36 mm. Upperside. Basal portion light greeny-blue as in f. fulgurata,
the distal border forming an irregular curve from just short of the end of the cell to the hind
margin in the discal line ; a blue spot in the cell may be free or contiguous with the basal blue.
The distal portion of the wing, i.e. a little more than half the area, is black, slightly paler glaucous
from apex to hind angle of border ; the subapical post-discal spots are large, light greeny-blue
or even whitish and are usually angular, apex directed inward, or V-shaped, followed by a series
of lunate marks to 1b, the ends of the marks often touching the glaucous border marks. Hind
wing with a large pale bluish-green basal patch greyish at the inner fold, not reaching the costa on
the upper border but here represented by blue marks in the discal and post-discal lines, the marks
may be free or joined to the patch; the outer border of which may be evenly or irregularly
curved ; border of wing black, widest at upper angle, decreasing gradually towards anal angle ;
submarginal linear marks usually clear, bluish or greeny-blue, whitish at the double spot at anal
angle. Marginal border reddish to between tails, then mixed with olive-green to anal angle,
internally accentuated by brighter green ; double spots at anal angle black ; extreme margin
black. Tails rather slender, upper 5 mm., lower 3—4 mm., reddish or greenish with black edging.
Underside. Base and disc of fore wing usually strongly satiny, paler at upper part of apex but
border in curvature reddish-brown ; a slight indication of an interrupted darker band beyond

EnTom. 18, 3 7 (1)

94 V.G. L. VAN SOMEREN

post-discal line terminating in black spots or in a line in 1b, with a double tornal black mark
beyond. Basal black lines may be fine or obscured. Hind wing, ground colour rather browner
than fore, with a slight indication of satiny areas toward base of costal area ; basal black lines
thin or may be obscured ; post-discal row of reddish and olive lunate marks usually distinct or
may be faintly indicated ; a pale triangular mark usually present above anal angle ; beyond the
post-discal line there may be some greyish-blue scaling internal to the submarginal series of
contiguous lunate greyish or lilac marks ; marginal border reddish then olive-ochre above anal
angle.

Variation. (a) A minor variation in which the fore wing outer border is broadly greyish,
especially at apex, almost reaching the post-discal row of light spots. The hind wing also
exhibits a greying of the distal border.

(b) In this variation the basal blue of fore and hind wings is a deep greeny-blue and rather
restricted in extent; the distal half of the fore wing is velvety black, the subapical post-discal
spots greenish-blue and the remainder of the series greenish and hardly visible. In the hind
wing the discal blue does not extend beyond area 5 on upper border. The underside is darker
than usual and more mahogany in colour so that the lines and bars are obscured.

Range: The type was described from Pungo Andongo in Northern Angola. The
species ranges eastward into Katanga (south), the two portions of Zambia and the
N.W. of Rhodesia; the distribution is thus somewhat restricted.

In the introductory remarks mention was made of the confusion surrounding this
species, largely due to the lack of field knowledge of the species concerned. Dollman
was the first to study fulgurata along with manica in the field, and his observations
were made public by Riley (1921). However, we note that Poulton (1925 : 541)
considered mima Riley, a female form of fulgurata, to be the same as female form
coryndont Rothschild, which is a female form of Charaxes viola phaeus. The mistake
is understandable, for these female forms are confusingly alike, but on the evidence
of Dollman’s careful observations the mistake should not have been made.

It will be seen from the distribution map E that fulgurata flies in some of the terri-
tory occupied by Charaxes manica and Charaxes viola phaeus and also Charaxes
bohemani. It is generally believed that both sexes of bohemani act as “‘ models ”’
for these very similar species. Both sexes of bohemani are equally common and
one would expect them to be equally effective as ““models’’. It is interesting to
note that the female form mima of fulgurata is by far the commoner, the proportions
being 20 mima to 1 fulgurata. All 15 females taken or bred by Dollman are f. mima.

Charaxes gallagheri van Son
(Pl. 15, figs. 116-118)

Charaxes gallagheri van Son, 1962 : 146 [g.]
Charaxes gallagheri van Son; van Son, 1963: 9 [9].

Recent research has revealed that the female of this species was actually described
in 1957 by van Someren & Jackson but was erroneously placed to Charaxes aubynt
australis. The specimen had been taken by B. D. Barnes (1934) in the Umtali area
of the Vumba district in Rhodesia, thus actually outside the range of aubyni australis
which is now restricted to Malawi and southern Tanzania.

Vide comments under that species.

REVISIONAL NOTES ON AFRICAN CHARAXES 95

Very few examples of this interesting species had been obtained up to the time
Dr. G. van Son published his descriptions, and all but one were males. Males have
now been taken in some numbers by various collectors, but only a few females have
been trapped, mostly by Dr. C. B. Cottrell. All specimens have come from a rela-
tively small area of central-eastern Rhodesia in the triangle Salisbury-Umtali—Fort
Victoria, and mostly from the Odzi River area.

The original descriptions are very full and I propose to give only a condensed
version to accompany the illustrations which form an integral part of the present
series of Revisions.

Mate. Fore wing length 32-33 mm., shape rather falcate, outer margin strongly incised and
apex pointed, thus somewhat like Charaxes viola. Upperside. Ground colour blue-black with
bluish-green sheen at base, costa slightly brownish. A subcostal blue spot toward end of
cell ; in the discal row an elongate whitish mark in 6 with a smaller spot in 5 followed by a more
obscured spot in 3 and an indication of a spot in 2 ; post-discal series complete : three subapical
spots in a row all white, spots in 4—3 set in, thus forming an angle with subapical row, also white,
spots in 1b—2 larger, less defined and bluish ; margin with large glaucous internervular marks
from apex to hind angle, that in 1b double; fringe white between veins. Hind wing ground
colour blue-black, slightly greyer on inner fold ; costa with a large white mark in discal line with
a smaller spot below ; post-discal row of contiguous greenish spots distinct from 5 to above anal
angle ; submarginal row of linear whitish spots edged blue distally, complete from upper angle
to anal angle ; admargin with greenish-blue lunate marks from anal angle to above upper tail,
where there is an admixture of dull reddish scales ; margin black, with white fringe between
veins ; tails black with central greenish line, upper 5 mm., lower 4mm. Underside. Fore wing
ground colour silvery-greyish, slightly brownish on distal third, especially in the curvature of the
margin. The basal markings are: three black dots towards base of cell, a zigzag black line
just beyond mid area of cell, end of cell narrowly black, two slight black lines basad in 5, one
sub-basad in 4, two lines well separated sub-basad in 1b and 2. Post-discal row of spots of
above indicated below and outlined proximally in black in 1b-3; submarginal black present
from 4 to tornus, strong in 1b and 2 but fading out above; margin with fuscous shading in
curvature. Hind wing ground colour silvery-greyish with fine black lines in basal area and on
discal line on inner border of the silvery bar which extends from costa to above anal angle where
the bar expands into a triangle, followed by a zigzag brownish post-discal bar carrying fine black
crescentic lines ; submarginal row of whitish marks complete from upper angle to above anal
angle ; margin greenish at anal angle to upper tail then slightly reddish, and with black dots
between especially marked at anal angle.

The foregoing description is based on a specimen taken by H. J. Duke at Chriten Bank, and
is almost identical with the type (Pl. 15, fig. 116).

Variation. The fore wing spots may be smaller and more strongly blue and those of the hind
wing smaller and bluer, the costal spots smaller ; the marginal glaucous marks less strong (Pl. 15,
fig. 117). Conversely, a specimen may have the fore wing spots larger and whiter, and the glau-
cous marginal spots very marked ; the hind wing pattern stronger on a more greeny-black ground.

FEMALE. Fore wing length 40 mm., apex rather pointed, outer margin slightly incised, more
so than in Charavxes ethalion 9 f. ethalion, to which it bears a superficial resemblance ; it resembles
more the female of Charaxes aubyni australis in shape and pattern. Upperside. Ground colour
of fore wing black, slightly more brownish at base, with some greenish sheen in side light. There
is a bluish spot in upper portion of cell end which may be very conspicuous (as in the type), or
obscured ; discal spots: one elongate in 6 followed by a smaller one in 5, no spot or only a sugges-
tion of one in 4, spots from 3 to hind margin increasing in size to long streak in 1a ; post-discal
spots comparatively large, three subapical rather transverse elongate, spot in 4 more rounded
and set in, spots in 3-2 elongate vertical, double spot in 1b contiguous with discal mark, all

96 V.G.L. VAN SOMEREN

spots creamy-white with slight bluish-green scaling on distal ends in 1b and 1a; marginal
border without marks except for obscure greyish-blue streaks in Ib at tornus. Hind wing ground
colour as fore, or slightly blacker on distal third ; discal band wide, extending from costa to
inner fold above anal angle, fairly uniform in width but tapering above anal angle, white or
creamy-white with greeny-blue scaling on borders; there is an obscure bluish spot subcostal
and more basad to the band. Submarginal linear whitish marks complete from upper angle to
anal angle, the double spot at anal angle with some bluish-violet on distal edge ; marginal lunules
greenish from anal angle to upper tail, then with admixture of reddish to 5. Tails black with
median whitish streak ; edge of wing black with some white cilia, especially toward anal angle ;
upper 7 mm., lower 5 mm. Underside. Silvery-greyish as in the male but with rather more
brownish in spaces between discal and post-discal lines of whitish spots and browner in marginal
curvature of fore wing. Black spots and lines as in the male but larger. Hind wing silvery
greyish at base followed by a more brownish band proximal to the conspicuous white discal band ;
the distal border of the band also brownish bordered by the zigzag line of contiguous greenish-
grey lunate marks outlined in black proximally and by dull reddish distally ; submarginal row
of silvery-grey marks complete but rather diffuse but with black marks distally extending from
upper angle to anal angle ; above upper tail the marginal lunules are reddish, then greenish to
anal angle.

This description is based on a specimen from Chriten Bank taken by C. B. Cottrell.

Range: Apparently confined to central-eastern Rhodesia.

The position of gallagheri within the general etheocles s.1. complex is uncertain. The
underside ground colour and general pattern is reminiscent of Charaxes contraria
Weymer of the Tanzania and Kenya Coast belt ; and even more so of Ch. martini
described below of S. Malawi, but the uppersides are quite different. The pattern
above resembles somewhat that of Charaxes guderiana, which lacks the white costal
spot of hind wing ; there is also a superficial resemblance to Charaxes etesipe tavetensis,
which usually has the white costal mark, but these species should not be confused.

The species has not yet been bred.

Charaxes martini sp. n.
(Pl. 16, figs. 130, 131)

The recent discovery of this species adds yet another to the ‘“‘ Complex of Black
Charaxes”’ (males) of Eastern Africa.

In general shape it resembles Charaxes viola phaeus and Charaxes contraria Weymer,
the fore wings being pointed and falcate. The upper side is black with very limited
markings ; the underside is silvery greyish in ground colour, thus totally different to
viola phaeus, but resembling contraria and the recently described Charaxes gallaghert
van Son in this respect.

Mate. Fore wing length 30 mm., apex pointed, and falcate. Upperside. Ground colour
deep black with a slight greenish sheen toward base ; pattern limited to a very small blue spot, in
subcosta beyond end of cell, one distinct though small white spot in mid-subapical row, with
just the suggestion of bluish marks above and below in this line ; margin of wing with traces of
internervular greenish streaks, extreme edge with white cilia widely broken by black on veins.
Hind wing, ground colour deep black but with a strong greenish suffusion over the lower part of
the disc, directly internal to a row of strongly marked greenish post-discal spots, largest in 2
decreasing in size to 5; this is followed by a very distinct row of white elongate marks, double at

———

REVISIONAL NOTES ON AFRICAN CHARAXES 97

anal angle and extending to upper angle; border with a wide band, a mixture of reddish and
greenish scaling above upper tail, then olive to anal angle ; extreme margin black with slight
white fringe between veins. Tails of about equal length, 4-5 mm.; the anal lobe rather pro-
nounced ; margin of wing bluntly dentate. Underside. Ground colour silvery light greyish,
the black marks in basal half of fore wing strong ; the discal bar silvery proximally outlined in
black ; beyond a broad brownish band from costa to hind margin, first faintly then more heavily
black lined in 1b opposite the tornal black mark, which is elongate and extends slightly into 2 ;
the concavity of wing shaded brownish. Hind wing, ground colour silvery, black lines in basal
area distinct ; a long thin line from costa, double in cell, runs from costa towards the anal angle,
then curves up to the inner fold ; this line borders the discal silvery bar which runs from costa
to above anal angle, which in turn is followed by a brownish band, on the distal side of which are
pale olive-green contiguous lunate marks in the post-discal line, bordered distally with brownish
and carrying the submarginal series of white to lilac linear marks accentuated distally in black ;
border reddish above upper tail, then pale olive to anal lobe with double black mark.

Holotype male. Matawi: Mlosa stream, Mlange, 2,500 ft., 23.ix.1962 (P. T.
Martin). To be deposited in British Museum (N.H.).

In some respects this insect resembles Ch. contraria Weymer, of the coastal belt of
Tanzania and Kenya and may eventually prove to be a subspecies of it, but until
more examples are obtained, including females, it seems best to describe it as a
species. It should be noted that contraria has previously been treated as a subspecies
of Charaxes catachrous of Congo—Uganda, but this now appears incorrect, for reasons
which will be given when the etheocles group is dealt with.

Range: Only known from the foot hills of Mt. Mlange (Mlosa Stream), Malawi.

SYSTEMATIC. LIST

Charaxes berkeleyi van Someren & Jackson*

Charaxes berkeleyi van Someren & Jackson, 1957. Type locality: Kenya, Ngong
area. Range: In Kenya, mostly east of the Rift Valley.

Charaxes aubyni van Someren & Jackson

Charaxes aubyni aubyni van Someren & Jackson, 1952. Type locality: Dabida Mt.
100 miles W.N.W. of Mombasa. Range: Kenya, Teita, Mbololo
and Sagala hills and Tanzania on Mt. Kilimanjaro, Monduli,
Usambara range and Unguru Mts.

2 f. ochrefascia van Someren & Jackson, 1957. Type locality: Teita hills.

ecketti van Someren & Jackson, 1957. Type locality: Kenya, Ngong.
Range: Kenya in Nairobi-Ngong and Kikuyu areas to the Aber-
dares, Mt. Kenya, Mau, Sotik and Upper Mara River.

2 f. ochretincta van Someren & Jackson, 1957. Type locality: Kenya,
Ngong.

australis van Someren & Jackson, 1957. Type locality: N. Malawi,
Nyamkowa Mt. Range: Malawi, Mt. Mlange etc., Blantyre,
Limbe, Zomba and Nyika plateau. In Tanzania in the southern
highlands.

4 See footnote on p. 74.

98 V.G. L. VAN SOMEREN

Charaxes marieps van Someren & Jackson

Charaxes mariepbs van Someren & Jackson, 1957. Type locality: Transvaal,
Mariepskop. Range: Confined to eastern Transvaal.

Charaxes karkloof van Someren & Jackson

Charaxes karkloof karkloof van Someren & Jackson, 1957. Type locality: Natal,
Karkloof. Range: Natal, in high forests of Karkloof, Port
St. John’s and Stutterheim.
capensis ssp.n. Type locality : South Africa, van Staaden’s Pass. Range:
Only known from area of type locality.

Charaxes alpinus van Someren & Jackson
Charaxes alpinus alpinus van Someren & Jackson, 1957. Type locality : Rhodesia,
Vumba Mts. Range: Only known from the higher elevations of
the Vumba Mts., Rhodesia.
nytkensis ssp. n. Type locality: Malawi, Nyika plateau. Range: Only
recorded from area of type locality.

Charaxes manica Trimen
Charaxes manica Trimen, 1894. Type locality: Mozambique, Manica. Range:

Occurs in many localities in Mozambique, Rhodesia, Zambia,
Malawi, Tanzania, Congo Rep. and Angola.

2 f. pseudophaeus van Someren & Jackson, 1957. Type locality : Mozam-
bique, Beira.

2 f. chintechi van Someren & Jackson, 1952. Type locality: Malawi,
Chintechi.

2 f. protomanica forma n. Type locality: Malawi, Chintechi.

Charasxes fulgurata Aurivillius

Charaxes fulgurata Aurivillius, 1889. Type locality: Angola, Pungo Andongo.
Range: Angola, Congo Rep., Zambia and N.W. Rhodesia.
Synonym: Charaxes etheocles etheocles 2 f. lunigera Rothschild in Rothschild
& Jordan, 1900. Delagoa Bay, Zomba and Taveta.
2 f. mima Riley, 1921. Type locality: Zambia, Solwezi.

Charaxes gallagheri van Son

Charaxes gallagheri van Son, 1962. Type locality: Rhodesia, Mapembi (Umtali
dist.). Range: Confined to central-eastern Rhodesia.

Charaxes martini sp. n.

Charaxes martini sp.n. Type locality: Malawi, Mlanje. Range: Only known from
the type locality.

REVISIONAL NOTES ON AFRICAN CHARAXES 99
ACKNOWLEDGEMENTS

The task of compiling these revisions has not been an easy one, and I am greatly
indebted to the many who have assisted me with specimens, photographs and valuable
information. I wish especially to place on record my thanks to members of the
Entomological Department of the British Museum (N. H.), particularly Mr. N. D.
Riley and Mr. T. G. Howarth; to Mr. Ernest Taylor of the Hope Department of
Entomology, Oxford University Museum; Dr. G. Bernardi of the Paris Museum ;
Dr. G. van Son of the Transvaal Museum, Pretoria and Dr. Elliot Pinhey of the
National Museum, Bulawayo; to Dr. C. B. Cottrell of the University College,
Salisbury ; to Mr. J. D. Handman of Malawi ; to Mr. K. M. Pennington of Balgowan,
Natal ; Mr. J. Lawson of the Natal Museum, Durban ; Dr. C. H. McCleery of Njombe,
Tanzania ; Dr. Arthur Rydon of Arusha, Tanzania; Mr. T. H. E. Jackson of Kitale,
Kenya; Mr. R. H. Carcasson of the National Museum, Nairobi, Kenya, for access to
the material taken by the Japanese Expedition to Kigoma, Tanzania; to Mr.
Barton Eckett and Mrs. Joan Wedekind for Zambian specimens and lastly to Dr. A.
de Barros Machado of Museum do Dundo, Luanda, Angola, for useful Angolan
material.

REFERENCES

References included in van Someren (1963, 1964) are not listed below.

Butter, A. G. 1900. On two consignments of butterflies collected by Mr. Richard Crawshay
in the Kikuyu Country of British East Africa in 1899 and 1900. Proc. zool. Soc. Lond.
911-945, pl. 58.

CARPENTER, G.D.H. 1945. Notes on Chavaxes (Lep. Nymphalidae) in the Hope Department

of Entomology, University of Oxford. Proc. R. ent. Soc. Lond. (B) 14: 81-88.

1946. The geographical distribution of the forms of the African Nymphaline butterflies
Charaxes etesipe Godart and Ch. penricet Rothschild. Tvans. R. ent. Soc. Lond. 98 : 91-104.
Dewitz, H. 1887. Westafrikanische Tagschmetterlinge. Nova Acta Acad. Caesar. Leop.

Carol. 50 (4) : 367-372, pl. 17.

Fox, R. M. 1963. New African butterflies. Ann. Carneg. Mus. 36 : 213-224.

GopartT, J. B. 1824. In Encyclopédie Méthodique, 9. Histoire Naturelle. 828 pp. Paris.

GrosE-SMITH, H. 1885. Description of two new species of butterflies. Entomologist’s mon.
Mag. 21 : 247-248.

GrosE-SmiTH, H. & Kirsy, W. F. 1887-1892. Rhopalocera Exotica,1 (1). London.

Hatt, A. 1930. New forms of Tropical Rhopalocera. Entomologist 63 : 278-279.

Hewitson, W. C. 1863. Illustrations of new species of Exotic Butterflies 3 : 1-120, pls. 1-60.
London.

Latuy, P. I. 1906. A new species of Chavaxes from Rhodesia. Entomologist, 39 : 125.

1925. Notes sur les Charaxes de la collection de Madame G. Fournier. Encyclop. ent. Sér.
3, B 1: 93-97, pls. 2-3.

Lucas, H. 1872. Description de quelques Lépidopteres appartenant aux genres Charaxes et
Cyligramma et provenant du voyage de A. Grandidier 4 Madagascar. Annis Sci. nat. (5)
15 : 1-4.

Pinuey, E.C.G. 1948. Records of Southern Rhodesian butterflies (Rhopalocera). Occ. Pap.
natn. Mus. Sth Rhod. 15 : 276-341.

Poutton, E. B. 1918. New mimetic female forms of Chavaxes ethalion Boisd., and etheocles

Gr., and bred male-like females of the latter. Proc. ent. Soc. Lond. 31 : 79-84.

1932. The geographical races of Charaxes ansorgei Rothsch. and their mimetic associations.

Proc. ent. Soc. Lond. 7 (1) : 6-9.

100 V.G.L. VAN SOMEREN

Poutton, E. B. 1933. The geographical races of Chavaxes ansorget Rothsch. (Lep. Rhopalocera).
Stylops 2 (1) : 12-16, pl. 1, 2.

RiLtey, N. D. 1921. Notes on the Rhopalocera of the Dollman Collection. Tvans. ent. Soc.
Lond. 1921 : 234-259.

ROTHSCHILD, W. 1894. Some new species of Lepidoptera. Novit. zool. 1: 535-540.

1897. Descriptions of some new species and subspecies of Lepidoptera. Novit. zool. 4:

179-184.

STICHEL, H. in Bryxk, F. 1939. Lepidopterorum Catalogus, Nymph. I : 375-542.

STONEHAM, H. F. 1932. Two new races of Charaxes pollux Cr., from Kenya, British East
Africa. Bull. Stoneham Mus. no. 9: 1-2.

StoracEe, L. 1949. Spedizione zoologica del Marchese Saverio Patrizi nell Basso Giuba e
nell’Oltregiuba, VI-VIII. 1934. Lepidotteri diurnini. Annali Mus. civ. Stor. nat.
Giacomo Doria, 64 : 12-29.

TaLpot, G. 1932. Four new forms of Rhopalocera. Bull. Hill Mus. Witley, 4 (3) : 288-289.

TRIMEN, R. 1894. Ona collection of Butterflies made in Manica, Tropical South-east Africa
by Mr. F. C. Selous, in the year 1892. Proc. zool. Soc. Lond. : 14-82, pls. 4-6.

UncemacH, H. 1932. Contribution a l'étude des Lépidoptéres d’Abyssinie. (Pt. 1,
Rhopalocéres). Mém. Soc. Sci. nat. phys. Maroc, 32 : 1-122, 2 pls.

VAN SOMEREN, V.G.L. 1964. Revisional notes on African Charaxes, Part II. Bull. Brit. Mus.
(Nat. Hist.), Ent. 15 : 183-235, 23 pls., 4 maps.

VAN SOMEREN, V. G. L. & Jackson, T. H. E. 1952. The Charaxes etheocles-ethalion complex :
A tentative reclassification of the group (Lepidoptera, Nymphalidae). Tyvans. R. ent. Soc.
Lond. 103 : 257-284.

1957. The Charaxes etheocles-ethalion complex (Lepidoptera : Nymphalidae) Supplement

No. 1. Ann. Transv. Mus. 23 (1) : 41-58, I map, pls. 3-4.

VAN SOMEREN, V. G. L. & Rocers, K. St A. 1932. Butterflies of Kenya and Uganda. J/ E.
Africa Uganda nat. Hist. Soc. 42-43 : 141-172, pls. 108-121.

vaAN Son, G. 1962. New African Butterflies, 1. J. ent. Soc. sth Afvy. 25 (1) : 140-148, pl. 2.

1963. New African Butterflies, 2. Novos Taxa ent. 29: 1-11, I pl.

ERRATUM

Bull. Brit. Mus. (Nat. Hist.) Ent. 13 (7) : 240
For AURIVILLIUS, C., in SEITz, A., 1925, read AURIVILLIUS, C., in SEITZ, A., I9II.

INDEX
Synonyms in italics
abyssinicus, 61, 69 chintechi, go, 98 |
alpinus, 84, 98 cytila, 86
antanala, 68 |
aubyni, 74, 77, 97 dealbata, 66, 69

australis, 79, 97 ecketti, 78, 97

etesipe, 59, 69
etheta, 59, 69
evansi, 74 (footnote), 75

berkeleyi, 74, 97

cacuthis, 68, 70
caeruleotincta, 69
capensis, 83, 98 flavus, 69
castoroides, 69 fulgurata, 92, 98

ei

INDEX IOL

gallagheri, 94, 98 ochretincta, 79, 97
gordoni, 63, 69
paradoxa, 67, 70

karkloof, 81, 98 patrizii, 62, 69
peculiaris, 65, 69
lunigera, 93, 98 pemba, 64, 69
pembanus, 70, 74
manica, 86, 98 penricei, 65, 69
marieps, 80, 98 protomanica, 91, 98
maritima, 52, 54 pseudophaeus, 89, 98
martini, 96, 98
melloni, 52, 53, 54 tanganyikae, 67, 70
meru, 53, 54 tavetensis, 62, 69

mima, 93, 98

usambarae, 73, 74
nyikensis, 85, 98

violetta, 47, 50, 54
ochrefascia, 97 violetta transitional form, 51

ENTOM. 18, 3. 7 (2)

PICAnihet
Charaxes violetta violetta Grose-Smith

Fics. 1, 2. Type ¢ (Mozambique: Delagoa Bay), B.M. (N.H.) upper and underside.

I'ies. 3, 4. Topotypical g and 2 (Mozambique: Delagoa Bay), upper and undersides.

Fics. 5, 6. Type 2 (Mozambique: Delagoa Bay), B.M. (N.H.) upper and underside.

Fics. 7, 8. Topotypical g and 9 (Mozambique: Delagoa Bay), upper and undersides. Note
variation on undersides.

I

PLATE

Bull. B.M. (N.H.) Entom. 18, 3

PLATE 2
Charaxes violetita Grose-Smith

Fic. 9. ¢ transitional (Vumba Mts.) upper and underside.

Fic. 10. ¢g (Malawi: Monkey Bay, 1,600 ft.), Handman, upper and underside.

Fics. 11, 12. ¢ (S. Tanzania: Newala, north of Ruvuma River), upper and undersides.
Fic. 13. @ transitional (Vumba Mts.) upper and underside.

Fics. 14, 15. @ (Malawi: Nkata Bay, 1,900 ft.), Handman, upper and undersides.

Fic. 16. @ (Cent. Tanzania: Kimboza Forest nr. Morogoro), upper and undersides.

PG Aer: 2

Bull. B.M. (N.H.) Entom. 18, 3

PLATE 3
Charaxes violetta Grose-Smith and subspecies

Fic. 17. vtoletta Grose-Smith, ¢ (Tanzania : Iringa Dist., Kitonga Gorge), upper and under-
side.

Fic. 18. violetta Grose-Smith, 2 (Arusha, N.W. of Kilimanjaro), upper and underside.

Fics. 19, 20. maritima ssp. n., Type ¢ (Kenya Coast, Rabai Hills) and ¢ (Usambara Hills,
Lower forest below Amani), upper and undersides.

Fic. 21. vtoletta Grose-Smith, 2 (Tanzania : Iringa Dist., Kitonga Gorge), upper and under-
side.

FIGS. 22, 23. maritima ssp. n., 29 and Type 9 respectively (Kenya Coast: Rabai Hills),
upper and undersides.

Fic. 24. maritima ssp. n., 9 (Usambara Hills, Lower forest below Amani), upper and underside,

PLATE 3

Bull. B.M. (N.H.) Entom. 18, 3

PLATE 4
Charaxes violetta Grose-Smith and Ch. etesipe (Godart)

Fics. 25, 26. violetta meru ssp. n., 3 (Mt. Kenya, Lower forest, Meru), upper and undersides.

Fic. 27. violetta melloni Fox, Holotype g¢ (Tanzania : Nguru Mts., Morogoro Dist.), upper and
underside. :

Fic. 28. etlesipe etesipe (Godart), topotypical g (Sierra Leone), upper and underside.

Fics. 29, 30. vtoletta meru ssp. n., 9 (Mt. Kenya, Lower forest, Meru), upper and undersides.

Fic. 31. violetta melloni Fox, Allotype 2 (Tanzania: Nguru Mts.), upper and underside.

Fic. 32. etesipe etesipe (Godart), topotypical 2 (Sierra Leone), upper and underside.

Bull. B.M. (N.H.) Entom. 18, 3 PLATE 4

PLATE 5
Charaxes etesipe (Godart) and subspecies

Fic. 33. eltesipe (Godart), g (Uganda: Katera forest, W. side L. Victoria), hind wing costa
with large white patch ; post-discal blue band broad, upper and underside.

Fic. 34. etesipe (Godart), ¢ (S.W. Uganda : Kayonza forest), hind wing costa with one small
costal spot ; post-discal blue band narrow, upper and underside.

Fic. 35. etesipe 9 £. caeruleotincta Carpenter, Type 9 (Uganda: Entebbe). (Hope Dept.,
Oxford).

Fic. 36. abyssinicus Rothschild, g¢ (Abyssinia: Youbdo, Ungemach) upper and underside.
(Photo Paris Museum).

Fic. 37. etesipe 9 f. casteroides Poulton, (Uganda: Jinja), fore wing bar strongly ochre ; hind
wing band white at costa but ochre with greenish distal border, upper and underside.

Fic. 38. etesipe (Godart), 9 (Nigeria: Calabar), a similar form to caeruleotincta Carpenter,
upper and underside.

Fic. 39. etesipe 2 f. caeruleotincta Carpenter, Paratype 2? (Uganda; Entebbe). (Hope Dept.,
Oxford).
’ Fic. 40. abyssinicus Rothschild, 9 (Abyssinia: Youbdo, Ungemach), upper and underside.
(Photo Paris Museum).

-

Bull. B.M. (N.H.) Entom. 18, 3 PLATE 5

PLATE 6
Chavaxes etesipe (Godart) subspecies

Fic. 41. tavetensis Rothschild, g (Kenya: Teita Hills, A. Rydon), fore wing spots small ;
hind wing blue band rather restricted and not extended towards tails, upper and underside.

Fic. 42. tavetensis Rothschild, g¢ (Kenya: Rabai Hills), hind wing band extended toward
area of tails; fore wing spots small; marginal spots large, upper and underside.

Fic. 43. govdoni van Someren, topotypical g (Mt. Kenya, Lower Meru forest), upper and under-
side.

Fic. 44. pembassp.n., ¢ (Pemba Is. north of Zanzibar, A. Rydon), fore wing strongly acumi-
nate ; hind wing blue band broad, upper and underside.

Fic. 45. tavetensis Rothschild, 9 (Kenya: Teita Hills, A. Rydon), fore wing bar ochre and
white ; hind wing band white, upper and underside.

Fic. 46. tavetensis Rothschild, 9 (Kenya: Rabai Hills), fore wing bar creamy, upper spots
white ; hind wing band white with slight greeny-blue borders, upper and underside.

Fic. 47. govdoni van Someren, topotypical 2 (Mt, Kenya, Lower Meru forest), upper and
underside.

Fic. 48. pembassp.n., 2 (Pemba Is.), fore wing bar white ; hind wing band white with greeny-
blue on borders, upper and underside.

Bull. B.M. (N.H.) Entom. 18, 3 PLATE 6

PLATE 7
Chavaxes penricei Rothschild and subspecies

Fic. 49. penricet Rothschild, $ (Zambia: Mumbwa, Wedekind), a typical ¢ with large fore
wing white spots, those in ra—1b slightly shaded bluish proximally ; fore wing marginal spots
large ; hind wing bar almost straight on outer edge, blue tinged on borders, upper and underside.

Fics. 50 and 51. penricet Rothschild, g¢ (Tanzania: Songea Dist.), typical specimen and
specimen with blue suffusion and reduction of hind wing band, upper and undersides.

Fic. 52. dealbata Joicey & Talbot, $ (N.W. Angola: Pungo Adongo), fore wing upper spots
large and white, those in 1a—1b blue, hind wing bar white at costa, rest solidly blue, upper and
underside.

Fic. 53. penricei Rothschild, 9 (Zambia: Mumbwa, Wedekind), fore and hind wing bars
white, the latter with slight blue scaling on borders, upper and underside.

Fic. 54. penricei Rothschild, 9 (Tanzania: Songea Dist.), with creamy fore wing bar,
upper and underside.

Fic. 55. penricei ab. flavus Lathy, 9 (Malawi: Chintechi), both wings with yellow-ochre
bars, (Photo Paris Museum).

Fic. 56, g (Zambia: Lundi), fore wing spots large and white, those in 1a—1b blue, white in
centre ; hind wing costal spot large and white, discal bar bluish-white, elongate in 2, upper and
underside.

(Figs. 52 and 56 represent a blue variation of penricei, undersides with strongly contrasting
white patterns).

PLATE 7

Bull. B.M. (N.H.) Entom. 18, 3

PLATE 8
Charaxes penricei Rothschild subspecies, Ch. pavadoxa Lathy and Ch. cacuthis Hewitson

Fic. 57. penricei dealbata Joicey & Talbot, Type g (Portuguese Congo : Kibokolo, do Zombo),
upper and underside, compare with specimen from Pungo Adongo, Plate 7, Fig. 52. Photos
B.M. (N.H.) Nos. 36078 and 36079.

Fics. 58-60. penricei tanganyikae ssp. n., Fig. 58 g (Kigoma, E. side L. Tanganyika, in forest),
fore wing spots large ; hind wing bar placed more toward discal line, Jap. Exped. 12/61. Fig. 59
3 (Kungwe, E. side L. Tanganyika), very similar to preceding specimen, hind wing costal white
spot larger, Jackson, 2/54. Fig. 60 9 (Tanzania: Geita Dist., Geita, S.W. of L. Victoria), fore
wing discal bar white ; hind wing bar with slight bluish on borders, Jackson, 10/37, upper and
undersides.

Fics. 61 and 62. pavadoxa Lathy, g and 2 (Comoro I.), upper and undersides. Photos
B.M. (N.H.) Nos. 36149 and 36150; 36153 and 36154.

Fics. 63 and 64. cacuthis Hewitson, g and 9 (Madagascar), upper and undersides.

Bull. B.M. (N.H.) Entom. 18, 3 PLATE 8

PLATE 9
Chavraxes usumbarae van Someren & Jackson and Ch. pembanus Jordan

Fics. 65 and 66. wusambarae van Someren & Jackson, topotypical gg (Usambara Range,
Amani), upper and undersides.

Fics. 67 and 68. pembanus Jordan, Type g (Pemba I.) upper and underside. Photos
B.M. (N.H.) Nos. 34028 and 34029.

Fics. 69 and 70. usambarae van Someren & Jackson, Allotype 2 and Paratype 2, upper and
undersides. Photos B.M. (N.H.) Nos. 35019 and 35020; 35015 and 35016.

Fics. 71 and 72. pembanus Jordan, topotypical g and Neallotype 2 (Pemba I., Rydon) upper
and undersides.

PLATE 9

Bull. B.M. (N.H.) Entom. 18, 3

PLATE 10
Charaxes aubyni Poulton and subspecies

Fics. 73-75. aubyni Poulton, typical g, Q and Q f. ochrefasciata respectively (Kenya:
Teita Hills), upper and undersides.

Fic. 76. australis van Someren & Jackson, typical ¢ (Malawi: Mlosa stream, Mlange,
Handman) upper and underside.

Fics. 77-79. ecketti van Someren & Jackson, typical g and 9 (Kenya: Karen-Ngong) and
Q f. ochretincta, underside colour tinged brownish, respectively, upper and undersides.

Fic. 80. australis van Someren & Jackson, Neallotype 2 (Malawi: Mlange, Blantyre, Limbe,
Handman & Martin), upper and underside.

PAT E50

Bull. B.M. (N.H.) Entom. 18, 3

PLATE 11

Charaxes berkeleyi, van Someren & Jackson, Ch. karkloofi van Someren & Jackson and Ch. marieps van Someren & Jackson

Fics. 81-83. berkeleyi van Someren & Jackson, g and 2 99, typical, (Kenya: Karen-Ngong), note strong marginal lunules in
fore wing and borders in hind wing of g ; uppermost ? underside strongly patterned, lower 9 (bred from same family) with underside
suffused brownish, upper and undersides’. :

Fics. 84 and 85. karkloof karkloof van Someren & Jackson, topotypical ¢ and 2 (Natal: The high mist forests of Karkloof and
Howick), upper and undersides. Pennington Coll.

Fics. 86 and 87. karkloof capensis ssp. n. Holotype g and Allotype 2 (East Cape Province: van Stadens Dist.), upper and
undersides.

Fics. 88 and 89. marieps van Someren & Jackson, $ and 2 Paratypes, topotypical (East Transvaal : Mariepskop) upper and
undersides. Kenway Coll. Durban Museum.

5 See footnote on p. 74

€ ‘gr ‘moyuyq (‘H'N) ‘Wd ‘11

Il a4LV Id

PLATE 12
Charaxes alpinus van Someren & Jackson and subspecies

Fics. 90, 91. alpinus van Someren & Jackson, ¢ and topotypical 9 (High forests on Vumba
Range), ¢ typical, fore wing rather pointed, slightly acuminate, falcate ; hind wing tails very
short and stout ; 9 fore wing discal bar comparatively restricted and narrow ; post-discal spots
free to 1b, spots white ; hind wing band broad at costa, tapering toward anal angle, tails robust ;
colour white with slight greyish-blue on borders, upper and undersides.

Fics. 92-97. mnyikensis ssp. n., Holotype ¢ and Allotype 2 (Nyika Plateau, Cottrell leg.) and
6 and 9 Paratypes respectively, upper and undersides.

PLATE $12

Bull. B.M. (N.H.) Entom, 18, 3

PLATE 13
Charvaxes manica Trimen and forms

Fic. 98. ¢ (Malawi: Nkata Bay), fore wing with well marked blue spots subcostal ; hind wing with indication of post-discal
greeny line; submarginal spots very small, punctiform, upper and underside.

Fic. 99. @ £. manica, (Malawi: Limbe), basal areas pale greeny blue ; fore wing post-discal spots strongly marked, upper and
underside.

Fic. 100. @ f. manica, (P.E.A.: Amatongas, nr. Beira), basal areas blue, but upper part of fore wing with dusky scaling ;
post-discal spots clear ; hind wing submarginal spots clear, upper and underside.

Fic. 101. Q f. manica, (Zambia: Mumbwa), basal areas strongly blue ; post-discal spots on fore wing reduced to one subapical ;
hind wing submarginal line obscured, upper and underside.

Fic. 102. @ f. manica, (Zambia: Mumbwa), basal areas dark blue, not extending to costa of hind wing ; hind wing submarginal
spots obscured, upper and underside.

Fic. 103. @ f. manica, (Zambia: Mumbwa), basal areas pale greeny-blue ; fore wing white bar very broad on hind border ;
submarginal line of hind wing clear, upper and underside.

Fic. 104. Q f. manica, (P.E.A.: Dondo-Amatongas), basal areas blue with some white dusting distad, upper and underside.

Fic. 105. Qf. manica, (Malawi: Nkata Bay), upper part of fore wing basal area black scaled ; post-discal spots obscured except
two subapical; hind wing patch mauvy-blue, upper and underside.

Fic. 106. @ f. pseudophaeus, (Malawi: Nkata Bay), basal areas mauvy-blue ; fore wing blue extended distad in 1a-1b, post-
discal spots clear to 2; hind wing marginal black border narrow but submarginal linear marks distinct and conjoined, upper and
underside. See also Plate 14, Fig. 115.

€ ‘gt -mowuyq (‘H'N) ‘Wd 1"

€i AaLVId

PLATE 14
Charaxes manica Trimen forms

Fics. 107, 108. @ f. chintechi van Someren & Jackson, Holotype and Paratype, upper and undersides. Photos B.M. (N.H.)
Nos. 34007 and 34008 ; 34011 and 34012.

Fics. 109, 110. @ f. chintechi van Someren & Jackson, (P.E.A.: Dondo) and (P.E.A.: Amatongas), basal areas of wings
suffused with greenish-blue scaling ; distal part of fore wing cell blackish ; discal and post-discal spots orange, the former shading to
bluish-white on hind margin; hind wing patch bluish-green with white discal bar showing through, these specimens represent a
primitive stage toward forms manica and pseudophaeus, upper and undersides. Coll. Pennington.

Fics. 111-113. Q f. protomanica f. n., Holotype (Malawi: Chintechi. B.M. (N.H.)), Paratype (P.E.A.: Amatongas.
Coll. Pennington) and (Nyasaland: Chintechi) respectively, fore wing spots orange, discal bar white ; base of wing greeny-blue ;
hind wing basal area greeny-blue ; discal patch greeny-blue with white discal bar showing through, upper and undersides.

Fic. 114. @ f. protomanica f.n., (Malawi: Chintechi. B.M. (N.H.)), shows merging of spots beyond cell with curved white bar ;
hind wing patch strongly blue-green ; fore wing post-discal spots orange ; a more advanced stage in the development of the manica
pattern but retaining the primitive spots in the fore wing, upper and underside.

Fic. 115. Q f. pseudophaeus, (P.E.A.: Dondo-Amatongas), basal areas greeny-blue, upper and underside.

€ ‘gr -mowuq (‘H'N) ‘W'd ‘1

bl ALVId

PLATE 15
Charaxes gallaghert van Son and Ch. fulgurata Aurivillius

Fics. 116-118. gallagheri van Son, Type 3, ¢ var. and Neallotype 9 (Rhodesia : Mapembe,
nr. Odzi R., H. Cookson), upper and undersides. Photos D. Cookson.

Fics. 119 and 120. fulgurata Aurivillius, g and 9, the original figures published by Dewitz
as ‘ ephydra var.’ on which Aurivillius founded Charaxes fulgurata, upper and undersides.

Fic. 121. fulgurata 9 f. fulgurata Aurivillius (/unigera Rothschild), 9 (Zambia: Chisamba),
basal half of fore wing and disc of hind wing greenish-blue, the blue extending well over the
hind part of the fore wing white bar; margin of wing strongly glaucous ; submarginal line of
hind wing strong, upper and underside.

Fic. 122, fulgurata 9 f. fulgurata Aurivillius, 9 (Zambia: Chisamba), basal area of fore wing
and disc of hind wing greeny-blue ; white bar of fore wing well developed ; distal half of wing
very black, but margin glaucous; hind wing submarginal line obscured, upper and underside.

Fic. 123. fulgurata 9 f. fulgurata Aurivillius, 9 (Zambia: Chisamba), basal areas darker
greeny-blue with some dark scaling in cell of fore wing; margin glaucous; lower post-discal
spots obscured, submarginal spots hardly visible but border strong, upper and underside.

PLATE 15

Bull. B.M. (N.H.) Entom. 18, 3

116

117

122

PLATE 16
Charaxes fulgurata Aurivillius and forms, and Ch. martini sp.n.

Fics. 124, 125. fulgurata Aurivillius, 3, and 9 f. mima Riley (Zambia: Solwezi Dist., bred
Dollman), upper and undersides. Photos B.M. (N.H.) Nos. 34030 and 34031 ; 34013 and 34014.

Fic. 126. fulgurata f.f. mima Riley, 2 (Zambia: Chisamba, Cottrell Coll.), basal area of fore
wing and disc of hind wing darker greeny-blue ; subapical blue spots of fore wing strong but
lunules faint ; submarginal spots of hind wing distinct, upper and underside.

Fic. 127. fulgurata f.{. mima Riley, 2 (Zambia: Chisamba), basal area of fore wing and disc
of hind wing bright greeny-blue ; margin of fore wing very glaucous ; subapical marks and lunules
well marked ; some extension of greenish scaling distad in ta-1b; hind wing marginal line
strong, upper and underside.

Fic. 128. fulgurata Aurivillius, g¢ (Zambia: Kabompe/Bulwale, Cottrell Coll.), with upper-
side markings well developed, especially the post-discal green line in hind wing, but with weakly
developed pattern on underside, upper and underside

Fic. 129. viola 9 f. (Zambia: Mankoya, Cottrell Coll.), base of fore wing and disc of hind
wing greeny-blue ; discal and post-discal spots in upper half of fore wing orange tawny ; sub-
marginal spots in hind wing strong; border well developed ; c.f. f.f. chintechi of Ch. manica,
upper and underside.

Fics. 130 and 131. martini sp. n., Holotype $ (Malawi: Mlange, Mlosa stream), upper and
underside.

PLATE 16

Bull. B.M. (N.H.) Entom. 18, 3

?
" o Pom
“ i
?
-
1s

A LIST OF SUPPLEMENTS
TO THE ENTOMOLOGICAL SERIES
OF THE BULLETIN OF
THE BRITISH MUSEUM (NATURAL HISTORY)

zt. Masner, L. The types of Proctotrupoidea (Hymenoptera) in the British
Museum (Natural History) and in the Hope Department of Entomology, Oxford.
Pp. 143. February, 1965. 5.

2. Nixon, G. E. J. A reclassification of the tribe Microgasterini (Hymenoptera :
Braconidae). Pp. 284; 348 Text-figures. August, 1965. {6.

3. Watson, A. A revision of the Ethiopian Drepanidae (Lepidoptera). Pp. 177;
18 plates, 270 Text-figures. August, 1965. {£4 4s.

4. SANDS, W. A. A revision of the Termite Subfamily Nasutitermitinae (Isoptera,
Termitidae) from the Ethiopian Region. Pp. 172; 500 Text-figures. October,
1965. £3 5s.

5. AHMAD, I. The Leptocorisinae (Heteroptera: Alydidae) of the World. Pp. 156;
475 Text figures. November, 1965. {£2 I5s.

6. OxaDA, T. Diptera from Nepal. Cryptochaetidae, Diastatidae & Drosophilidae.
Pp. 129; 328 Text-figures. £3.

7. GILIOMEE, J. H. Morphology and Taxonomy of Adult Males of the Family
Coccidae (Homoptera: Coccoidea). Im press.

8. FLETCHER, D. S. A revision of the Ethiopian species and a check list of the
world species of Cleora (Lepidoptera: Geometridae). In press.

PRINTED IN GREAT BRITAIN BY ADLARD & SON LIMITED, BARTHOLOMEW PRESS, DORKING

INDO-ORIENTAL HORAGINI
(LEPIDOPTERA : LYCAENIDAE)

C. F. COWAN

BULLETIN OF
THE BRITISH MUSEUM (NATURAL HISTORY)
ENTOMOLOGY Vol. 18 No. 4
LONDON: 1966

° AS nes

INDO-ORIENTAL HORAGINI
(LEPIDOPTERA : LYCAENIDAE)

By

C.F. COWAN \,,

Little Gaddesden House, Berkhamsted, Herts., England

Pp. 103-140 ; 3 Plates

BULLETIN OF
THE BRITISH MUSEUM (NATURAL HISTORY)
ENTOMOLOGY Vol. 18 No. 4
LONDON : 1966 :

THE BULLETIN OF THE BRITISH MUSEUM
(NATURAL HISTORY), instituted in 1949, ts
issued in five series corresponding to the Departments
of the Museum, and an Historical series.

Parts will appear at irregular intervals as they become
veady. Volumes will contain about three or four
hundred pages, and will not necessarily be completed
within one calendar year.

In 1965 a separate supplementary series of longer
papers was instituted, numbered serially for each
Department.

This paper is Vol. 18, No. 4 of the Entomological
series. The abbreviated titles of periodicals cited
follow those of the World List of Scientific Periodicals.

© Trustees of the British Museum (Natural History) 1966

TRUSTEES OF
THE BRITISH MUSEUM (NATURAL HISTORY)

Issued 10 June, 1966 Price £1 Is.

INDO-ORIENTAL HORAGINI
(LEPIDOPTERA : LYCAENIDAE)

By C. F. COWAN

SYNOPSIS

The isolated but closely allied genera Horaga Moore and Rathinda Moore fly in the triangle
India-South China~Papua. They have a characteristic appearance and highly aberrant
genitalia. They contain only eight species with about forty subspecies, to which four new ones
are now added. All published names are discussed and a number of species and subspecies are
illustrated for the first time, as are the ¢ genitalia of all species, complete and to the same scale.

Full references are given for all taxa, and a catalogue is included of the specimens in the
British Museum (Natural History).

CONTENTS

Page
INTRODUCTION. : : ; : : ; ; : ; z fLO§
STATUS OF THE TRIBE . : : , 5 ; : ‘ ; e) 206
EARLY STAGES . ; ‘ . , ; E ; ; ; £OT.
SCOPE OF THE SURVEY . : : ; ‘ ; : ; ; . 108
EXTRANEOUS TAXA . : F ‘ : : : ; ‘ . 108
ACKNOWLEDGEMENTS . : ; ; a ; : : ‘ + OS
Rathinda MooRE : ; : ; ; : ‘ : , PeerOO)
R. amor (Fabricius) : ; : : : ; : , +), £09

Horaga Moore: Name; Pattern; g Genitalia; Sexual Insignia; Range ;
Keys to the Species ; ; : : 4 : : 110
H. onyx (Moore) ‘ : ‘ . : ; : ‘ : . 525
H. syrinx (Felder) . : A : A : : : ; <" (E25
H. albimacula (Wood-Mason & de Nicéville) . : , : ; ramet 7
H. amethysta H. H. Druce . ; F : : : ‘ : .? eae
H. selina Grose Smith , ; : : ; i : ‘ o OE34
H., lefebvrei (C. & R. Felder) A i ; : ; ‘ ‘ ieeras
H.vavasanaSonan . : ; ; : ° F é : na 136
SYNONYMIC LIsT . ; , : a : ; ; 5 : Zh 530
REFERENCES ‘ : . : : ‘ : ‘ ‘ ‘ 7 also
INDEX : ; ‘ ; ‘ ; : ; : j : <= 540

INTRODUCTION

THERE are three small Indo-Oriental generic groups or tribes in the Theclinae sub-
family of Lycaenidae which I refer to as the Horagini, the Cheritrini, and the
Marmessini (tribb. n.). In the course of a current study of the last, problems
involving species of the other two have repeatedly arisen, and the present survey
has been compiled. Separate articles are planned for the Cheritrini and the Mar-
messini, but some repetition will be avoided by covering all three in the general
remarks below on their status, and the full list of references appended should require
only minor additions later.
ENTOM. 18, 4. 9

106 C. F. COWAN

There are 61 names involved in the nomenclature of the 2 genera Horaga and
Rathinda. They have never yet been comprehensively surveyed though Seitz
(1926) discussed 40 (figuring 12) out of the 52 then current. One of those omitted
by Seitz was syrinx (Felder, 1860), the second senior taxon.

Seitz’ work was based on that of Fruhstorfer, who had published two important
preparatory articles in I912 and 1914. These articles demonstrate one great
difficulty in the genus Horaga ; that of distinguishing between the various races of
onyx, which now appears as a mainly northern species, and those of the much more
southerly syrinx (at that time known as moulmeina). In 1912, as will be seen in the
synonymy under each name, Fruhstorfer listed 15 subspecies for onyx, including
moulmeina. Two years later he had moved to the other extreme, allotting moulmeina
13 and leaving onyx with only 3. Seitz listed them all together, and so most of them
remained till 1941.

Corbet (1941) dealt with 30 names, including 3 not mentioned by Seitz, in his
appraisal of Horaga from the Malayan viewpoint. He recognized the importance of
syrinx and, distinguishing between it and onyx, arranged their respective subspecies
almost as they are below.

STALUS OF “LAE TRIBE

Each of the three tribes Horagini, Cheritrini and Marmessini forms a compact
entity quite distinct anatomically from all other Lycaenidae. Each is smooth-
eyed, has smoothly scaled palpi, antennae with gradual cylindric club, and a lobed
hind wing with three tails of which the central one, that at vein 2, is the longest.
These characters are also shared by the small genera Eooxylides de Nicéville and
Neomyrina Distant which, however, are both of unmistakable appearance and have
more normal Lycaenid genitalia, while the fore wing vein 5 of Neomyrina consistently
originates much closer to vein 6 than to vein 4 instead of mid-way between them.

The Horagini differ in appearance from the other two tribes, as also from Eooxylides
and Neomyrina, in wing pattern and in having the hind wing tails all filamentous,
the termen not being extended along any ; whereas the others have the tail at the
end of vein 2, at least, broad-based, both the wing margin and the vein being produced
along the vein for some distance.

The ¢ genitalia of each tribe are highly aberrant and quite characteristic. They
are the most positive feature for identifying the allegiance of a doubtful g. They
are small, compact and robust in the Cheritrini, whose stout aedeagus is charged
with at least one cornutus and each of whose valvae carries dorsally a long rearward-
curved horn. They are large, flat, simple, and loosely assembled in the Marmessini.
In both these tribes the normal brachia or falces of the tegumen are entirely obsolete.

The Horagini, by contrast, has a heavy, elongate and most elaborate ¢ genitalia
without cornuti or horned valvae, but incorporating long and specialized brachia
(Pls. 2 and 3) which are unique among Lycaenidae. They are not the usual simple
paired hooks curved rearwards, but in Rathinda are very long with a central spicule,
and curve forwards to the bases of the valvae where they curl round for the tips to
point rearwards. In Horaga they are folded ventrally above the valvae and are

INDO-ORIENTAL HORAGINI 107

asymmetric ; one, always the left, is simple and more or less falciform, while the
right is modified in each species, spectacularly so in three.

The wing structure has been discussed by Moore (1881), de Nicéville (1890),
and others in detail and will not be mentioned further except to emphasize that no
feature can be found in it reliably to separate the two genera. Rathinda can readily
be distinguished from Horaga by wing pattern and by ¢ genitalia, but not by shape
or venation ; early definitions on the last basis for Indian forms resulted in the
quite plausible description of a new Rathinda in the Philippines which in fact was an
already known indigenous Horaga.

Wing venation is a valuable aid to the separation of genera, but there comes a
point when individual or specific variation in some detail exceeds generic, and it
becomes necessary to define recognizable genera on other grounds to avoid sinking
them. The latter step should not be taken till far more is known about early stages,
anatomy, courtship procedure and the significance of sexual insignia (‘‘ secondary
sexual characters ”’) (cf. Varley, 1962). Such a situation was met in the Pierid genus
Saletara Distant which (vide Cowan, 1955) had to be upheld distinct from Apfias
by the unusual ventral tuft of the 2 when venation differences proved unreliable,
and the recent merger of four good Lycaenid genera into Jacoona Distant on the
grounds that no external feature could be found to distinguish their 99 is regrettable ;
the $¢ are so divergent, and habits and habitats so varied.

So whereas Rathinda could, on wing venation and shape and most other superficial
characters, well be submerged in Horvaga, which has page priority, I preserve it.

EARLY STAGES

The larva and pupa of Rathinda amor were described and well figured by Moore
(1881 : 71, pl. 34, figs. rb) mistakenly under the name Sfalgis epius (whose correct
early stages were illustrated by Aitken, 1894). Another illustrated account was
given by Davidson, Bell & Aitken (1896 : 389, pl. 5, figs. 7, 7a). Sevastopulo (1935,
and 1947 : 577) added further food-plants.

Horaga onyx was similarly treated by Mackinnon & de Nicéville (1898 : 387, pl. 5,
figs. 18a, 18b), and on the following page H. albimacula (viola).

In all these accounts only the mature larva and the pupa are dealt with. Bell
(1919 : 754) gave a most valuable description of the full life history of Rathinda amor,
giving its habits and listing food-plants, after which he summarized all known about
H. onyx and H. albimacula.

The known larvae of the tribe are very similar in form as might be expected, and
unlike other Lycaenidae with from eleven (Horaga) to fifteen (Rathinda) triangular
fleshy dorsal and lateral horns. The resulting prickly appearance of the larva
inspired the colloquial name “‘ Monkey Puzzle” for Rathinda, after the spinose
Chilean conifer Avaucaria. This species feeds on the flower and leaf buds of a large
range of shrubs including Rubiaceae, Myrtaceae, Dipterocarpeae, Euphorbiaceae,
Loranthaceae and Sapindaceae, while the only food-plant recorded for Horaga is
Coriaria nepalensis.

The pupae, which stand erect from the upper surface of a leaf or from a twig, are

108 Cc. F. COWAN

stout and strongly incurved ventrally. They are firmly fixed at the tail but other-
wise free, and are capable of making a clicking noise.
There is no record of association with ants, whose presence is not necessary for
survival.
SCOPE OF THE SURVEY

All the published names in the tribe are discussed below, giving references and the
representation of each in the collections of the British Museum (Natural History),
which is abbreviated hereafter to B.M. (N.H.).

The Synonymic List of the complete tribe is in the same sequence as the main
discussion, gives the range of each subspecies, and acts as a summary.

Photographic illustrations are given on the Plates of at least one ¢ specimen of
each species, and of the complete 3 genitalia to scale of all species.

EXTRANEOUS TAXA

The following seven names have from time to time been included incorrectly in
Horaga and will not be mentioned further :

achaja Fruhstorfer, 1912 : 233. Anatomically in the Marmessini; to be dealt
with in due course.

akara Ribbe, 1926:83. Assumed a misspelling for anava Fruhstorfer. See
entry below under celebica Ribbe.

andamana Moore, 1877: 589. A subspecies of Hypolycaena erylus (Godart), not a
synonym of Horaga rana de Nicéville.

araotina Evans, 1933 : 413. A name which will be dealt with in Marmessini.

celebica Ribbe, 1926: 83. ‘‘ Horaga? ... ahnlich H. akava ...”. From further
lengthy description undoubtedly not a Horaga, probably either Hypolycaena
or Chliaria.

himeros Fruhstorfer. Nomen nudum? Probably a label name only, for Rathinda
amor in Ceylon. ¢ and 9 “types” in B.M. (N.H.).

inart Wileman, 1908: 325. Described as a Tajuria species. Applies to the
Formosan subspecies of Chliaria kina Hewitson, see Corbet (1940 : 90).

ACKNOWLEDGEMENTS

I am grateful to the Trustees of the British Museum (Natural History) for per-
mission to examine the National Collections and Libraries, and to the Keeper of
Entomology and his staff, particularly those at Tring where the Lycaenidae are at
present housed, for their ever-ready help. I am especially indebted to Mr. N. H.
Bennett of the Staff for his patient assistance in the dissection of specimens and
advice from his long experience of Lycaenidae, and to Mr. G. E. Tite for his continual
encouragement.

I would like also to thank Colonel J. N. Eliot, Mr. G. C. Stubbs, and the Hope
Department of Entomology, University Museum, Oxford for the generous loan of
specimens from their collections, and finally Mr. J. V. Pearman, without whose
extra-low-power eyepiece the very large genitalia would not so easily have been
photographed.

INDO-ORIENTAL HORAGINI 109

RATHINDA Moore, 1881

Rathinda Moore, 1881: 99. ‘‘ Type R. amor”? (ibid.).
Cupido Hiibner, 1819: 77 [praeocc.]. Type Papilio amor Fabricius, fixed by Hemming,

1960 : IO.

The gender of the genus must be fixed. The derivation is obscure. It may reflect
the sombre colour by coupling the words Hindi and vat (Urdu; gender feminine)
meaning night. Less probable is a connection with vatha, the Brahman processional
car with red decoration, or with vathin, the ancient Brahman priesthood. But it may
well be fanciful ; Moore had earlier in the same publication (p. 56) introduced the
name Rahinda for a new Nymphalid genus. All things considered, it should be
regarded as feminine.

Moore diagnosed the genus on the page following his Horaga, but defined no actual
difference. The fore wing costal veins are more widely separated, and the hind wing
termen more dentate, than is the case with Horaga species in India, but not with
some from further east.

The hind wing pattern is distinctive and unusual in having a more or less complete
submarginal band of reddish spots on the plain dark brown upperside, and a complex
pattern of black strigae and whitish stipple on the buff underside. The white band
which bisects the brown fore wing on both surfaces is arcuate from mid-costa to
the outer end of the dorsum, not straight to mid-dorsum as in Horaga underside.

The 3 genitalia are as aberrant as those of Horaga, and of similar stamp, but the
long reflexed brachia with their central spine are quite distinct.

The differences between the two genera may be summarized as follows :

RATHINDA

Hind wing upperside with red markings ; no blue on the upperside ; underside base of fore
wing and most of hind intricately patterned with black and whitish on buff; white band on
underside of fore wing arcuate from mid-costa to near tornus.

3d brachia symmetric, each with a median spine, and excessively long; 1} times length of
vinculum and the excess length curled round in the anterior, ventral, end of the vinculum.

No ¢ insignia.

Small; confined to Ceylon and India south of the Himalaya.

HORAGA

Upperside often with blue or violet, but never red ; underside ground colour more or less
uniform and markings simple ; fore wing underside white band inner edge straight, from mid-
costa to mid-dorsum (or otherwise in the Philippine Jefebvret and the Formosan rarasana).

$6 brachia asymmetric, and not longer than the vinculum.

With or without ¢ insignia.

Mostly larger ; all seven species occur east of India, three reach there, of which two are found
in Ceylon.

Rathinda amor (Fabricius)

(PL tte, TOs, Pl 2) ties, £73'b)

Papilio amor Fabricius, 1775 : 518, No. 321. India Orientali.
Papilio triopas Cramer, 1780 : 64, pl. 320, figs. G, H. Coromandel.
(The misspelling tviopus has frequently appeared.)

IIo Cc. F. COWAN

Also figured by Seitz, pl. 146B, fig. fr (Q) ; by Woodhouse (1952 : pl. 20, figs. rr,
12) ; and many others.

A species somewhat variable in size, markings and underside colour according to
season, but with no constant geographic differences.

Fore wing length 3 10-14 mm., ? 12-16 mm.

It has been said not to occur above 2,500 feet, but Woodhouse (1952 : 138) gives
it up to 3,000 feet in Ceylon and I have found it at that elevation in South India
round Bangalore and Nandi Drug.

B.M. (N.H.), 106 3, 146 9, Ceylon, Peninsular India, East Pakistan, and Assam.

HORAGA Moore, 1881
Horaga Moore, 1881: 98. ‘“‘ Type H. onyx.” (ibid.).

Gender of the Genus

I deduce that the name is an allusion to the characteristic underside white stripe
which, with its dark edging, gives a trompe l’oeil slit-like or gashed effect. The Greek
feminine noun pawyas means a chink or rent, and the genus is of feminine gender.

Hewitson (1863 : 35), dealing with the Philippine species lefebvrei, at that time
in the genus Myrina, remarked that Boisduval had proposed a new genus for it.
No such name has been traced.

Generic Pattern

The basic wing pattern of the genus is simple. The upperside is dark brown to
black with an ovate white fore wing discal patch ; the wing bases often being broadly
blue or violet. The underside is ochreous to olive brown with a white stripe across
both wings from mid-costa on the fore wing to near the hind wing tornus, where it
becomes overlaid with shining metallic green scales and is sharply angled inwards
to end at mid-dorsum ; this stripe is narrowly dark brown along its outer (fore wing)
or inner (hind wing) edge. The hind wing tornus bears two large black spots, one
marking the lobe, the other at the end of space 2, which are separated by a large
black-dusted pale grey space. Interior to these markings is a series of metallic
green lines.

The basic pattern is slightly modified for each species, but is only departed from
materially on the undersides of the two eastern species lefebvvet and rarasana.

3 Genitalia

The Felders, when naming /efebvrez in 1862, remarked on the long slender abdomen
of the g, and its striking resemblance to the equally aberrant but quite unrelated
species of the Miletini. In Hovaga the length is caused by the elongation and
inclination of the vinculum, which lies almost parallel to the abdominal axis and the
valvae. Only in onyx and syvinx do the genitalia even approach normal size ; in
the smallest species, amethysta, they are some 5 mm. long, comparable with half the
fore wing length. The tegumen terminates in small paired subtriangular uncal
plates. The aedeagus is simple and relatively small.

INDO-ORIENTAL HORAGINI III

The left brachium or falx is in all species a more or less simple, curved, evenly
tapered spike, crossing the other like half-folded-arms between the tegumen and the
valvae from left to right.

The right brachium is not greatly differentiated from the left, nor does it vary
much, among the more widespread western species. In the difficult onyx-syrinx
complex, which has the more compact build, it is gradually flattened into a rounded
blade, ending in an abrupt point which springs from the inner edge.

The right brachia of albimacula and the eastern varasana, whose vinculum and
tegumen are about twice as heavy, valvae longer and aedeagus relatively slimmer,
are very similar.

Two species have a spur near the base of the right brachium. In the Philippine
lefebvret the brachium is darkly sclerotized for the basal third, where it emits a
sharp spine ; it is then deflected inwards as a long, broad, square-shouldered blade ;
and its terminal third is a stout curled spike, very brittle and easily broken off (it
was broken in three out of six specimens examined). H. selina from Sulawesi
(Celebes) has a prominent handle-like projection from the extreme base of the right
brachium, which is then angled, and terminates in a broad square bifurcation.

The armature of the rare amethysta is even longer, and the right brachium crosses
over to end in an abrupt saucer-shaped disc which it holds in a horizontal plane
parallel to the wings when set. The function of this frying-pan affair defies conjec-
ture. It also is brittle and frequently missing.

The valvae of all species are more or less densely covered on their inner faces with
a vesture of fine hairs which may be spinose in the basal portion. They are blade-like
and tapered, thick along the curved ventral edge, very thin dorsally, round-tipped,
and specifically distinct.

Sexual Insignia

The 3 insignia (‘‘ secondary sexual characters ’’) of the genus, when present, are
modest. None are evident in albimacula, amethysta, and selina, and rarasana has
only a small indistinct dark grey brand lying astride vein 1, about a third of its
length from the base, on the fore wing underside.

Both onyx and syrinx have the fore wing dorsum slightly bowed outwards instead
of being quite straight, and they bear a long oval ochreous brand in a larger white
polished area about the centre of vein r. Associated with this brand, on the upper-
side of the hind wing, is a dense patch of very fine, long, dark brown hairs rising from
the basal half of space 7. These hairs, which are somewhat fugitive, are normally
erect but may be flattened by the fore wing in set specimens. They have not been
noted before and are obscure, being exactly concolorous with the basal half of the
hind wing costa, but can be clearly seen when viewed in silhouette from the side at
wing level. They are much finer than the ordinary recumbent hair-scales found on
the wings of both sexes.

On the underside of the fore wing of the ¢ lefebvrei there is a similar but obscure
white brand, and the hind wing upperside has a very dense tuft of slightly paler and
much more concentrated, so less obscure, greyish brown hairs rising from the upper

112 C. F. COWAN

half of the cell. The 3 palpi of this species are of a greasy translucent cream colour,
whereas those of the 9 are of the normal white and black.

Geographical Range

Horaga species inhabit India (south and east from Himachal Pradesh), Ceylon, and
East Pakistan, southwards through Malaysia and Vietnam to Formosa and New
Guinea. The largest is confined to Formosa, one species is known only from the
Philippines, and one only in Sulawesi. The remaining four species range wider.
None is common, though they often favour open country.

In listing subspecies a west to east sequence will be followed.

KEYS TO THE SPECIES

A Key for the separation of the Hovaga species by superficial characters is given below,
followed by one based on the ¢ genitalia. These cannot be combined, aberrant tendencies from
either aspect not always coupling with those from the other.

KEY FOR THE VISUAL SEPARATION OF HORAGA SPECIES

(Pl. 1)
I Underside with the normal generic pattern, the ochreous colouring similar on either
side of the median band on both wings. Widely distributed. ° ‘ 2
Underside with the hind wing at least abnormal. Restricted (eeabution ° 5

2 Larger in each locality. Hind wing underside white band distinct. Upperside in
most areas with much blue or violet. ¢ fore wing dorsum slightly bowed and an
ochreous brand on the underside near the centre of vein 1 ; hind wing upperside
with an obscure subcostal area of long brown hairs. Very variable asta

outside the tropics, and geographically . ; 3
— Smaller. Hind wing underside median band obsolescent, often leaving only the inner
dark line. Upperside much less blue, often none. No ¢ insignia ; 4

3 Usually slightly smaller, duller, and with more rounded wings in each locality. The
underside white band broader on both wings, more abruptly dilated just below the
fore wing costa to which it, or at least the dark line at its outer edge, usually
extends ‘ . onyx
— Slightly larger, brighter, the 3 fore wing apex pointed and the termen straighter.
Underside brighter, the fore wing white band not reaching above vein 7 and, in
Malaya, Sumatra and Borneo, distinctly narrower and even : : syrinx
4 Upperside all or mostly dark brown, with the fore wing white patch extensive as usual
and the violet areas normally very restricted. Underside with the fore wing white
band dilated as usual, at least 3 mm. wide in space 3. Fore leg tibia plain, or at
most one dark ring . ‘ albimacula
-— Upperside mostly violet, with regular ‘dull brown. borders: the fore wing discal spot
reduced to seldom more than a bar at the cell-end, though the band may show
through from below. Underside fore wing white band narrow, seldom over I mm.
wide, and straight. Fore leg tibia, as in onyx and syrinx, with two dark rings
between the dark joints . : ‘ amethysta
5 Underside with the fore wing normal and the hind wing normal up to the median band,
beyond which are an irregular metallic green line and two black spots in spaces 2
and 6, followed by a broad area to the termen minutely mottled ash-grey and white.
No blue on the upperside. Confined to Sulawesi : , selina

INDO-ORIENTAL HORAGINI

— Underside both wings abnormal .

6 Upperside dark brown, usually with some sparse basal blue scales, and the fore witig
white patch large, oblique, and shifted in to the centre of the wing. Underside
chestnut brown with dark-edged white markings which differ slightly between the
sexes ; on the fore wing a costal streak from the base under vein 12 and a semicircular
or subtriangular area based on the dorsum with apex at the cell-end; on the hind
wing a subcostal streak from the base, and ovate spots at the cell-end (larger,
triangular and filling the cell in the 9), postdiscal across spaces 4 and 5 and at
mid-costa (all conjoined in the 9); two metallic green lines, one curving from the
base below the cell to mid-vein 4, the other submarginal. Fore leg tibia mainly
black. Male palpi translucent vaseline-yellow, and the hind wing with a prominent
tuft of grey-brown hairs overlying the base of the cell. Confined to the Philippines

113

lefebvrei

— Upperside black; cilia of termen prominently white, an oblique white band on the
fore wing from near mid-costa towards the tornus, and a rectangular shining violet
subcostal area filling most of spaces 4 to 6 on the hind wing. Underside chalk-white;
both wings with ochreous narrow cell-end bars and broader postdiscal and terminal
bands, the last carrying on the hind wing a complete submarginal series of metallic

green lunules and a large black tornal spot in space 2._ Confined to Formosa. rarasana

KEY TO THE DIFFERENCES IN ¢ GENITALIA HORAGA
(Pls. 2 and 3)

1 Conventional for the genus; right brachium or falx simple, curved; uncus lobes
broadly triangular; valvae tapered, curved, and hirsute along ventral half of inner
face : ‘ . ; ; ; : ‘ j ;

— Right Beenie ainonnet compound; uncus lobes stunted; valvae less tapered,
inner face more completely hirsute .

2 Smaller, overall length excluding aedeagus under 2m, and not visibly affecting
length of abdomen. Valvae basal half broad, distal taper more abrupt .

— Abdomen, as in all remaining species, pra. elongate, genitalia length excluding
aedeagus at least 3 mm.

3 Valvae basally truncate, distal third enpetea te the tip i eriglen xe abou 207, aig
hairs and anterior spines clothing only the posterior half of the inner face; usually
but not invariably with a distinct tooth mid-way along the ventral margin.
Aedeagus stout (Pl. 2, fig. 18).

—- Valvae cued of uniform width, distal quarter hapenay end een peved ARASS
almost 90°; distal three-quarters of inner face hairy; ventral margin simple.
Aedeagus slim. (cf. Fruhstorfer, 1912 : 232, fig. 3, from Java) (Pl. 2, fig. 19)

4 Uncus lobes broad as in onyx and syvinx. Valvae not projecting beyond uncus;

N

onyx

syrinx

with a slight but even curve. (Pl. 2, fig. 20) . ; albimacula

—- Uncus lobes long and narrow. Valvae project well evened end ae uncus and are

narrow, tapered, and distinctly sinuate (Pl. 3, fig. 21) ' . Yrarasana

5 Right brachium ends ina disc. Uncus lobes small and ovate (PI. 3, Ae. 33)

— Right brachium sharp-ended; equipped with a sub-basal is bans Uncus lobes
stalked ‘

6 Right brachium with the Garibiniat half neal Bitareate (Pl. 3; fig. 23)
— Right brachium centrally flattened and shouldered before a single terminal curled

. selina

amethysta

6

spike (Pl. 3, fig. 24) ‘ é : ; i ; : ‘ : . lefebvrei

114 C. F. COWAN

Notes: (i) The valvae of five species are well illustrated to scale by Corbet (1941 :
49). That of onyx is one without the central tooth. Also shown with them, but
reversed in aspect so concealing its allegiance to the Marmessini, is that of achaja
Fruhstorfer.

(ii) Three illustrations are given in Corbet (1956: pl. 16). These are not to scale,
otherwise that of albimacula would be by far the largest on the plate.

(iii) Shir6zu (1960 : 31T, 313, figs. 345, 346) gives admirable illustrations of rarasana,
albimacula (as anytus), and onyx to scale. That of onyx shows the central tooth of
the valva.

HoRAGA ONYX
(Pl. 1, figs. 1-5; Pl. 2, fig. 18)

Though it has the greatest range in longitude, from N.W. India to South China
and Formosa, this species is well surpassed in southern latitudes by syrinx. It is
the common species in the north but becomes much rarer at the equator and does
not extend beyond Sambawa and Borneo. Its full range on the Asian mainland will
be interesting to ascertain. It would appear to have reached Formosa via Asia
rather than the Philippines where it is not known, but neither has it been recorded
between Thailand and Hongkong, nor from Hainan. In its northern, non-tropical
races it has well-marked seasonal forms.

The anterior tail is very short in Ceylon specimens, intermediate in Indian ones,
and thereafter normal.

The white band on the fore wing underside, or at least the dark posterior line
edging it, reaches the costa, whereas in syvinx it does not. This feature, well known
and reliable in Indian races, breaks down in specimens from Malaya, Sumatra and
Borneo where the band becomes extremely narrow and even in syrinx, “ sausage-
shaped ” in 99 and linear in some 3g, while in onyx it is also narrowed but continues
to be anteriorly dilated. Then in Java, as far south of the equator as Ceylon is
north of it, the broad-banded features return to both species.

A series of seven Hovaga ex coll. Rothschild taken by Hagen on the island of
Banka in 1891 contained two onyx and four syrinx 3d (confirmed by dissection)
and cast an interesting light on the appearance of the two species there. They
conform to the respective Sumatran subspecies.

H. onyx cingalensis Moore

Horaga cingalensis Moore, 1883 : 525. Ceylon.
H. onyx cingalensis Moore ; Fruhstorfer, 1912 : 232; I914: 35.

Figured by Woodhouse (1952 : 138, pl. 20, figs. 20, 21.)

This is quite the brightest of the onyx subspecies. The ¢ upperside is of a deep clear blue
which fills most of the fore wing cell; the fore wing apical border is black and tapers to a point
at the tornus; the white spot is clear, quadrate, comparatively small, and usually with some
blue between it and the terminal border. The hind wing is blue up to the costal border above

INDO-ORIENTAL HORAGINI II5

vein 6, and out to the almost linear terminal border ; the white median band is faintly visible
by transparency from below.

The 9 upperside is duller, rather violet-blue, with a slightly larger white patch and broader,
more diffuse, dark borders.

The underside is olive-brown as in onyx wet season form, with well-marked white bands and
tornal black spots.

The size is average for the species ; fore wing length 13-16 mm.

South Indian and some North-West Indian specimens are intermediate between this and the
next subspecies, having broader black borders but the same bright blue colour in the ¢.

B.M,. (N:H.) 3 Holotype; 31.0) 12°9, Ceylon; 22.4):23 93'S. India.

H. onyx onyx (Moore)
(Ebr fipsst 2)

Thecla onyx Moore, 1857 : 30 as Boisduval M.S. ‘‘ Moulmein ”’.
Horaga onyx (Moore) ; Moore, 1882: 248. ‘‘ Himalaya not Burmah.”
H. onyx onyx (Moore) ; Fruhstorfer, 1912 : 232.

H. onyx onyx (Moore), f. avta Fruhstorfer, 1914: 34. Assam.

Figured by Seitz in Vol. r (1908: pl. 72, figs. cr, c2). In Vol. 9 (1926: 981) he
says these figures are of cingalensis, but they are much nearer onyx.

This is the type-species of the genus, and its senior taxon. Though the name
had been current with various authors for at least ten years before 1857, Moore’s
was the first description. _

Moore described his single specimen as “ pale ferruginous brown ”’ on the underside,
and from “‘ Moulmein—presented by the Trustees of the British Museum ”’ to the
East India Company Museum. The East India Company collection was returned
to the B.M. (N.H.) in 1860.

In his subsequent amplification Moore said that the type specimen was a 4, that
the fore wing underside white band reached the costa, and that it was “ now in the
British Museum. Its locality label is Himalaya, not Burmah’’.

The specimen standing as the type in the B.M. (N.H.) is a 9, has a bright ochreous
underside, a fore wing band ending far short of the costa, and a label reading “ E.
Indies’’. It thus agrees neither in sex, colour, markings nor data with Moore’s
description. But a search through the main collection disclosed a 3 agreeing well
with the original description, labelled ‘‘ Himalaya” and “ E.I.C. 60-15. After
careful rechecking I am satisfied that this is the true holotype of Thecla onyx Moore,
and I have inserted it in the B.M. (N.H.) Type Collection, suitably labelled. The
currently accepted nomenclature is not disturbed.

Both sexes of this well known subspecies are variable in size, colour and markings.

A valuable series of 28 gj and 11 9 from the Naga Hills, Assam, ex coll. Tytler,
has full data labels and thus elucidates the pattern of the seasonal variation in this
region, which is paralleled in other non-tropical areas. Four representative $3
have been dissected and all have the dentate clasp.

The wet season form (f. onyx Moore) is larger (fore wing length 13-16 mm.), dark and well
marked. The upperside is much darker and duller than cingalensis ; both wings usually have

116 C. F. COWAN

broad terminal borders, with the blue dull and often suffused with violet ; the fore wing black
costal border fills at least the anterior half of, and often all, the cell. The underside is compara-
tively dark olive brown, with bold markings.

The dry season form (f. avta Fruhstorfer) is smaller (fore wing length 12-14 mm.) and much
paler, dull blue with narrower greyish borders and the white fore wing patch more extensive
on the upperside. On the equally paler underside the outer halves of each wing become light
grey, and the tornal spots are reduced to dots and yellow crowned.

B.M. (N.H.) g Holotype, Himalaya ; 143 3, 99 9, North India (Kumaon-Sikkim),
Assam (including the holotype of avta Fruhstorfer), Burma, Thailand.

H. onyx rana de Nicéville

Horaga vana de Nicéville, 1889 : 283, pl. 14, fig. 10. South Andaman Is.
H. ouyx [sic] rana de Nicéville ; Fruhstorfer, 1912 : 232.

This is a most distinct large race with intensely contrasted colouring above and
below. It was also figured by Swinhoe (1911: pl. 707, figs. 4, 4a, 4b). Fore wing
length 15-17 mm.

The upperside is black, with the fore wing discal white spot smaller than usual (so reminiscent
of mainland syvinx which is not recorded from these islands) ; the $ bears a dusting of bright
blue over the greater portion of the hind wing and usually the dorsal area of the fore wing. The

underside is rich dark chocolate-brown with broad white median bands, showing some seasonal
variation, particularly in the intensity of the hind wing terminal markings.

B.M. (N.H.) 17 3, 15 9, Andaman Is.; 1 4, Nicobar Is., r 3, ‘‘ Burma”’.

H. onyx zuniga Fruhstorfer

B. [sic, for H.] onyx zuniga Fruhstorfer, 1912 : 233. Nias.
H. moulmeina zuniga Fruhstorfer ; Fruhstorfer, 1914 : 35.
H. onyx zuniga Fruhstorfer ; Corbet, 1941 : 49.

Figured by Seitz; pl. 157, figs. 15, 16, 3; 17, 9.

This is another distinctive island race, but small and pale in very marked contrast to its
neighbour vana.

The upperside is pale dull violet with narrow brown borders, the white bands visible from
below by transparency ; the fore wing white patch is small and diffuse, and clear of the dull
brown border. The underside is rather dark ochreous grey with clear white bands.

Fore wing length ¢ 13 mm., 2? 14 mm.

B.M. (N.H.) g Holotype, 6 3, r 9, Nias Is.

H. onyx sardonyx Fruhstorfer
(Pl. 1, fig: 3 > Pi 2; fig. 18)

H. moulmeina sardonyx Fruhstorfer, 1914: 33. N.E. Sumatra.
H. onyx sardonyx Fruhstorfer ; Corbet, 1941 : 49.

Not previously figured.

INDO-ORIENTAL HORAGINI LL7,

It has often been remarked that insect forms from Sumatra and Malaya, and toa
lesser extent Borneo, are similar. This may be the case with onyx, but it seems so
scarce in the area that there is insufficient material to decide whether the one sub-
specific name available can be applied to all. Provisionally I do so.

True N.E. Sumatran sardonyx is represented in B.M. (N.H.) by the three specimens
mentioned by Fruhstorfer (l.c.), and also by de Nicéville & Martin (1896 : 479) who
remarked “‘ from Selesseh to Bekantschan (i.e. at 2,500 ft., between Medan and the
Central Battak Mountains) ... very rare, as Dr. Martin has not obtained more than
4 specimens in I3 years ’’.

The fore wing upperside has a very quadrate white patch set in a broad black apex and
termen and resting on the inner half of vein 2, the black border running inwards under vein 2
to below the centre of the white patch ; the blue area internal to the border and the patch just
enters the lower (dorsal) half of the cell. The hind wing costa above vein 6 is brown as usual, the
remainder of the wing being basally blue shading evenly to a dull brown termen with the usual
white submarginal and black marginal lines. The underside is dull ochreous grey; the fore
wing white band, though not extending above vein 7 with the dark line, is sharply widened
below it, to measure 3 mm. at vein 3.

Two $9 from Banka Island, over 600 miles away off S.E. Sumatra, are very similar,
but the intervening and the western forms from this vast area seem to be entirely
unknown.

A § and 2 from North Borneo are similar, but paler below with broader bands,
that on the fore wing reaching the costa, while the upperside white patch is more
diffuse and extends just below vein 2. There is no similarity between these two
specimens and corniculum Druce (see under syrinx maenala).

The fore wing lengths of all the foregoing are ¢ 13, 9 15 mm.

Specimens from Singapore tend to be smaller ($ rr-13, 2 13-14 mm.), darker, and
with less extensive white markings (PI. 1, fig. 3), but the anterior taper of the fore
wing underside band is still marked and serves to distinguish them from syrinx
maenala (Pl. 1, figs. 6, 9). These were taken flying in about equal numbers with
syrinx maenala, often on the same day, in 1937-38 and 1952-53, and the series of
each species include pairs taken in cop.

B.M. (N.H.) g Holotype, 2 3, N.E. Sumatra; 2 3, Banka (Hagen); 4 3, 1 9,
Singapore (Cowan), 1 3, 1 2, N. Borneo.

H. onyx fruhstorferi Corbet
(Pl. 1, fig..4)
H. onyx fruhstorferi Corbet, 1941 : 50. Central Java.

As is the case with syrinx, the transit to Java results in reduction of black colour
and extension of blue and white. The appearance is very reminiscent of the Ceylon
race,

Despite the large series of syrinx onychina from Java in B.M. (N.H.), there is
only the unique specimen of onyx fruhstorferi. The latter is much smaller (fore wing
length 13 mm.), with broader black borders (but not so broad as in sardonyx), and

118 C. F. COWAN

the blue areas slightly tinged with violet. The fore wing white patch extends well
below vein 2.
B.M. (N.H.) g Holotype, Central Java, 1,500 ft.

H. onyx akronyx subsp. n.
(Pl. 1, fig. 5)

The name is taken from the Greek axpov = furthermost, though axpovvé = tip
of fingernail. Sambawa marks the southern known limit of the species.

Matching fruhstorferi in size, akronyx 3 has a distinctly more pointed apex and straighter
termen to the fore wing.

The upperside blue colour is brighter and the white patch ovate rather than quadrate, while
the black borders are more restricted ; a white submarginal line appears at the fore wing tornus.
The holotype ¢ has a small black wedge anterior to the white patch at the base of vein 2 which is
lacking in the paratypes.

The underside is as in fruhstorferi but with slightly narrower white median bands.

The @ is similar to the ¢ but larger (fore wing 15 mm. against 13-14 mm.), paler, and with
more diffuse white markings.

B.M. (N.H.) g Holotype, 2 Allotype, and 2 g Paratypes, Sambawa (Doherty,
September 1891).

H. onyx moltrechti Matsumura

Horaga moltrechti Matsumura, 1919 : 604, pl. 47, fig. 9. Formosa.

Horaga asakurai Nire, 1920: 376. Formosa (Pulisha or Horisha).

H. onyx moltrechti Matsumura [syn. asakurai Nire] ; Shir6ézu, 1960 : 312, 313, fig. 346; pl. 67,
figs. 715-718.

Shirézu’s figure of the 3 valva and illustrations of the insects are excellent. I have
followed his synonymy.

Matsumura and Nire attribute the name moltrechti to Oberthiir but no justification
for this can be found.

This subspecies is probably as variable seasonally on the Asian mainland as onyx from India,
but the specimens available all approach the dry season form, compared with which the ¢ is
generally larger and of a richer blue and less black on the upperside, while the underside is pale
grey with obsolescent white bands and tornal black dots. The @ is heavily black-dusted above,
and ochreous brown below. Fore wing length 14—17 mm.

B.M. (N.H.) 2 g, Hongkong; 1 J, 2 9, Formosa.

HoRAGA SYRINX
(Pl. 1, figs. 6-9; Pl. 2, fig. 19)

Though not known west of Sikkim nor north-east of the southern Philippines, this
species ranges far south to New Guinea. It is the only species which is known to
straddle both Wallace’s and Weber’s Lines, and it shows interesting reactions to

INDO-ORIENTAL HORAGINI 11g

each which may be anticipated in other species. It also has a very distinctive
appearance in the “ Neomalayan”’ area of Sumatra—Malaya—Borneo.

It is sometimes very close to onyx in appearance, but can usually be identified
by its slightly larger size, a more pointed fore wing with straighter termen, and
brighter, more distinct, markings. The ¢ genitalia are as small as those of onyx,
and it has the same short abdomen.

H. syrinx sikkima Moore

Horaga sikkima Moore, 1883: 525. Darjiling.
H. moulmeina sikkima Moore ; Fruhstorfer, 1914 : 35.
H. syrinx sikkima Moore ; Corbet, 1941 : 50.

Nowhere well illustrated.

The description of this subspecies immediately follows that of moulmeina, of which
it has often been made a synonym. It appears sufficiently distinct, with a fair
amount of seasonal variation parallel to, but less pronounced than, that of onyx.

It is large (fore wing length 15-18 mm.), robust and dark. On the ¢ upperside the black fore
wing borders fill the cell and the outer quarters of spaces 1 ; often the base is black, leaving only
a small subdorsal area of dark blue; the white patch is nearly as large as in onyx but crossed by
black veins which give it a granular appearance ; the hind wing costal border is broad and the
terminal diffuse, leaving a discal area of dark blue more or less crossed by black veins. The 2
is duller overall, with the white patch diffuse. The underside is bright ochreous yellow, and the
fore wing white median band is strongly dilated centrally.

B.M. (N.H.) Holotype, Darjiling; 17 ¢, 12 9, Sikkim, Bhutan, Assam.

H. syrinx moulmeina Moore

Horaga moulmeina Moore, 1883 : 525. Moulmein.

H. onyx moulmeina Moore ; Fruhstorfer, 1912 : 232.

H. moulmeina moulmeina Moore ; Fruhstorfer, 1914 : 35.
H. syrinx moulmeina Moore ; Corbet, 1941 : 50.

Nowhere well illustrated, this subspecies is transitional to the Malayan one. It is smaller
than stkkima (fore wing length 14-16 mm.) ; the upperside white patch is more compact, seldom
showing in the cell nor below vein 2 in the ¢; the blue areas are brighter and more extensive,
usually filling the fore wing base and reaching the hind wing termen, where the border is replaced
by some black marginal spots in spaces 2 to 5. The underside is distinctly darker ochreous and
the median bands narrower.

B.M. (N.H.) g Holotype, Moulmein; 4 3, S. Burma to Mergui. The specimen
reported by Godfrey (1930 : 346) under H. halba from Renong (S. Thailand) is
probably this.

H. syrinx artontes Fruhstorfer
(Plex, 2g27)

Horaga affinis arvtontes Fruhstorfer, 1912 : 233. Nias Is.
H. syrinx artontes Fruhstorfer ; Corbet, 1941 : 50.

Not previously illustrated.

ENTOM. 18, 4 10

120 C. F. COWAN

This remarkable race does not follow the trend of onyx zuniga, but is large and silvery, like
species of Marmessus in this remote island. The unique ¢ specimen is as large as any from
Neomalaya (fore wing length 17 mm.). The upperside is pale shining grey-blue with very
narrow dark grey borders ; the white spot is small and narrow, edged with black and well clear
of the border ; the hind wing white median band shows through from below, and the terminal
white and black lines and white cilia are very prominent.

The underside is normal, the ground colour being rather dark brownish ochreous.

B.M. (N.H.) g Holotype, Nias (no further data).

H. syrinx maenala (Hewitson)
(Pl. 1, figs. 6, 9)

Myrina maenala Hewitson, 1869: Suppl. 7, pl. 3, figs. 85, 86. Borneo.
Horaga halba Distant, 1886: 460, pl. 44, fig. 23. Penang (9). syn. n.
Horaga corniculum H. H. Druce, 1895: 611, pl. 34, fig. 8. Kina Balu. syn. n.
Horaga affinis H. H. Druce, 1895 : 611, pl. 34, fig. 9. Kina Balu and Labuan.
. onyx corniculum Druce; Fruhstorfer, 1912 : 232.

. affinis affinis Druce ; Fruhstorfer, 1912 : 233.

. moulmeina (2?) corniculum Druce; Fruhstorfer, 1914 : 35.

. moulmeina halba Distant ; Fruhstorfer, 1914: 35.

. maenala (Hewitson) ; Seitz, 1926 : 982.

. onyx halba Distant ; Corbet, 1941 : 49.

onyx corniculum Druce ; Corbet, 1941 : 49.

. syrinx maenala (Hewitson) ; Corbet, 1941 : 50.

. syvinx affinis Druce ; Corbet, 1941 : 50.

THESE es

The authors’ illustrations quoted above are good. Seitz copies those of corniculum
(pl. 147, figs. b8, cr) and affinis (pl. 147, figs. c2), but not very clearly.

The appearance of onyx sardonyx throughout the Neomalayan area (Sumatra—
Malaya—Borneo), and its differentiation from syrinx maenala, have already been
discussed, and my illustrations on Pl. 1 are designed to illustrate the Malayan forms
of both in Singapore. The latter species also presents a fairly homogeneous, but
variable appearance in the area. Its distinctive characteristics are the dark pointed
fore wing of the ¢, and the greatly reduced white markings—more so than in omyx—of
both sexes.

The variability and lack of material combined to confuse the nomenclature. The
three names given to individual Bornean specimens respectively denote the two
distinctive extreme ¢ forms, and the normal one (maenala, with an obsolete white
patch, the unique corniculum with a large one, and the intermediate affinis).
Distant’s figure of halba, a 9 from Penang with the narrow even fore wing median
band, suggested synonymy with form affinis, and this was confirmed by the acci-
dental discovery of the holotype (Cowan, 1966).

The normal form is not far from moulmeina, but even.smaller (fore wing length
3 13-15, 213-16 mm.), and distinctly darker. The ¢ upperside is bright shining
blue with the apical half of the fore wing quite black and the white spot very small,
sullied, and black-veined. The white spot in the 9 is clear, but inwardly black-
edged.

The name affinis has on occasion been wrongly accredited to Staudinger.

INDO-ORIENTAL HORAGINI 121

The B.M. (N.H.) series are subdivided for clarity :
f. maenala 3 Holotype, Borneo. 1 3, Langkawi Is. (Stubbs).

f. affinis I 9, Langkawi Is. (Miller). 4 3, 5 9, Penang (Kerr, 9 Holotype of halba
Dist. ; Adams, Lakatt and Pambu). 23,19, Malaya. 2 4,2 9, Singapore (Cowan,
includes pair taken in cop.). 292, N.E.Sumatra ; Selesseh and Bekantschan (Martin).
(These do not pair off with the 3 3 sardonyx already discussed). 4 3, 1 2, Banka
(Hagen). 11 3,29, Borneo; Kina Balu and Brunei.

H. syrinx onychina (Staudinger)

Sithon onychina Staudinger, 1889: 113. Java.

Horaga holothura Swinhoe, 1894: 430. E. Java; Malang. syn. n.

. onyx holothura Swinhoe ; Fruhstorfer, 1912 : 232, “‘ probably W. Java”’.
. onyx onychina (Staudinger) ; Fruhstorfer, 1912 : 232. E. Java.

. moulmeina holothura Swinhoe ; Fruhstorfer, 1914: 35. W. Java.

. moulmeina onychina (Staudinger) ; Fruhstorfer, 1914: 35. E. Java.

. syvinx onychina (Staudinger) ; Corbet, 1941: 50. Java.

. syvinx holothura Swinhoe ; Corbet, 1941: 50. E. Java.

cumiseasemstiseass

Well illustrated by Seitz (pl. 157, figs. i2 3, 13, 14 2) ; also, but surprisingly poorly,
by Piepers & Snellen (1918 : 103, pl. 27, figs. 164a, b). The misspelling holothuria
has often occurred.

Fruhstorfer and others believed that different subspecies fly in East and in West
Java. Icannot find this proved for our group of Lycaenidae. There is some varia-
tion in H. syrinx here, but Kalis (1933 : 85-86) investigated the parallel but even
more pronounced case of Marmessus ravindra Horsfield in the island and found that
variation was seasonal rather than geographic ; that a large race flew throughout
the island, with smaller individuals in part of the year in the eastern portion.

In the several long series of the species in the B.M. (N.H.), including Fruhstorfer’s,
many with coloured labels for East, West and Mid-Java, no seasonal data are given,
and locality seems to have no influence on appearance.

High mountains and volcanoes are evenly distributed throughout the island, so
are unlikely to affect the issue.

Staudinger referred to the ‘
Swinhoe when describing his four specimens from Malang as “‘ expanse 1-4 inches
(equivalent to fore wing length over 18 mm.). Swinhoe’s type specimen, however,
is definitely small, with fore wing length only 15 mm.

I regard the Javan subspecies of H. syrinx as onychina Staudinger, with synonym
holothura Swinhoe, which name cannot be applied to West Javan specimens but
there may be a case for using it as a seasonal form name in East Java.

‘magnificent ’’ large form in his description, as did

a)

The normal form throughout the island is in the ¢ brilliant bright blue with the inner edge of
the very narrow apical border semicircular from base to tornus on the fore wing, and the white
discal patch large, diffuse and ovate. The 9 upperside is similar, but the blue areas are strongly
tinted with violet and the hind wing white band shows through from below. The median
bands on the underside are complete and very broad. Fore wing length is 15-17 mm.

122 C. F. COWAN

B.M. (N.H.) 68 3, 19 9, West, Central and East Java (including the ¢ holotype of
holothura from E. Java).

H. syrinx privigna Fruhstorfer stat. n.

Horaga privigna Fruhstorfer, 1897 : 7, 113. Lombok.
H. onyx privigna Fruhstorfer ; Fruhstorfer, 1912 : 232.
H. moulmeina privigna Fruhstorfer ; Fruhstorfer, 1914 : 35.

Nowhere illustrated. Described from a single 9.

This subspecies is very similar to Javan specimens; fore wing length 17 mm. The white
discal patch on the upperside is slightly smaller, compressed by the broader dark borders, which
are also wide on the hind wing. The fore wing underside white band was correctly described
as longer and narrower than in Java, reaching the costa, but Seitz (p. 982) transposed this to
short and broad.

Bali 29 are similar to privigna on the upperside but have the broad band below like Javan ones.

B.M. (N.H.) 2 Holotype, Lombok ; 3 9, Bali.

H. syrinx decolor (Staudinger) stat. n.

Sithon onyx var. decolory Staudinger, 1889: 112. Palawan.
H. onyx decolor (Staudinger, 1898 [sic]) Fruhstorfer, 1912 : 232.
H. moulmeina decolor (Staudinger) ; Fruhstorfer, 1914 : 35.

Nowhere illustrated. Described by comparing it with “‘ onyx Moore from Sikkim
and Amboina ”’.

The upperside blue is very much reduced, particularly in the g. The underside is deep olive
yellow, the hind wing with a series of black terminal spots above the tornal ones. Probably
quite close to paulla (see below). Fore wing length 17 mm. (9).

B.M. (N.H.) r 9, Palawan.

H. syrinx joloana Fruhstorfer stat. n.

Horaga onyx joloana Fruhstorfer, 1912 : 232. Jolo Is.
H. moulmeina joloana Fruhstorfer ; Fruhstorfer, 1914 : 35.

Nowhere illustrated. Probably very like the next subspecies, paulla.
Fruhstorfer’s brief description of this race is that it differs from decolor in being
more violet-blue above, with the fore wing white spot larger.

H. syrinx paulla Fruhstorfer stat. n.

Horaga onyx paullus Fruhstorfer, 1912 : 232. Bazilan Is.
H. moulmeina paullus Fruhstorfer ; Fruhstorfer, 1914 : 35.

Nowhere illustrated. The holotype was so small that a dissection was made to
prove its species. It is a good syrinx (Pl. 2, fig. 19). The fore wing length is 13 mm.

INDO-ORIENTAL HORAGINI 123
Translating Frustorfer’s original description :

“Smaller than decolor (Seitz wrongly says ‘ larger’), upperside darker blue
with very wide black apical border; a large rather quadrate clear white
discal patch ; below greenish yellow with a light brown apical wash and the
white median band broken in the centre, its hind portion suffused with
greenish.”

As this and the two preceding subspecies are so little known and perfunctorily
described, and since their territory is a nodal one between those of Borneo westwards
so far discussed, those of the Philippines, and those ranging south and east through
Celebes and Halmahera, a fuller description of this key specimen is given. The
possibility of its being a chance dwarf should be borne in mind.

The fore wing above is black, with a shining blue dark-dusted area confined to a semicircle
with diameter on the basal half of the dorsum, and barely entering the cell; the white discal
patch being clearly defined on the black ground, 24 mm. wide by 3} mm. along its straight inner
edge which lies obliquely from the origin of vein 5 to below vein 2. The hind wing is black with
some blue dusting in the cell and just beyond, a prominent white anteterminal line, black
terminal line, and white cilia. The tails are normal.

The rich ochreous brown underside has some postdiscal dark grey shading external to the
white fore wing band which is short and broad, measuring as on the upperside ; the hind wing
band is narrow and broken in the centre ; above the tornal black spot in space 2 are three well
defined submarginal black spots crowned with white lunules at the ends of spaces 3, 4, and 5;
the normal tornal metallic green scaling runs into space 3.

B.M. (N.H.) 3 Holotype, Basilan (Doherty, February/March 1808).

H. syrinx camiguina Semper stat. n.

Myrina ciniata Hewitson 2 var. Mindanao ; Hewitson, 1869 : Suppl. 6, pl. 3, fig. 84.
Hovraga camiguina Semper, 1890: 216. Camiguin Is.
H. ciniata camiguina Semper ; Fruhstorfer, 1912 : 233.

There are two Camiguin Is. (at least). One is just north of Luzon. This one is
just north of Mindanao.

Semper described this subspecies by comparing it with Hewitson’s 1869 figure of
the underside of his 2 variety from Mindanao of ciniata. His two 99 differed in
having a larger, unmarked, almost semicircular white spot on the fore wing upperside
(for which he must have used Hewitson’s 1863 fig. 3r on pl. 14), and a narrower,
more regular white median band on the hind wing underside.

The fore wing length is given as 15 mm. which, though compatible with the ¢
syrinx paulla, is small for other 9° in the region.

Till further material of both sexes is available this name must stand for Mindanao
specimens as well.

The name has on occasion been misspelt caminguina.

124 C. F. COWAN

H. syrinx permagna Fruhstorfer

Horaga ciniata permagna Fruhstorfer, 1912 : 233. Toli Toli, N.W. Celebes.
H. syrinx permagna Fruhstorfer ; Corbet, 1941 : 50.
(‘‘ Myrina ciniata g’’ Hewitson, 1869 : Suppl. 6) (no loc.).

Figured by Seitz: pl. 158, fig. a3 (2 nec ¢ ?).

Fruhstorfer described permagna by comparing it with “‘ ciniata Hew. ; type from
South Celebes ’’, than which it was larger and with wider white markings above and
below. He did not indicate its sex, but his description, and Seitz’ illustration in
particular, fit a 9.

There are two 9g in B.M. (N.H.), one ex colls. Boisduval and Rothschild labelled
“Celebes ’’, the other ex coll. Hewitson labelled ‘‘ Celeb.’’ and, in a later hand,
“Calabar’’. The latter stood as the type of cimiata Hew., an untenable position
(q.v. below) but one which had apparently misled Fruhstorfer and others. The
two are identical and, with a fore wing length of 18 mm., a good match for the 9
permagna, which name I apply to cover all Celebes until further material is obtained
from different parts of this vast and interesting complex of peninsulae.

Wallace (1869 : i, 280) noted that Celebes insect races, particularly in butterflies,
have exceptionally large and elongate pointed forewings. This feature is apparent
in these two 94.

There is no blue on the upperside of this subspecies, which is intense black in the J, less intense
in the 2, with a comparatively small but clear white fore wing patch ; the hind wing has some
pale shading at the cell-end and mid-costa. The underside is marked with terminal spotting
on the hindwing as in paulla.

B.M. (N.H.) 2 3, Celebes ; 2 9, Palos Bay, W. Celebes (Doherty).

H. syrinx samoena Grose Smith stat. n.
Horaga samoena Grose Smith, 1895: 513. Batchian (2 nec Q).

Grose Smith described under this name a very small 3 (fore wing length 14 mm.)
and three 99 (17-18 mm.). His two type specimens and the two syntypes (no actual
holotype having been designated) are in the B.M. (N.H.). The discrepancy in size is
significant ; the 92 are compatible with syvinx but the 3, which has no sexual
insignia, is not, and I bracket it with and discuss it later under cimzata for reasons
there given.

I select from his two types the 9 as the LECTOTYPE of samoena Grose Smith.
His description is accurate. The upperside is dark brown, unmarked except for
the oblique white patch which is ovate, sullied, and crossed by two dark veins, and
the dull whitish submarginal line on the hind wing. The underside is normal for
the species, with the inclusion of the diminishing series of subterminal black dots,
each with white lunule, at the ends of spaces 2 to 6 of the hind wing.

The fore wing length is 18 mm. In size and markings it closely resembles Seitz’
figures for permagna upperside (1926 : pl. 158, fig. a3) except that the white patch is
small and dusky ; and for cimiata underside (l.c., fig. a2) except for its greater size.

INDO-ORIENTAL HORAGINI 125

A fourth 2 taken by Waterstradt has rather wider and clear white markings but
is otherwise similar.

B.M. (N.H.) 2 Lectotype and r 9, Batchian (Doherty, March 1892). 1 9, Halma-
hera (Doherty, August 1892). 1 9, Batchian (Waterstradt).

H. syrinx syrinx (Felder)

Myrina syrinx C. Felder, 1860: 452. Amboina.

M. onyx Moore; syn. syrinx Felder ; Hewitson, 1865 : 40, line 3.
H. syrinx (Felder) Moore, 1883: 525. ‘‘ confined to Amboina ”’.
H. syrinx (Felder) ; Fruhstorfer, 1897: 115.

H. syrinx syrinx (Felder) ; Corbet, 1941 : 50.

These are the only references to this name, and it has not been illustrated.

6 upperside lustrous bright blue with a purple sheen ; the fore wing costa and outer third,
hind wing costa and inner margin all black ; the fore wing discal white patch large and clear ;
a hind wing subterminal row of black dots followed by the white and the black marginal lines
prominent before the white cilia ; traces of the white median band are visible on the hind wing
from below, and the lobe is marked with blue metallic scales.

Q upperside dark brown, the wing bases suffused with pale purple-blue which may reach the hind
wing termen ; the fore wings with a shortened discal band. Underside ochreous with a white
discal band on all wings; the hind wings with four terminal black spots inwardly edged with
shining blue.

The description of the 3 upperside is from a Ceram specimen in good condition.
That of the 2 and the underside is translated from Felder’s with my addition in
italics, and agrees with 2 specimens before me. He called the white patch on the
fore wing upperside ‘‘ abbreviata’’ because there were at the time no comparable
species known ; as a discal band in the normal sense it is short, but compared with
the allied species we now know the patch is long, from vein r to the upper apex of
the cell.

Felder’s type specimen, so marked ex coll. Rothschild, survives in very battered
condition with three wings parts of which are transparent, but traces of the blue
colour are left. The abdomen attached to her looked strange and, on dissection,
proved to be that of a totally unrelated J. This specimen has only recently been
found, but it proves that Corbet was correct in his diagnosis of the taxon, and corro-
borates his remark on the omission of any reference to the blue colour by Felder.
Further evidence that blue was present may be adduced from Hewitson’s rejection
of syrinx as being a synonym of onyx Moore. This was an unwarranted confusion
of localities which Moore tried to rectify jointly with the similar error in the case of
ciniata Hewitson, when he averred that syvinx Felder was confined to Amboina and

ciniata to Batchian. These two valid corrections unfortunately remained over-
looked.

B.M. (N.H.) 2 Holotype, Amboina. 1 9, Amboina (Doleschall & Martin, 1801 ;
ex colls. Van der Poll and Adams). 1 4,5 9, Central Ceram (C.F. & J. Pratt, 1919).

126 Cc. F. COWAN
H. syrinx schoutensis Joicey & Talbot stat. n.
(Pl. x, fig. 8)

Horaga schoutensis Joicey & Talbot, 1916: 79. Schouten Is.
H. onyx schoutensis Joicey & Talbot; Seitz, 1926 : 982.

Here illustrated for the first time. The Schouten Islands referred to are those off
north-western, not central, New Guinea.

This is an interesting extension of the species, apparently well established and
widespread. The size (fore wing length 16-18 mm.) and ¢ genitalia entirely conform
to the syrinx stamp.

The ¢ upperside is black, with the white patch semicircular and large ; the basal half of the
dorsum is pale shining blue, which colour reappears diffusely and to a variable extent in the hind
wing cell. The Q is duller; dark grey with more white and little blue.

The underside is ash-grey, slightly ochreous at the hind wing tornus, and the white median
bands broad and outwardly diffuse.

B.M. (N.H.) g Holotype, Biak Is. (A.C. & F. Pratt, June 1914, labelled ‘‘ schouten-
sae”’). 5 3, i 9, Roon Is. (C.F. & J. Pratt, July 1920). x 3, Ron Is. (Doherty,
July 1897). 1 3, Dorey (Doherty, April 1897). 2 3, Mefor Is. (Doherty, June 1897 ;
C.F. & J. Pratt, August 1920—the specimen illustrated). 1 9, Ambabaki (Laglaise).
I g, 1 9, Hydrographer Mts. (Eichhorn Bros., April 1918).

Note: these localities are, from west to east :

Ambabaki ; central on the north-facing coast of the N.W. peninsula of New Guinea
at Lat. 4° S., Long. 133° E.

Dorey ; Manokwari, just south of Cape Manori; 1° S., 134° E.

Mefor Is. ; west entrance to Great Geelvink Bay; 1° S., 135° E.

Se Is. }S.W, quarter of Great Geelvink Bay ; 24° S., 1344° E.

Biak Is. ; east entrance to Great Geelvink Bay ; 1° S., 136° E.

Hydrographer Mts. ; over 1,000 miles away in Papua; 9° S., 1484° E.

HOoRAGA ALBIMACULA
(Pl, x, figs. 17,22; PL 2, fig; 20)

Save that it has not yet been recorded from Sumatra and Nias, this little species
is known so far from all the territory of onyx, and from all that of syrinx except
south-east from Borneo, in which island it seems very rare. This is a new concept
of the range of the species ; it has only recently been recognized that it has a sub-
species in Ceylon and India, and it is now extended to Bali, Celebes, the Philippine
area and Formosa.

The wings appear comparatively narrower than those of onyx and syrinx, and
less rounded than in amethysta. On the hind wing underside the white band is
obsolescent, leaving usually only the dark median line which, as in amethysta, is
prominent and often heavily overlaid with metallic green scales. The upperside

INDO-ORIENTAL HORAGINI 127

is normally very dark brown, but some subspecies are blue or purple-blue and a few
are interesting in having occasional purple-blue forms. The white patch on the
fore wing upperside is always prominent.

It is possible but unlikely that albimacula might have to give way to cinzata as the
specific name for a senior subspecies in Batchian. This is discussed later.

The species is remarkable in having so many races described from unique speci-
mens. I add two more, in Bali and Celebes.

H. albimacula viola Moore

Horaga viola Moore, 1882 : 248. Dharmsala (Kangra, N. India).
H. albimacula viola Moore ; Woodhouse, 1952 : 138, pl. 20, fig. 22.

Illustrated by most authors on Indian species.

The sexes are alike, though the occasionally has traces of purple-blue basal scaling on the
upperside. This is very dark brown, and the usual white fore wing patch is sometimes suffused
with pale orange. The underside ground colour is dark ochreous brown.

The size is variable ; fore wing length 103-14 mm. (9 mm. in some very small 3).

There are three specimens labelled types in the B.M. (N.H.) Type Collection ;
(1) a brown ¢ labelled ‘“‘ Sikkim 1886, O. Moller.’’, ex coll. Elwes; (2) a brown 9
labelled ‘“‘ Horaga viola g Type, Moore” and “‘ Kangra”’ (this is bodyless but I
diagnose sex by size and wing profile) ; and (3) a blue 9 labelled as last but “‘ 9 Type ”’.
The first by its date and locality cannot be a Type; Moore’s description gave
‘“ Dharmsala ; type in B.M.”’, Dharmsala being in Kangra. The second and third
agree with the original description of the respective sexes, the former is a viola as
currently recognized and the latter an onyx, and both are 99.__I now select the second
of the three, the specimen labelled and described by Moore as Horaga viola 3, to be
the LECTOTYPE 2 of Horaga viola Moore (1882: 248). It agrees with my descrip-
tion above, having no trace of blue colour on the upperside. The current nomen-
clature is not disturbed.

B.M. (N.H.) 2 Lectotype, Kangra; 29 3, 9 9, S. India, Sikkim, Assam, Burma ;
(also known from Ceylon).

H. albimacula albimacula (Wood-Mason & de Nicéville)

Sithon albimacula Wood-Mason & de Nicéville, 1881 : 249. Andamans.
Horaga albimacula (Wood-Mason & de Nicéville) ; de Nicéville, 1889 : 284, pl. 14, fig. 9.
H. albimacula albimacula (Wood-Mason & de Nicéville) ; Corbet, 1941 : 48.

The illustration quoted is good, and well shows how this very dark and richly
marked subspecies faithfully repeats the features of its compatriot onyx rana, figured
with it.

The upperside is black with a large white patch, and the centre and disc of the hind wing are
deep shining violet, leaving a regular terminal border. The underside ground colour is rich
deep brown.

The fore wing length is 1o-12 mm.

B.M. (N.H.) 9 g, x 2, Andaman Is.

128 C. F. COWAN

H. albimacula malaya Corbet
Horaga albimacula malaya Corbet, 1941 : 48. Singapore.

Nowhere illustrated, this subspecies is very similar to viola but in both sexes all
wings are often heavily sprinkled in the basal areas with purple-blue scales, and the
white fore wing patch is slightly narrower. The fore wing length is 11-12 mm.

Described from Singapore, it flies there with onyx sardonyx and syrinx maenala,
and is also known from peninsular Malaya.

B.M. (N.H.) $ Holotype, 2 Allotype, Singapore, and 2 3g, Singapore (Cowan),
one with and one without blue scaling.

H. albimacula anara Fruhstorfer stat. n:

Horaga anarva Fruhstorfer, 1898: 180. East Java.
H. anytus anara Fruhstorfer ; Seitz, 1926 : 982, 1116, pl. 157, fig. ig.

Seitz’ figure is good. Also illustrated (as anytus) by Piepers & Snellen (1918 : 103,
pl27; fig 165).

This race is very like viola but paler below, and the fore wing white areas on both
surfaces are wider. Small; fore wing length 1m mm.

B.M. (N.H.) g Holotype, Lawang, East Java.

H. albimacula violetta subsp. n.
PPL hit tte. Pio. ea)

This is a surprising and very distinct specimen from Bali. It is large for the
species ; fore wing length 14 mm.

The ¢ upperside is pale violet-blue of a powdery appearance ; the dark brown costal border
almost fills the fore wing cell, continuing round the apex to form terminal borders 2 mm. wide
on both wings ; the white fore wing patch is large and semicircular. The underside is exactly
as in viola except for the much wider fore wing band.

B.M. (N.H.) 3 Holotype, Bali (low country, Doherty, April 1896).

H. albimacula bellula Fruhstorfer stat. n.
Horaga bellula Fruhstorfer, 1897: 114. Sambawa.
H. anytus bellula Fruhstorfer ; Fruhstorfer, 1912 : 233.
Figured by Seitz: pl. 158, fig. a8 (underside).
This is a reversion to the black and white viola pattern, very like anara from Java
but darker below and with less extensive white areas. Fore wing length 12 mm.

H. albimacula chalcedonysx Fruhstorfer

Hovraga chalcedonyx Fruhstorfer, 1914 : 33. ‘‘ Sintang, S.W. Borneo ’’.

Hovaga onyxitis Fruhstorfer (1914: 34). ‘‘ Sintang, W. Borneo’’. stat. & syn. n.
. moulmeina chalcedonyx Fruhstorfer ; Fruhstorfer, 1914 : 35.

. onyx chalcedonyx Fruhstorfer ; Seitz, 1927 : 1116.

. anytus onyxitis Fruhstorfer ; Seitz, 1927 : 1116, pl. 157, fig. i8.

. albimacula (?) chalcedonyx Fruhstorfer ; Corbet, 1941 : 49.

samsemseass

INDO-ORIENTAL HORAGINI 129

The figure by Seitz is of the male upperside.

Fruhstorfer described chalcedonyx, immediately followed by onyxitis, both from
“one male in coll. Fruhstorfer’”’ and from the same locality. The type specimen
of chalcedonyx, ex coll. Fruhstorfer and with his labels on, agreeing with his des-
cription, is in the B.M. (N.H.) andisaQ. It is very slightly larger and with a little
more blue than, but otherwise identical with, the 9 allotype of malaya from Singapore,
and it is clear that chalcedonyx is the West Borneo coastal subspecies of albimacula.
Seitz’ figure and Fruhstorfer’s description of onyxitis can only refer to the 3 of this
subspecies, showing it to be intermediate between malaya and violetta, having a
black and white fore wing upperside and a “ dark violet-blue hind wing with a
moderately broad black border which tapers somewhat tornally ”’ (translated from
Fruhstorfer).

Fruhstorfer himself surmised that this would prove a race of albimacula (= anytus).
It differs remarkably from the next, from montane North Borneo, which has no blue.

B.M. (N.H.) 2 Holotype ; Sintang, West Borneo.

H. albimacula albistigmata Moulton

Horaga albistigmata Moulton, 1912: 159. Sarawak.
H. albimacula albistigmata Moulton ; Corbet, 1941 : 49.

Nowhere illustrated, this subspecies is in all respects like a small dark viola except
that the fore wing white spot is smaller on both surfaces. The fore wing length is
104 mm.

B.M. (N.H.) g Holotype; Madihit Hills, East Sarawak (May IgI1I).

H. albimacula anytus (Staudinger) stat. n.

Sithon anytus Staudinger, 1889 : 113, pl. 1, fig. 12. Palawan.

Yet another unique specimen named ; this time figured by the author by photo-
graph, leaving no doubt as to its status.

The upperside is black and white as in viola, anara and albistigmata, with the white
patch small as in the last. It differs from all these, and follows the generic trend in
the area, by bearing a terminal series of black spots on the hind wing underside.

The fore wing length is 1m mm.

H. albimacula taweya subsp. n.
(PE he Tt)

Another large new subspecies. The unique ¢ lay in the series, very similar on the
upperside, of H. selina, and was caught with it at Taweya by Doherty on his last
visit to Celebes.

The ¢ upperside is brownish black, unmarked except for the fore wing white patch and the
indistinct whitish submarginal line on the hind wing; the white patch is large, the outer edge
semicircular and diffuse, the inner upright not oblique, running straight from the upper apex
of the cell to just beyond the centre of vein 1.

130 Cc. F. COWAN

The underside is ochreous brown, only slightly darker than in violetia (my illustration is too
dark), with the fore wing white band as on the upperside; the band on the hind wing is
obsolescent as usual and the inner dark brown median line is edged outwardly with metallic
green scales along its posterior half; similar green scales crown the series of black spots which,
as in anytus and the Celebes race syvinx permagna, decrease along the termen from the large
one in space 2 to the indistinct one in space 6.

The fore wing length is 14 mm.

B.M. (N.H.) ¢ Holotype, “‘ Taweya, north of Palos Bay ’”’, Celebes west coast
(Doherty, August-September 1896).

H. albimacula bilineata Semper stat. n.
Horaga bilineata Semper, 1890: 216. ‘‘ Panaon, S.E. Mindanao ”’.

Nowhere figured. Described, like his camiguina, from two 929, which are distinctly
smaller ; fore wing length 13-14 mm.

The upperside is dark brownish grey with a comparatively large white patch. The underside
is ‘‘ ash-grey ”’, slightly darker at the fore wing apex ; the median band wide on the fore wing
and nearly obsolete on the hind, where it is edged darker on both sides and marked with metallic
green scales throughout its length ; similar scales crowning the full series of terminal spots.

I cannot find “ Panaon, S.E. Mindanao’”’ on the map. There is an island of that
name off N.E. Mindanao, between it and Leyte, which Semper mentions in his
appendix under “‘ Fiinfter Bezirk ’’ (p. 363), Mindanao itself being in his ‘‘ Sechster
Bezirk” *,

H. albimacula triumphalis Murayama & Sibatani stat. n.
Horaga anytus triumphalis Murayama & Sibatani, 1943: 40, pls. 5 & 6, figs. 8. Formosa,

Well figured by Shirézu (1960 : 313, fig. 346; pl. 67, figs. 719-722).

A pleasing closure to the range of this species. Shirézu’s illustrations show this
subspecies to have the ¢ genitalia characteristic of the species, and an appearance
near to the bluish forms from Malaya. The white bands on the underside are
slightly better marked on the hind wing but narrow on the fore wing.

Fore wing length 12-13 mm.

HoRAGA AMETHYSTA
(Pi. x, fig. 13; Pl. 3; fig. 22)

Very little is known of this extremely rare and elusive species with the monstrous
and specialized genitalia.

First found in Borneo, several specimens are known from different sources there,
but it was not till 1941 that Corbet gathered various examples from Malaya and
recognized that it also occurred in Java and Nias. Western forms are small but it
increases in size eastwards. It has been taken on mountain peaks of over 5,000 feet,
through all elevations down to practically sea level, and is probably, like other
species of the genus, an “ open hilltop’ or “ jungle canopy ”’ insect.

INDO-ORIENTAL HORAGINI 131

Seitz does not illustrate, nor even mention, amethysta. I figure the smallest sub-
species, 7sna, from Nias. The only other illustrations have been Druce’s originals
of 1902.

The species is distinctive on the upperside in the small size of the fore wing white spot, which
is at most as deep as the width of the cell, and on the underside in the narrow linearity of the
median bands and the prominence of the metallic green markings. The upperside violet colour
is also of a different, uniformly shining, quality to the dusted bluish of other species.

The wings are rounded, and fore wing veins 11 and 12 tend to be more strongly bowed towards
each other than usual ; in one 9 specimen ex coll. Cator from Sandakan (“ S’kan ’’) they palpably
touch.

The possibility that the name cimiata may have to be adopted as the senior specific
name is discussed after subspecies amethysta. I insert it as a floating subspecies
below, awaiting positive evidence from Batchian for its correct allocation in the
classification.

H. amethysta purpurescens Corbet
Horaga amethystus purpurescens Corbet, 1941: 47. Malaya, S. Burma.

This name has on occasion been misspelt purpurascens.
The holotype and allotype specimens which were in the F.M.S. Museum, Malaya,

until the 1939-45 war, must be presumed lost, but there survive in B.M. (N.H.)
three females of Corbet’s original type-series.

The upperside is deep lustrous violet with dark brown borders up to 2 mm. wide in the 4,
whose fore wing white spot is obsolescent ; in the 9 the borders are broader and diffuse, expand-
ing at the fore wing apex to reach the white spot, which is narrow and placed across the cell-end.

The rich ochreous underside fades to ash-grey in worn specimens, which also may lose the
metallic green scales on the hind wing band.

This subspecies is similar to the Bornean one, but the borders tend to be narrower and the
white fore wing spot slightly larger.

Burmese specimens are smaller, have still narrower borders and a larger spot, but these
differences are slight.
The fore wing length is 12-14 mm.

B.M. (N.H.) 5 9, Mergui (Evans), Renong (2) (Doherty), Malaya (2) (Corbet,
Gunnery). Other Malayan specimens in colls. Eliot and Stubbs were taken in the
Cameron Highlands, Frasers Hill, on Mount Ophir summit, in the plains, and on
Pulau Parit, Kerimun Is. (west of Singapore).

H. amethysta isna Corbet
(PL ao fie 134 OP 341022)
Horaga amethystus isna Corbet, 1941 : 48. Nias Is.

An interesting specimen in good condition is small and pale, recalling its compatriot onyx
zuniga. Fore wing length 114 mm.

132 C. F. COWAN.

The upperside of the unique ¢ is lavender rather than violet, with a prominent round white
spot at the end of the fore wing cell, below which the underside band shows through ; the borders
are regular, narrow, and obsolete at the costa. The white band on the fore wing underside is
relatively wide (1 mm.).

B.M. (N.H.) 3 Holotype, Nias Is.

H. amethysta overdijkinki Corbet
Hovraga amethystus overdijkinki Corbet, 1941: 48. W. Java.

The unique @ is of a much bluer tint than in all other specimens of this species, but still uniform
over all the upperside. The brown borders are narrower and more regular than in all other 99.
The fore wing white spot is larger and elongate, just extending into spaces 3 and 5. The fore
wing length is 124 mm.

B.M. (N.H.) 2 Holotype, Mt. Halimon, Soekaboemi, W. Java.

H. amethysta amethysta H. H. Druce
Horaga amethystus H. H. Druce, 1902: 118, pl. 11, figs. 4, 5. N. Borneo.

The 3 upperside is deep shining violet with regular narrow brown borders and a minute white
fore wing spot. The 9, with a larger diffuse spot, has broader borders than any other subspecies
so far. The underside fore wing median band is entire but linear in the g, straight and $ mm.
wide in the 9; the hind wing band as usual is linear and completely overlaid with shining green
scales.

Fore wing lengths are 12 mm. (g), 13-14 mm. (9).

B.M. (N.H.) g§ Holotype, British N. Borneo. (The female allotype from “ N.
Borneo, probably East Coast Residency ”’ (Pryer) is in coll. Hope Dept., Oxford.)
2 9, N.E. Borneo, Sandakan.

H. ciniata (? amethysta) ciniata (Hewitson)

Myrina ciniata Hewitson, 1863 : 35, pl. 14, figs. 30, 31. ‘‘ Batchian and India ’’.
M. ciniata Hewitson ; Hewitson, 1869: supplement 6. 4, no loc.

Horaga ciniata (Hewitson) Moore, 1881: 99. ‘ Ceylon ’’.

H. ciniata (Hewitson) ; Moore, 1883: 525. ‘‘ confined to Batchian ”’.

H. ciniata (Hewitson) ; de Nicéville, 1890: 417. Batchian.

Hovraga samoena Grose Smith, 1895 : 513 (g nec 9). Batchian.

. ciniata (Hewitson) ; Fruhstorfer, 1897: 114-5. N. & S. Celebes.

. ciniata (Hewitson) ; Swinhoe, 1911: 12. Celebes.

. ciniata (Hewitson) ; Fruhstorfer, 1912 : 233. ‘‘ Type aus Sud Celebes ”’.
. ciniata (Hewitson) ; Seitz, 1926: 982, pl. 158, figs. ar, a2. S. Celebes.

. syvinx ciniata (Hewitson) ; Corbet, 1941: 50. S. Celebes.

Suseise meee

There has been unfortunate confusion and uncertainty over the identity and
application of the name ciniata which even now cannot be fully resolved for lack of
material.

Hewitson described specimens varying in size from 0-9 to 1-3 inches’ expanse
(equivalent by his method to fore wing lengths 114 to 164 mm.) from Batchian and

INDO-ORIENTAL HORAGINI 133

India, figuring the upper and undersides of a clearly identifiable 9, though he did
not state its sex. This 9 I have located in the B.M. (N.H.) main collection, with
Hewitson’s labels reading “‘ Batchian ”’ and, glued underneath, two scraps “ ciniata ”’
and ““Ba...’. It agrees perfectly with the 1863 description and illustrations, with
a fore wing length 15 mm.

In 1869 Hewitson said that his original figure “ was from a female. I have since
received the male, which does not differ from it except in its greater size and in the
more acute apex of the anterior wing ’’. He increased the size range of the species
now to 0:g-1'4 inches (fore wing length 113-18 mm.).

Standing in the B.M. (N.H.) Type Collection as the type specimen of ciniata Hew.
is a g, with Hewitson labels reading ‘‘ Calabar ’’ and, glued below, a scrap reading
“Celeb.” This 3 has a fore wing length 18 mm., and the fore wings are very pointed.
I am confident that this is the ¢ to which Hewitson referred. It is clear that it is
not conspecific with the much smaller 92 which Hewitson originally had named
ciniata. Hewitson gave no more information on it than my quotation above, and
no locality, though it undoubtedly came from Celebes.

Moore had obviously been studying this problem when, in 1883, in correcting his
own earlier error referring to Ceylon, he said firmly that cimzata was confined to
Batchian (and syvinx to Amboina). He was followed by de Nicéville, but by 1897
the false g seems to have appeared and misled all later authors.

No other reference to this subject can be found, nor any explanation for the quite
unwarrantable appearance of this J in the status of a type specimen, and only a
ruling by the International Commission on Zoological Nomenclature could now
legalize such status. I do not consider a request for such a ruling justifiable. The
original 1863 description of the small 2, though confusing localities, was good, and
so was the illustration of the 9 from Batchian, which has been reproduced by Seitz,
and the J, from a different locality, has never been described nor illustrated. I have
therefore placed and discussed the ¢ already, with a second similar one from Celebes,
where they belong under syvinx permagna.

I designate the 2 I have referred to, ex coll. Hewitson from Batchian, to be the
LECTOTYPE of Myrina ciniata Hewitson (1863). Hewitson’s description and
illustrations, the latter reproduced by Seitz, are good except that the size should
be given as fore wing length 15 mm. (instead of expanse 0-9 to 1-3 inches). The
specimen is in the B.M. (N.H.) Type Collection.

Under syrinx samoena I have already discussed Grose Smith’s specimens of that
name and segregated his unique small J with the violet-centred hind wing as being
different. It has no abdomen, palpi nor fore legs, but its wings do look masculine.
Nor has it sexual insignia, so assuming it is a ¢ it must be either a subspecies of
albimacula, one of amethysta, or a new species. The last is unlikely ; the size fits
either of the alternatives, and both the deep violet colour and the reduced fore wing
white markings indicate amethysta. Its 14 mm. fore wing matches well the 15 mm.
of Hewitson’s 9 ciniata, and they may well be conspecific.

More cannot be decided until fresh material of both sexes of all species is available
from Batchian. If they are conspecific, the name ciniata will be the senior taxon in

134 C. F. COWAN

the species I have dealt with as amethysta (or possibly albimacula). If they are
distinct, the 2 will still bear the senior name ciniata to the appropriate species, but
the $ will represent a new species or subspecies, as I have already fixed the name
samoena to apply to the Batchian race of syrinx.

It must be remembered that neither albimacula nor amethysta is yet known from
the Batchian side of Wallace’s Line although both approachit. The effect of crossing
it is unpredictable and often spectacular, though in the known case of syrinx it is
only moderate.

B.M. (N.H.) 2 Lectotype, Batchian (ex coll. Hewitson). 1 g (?), (the “ g Type’”’
of samoena Grose Smith).

Horaga selina Grose Smith
(Pl. 3; fies £1. 3 fig) 25)
Horaga selina Grose Smith, 1895 : 513. ‘‘ South Celebes ’’.

An interesting species not listed by Seitz. Nearly half the 20 specimens in B.M.
(N.H.), most ex coll. Rothschild, were collected by Doherty in 1891, the rest in
1896; all found by the same man in the one restricted locality. The five year
interval allows one to hope the species survives there still.

The sexes are alike and the figure on Pl. 1 shows their appearance well. The upperside is very
dark brown and white ; the fore wing patch long, tapered and oblique, often coated with orange
scales. The underside white band is narrow and regular, and followed on the hind wing by a
straight ochreous line and an irregular metallic green one before the distinctive wide postdiscal
black and grey irrorated area. The fore wing length is 15-18 mm.

In effect the submarginal markings of the hind wing underside are shifted in to the discal area,
and the outer third is “‘ filled in ’’ with the irrorations. It is interesting that this phenomenon
is paralleled in the quite unrelated but also 3-tailed Semanga superba (Druce), a more widespread
species with the fore wing and the basal half of the hind wing plain brown but whose “ terminal
red spots and metallic lunules ’’ are postmedian, in a marbled grey area extending to the termen.

B.M. (N.H.) 3 Holotype, 2 Allotype, 6 9, Celebes, west coast ; Taweya, north of
Palos Bay (Doherty, August-September 1891) ; 5 3, 7 2, the same (August-Septem-
ber 1896).

HORAGA LEFEBVREI

(Pls, hes 3§ >. PLS; hig 2a)

This aberrant Philippine species is the only one with marked sexual dimorphism
on the underside, a fact which has led to some synonymy. Veins 1r and 12 of the
fore wing are not so close as in other species, which caused one of the synonyms to
be made in Rathinda.

The upperside is like that of selina but the white patch is ovate and more central
on the fore wing, while the sub-basal area is more or less dusted with pale blue scales,
particularly in the ¢.

The underside of the 3 is shown on PI. r (a specimen from Mindoro where the hind
wing ground colour is uniformly ochreous). The ground colour normally is dark

INDO-ORIENTAL HORAGINI 135

brown shading to chestnut at the fore wing apex and to ochreous in the dorsal half
of the hind wing; the white costal streaks, central spot, and chain of postdiscal
markings (which dorsally become overlaid with metallic scales) on the hind wing
are all heavily ringed with dark brown ; and there is a complete submarginal series
of metallic green lunules capping black terminal spots or dots. The subtornal spots
on either side of the end of hind wing vein 2 are well marked and both black.

The @ differs on the underside by having much more extensive white markings
on both wings ; the hind wing central spot becomes subtriangular, one apex extending
to the base, and the postdiscal chain is more or less conjoined.

Either sex (B.M. (N.H.) has one such of each from Luzon and a male from Min-
danao) may have the whole under surface shot with a purple gloss, and the metallic
markings blue instead of green.

The ¢ has a distinctive plume of suberect soft brown and grey hairs along the upper
edge of the hind wing cell on the upperside, and its palpi are clothed with two forms
of translucent scales giving them a greasy cream appearance below the black tip.

H. lefebvrei lefebvrei (C. & R. Felder)

Myrina lefebvrei C. & R. Felder, 1862: 291. Luzon.
Horaga lefebvrei (Felder) Semper, 1890 : 215.
Rathinda cuzneri Schultze, 1907 : 361, pl. 1, fig. 1. Luzon. comb., jf. & stat. n.

The name has on occasion been misspelt lefevrez.

Figured also by Seitz: pl. 158, figs. a4 (j nec 9), a5

The appearance has been described in the paragraphs above. The fore wing
lengths are 14-17 mm. (9), 14-18 mm. (9).

The Felders’ original good descriptions were of a normal g with minimal white
markings on the underside, and a normal 2 with these markings large. The type
specimens agree.

Schultze described fully, and figured by photograph, a normal 2 with maximum
white on the underside, saying that the § was similar. We may avoid absolute
synonymy by designating the aberrant g with enlarged white markings as 3 form
cuznert (Schultze) f. & stat. n.

B.M. (N.H.) ¢ holotype, ? Allotype, 3 3, 4 9, Luzon.

H. lefebvrei osma Fruhstorfer

Horaga lefebvrei osma Fruhstorfer, 1912 : 233. Mindanao.
Q f. melevra Seitz, 1926 : 983, pl. 158, fig. a7. syn. n.

Figured by Seitz; pl. 158, figs. a6 (as asma), a7 (as melera).

Fruhstorfer tersely described this race as differing from lefebvrez in having the fore
wing white spot smaller and the terminal borders darker brown on both wings below.
These differences are very slight, but comparing series from Luzon and Mindanao
there is a trend in the latter for all underside white markings to be smaller. The
size is as lefebvrei, but a dwarf male in B.M. (N.H.) has a fore wing length of only
II mm.

ENTOM. 18, 4 It

136 Cc. F. COWAN

‘

Seitz described 2° form melera as “ with very enlarged white spots and a much
reduced brick-red area on the hind wing underside’’. Again, as his illustration
shows, this is the normal 9, and melera can only be regarded as a subjective synonym
of osma unless, as in Luzon, a Mindanao 3 form with enlarged white markings is
found. Fruhstorfer’s type specimens are a small and poorly marked 4, and a
normal 9 exactly like Seitz’ illustration of melera.

B.M. (N.H.) 3 Holotype, 2 Allotype, Mindanao; 12 g, 12 2, Mindanao (mostly
Lanao Plain, June 1914, Wileman).

H. lefebvrei osmana subsp. n.
(Pl. 1, fig. 15; Pl. 3, fig. 24)

Both sexes in Mindoro show differences from the Luzon and Mindanao races which,
though slight, are constant and exceed those between the latter.

The fore wing length is 13}-154 mm., slightly smaller than Jefebvvei and osma.

The fore wing upperside white area is reduced from an ovate to a narrow, almost rectangular,
oblique band 2-2} mm. wide in the g, 3 mm. in the Q, and blue scaling at the base is minimal.

On the underside all white markings are much reduced in both sexes, and the hind wing ground
colour is uniform rufous ochreous instead of shading to dark brown in the apical half; the dark
brown edging to the white markings is prominent and catenulate.

B.M. (N.H.) 3 Holotype, 2 Allotype, Mindoro (Platen). 4 3, 2 2, Mindoro (Platen;
Everett, December 1894 ; Staudinger).

Horaga rarasana Sonan
(Ploy, eto Ph ae 2x)
Horaga varasana Sonan, 1936: 207, pl. 14, fig. 4. Formosa.

This species has also been well illustrated by Shirdzu (1960 : 311, fig. 345; pl. 67,
figs. 711-714). It is most distinctive.

The illustration on Pl. 1 shows the pure black and clouded white fore wing and the black and
shining violet hind wing upperside ; and the underside which is chalk white with dark brown
cell-end and postdiscal bands, beyond which the terminal borders are dark ochreous on the fore
wing and bright ochreous on the hind, the latter carrying the bold metallic green lunulate
markings.

The fore wing length is 17 mm. (3), 20 mm. (9).

The fore wing costal veins 11 and 12 are only slightly bowed towards each other in this species.

B.M. (N.H.) 2 3, 1 9, Formosa; Mt. Rara, July 1964 (Murayama).

SYNONYMIC LIST OF THE HORAGINI
RATHINDA Moore, 1881
syn. CUPIDO Hibner, 1819 [praeocc. |
R. amor (Fabricius, 1775) Ceylon, India, E. Pakistan.
syn. triopas (Cramer, 1780)

INDO-ORIENTAL HORAGINI 137

HORAGA Moore, 1881
H. onyx cingalensis Moore, 1883
onyx (Moore, 1857)

f. arta Fruhstorfer, 1914
vana de Nicéville, 1889
zuniga Fruhstorfer, 1912
sardonyx Fruhstorfer, 1914
Sruhstorferi Corbet, 1941
akronyx subsp. nov.
moltrechti Matsumura, 191g
syn. asakurat Nire, 1920
H. syrinx stkkima Moore, 1883
moulmeina Moore, 1883
artontes Fruhstorfer, 1912
maenala (Hewitson, 1869)
synn. & ff. :
halba Distant, 1886
affinis H. H. Druce, 1895
corniculum H. H. Druce, 1895
onychina (Staudinger, 1889)
syn. holothura Swinhoe, 1894
privigna Fruhstorfer, 1897
decolor (Staudinger, 1889)
joloana Fruhstorfer, 1912
paulla Fruhstorfer, 1912
camiguina Semper, 1890
permagna Fruhstorfer, 1912
samoena Grose Smith, 1895
syrinx (C. Felder, 1860)
schoutensis Joicey & Talbot, 1916
H. albimacula viola Moore, 1882
albimacula (Wood-Mason &
de Nicéville, 1881)
malaya Corbet, 1941
anara Fruhstorfer, 1898
violetta subsp. nov.
bellula Fruhstorfer, 1897
chalcedonyx Fruhstorfer, 1914
syn. onyxitis Fruhstorfer, 1914
albistigmata Moulton, 1912
anytus (Staudinger, 1889)
taweya subsp. nov.
bilineata Semper, 1890

Ceylon, S. India.

N. India, E. Pakistan, Burma,
Thailand.

(dry season form).

Andaman Is.

Nias Is.

Sumatra, Malaya, Banka, Sarawak.

Java.

Lombok, Sambawa.

Formosa, Hong Kong.

N. India, E. Pakistan, N. Burma.

S. Burma, S. Thailand.

Nias Is.

Sumatra, Malaya, Banka,
Sarawak, Kalimantan, Sabah.

Java.

Bali, Lombok.

Palawan.

Jolo Is.

Bazilan Is.

Camiguin Is., Mindanao.
Sulawesi.

Batchian Is.

Amboina, Ceram.

N.W. Irian—Papua.
Ceylon, India, E. Pakistan, Burma.
Andaman Is.

Malaya.

Java.

Bali.

Sambawa.

Kalimantan (west coast).

Sarawak.

Palawan.

Sulawesi (west coast).
Mindanao,

138 C. F. COWAN

triumphalis Murayama & Formosa.
Sibatani, 1943
H. amethysta purpurescens Corbet, 1941 S. Burma, Malaya.
isna Corbet, 1941 Nias Is.
overdiujkinki Corbet, 1941 Java.
amethysta H. H. Druce, 1902 Sarawak, Sabah.
H. ciniata (? syn. amethysta) ciniata (Hewitson, Batchian Is.
1863)
H. selina Grose Smith, 1895 West Sulawesi.
H. lefebvrei osma Fruhstorfer, 1912 Mindanao.
syn. melera Seitz, 1926
osmana subsp. nov. Mindoro.
lefebvrer (C. & R. Felder, 1862) Luzon.
3 f. cuznert (Schultze, 1907)
H. ravasana Sonan, 1936 Formosa.
REFERENCES

AITKEN, E. H. 1894. The larva and pupa of Spalgis. J. Bombay nat. Hist. Soc. 8 : 485-487,
2 pls.

BELL, T. R. 1919. The common Butterflies of the Plains of India (23). J. Bombay nat. Hist.
Soc. 26 (3) : 750-769.

CorBET, A.S. 1940. A revision of the Malayan species of Chliaria Moore. Proc. R. ent. Soc.
Lond. (B) 9 : 90-92, figs. 1-6.

1941. A Revision of the Malayan species of Hovaga Moore. Proc. R. ent. Soc. Lond. (B)

10 : 46-50, figs. 1-6.

1956. (in CorBET, A. S. & PENDLEBURY, H. M.) The Butterflies of the Malay Peninsula

Edn. 2. 11 + 537 pp., 159 figs., 55 pls. Edinburgh.

Cowan, C. F. 1955. The status of the Genus Saletava and its Species. Bull. Raffles Mus.
25 : 171-184, I map.

1966. The discovery of some Type Specimens of W. L. Distant’s. Proc. R. ent. Soc. Lond.

(B) 35: in press.

CRAMER, P. 1780. De uitlandsche Kapellen ; Papillons exotiques (4). 252 pp., pls. 289-400.
Amsterdam.

Davipson, J., BELL, T. R., & AirkEN, E.H. 1896. The Butterflies of North Kanara District,
Bombay (2). J. Bombay nat. Hist. Soc. 10 (3) : 372-393, 2 pls.

Distant, W. L. 1882-1886. Rhopalocera Malayana. 16 + 481 pp., 44 pls. London.

DouHERTY, W. 1891. The Butterflies of Sumba and Sambawa. J. Asiat. Soc. Bengal 60
(2) : 141-197.

Druce, H. H. 1895. A Monograph of the Bornean Lycaenidae. Proc. zool. Soc. Lond.

1895 : 556-627, pls. 31-34.

1902. On new and little-known Butterflies of the Family Lycaenidae. Proc. zool. Soc.
Lond. 1902 (2) : 112-121, pls. 11, 12.

Evans, W. H. 1925. The identification of Indian Butterflies (8). J. Bombay nat. Hist. Soc.
30 (4) : 756-776, I pl.

1927. The Identification of Indian Butterflies. 13 + 302 pp., 32 pls. Madras.

1932. The Identification of Indian Butterflies. Edn. 2. 10+ 454 pp., 32 pls. Madras.

1933. Some little-known or apparently unrecorded Lycaenidae and Hesperiidae from
the Malay Peninsula. J. fed. Malay St. Mus. 17: 406-417.

Fapricius, J. C. 1775. Systema Entomologiae. 32 + 832 pp. Flensburgi & Lipsiae,

INDO-ORIENTAL HORAGINI 139

FELDER, C. 1860. Lepidopterorum Amboinensium a Dre. L. Doleschall collectorum Species
novae. Sber. Akad. Wiss. Wien. 40 : 448-462.

— & FELDER, R. 1862. Lepidoptera nova a Dre. C. Semper in insulis Philippinis collecta
(2). Wien. ent. Mon. 6: 282-294.

FRUHSTORFER, H. 1897. (a) Aufzahlung der von mir auf der Insel Lombok im Jahre 1896

gefangenen Rhopaloceren. Berl. ent. Z. 42:1-14. (b) Neue Rhopaloceren aus dem

Malayischen Archipel. idem : 113-118.

1898. Neue Lepidopteren aus Asien. idem 43 : 175-199.

1912. Uebersicht der Lycaeniden. idem 56: 197-272, figs. 1-4.

1914. Neue Lycaenidae. Dt. ent. Z. Iris 28 : 33-35.

GopFrREy, E. J. 1930. A revised List of the Butterflies of Siam, with notes on Distribution.
J. Siam Soc. (nat. Hist. Suppl.) 7 (4) : 203-397, I map.

GRosE SMITH, H. 1895. Descriptions of new Species of Butterflies captured by Mr. Doherty
in the Eastern Archipelago (2). Novit. zool. 2 : 505-514.

HEMMING, F. 1935. On the Dates of publication of Hewitson, W. C., ‘ Illustr. diurn. Lep.

Lycaenidae’. Ann. Mag. nat. Hist. (10) 15: 117-120.

1960. Principles of Selection of Type-Species; Selection of Type-Species for seven

Genera of the Family Lycaenidae. Annot. Lep. 1: 7-11.

HeEwitson, W. C. 1863-78. Illustrations of diurnal Lepidoptera, Lycaenidae. 6 + 228 + 47
pp.,92 + 8+ 7 pls. London.

HUsner, J. 1819. Verzeichniss bekannter Schmettlinge (1816-1827). 431 + 72 pp. Augs-
burg.

Joicey, J. J. & TaLsBot, G. 1916. New Lepidoptera from the Schouten Islands. Tvans. ent.
Soc. Lond. 1916 : 65-83, 4 pls.

Kaus, J. P. A. 1933. Bijdrage tot de Kennis van de Lepidoptera van Nederlandsch-Indie.
Tijdschr. Ent. 76 : 47-86.

MackINNoN, P. W. & DE NICEVILLE, [C.] L. [A.] 1898. The Butterflies of Mussoorie in the
western Himalayas. (2). J. Bombay nat. Hist. Soc. 11 (3) : 368-389.

MatsumurRA,S. 1919. Thousand Japanese Insects, Addit.3. 4 + pp. 475-742, + 34 +34 2;
pls. 26-53. Tokyo.

Moore, F. 1857. Catalogue of the lepidopterous Insects in the Museum of the Honourable East

India Company. 5 + 278+ 4+ 11 pp., 12 + 6 pls. London.

1877. The lepidopterous Fauna of the Andaman and Nicobar Is. Proc. zool. Soc. Lond.

1877 : 580-632, 3 pls.

1881. The Lepidoptera of Ceylon (1). 12+ 190 pp., 71 pls. London.

1882. List of the Lepidoptera collected by the Rev. J. H. Hocking, chiefly in the Kangra

District, N.W. Himalaya. Proc. zool. Soc. Lond. 1882 : 234-263, 2 pls.

1883. Descriptions of new Asiatic diurnal Lepidoptera. Proc. zool. Soc. Lond. 1883:

521-535, 2 pls.

Mou ton, J.C. 1912. A List of the Butterflies of Borneo with Descriptions of new Species (3).
J. Straits Br. R. Asiat. Soc. 60 (1911) : 73-177, 1 pl.

Murayama, S. & SIBATANI, A. 1943. Notes on some Formosan Species of the Subfamily
Theclinae, with descriptions of new Forms. Tvans. Kansai ent. Soc. 13: 37-56, 4 pls.,
text figs.

NICEVILLE, [C.] L. [A.] de. 1887. On new or little-known Butterflies from India. Proc. zool.
Soc. Lond. 1887 : 448-467, 2 pls.

——1889. New Butterflies from the Indian Region. J. Asiat. Soc. Bengal 57 (2) (4) : 273-293,

3 pls.

1890. The Butterflies of India, Burmah and Ceylon (3). 12 + 503 pp., Ftsp. + pls. 25-29.

Calcutta.

—— & Martin, L. 1896. A List of the Butterflies of Sumatra with special reference to the
Species in the northeast of the Island. J. Asiat, Soc. Bengal 64 (2) (3) (1895) : 357-555.

140 Cc. F. COWAN

NrrRE, K. 1920 (in Japanese). On new Species and Subspecies of Butterflies native to this
Country. Zool. Mag., Tokyo 32 : 373-377.

PIEPERS, M. C. & SNELLEN, P.C. T. 1918. Rhopalocera of Java (4), Evycinidae & Lycaenidae.
44 + 114 pp., pls. 19-27. The Hague.

RIBBE, C. 1926. Neue Lycaenenformen, hauptsachlich von Celebes. Ent. Mitt. 15 : 78-o1.

SCHULTZE, W. 1907. New Lepidoptera of the Philippine Islands. Philipp. J. Sci. 2 (5):
361-375, I pl.

SEITz, A. 1907-09. Gross-Schmetterl. Evde (1); Palaearktischen Faunengebietes (1). Die

Palaearktischen Tagfalter. 379 pp., 89 pls. Stuttgart.

1914-27. idem (2); Exotische Fauna (9). Indo-Australischen Tagfalter (97-150); Ery-

cinidae, Lycaenidae, Hesperiidae. pp. 767-1197, pls. 138-175. Stuttgart.

SEMPER, G. 1890. Die Schmetterlinge der Philippinischen Inseln (1). Tagfalter (1886-92).
380 pp., 49 pls. Wiesbaden.

SEVASTOPOULO, D. G. 1935. The early Stages of Indian Lepidoptera Entomologist’s Rec.

J. var. 47 : 49-50.

1947. The early Stages of Indian Lepidoptera (17). J. Bombay nat. Hist. Soc. 46.

575-586.

SHIROzU, T. 1960. The Butterflies of Formosa in Colour. 8 + 479 pp., 479 text figs., 76 pls:
Osaka.

SONAN, J. 1936. Notes on some Butterflies from Formosa (4). Zephyrus 6 : 205-216, 3 pls.

STAUDINGER, O. 1889. Lepidopteren der Insel Palawan. Dt. ent. Z. Iris 2: 3-180, 2 pls.

SWINHOE, C. 1894. New Species of eastern Lepidoptera. Ann. Mag. nat. Hist. (6) 14: 429-

443-

1911-12. Lepidotera Indica (9). 8 + 278 pp., pls. 706~-756. London.

VARLEY,G.C. 1962. Plea for a new Look at Lepidoptera... the Scent Organs of male Moths.
Trans. Soc. Brit. Ent. 15 (3) : 29-40.

WaLtace, A. R. 1869. The Malay Archipelago. Edn. 2, 2 Vols. 312, 341 pp. London.

WiLeMAN, A. E. 1908. New and unrecorded species of Rhopalocera from Formosa. Annotnes
z00l. jap. 6 (5) : 307-335.

WoopHovwsE, L.G.O. 1952. The Butterfly Fauna of Ceylon. Edn. 2. 32 + 231 pp., I map,
1 + 55 pls. Colombo.

Woop-Mason, J. & DE NIcEvILLE, L. 1881. Second List of rhopalocerous Lepidoptera from
Port Blair, Andaman Islands. J. Asiat. Soc. Bengal 50 (2) (4) : 243-262, 1 pl.

INDEX
Indexed below are the text references to the taxa only.
Synonyms in italics.

achaja, 108, 114 Parte, 115, £37
affinis, 120, 137 *artontes, I19, 137
akara, 108 asakurat, 118, 137
*akronyx, 118, 137

albimacula, 107, 112, 113, 126, 127, 137 - bellula, 128, 137
albistigmata, 129, 137 bilineata, 130, 137
amethysta, I12, 113, 130, 132, 139

*amor, 107, 109, 136 camiguina, 123, 137
anara, 128, 137 celebica, 108
andamana, 108 chalcedonyx, 128, 137
anytus, 129, 137 cingalensis, 114, 137

araotina, 108 ciniata, 123, 124, 132, 138

INDEX 141

corniculum, 120, 137 osma, 135, 138
Cupido, 109, 136 *osmana, 136, 138
cuzneri, 138 overdijkinki, 132, 138
decolor, 122, 137 paulla, 122, 137
permagna, 124, 137
*fruhstorferi, 117, 137 privigna, 122, 137

purpurescens, 131, 138
*halba, 120, 137

himeros, 108
holothura, 121, 137
Horaga, I10

rana, 116, 137
*rarasana, 113, 136, 138
Rathinda, 109

inari, 108

: samoena, I24, 132, I I
*isna, 131, 138 4, 13 33, 137

*sardonyx, 116, 137
*schoutensis, 126, 137

*selina, 112, 113, 134, 138
sikkima, 119, 137

SViinn, -L¥2) F109. PEO, £25. 537

joloana, 122, 137
lefebvrei, 113, 134, 135, 138

*maenala, 120, 137

malaya, 128, 137 *taweya, 129, 137
meleva, 135, 138 triopas, 109, 136
moltrechti, 118, 137 triumphalis, 130, 138

moulmeina, 119, 137
viola, 107, 127, 137

onychina, 121, 137 *violetta, 128, 137
“onyx; 107, 112, 113, 114, 115,.137
onyxitis, 128, 137 zuniga, 116, 137

* Asterisked are the taxa illustrated on Plate 1.

PLATE 1

Upper and undersides of left pair of wings of ¢ specimens
of Rathinda and Horaga.

Fic. 1. Hovraga onyx onyx f. arta Fruhstorfer (dry season form). Imphal, 13 March 1912.

Fic. 2. H. onyx onyx f. onyx (Moore) (normal wet season form). Imphal, 18 July 1911.
Gen. Prep. NHB.1965/2506.

Fic. 3. *H. onyx sardonyx Fruhstorfer, Singapore, 25 April 1953. Gen. Prep. NHB.1965/
2527;

Fic. 4. *H. onyx fruhstorferi Corbet, Holotype. Java. Gen. Prep. ASC Horaga 18a.

Fic. 5. *H. onyx akronyx subsp. nov. Holotype. Sambawa, Sep. 1891. (right pair of
wings reversed). Gen. Prep. NHB.1965/2503.

Fic. 6. H. syrinx maenala (Hewitson) f. maenala. Singapore, Jan. 1938.

Fic. 7. *H. syrinx artontes Fruhstorfer, Holotype. Nias.

Fic. 8. *H. syrinx schoutensis Joicey & Talbot, Mefor Is., N.W. New Guinea, Aug. 1920.

Fic. 9. H. syrinx maenala f. halba Distant, Singapore, 11 Dec. 1936.

Fic. 10. Rathinda amor (Fabricius) Hosur, near Bangalore, S. India.

Fic. 11. *H. albimacula taweya subsp. nov. Holotype. W. Celebes, Aug. 1896. Gen.
Prep. NHB.1965/2471.

Fic. 12. *H. albimacula violetta subsp. nov. Holotype. Bali, April 1896. Gen. Prep.
NHB. 1965/2502 (see Pl. 2, fig. 20).

Fic. 13. *H. amethysta isna Corbet, Holotype. Nias. Gen. Prep. NHB.1965/2501 (see
a sec 22):

Fic. 14. *H. selina Grose Smith, Holotype. W. Celebes, Aug. 1891.

Fic. 15. *H. lefebvrei osmana subsp. nov. Holotype. Mindoro. Gen. Prep. NHB.1965/2462.

Fic. 16. H. vavasana Sonan. Mt. Rara, Formosa, July 1964.

* Subspecies or species illustrated for the first time.

Bull. B.M. (N.H.) Entom., 18, 4 PLATE 1

15 | 16
ggg
3 4 5} 6 7 8

ni Ui

CENTIMIETRES

_ Scale: -

PLATE 2

3 genitalia of Rathinda and Horaga

In all cases except Fig. 17, the left valva has been removed and the
lateral aspect from the left is shown. Scale, see Pl. 3.

Fic. 17. Rathinda amor (Fabricius). (a) Genitalia complete, left lateral aspect. (b) Stereo
Pair of the ventral aspect, with the aedeagus and both valvae removed, designed to show, so
focused on, the brachia ; the anal tube is just visible. Separation of Pair, 1-5 mm.

Note in all three photographs the extreme length of the brachia or falces, causing the tips to
be recurved (one of them is visible in Fig. 17a). Note also the central spine on each brachium.

Specimens ; (a) Orissa, Nov. 1913. (b) Ceylon.

Gen. Preps. ; (a) NHB.1955/1554. (b) NHB.1965/2508.

Fic. 18. Hovaga onyx (Moore). Note the stout aedeagus (in this case with vesica everted),
and angled valva (in this example showing the median tooth).

Specimen ; H. onyx sardonyx Fruhstorfer, Banka Is.

Gen. Prep. NHB.1965/2489.

Fic. 19. Hovaga syrinx (Felder). Note the longer, slimmer, aedeagus and the curved valva.
This specimen has the least curved valva of all syyinx examined ; normally the arc is a full 90°.

Specimen; H. syrinx paulla Fruhstorfer, Holotype. Bazilan, Feb. 1808.

Gen. Prep. NHB.1965/2517.

Fic. 20. Horaga albimacula (Wood-Mason & Nicéville). Note the much greater size though
the insect is smaller. Constant from India to Formosa, the Philippines and Sambawa.

Specimen ; H. albimacula violetta subsp. nov. Bali, Apr. 1896.

Gen. Prep. NHB.1965/2502 (specimen figured on Pl. 1, Fig. 12.)

Bull. B.M. (N.H.) Entom. 18, 4 PLATE 2

PLATE 3

3 genitalia of Horaga

In all cases the left valva has been removed and the lateral
aspect from the left is shown.

Fic. 21. Horaga vavasana Sonan. Note remarkable similarity in form to the three preceding
species.

Specimen ; Mt. Rara, Formosa. July 1964.

Gen. Prep. NHB.1965/2509.

Fic. 22. Hovaga amethysta Druce. Note the very elongate vinculum and valva, tiny aedea-
gus, and peculiar brachia ; the terminal bowl of the right brachium, here seen end-on, masks its
own stem and the pointed tip of the left brachium, the stem of which is clearly seen. A monstrous
appendage for such a small insect.

Specimen ; H. amethystaisna Corbet, Holotype, Nias, 1895. (specimen figured on PI. 1, Fig. 13).

Gen. Prep. NHB.1965/2501.

Fic. 23. Hovaga selina Grose Smith. The long simple left brachium is seen lying below
(anterior to) the bifid right brachium, which is furnished with a basal “‘ handle”’; the curved
tip of the left brachium appears between the arms of the right.

Specimen ; Taweya, Celebes west coast ; Aug. 1896.

Gen. Prep. NHB.1965/2511.

Fic. 24. Hovaga lefebvvei (Felder). The simple, curved left brachium contrasts with the
right which has 3 sections; a robust curved and pointed basal third, a broad thin central blade
here seen passing behind and over the left, and a sharp twirled tip here seen end-on and holding
the aedeagus.

Specimen ; H. lefebvrei osmana subsp. nov. Paratype; Mindoro.

Gen. Prep. NHB.1965/2467.

PLATE 3

Bull. B.M. (N.H.) Entom. 18, 4

HTH

HHA

co ‘t

NI

Scale of

imetres for

ill

Plates 2 & 3

M

ni

0 te

oA ase. Ae
* i
;

A LIST OF SUPPLEMENTS
TO THE ENTOMOLOGICAL SERIES
OF THE BULLETIN OF
THE BRITISH MUSEUM (NATURAL HISTORY)

I. MASNER, L. The types of Proctotrupoidea (Hymenoptera) in the British
Museum (Natural History) and in the Hope Department of Entomology, Oxford.
Pp. 143. February, 1965. £5.

2. Nixon, G. E. J. A reclassification of the tribe Microgasterini (Hymenoptera :
Braconidae). Pp. 284; 348 Text-figures. August, 1965. 6.

3. Watson, A. A revision of the Ethiopian Drepanidae (Lepidoptera). Pp. 177;
18 plates, 270 Text-figures. August, 1965. £4 4s.

4. SANDS, W. A. A revision of the Termite Subfamily Nasutitermitinae (Isoptera,
Termitidae) from the Ethiopian Region. Pp. 172; 500 Text-figures. October,
1965. £3 5s.

5. AHMAD, I. The Leptocorisinae (Heteroptera: Alydidae) of the World. Pp. 156;
475 Text figures. November, 1965. {£2 I5s.

6. OxapDA, T. Diptera from Nepal. Cryptochaetidae, Diastatidae & Drosophilidae.
Pp. 129; 328 Text-figures. £3.

7. GILIOMEE, J. H. Morphology and Taxonomy of Adult Males of the Family
Coccidae (Homoptera: Coccoidea). In press.

8. FLETCHER, D. S. A revision of the Ethiopian species and a check list of the
world species of Cleora (Lepidoptera: Geometridae). I” press.

PRINTED IN GREAT BRITAIN BY ADLARD & SON LIMITED, BARTHOLOMEW PRESS, DORKING

a “ — na
7 fe a {s 3
| 8 JUN 196

THE GENUS RHIPIDOCEPHALA “Yau «ss
(DIPTERA : ASILIDAE)

H. OLDROYD

BULLETIN OF
THE BRITISH MUSEUM (NATURAL HISTORY)
ENTOMOLOGY Vol. 18 No. 5

LONDON : 1966

\ 2)
6

8 JUN 196

SS

THE GENUS RHIPIDOCEPHALA
(DIPTERA : ASILIDAE)

BY

|
H. OLDROYD |.

British Museum (Natural History)

Pp. 143-172 ; 38 Text-figures

BULLETIN OF
THE BRITISH MUSEUM (NATURAL HISTORY)
ENTOMOLOGY Vol. 18 No. §
LONDON: 1966

THE BULLETIN OF THE BRITISH MUSEUM
(NATURAL HISTORY), instituted in 1949, 1s
issued in five series corresponding to the Departments
of the Museum, and an Historical sertes.

Parts will appear at irregular intervals as they become
veady. Volumes will contain about three or four
hundred pages, and will not necessarily be completed
within one calendar year.

In 1965 a separate supplementary series of longer
papers was instituted, numbered serially for each
Department.

This paper is Vol. 18, No. 5 of the Entomological
series. The abbreviated titles of the periodicals cited
follow those of the World List of Scientific Periodicals,

© Trustees of the British Museum (Natural History) 1966

TRUSTEES OF
THE BRITISH MUSEUM (NATURAL HISTORY)

Issued 7 June, 1966 Price Twelve Shillings

THE GENUS RAIPIDOCEPHALA

(DIPTERA : ASILIDAE)
By H. OLDROYD

SYNOPSIS

Rhipidocephala Hermann and Holcocephala Jaennicke are two genera of Asilidae that have been
recorded from both the Neotropical and the Ethiopian Regions. The differences between the
two genera are discussed and redefined, and reasons are given for the view that the Ethiopian
species belong to Rhipidocephala and the Neotropical species to Holcocephala. A wide view of
the genus Rhipidocephala is advocated, discounting differences of antennal structure, and
incorporating Pavoxynoton Janssens and Margaritola Hull as synonyms. Twenty-four species
are described, sixteen of them new, with a key to species, and figures of male and female genitalia.

In my key to the tribes and genera of African Asilidae (Oldroyd, 1963 : 7) I divided
the genera of the tribe Xenomyzini into two groups, with a constant difference in
wing-venation, but also differing in general appearance. The first group comprised
the six genera Oxynoton, Margaritola, Holcocephala, Oligopogon, Rhipidocephala and
Paroxynoton. Oligopogon stands apart from the rest, in body-shape, and in the
possession of acanthophorites in the female, and requires a separate study. The
other five genera are tiny, dark flies with broad, dark wings. Oxynoton Janssens,
1951 has a very distinctive, hump-backed development of the thorax, but the
separation of the other four is based chiefly upon characters of the antennae, and is
much less clear.

Understanding of the four genera, and of related forms in other regions, depends
upon a clear assessment of Holcocephala and Rhipidocephala. These were both
defined by Hermann (1926 : 154), who separated them in his key as follows :

Apical style one-segmented, drawn out into a point; moustache shrunk to mouth-margin;
mesonotum bare, or sometimes with scattered soft hairs. : , : g :

HOLCOCEPHALA Jfaennicke

Apical style two-segmented, with a brush of hairs at tip; moustache reaching almost to
bases of antennae ; mesonotum thickly covered with long, erect hairs : : ,

RHIPIDOCEPHALA Hermann

This gave the impression that antennal structure was the key to the separation of
these two genera, a lead that was followed by Hull (1962: 52) and by Oldroyd
(1963: 7). Hull’s description of Holcocephala refers to “a characteristic, shallow,
transverse groove, lying a short distance above the epistoma, which is absent from
such Old World genera... as Rhipidocephala’’, and study of Hermann’s generic
descriptions shows that Rhipidocephala was not, in fact, distinguished from Hoi-
cocephala upon antennal characters, but by details of the facial structure.

Text-figs. 2, 3 illustrate this difference, which is quite definite. In Holcocephala
the frons and face are deeply recessed, exaggerating the ‘‘ Goggle-eyed’”’ aspect

ENTOM. 18, 5. 13

146 H. OLDROYD

proper to the tribe Xenomyzini, and the face has a transverse groove which cuts off a
distinct epistomal ridge, to which the moustache is confined. The hairs of the
moustache are few and bristly, and the median swelling of the face has only fine
hairs. In Rhipidocephala (Text-fig. 2) the head is less withdrawn between the eyes,
the ocellar tubercle is huge and prominent, and in particular there is no transverse
groove across the face. The moustache often, though not always, extends over much

Fic. 1. Rhipidocephala semitestacea (Loew) 9.

THE (GENUS RATPIDOCEPHALA 147

of the facial lobe, and this area is always darker than its surroundings, and in most
species bears a bare, shining spot. Moreover, Holcocephala has the palpi relatively
long and slender, the proboscis stout and awl-shaped ; Rhipidocephala has smaller,
inconspicuous palpi and proboscis.

The antennae of Holcocephala show a high degree of constancy, with a conical
style and an apical spine. Within the scope of Rhipidocephala, as now defined, there
is much greater variation in antennal structure, and the question arises whether or
not the genus should be still further divided, or whether the variations should be
regarded as of specific value only. If the latter view is taken, then it is clear that
Margaritola Hull and Paroxynoton Janssens must also be merged with Rhipidocephala.

Among the African species examined during the present study, four types of
antennal structure can be recognized (Text-fig. 4, a-D). Species can be fairly clearly

Fics. 2, 3. Heads of: 2, Rhipidocephala; 3, Holcocephala. Showing generic differences in
frons, face and palpi; for antennae see text.

assigned to one or other of these types, though not without variation. Now Hermann
(1926 : 175) says that the genitalia of all the Rhipidocephala known to him are yellow
in colour, and that the ovipositor is conical, and divided into two lobes. This is so in
several species, but it is by no means universal in the genus. Text-figs. 6—38 show
some of the variants : many ovipositors are partly or wholly concealed ; and whether
exposed or concealed, some are yellow, some black, some bilobed, some not. More-
over there is no correlation between the type of ovipositor and the type of antenna.

It seems, therefore, that there is no good reason for subdividing Rhipidocephala on
the basis of either antennal or genital structure, and that one should retain the
clearly-defined genus indicated by the head-structure as shown in Text-fig. 2, reducing
Margaritola Hull and Paroxynoton Janssens to synonymy.

When defining Rhipidocephala, Hermann (1926 : 174) stated that it comprised the
African species previously assigned to Holcocephala, as well as two species from the
Neotropical Region : analis (Macquart), and flavipes Hermann.

148 H. OLDROYD

The case of analis is rather complicated. Macquart (1846: 69-70) based his
Discocephala analis upon three specimens from Colombia, and one “ exactly similar ”’
from Port Natalin Africa. The specific description says that the wing-venation is the
same as in Discocephala rufiventris, a synonym of Dasypogon abdominalis (Say, 1823),
and undoubtedly a Holcocephala from the figures of wing and head given by Macquart.
It seems, therefore, that the true analis Macquart is a Neotropical species of Holco-
cephala.

Hermann (1926:176) clearly based his account of Rhipidocephala on African
species, and stated that the genus comprised the African species previously assigned
to Holcocephala. The only evidence of any Neotropical species lay in two items : the
above statement by Macquart about analis; and the presence of Vienna in two
species of Rhipidocephala in the von Winthem collection, one labelled Mexico, the
other without locality, but presumed by Hermann to be from the same area.
Hermann regarded the latter, without locality, as being probably analis Macquart, and
made it the type-species of Rhipidocephala. The von Winthem specimen labelled
““ Mexico ”’ he described as a new species, R. flavipes Hermann.

Through the kindness of Prof. Max Beier I have been able to examine this von
Winthem material, and several points emerge. Firstly that the specimen labelled
“analis’’ cannot be the analis of Macquart because its venation is not the same as
that of Holcocephala rufiventris (Macquart). Therefore the type-species of Rhipido-
cephala must be :

Rhipidocephala analis Macquart; Hermann = angustior Oldroyd, sp. n. (see
below).

Under Art. 70(a) of the International Code of Zoological Nomenclature, this case of
a misidentified type-species must be referred to the Commission, requesting a use of
the Plenary Powers in support of choice number (1) : the nominal species actually
involved, which was wrongly named in the type-designation.

Before such a case can be made out for submission to the Commission, it will be
necessary to study not only the Neotropical but also the Oriental species involved in
this complex of genera.

The specimen in Vienna bears several labels, written and printed, but the word
“Mexico ’”’ appears only on a small printed label, apparently added at some later
date.

There is thus no evidence of any Rhipidocephala from S. America, and none of any
Holcocephala from Africa. Any distinction based upon antennal structure, such as
the one in my key (Oldroyd, 1963 : 7) should be abandoned, and the genus Rhipido-
cephala based upon the structure of the head. In this respect, and in the closed anal
cell, Seabramyia Carrera, 1958, is a variant of Holcocephala and not of Rhipidocephala.

As thus defined, Riipidocephala takes in all the mainland species that have been
described in the genera Rhipidocephala, Holcocephala, Paroxynoton and Margaritola.
From Madagascar, Holcocephala lambertoni Bromley is a normal Rhipidocephala, but
Rhipidocephala hyalipennis Oldroyd, 1959, is an Oligopogon.

It is a pity that the type-species selected by Hermann—angustior sp. n. (analis
Macquart ; Hermann)—belongs to a small group of species that differ in some

EHEVGENUS/ RAT PTDOGE PHA EA 149

respect from the majority of Rhipidocephala. They have no bare spot on the face,
the abdomen dorsally is not membranous to the extent shown in Text-fig. 5, and the
ovipositor is unusually long and stout (Text-figs. 32, 38). None of these characters is
constant. In particular, the dorsum of the abdomen shows varying degrees of
membranous development, and may have no more than a well-developed cleft
between segments one and two.

Fic. 4. Range of antennal structure within the genus Rhipidocephala, with examples of four
principal types : A—D.

Fic. 5. Rhipidocephala morio Hermann. Dorsum of abdomen, showing extent of membranous
area.

RHIPIDOCEPHALA Hermann, 1926

Rhipidocephala Hermann, 1926, Verh. zool.-bot. Ges. Wien 74 (1924/25): 174; Oldroyd, 1959,
Mém. Inst. scient. Madagascar (E) 11: 276; Hull, 1962, Robber Flies of the World : 64.

Type-species : Rhipidocephala angustior Oldroyd, sp. n., for analis Macquart ;
Hermann, by original designation.

Paroxynoton Janssens, 1953, Bull. Inst. r. Sci. nat. Belg. 29: 11, syn. n.
Type-species : Paroxynoton tigrinum Janssens, 1953, by original designation.
Margaritola Hull, 1958, Proc. ent. Soc. Wash. 60: 255, syn. n.

Type-species : Margaritola mirabilis Hull, 1958, by original designation.

150 H. OLDROYD

Head. As shown in Text-fig. 2, only slightly recessed at vertex, but with a very large and
prominent ocellar tubercle. Face smoothly rounded, not deeply recessed along eye-margins, and
without transverse groove delineating epistoma (cf. Holcocephala, Text-fig. 3). Moustache
consisting of scattered, relatively fine, bristly hairs, except in a few species, where they are
darker and stronger; these hairs often extending over lower four-fifths of face, though some-
times reduced to vicinity of epistoma. rons and face covered with tomentum, often darker on
frons, but sometimes unicolorous ; most species have an ovoid area that is bare, shining and
usually black, lying either dorsal to moustache, or with part of moustache growing fromit. Palpi
and proboscis relatively shorter than those of Holcocephala, and inconspicuous, though palpi have
long and rather stiff bristles.

The antennae, a feature of special interest in this genus, have already been discussed. They
are less constant in form than those of Holcocephala, and cover the four types shown in Text-fig. 4,
but with intermediates. Generally the second of the microsegments has long pile.

Thorax moderately convex, though slightly rounded in one or two species, never with an acute
angle as in Oxynoton francoisi Janssens. Mesonotum normally heavily covered with tomentum,
rarely with small bare spots (Text-fig. 1) or shining through thin tomentum; normally grey
with a dark brown pattern, consisting of a median stripe, broad and even divided anteriorly, but
tapering strongly posteriorly, flanked by two lateral spots. Small modifications of this pattern
are very constant, and make useful recognition features in practice, though difficult to define in a
key to species. Most species have long, erect, curved hairs, especially posteriorly and on scutel-
lum, which has no distinct marginal bristles. Pleura undistinguished, with a few hairs on meso-
pleuron, and with a row of stronger ones before halteres.

Abdomen remarkable for excavation of the dorsum into a membranous trough (Text-fig. 5).
Second tergite retains a narrow bridge across its middle, but otherwise tergites as far back as
seventh segment are broken into two widely separated sclerites. This membranous area of
abdomen seems to be confined to the two genera Holcocephala and Rhipidocephala, as here
defined. It is not always so fully developed, and in one or two species is restricted to first
segment, or first two; sometimes indicated merely by wrinkling of dorsal surface.

Genitalia of both sexes show good specific differences, with a wide range of shape and relative
size (Text-figs. 6-38).

Legs unremarkable, without swellings or special structures, but—with some variation—their
colour is of use in distinguishing species ; covered with short hairs and long, slender bristles, all
of which are normally pale yellow.

Wings (Text-fig 1) relatively broad, often expanded basally by broadening of costal and
axillary cells. Venation simple, primitive ; anal cell open or closed just before margin [closed
and stalked in Holcocephala), all other cells open on wing-margin.

Size rather constant at about 5-6 mm., both for body-length and for length of one wing.

KEy TO SPECIES

1 Face with a shining, bare spot, usually black (Text-fig. 2). Frons with few or no hairs
laterally, along eye-margins. Abdomen extensively membranous dorsally (Text-fig.

— Face entirely tomented, without any bare spot. Frons with long hairs laterally,
along eye-margins. Abdomen with membranous areas usually ae
developed, or confined to first twosegments . ; 19
2 Mesonotum without well-developed bare spots, though tomentum may be thin all
over, and then mesonotum appears rather shining, black. ‘ 3

— Mesonotum covered with bronze tomentum, which leaves bare and shining black the

humeri and two well-defined, round spots on each side. (Text-fig. 1)
semitestacea (Loew) (punctulata Wulp) (p. 152)

3 Mesonotum without the distinctive long, curved hairs, though a uniform covering of
short hairs may be present . ; ' : ‘ ‘ ‘ : : ‘ 4

Or i OO Se otk Gh

16

—

17

TLRE GENUS RHIPIDOCEPHALA

Mesonotum and scutellum with long, curved hairs, at least posteriorly, conspicuous in
side view .

Legs and abdomen reddish aie Mesonotum shining black through thin layer of
bronze tomentum, and with short ae hairs. Moustache includes bare area of

face . , mirabilis (Hull) (p.

Legs and abdomen entiely black or ashy grey. Mesonotum not at all shining black.
Moustache confined to mouth-margin, and arising from tomentum; bare area of face
lies dorsal to moustache. Z ; : :

Mesonotum humped, with chocolate- brown tomentum ; " darker stripes hardly

visible ; ; obscurata sp. n. (p.

Mesonotum not unusually humped, grey posteriorly, ih 3 distinct black stripes

anteriorly . : : . divestita sp. n. (p.

Legs conspicuously reddish yellow: at least i in part

Legs entirely black, except narrowly at knees

Femora uniformly black, contrasting strongly with reddish bases of tibise:
Femora partly or wholly reddish

Postscutellum grey, concolorous with mesonotum. ; tigrina (Janssens) (p.
Postscutellum brown, contrasting with mesonotum . inconspicua sp. n. (p.

Mesonotal pattern unique ; the 3 stripes reduced to a transverse row of three brown
spots, and mesonotum entirely grey both anteriorly and posteriorly

distincta sp. n. (p.

Mesonotum not so patterned :
Mesonotum yellow-grey, with short median stripe and a ‘cotspicuous brows iriangular
spot on each side before the suture. Hind femora almost entirely reddish yellow

fulva sp. n. (p.

Mesonotum dull brown with indistinct stripes. Hind femora reddish yellow, or black

with yellow tips ‘ : : 5 : : ; s :
Hind femora entirely reddish y ello ; “fore and middle femora black with yellow tips

flavipes Hermann os

All femora black with yellow tips
Mesonotum densely hairy, and mostly dull brow n, eeripes extending from front to rear

umbripennis (Loew) (p.
Mesonotum sparsely hairy, and stripes short F 3 tenera sp. Nn. (p.

Mesonotum almost uniformly ashy grey, with only a faint trace of the 3 darker stripes

zumpti sp. n. (p.

Mesonotum not uniformly ashy grey

Mesonotum dull black-brown, without any definite pier or irk only faistinee
stripes, and not noticeably more grey either anteriorly or posteriorly

Mesonotum with 3 distinct brown stripes, sometimes entirely grey either anteriorly o1 or
posteriorly . :

Antennae at least as isis as nee ficrosepinents saatively whe: Senin micro-
segment with a dorsal spine, but with only a short pile. Mesonotum not strongly

convex, little higher than vertex. ; d : caffra (Macquart) (p.

Antennae not longer than head, third segment eantivety short; second microsgment
with no dorsal spine, but with long hairs. Mesonotum strongly convex, extending

well above vertex, and with long hairs posteriorly. . - morio (Hermann) (p.

Thoracic stripes extending from anterior aes of mesonotum almost to posterior
margin ‘ :

Thoracic stripes shorter, leaving either anterior or - posterior third BE mesonotini grey

Moustache black. Mesonotum with a pair of parallel median stripes, replaced pos-
teriorly by a single stripe just before scutellum. Mesopleuron wholly yellowish

brown, forming a band from fore coxae to humeri : . signata (Hermann) (p.

ENTOM. 18, 5.

I51

162)

163)
17
18.

163)
13§

152 H. OLDROYD

— Moustache pale yellowish. Mesonotum with a heavily marked, solid, median stripe,
narrowing and disappearing posteriorly, with no median prescutellar stripe.
Mesopleuron mainly grey, brown only dorsally . ‘ congoiensis sp. n. (p. 164)

18 Mesonotum conspicuously marked, with three brown stripes coalescing in anterior
half; posterior half entirely grey with long, yellowish hairs: scutellum grey.

Genitalia black. ‘ . Scutata sp. n. (p. 166)
— Mesonotum grey anteriorly ; ; posteriorly, brown stripes show rather indistinctly;
scutellum brown. Genitalia reddish yellow, conspicuous. . . engelisp.n. (p. 166)
19 Costal margin of wing expanded by broadening of costalcell . é . a : 20
— Costal margin of wing not expanded ; costal cell narrow. 23
20 Sides of abdomen with recumbent as hairs. Female witha conspicuous, shining
ovipositor (Text-fig. 32) : ‘ fimbriata sp. n. (p. 167)
— Sides of abdomen without conspicuous, yellow hairs. Female with normal, in-
conspicuous Ovipositor . : F . ° : . 21

21 Ashy grey and brown species. Head with grey tomentum and long, brown hairs on
frons, antennae, moustache and occiput. Thorax with grey and brown tomentum
and long, brown hairs, especially in mesopleural and metapleural tufts
manicata sp. n. (p. 167)
— Black-brown species with hairs pale yellowish or white. Frons chocolate-brown,
contrasting with yellowish grey face. WHairs all pale. Mesopleural and meta-

pleural tufts also pale . P : : : : : : ‘ ‘ 22
22 Tibiae orange with black tips. Hairs of body, especially those of mesopleuron,
yellow : : speciosa sp. n. (p. 168)

— Legs all black. ais of boa Pian tone of eee silvery
doornensis sp. n. (p. 169)

23 Slender species, with legs entirely black quadrifaria Hermann (p. 169)
— Robust species, with tibiae and tarsi bright orange angustior sp. n. (p. 170)

Rhipidocephala semitestacea (Loew)

Discocephala semitestacea Loew, 1863, Wien. ent. Monat. 7: 11.
Rhipidocephala semitestacea (Loew) Hermann, 1926, Verh. zool-bot. Ges. Wien 7 : 178.
Holcocephala punctulata Wulp, 1899, Trans. ent. Soc. Lond. 1899 : 88, syn. n.

Loew’s paper of 1863 described specimens collected by C. Tollin at Bloemfontein.
The types are not in Vienna, and Horn’s list states that the Tollin material was
scattered. Some of it may be in the v. Roeder collection in Halle.

Loew’s original description of semitesatcea is short :

3 Fusca, fusco-pollinosa, opaca, thorace non vittato, abdomine pedibusque testaceis, tarsis apicem
versus fuscis, alis nigris ; antennae nigrae, stylo terminali breviusculo, in apice pilis minutis nigris
penicillato ; mystace pilisque palporum nigrorum dilute subfuscis : long corp. 2 lin. ; long. al.
24 lin.

Obs. Stylus antennarum penicillatus ut in Oligopogonibus, sed reliqua omnia ut in Discocephalis
genuinis, a quibus non separanda.

Hermann (1926 : 178) repeated the last observation, but had clearly not seen the
type, nor was it among the Rhipidocephala lent to me from the Vienna Museum by
Dr. Max Beier.

Specimens in the British Museum, determined by Engel, agree with the original
description as far as it goes. These specimens are distinct in having the black
mesonotum uniformly obscured by bright yellow tomentum, except for the humeri

LHE GENUS RALPIDOGEPHAELA 153

and for two small, round, bare, black spots on each side, exactly as in punctulata
Wulp. The latter is authentically represented in the British Museum by specimens
from Aden and Somaliland, and these are indistinguishable from the South African
specimens, even in genitalia.

Thus, although the type localities of the two species are so far apart that their
synonymy seems inherently improbable, it is not possible at the moment to find any
difference between them.

g Head. Frons and face uniformly covered with dense yellow tomentum, except for large,
bare, brownish area occupied by sparse yellow moustache. Antennae of type A (Text-fig. 4),
black, first two segments with yellow tomentum and yellow hairs. Proboscis and palpi black-
brown with yellow hairs. Occiput covered with yellowish grey tomentum and with yellow hairs
along eye-margins.

Thorax. Mesonotum shining black in ground colour, normally with uniform covering of dense
brassy yellow tomentum, except for humeri and two small, rounded spots on each side, which
are bare, shining black, and very conspicuous (Text-fig. 1). Clothing hairs yellow, long and
erect. Scutellum with brassy tomentum and yellow hairs. Pleura covered with greyish yellow
tomentum, a little browner anteriorly ; hairs bristly yellow, confined to anterior area of mesono-
tum, and to metapleural tuft in front of halteres.

Abdomen entirely bright orange, with brassy tomentum and long, yellow hairs. Male genitalia
as in Text-fig. 6. Tergites membranous as shown in Text-fig. 5, but being almost concolorous
with sclerites, this area is not obvious in dried specimens.

Legs. Coxae like pleura ; femora and tibiae yellow-brown ; tarsi darker. Hairs and bristles
yellow.

Wings uniformly dark brown. Halteres with large yellow knob.

Length of body 5 mm.; of wing 4°5 mm.

2 similar. Ovipositor compact, eighth tergite short, partly bare and shining, lamellae little
prominent.

Holotype of semitestacea ? in Halle; of punctulata in Hope Museum, Oxford.

Distribution. S.W. Arrica: Okahandja, ii-iili.19g28 (R. E. Turner) , Groot-
fontein, 1.1920 (R. W. Tucker) ; Waterburg, ii. 1920 (R. W. Tucker) ; Narebis, ii. 1921
(F. Barnard). Cape PROVINCE : Worcester, 17-31. viii.1928 (R. E. Turner).

TANGANYIKA : N.E. Kondoa, v.1955 (J. F. Lamerton)—this specimen forms a link
between the above S. African localities and the following :

SOMALILAND: Bohotle, 1903 (A. F. Appleton). ADEN: 16.iii.95 (C. G. Nurse).
Wulp’s original type material of punctulata, from Aden, was collected by Col. Yerbury.

Rhipidocephala mirabilis (Hull)

Margaritola mirabilis Hull, 1958, Proc. ent. Soc. Wash. 60 : 255; 1962, Robber Flies of the World :
66, comb. n.

Though this species was made the type of a new genus, it clearly belongs in
Rhipidocephala, as defined in the present paper. The antennae, of type A (Text-fig.
4) are not unique in the genus, and indeed the species is quite close to semitestacea.
Some specimens of semitestacea, especially those with the thorax greasy, could easily
be mistaken for mirabilis, and the single specimen of mirabilis in the British Museum
collection was originally labelled semitestacea by Engel.

154 H. OLDROYD

Q° Head. Frons and face with dense whitish tomentum, leaving a bare, shining black area
round ocellar tubercle, and a large, ovate area occupied by moustache. Ocellar bristles, and
those of sparse moustache, yellow. Antennae of type A (Text-fig. 4) black, with yellow hairs.
Proboscis and palpi also black with yellow hairs. Occiput covered with grey tomentum, more
densely white along eye-margins, and with short pale hairs.

Thorax. Mesonotum shining black, only narrowly tomented at extreme sides, on anterior
face, and on flat prescutellar area; clothed with yellow hairs that are moderately dense, but
uniformly shorter than those of semitestacea. Scutellum similar, i.e. shining black with yellow
hairs and a narrowly tomented rim. Pleura with grey tomentum, browner on mesopleuron.
Hairs fine, yellow, restricted to a patch on mesopleuron and a vertical row in front of halteres.

Abdomen entirely bright orange with yellow tomentum and short yellow hairs. Ovipositor
short, shining eighth tergite and lamellae almost concealed (Text-fig. 8).

Legs. Coxae and trochanters black in ground colour, and tomented like pleura. Femora,
tibiae and most of basitarsi orange ; tips of basitarsi and other tarsomeres black. Hairs and
bristles yellow.

Wings dark brown, becoming paler towards tip. Halteres large, orange.

Length of body 5 mm.; of wing 5 mm.

3g. Closely similar; genitalia as in Text-fig. 7.

Holotype in U.S. National Museum, Washington.

Distribution. Type locality is MocamBiqguE: Lourengo Marques. In B.M.
(N.H.) x 9, S. RuopeEsra : Sawmills, 12.iv.1920 (?collector). In South African
Institute for Medical Research 4 3, 3 2, TRANSVAAL : Rootberg & Letalel (F. Zumpt).

Rhipidocephala obscurata sp. n.

One of a small number of species in which the usual clothing hairs of the mesonotum
are lacking. In obscurata the stripes, too, are lacking, and the mesonotum is black-
brown with lighter brown rim. It thus resembles morio Hermann in general colora-
tion, but is distinguished by the absence of the mesonotal hairs, and by the genitalia.

6 Head. Frons and face black-brown in ground colour, but entirely covered with brown
tomentum except for the facial bare area. In this species the facial bare area is above the
moustache, which is reduced to a few yellow bristles on the epistoma (cf. discussion of Holco-
cephala, above). Antennae, however, of Rhipidocephala type D (Text-fig. 4), black, basal seg-
ments with pale hairs. Proboscis and palpi black with yellow hairs. Occiput brown with white
rim on eye-margin, and with short, fine, yellow hairs.

Thorax. Mesonotum entirely covered with dense tomentum, even over humeri ; more blackish
anteriorly, more brownish laterally and posteriorly. Hairs few and sparse, recumbent, and
hardly visible in side view. Pleura similar, heavily tomented, more blackish on dorsal half,
more greyish ventrally. Hairs yellow.

Abdomen. General colour ashy. Dark chocolate-brown at sides, more yellowish in mem-
branous areas dorsally. Each segment with a large area of grey tomentum laterally, but first
and most of second segments black-brown. 4g genitalia as in Text-fig. 9, shining black-brown.

Legs black, with yellow clothing hairs and yellow bristles.

Wings uniformly black-brown. Halteres yellow.

Length of body 5 mm.; of wing 5 mm.

@ closely similar. Eighth sternite is mostly shining, and forms a short, stout ovipositor.

Holotype 3. S. Ruopesta : Chavavuarea, 7 m. E. of Kariba, 6.11. 1956 (R. Goodier).
In B.M. (N.H.).

THE GENUS -RATPIDOGEPHALA 155

Paratypes: S. RHopEsta: Kariba Area, Chavavu area, 6.ii.1956, 19.1.1956,
292; Kessesse R. area, 20.i.1956, r g, I @ (all R. Goodier).

Fics. 6-13. Genitalia of Rhipidocephala ; those with a characteristic external appearance
are drawn in situ, those requiring dissection are drawn from slide-mounts: 6, semt-
testacea 3; 7, mirabilis 3; 8, mirabilis 2; 9, obscuvata g; 10, divestita, J; II, 12,
tigrina 3; 13, inconspicua g.

156 H. OLDROYD

Rhipidocephala divestita sp. n.

Another species with only insignificant bristles and hairs on the mesonotum,
divestita is distinguished from obscurata sp. n. by the distinct dark pattern of the
mesonotum and by the male genitalia.

$6 Head. Frons and face with yellowish white tomentum ; a brownish bar through bases of
antennae, and ocellar tubercle with brown tomentum. Face has a small, transverse bare area
lying dorsal to moustache, which is reduced to a few yellow bristles close to epistoma. Antennae
of type D (Text-fig. 4) black, basal segments with yellow hairs. Proboscis and palpi black with
yellow hairs. Occiput with white tomentum and white hairs.

Thorax. Mesonotum light grey, with a distinct black-brown pattern confined to anterior
half ; a broad median stripe, divided anteriorly, tapering posteriorly, and on each side of it a
triangular patch. Hairs reduced to a few posteriorly and on scutellum, which is uniformly grey.
Pleura black on dorsal half, grey ventrally.

Abdomen ashy grey like that of obscurata. Only middle, membranous areas of tergites a little
reddish. Base of each segment black-brown, but from third segment onwards with a pair of
large, grey lateral spots on each segment. Hairs short, yellow. Genitalia as in Text-fig. 10.

Legs black-brown, with yellow hairs and bristles.

Wings black. Halteres orange.

Length of body 5 mm.; of wing 5 mm.

Q similar. Ovipositor as in obscurata, short, stout, with eighth segment shining through thin
tomentum.

Holotype 3. ZANzIBAR: Mazi Maja, x—xii.1924 (H. J. Snell). In B.M. (N.H.).
Paratypes: ZANZIBAR: Mazi Maja, x—xii.1924 (H. J. Snell), t 3, 9 9.

Rhipidocephala tigrina (Janssens)

Paroxynoton tigrinum Janssens, 1953, Bull. Inst. yr. Sci. nat. Belg. 29:12, comb. n.

This species was originally described as type of a new genus, Paroxynoton, by
comparison with the same author’s earlier genus Oxynoton. Both genera are rep-
resented in the British Museum collection, P. tagrinum by paratypes. The specimens
show that, while Oxynoton francoisi Janssens is a very distinctive fly with a curiously
hump-backed mesonotum, P. tigrinum falls well within the concept of Rhipidocephala
as used in the present study. It is not even a particularly distinctive species of
Rhipidocephala, but is separated from allied species by the combination of characters
shown in the key.

6 Head. Face and frons shining black with ashy tomentum, mostly grey, but with a brown
band through bases of antennae, and ocellar tubercle with brown tomentum. Bare, shining black
area of face lunate, lying entirely dorsal to moustache, which is rather reduced, and lies near
epistoma ; bristles yellow. Antennae of type D (Text-fig. 4) but with second segment of style
thinner and more pointed, and with longer, sparser hairs in apical half; hairs of two basal seg-
ments yellow. Proboscis and palpi black with yellow hairs and bristles. Occiput with whitish
tomentum and pale yellow hairs.

Thorax. Mesonotum with dense grey tomentum, more whitish laterally and posteriorly, and
with prominent pattern of broad black-brown stripe, which ends abruptly posteriorly, flanked by
two large, triangular spots. Pale hairs short anteriorly, but long and dense posteriorly. Pleura
grey, without any obviously darker areas ; fine hairs and bristles yellow.

Abdomen. Dorsum extensively dull orange, not only membranous area but lateral tergites as
well; sternites more blackish. Hairs and bristles yellow. Male genitalia orange, large and
prominent, in dorsal view forming an arch (Text-figs. 11, 12).

THE GENUS RATPIDOGEPHALA 157

Legs. Femora black, except for narrowly red bases ; tibiae and basitarsi dull reddish, other
tarsomeres more blackish. Bristles and clothing hairs yellow, latter denser than in many
Rhipidocephala.

Wings dark brown, not appreciably lighter towards apex. Costal cell distinctly broadened.

@ similar. Ovipositor short and inconspicuous.

Holotype in Inst. R. Sci. nat., Bruxelles, Belgium.
Distribution. Known only from URUNbDI: Gihanga & Bubanza.

Rhipidocephala inconspicua sp. n.

Very similar to tugrina, including the structure of the antennae but separated by
the dark brown postscutellum and by the male genitalia. The existence of this
species supports the view that Paroxynoton Janssens is not generically distinct from
Rhipidocephala.

6 Head. Frons and face with yellowish grey tomentum, whiter on face, browner on frons ;
ocellar tubercle with dark brown tomentum. Bare facial area shining black, ovoid, immediately
dorsad of a rather sparse yellow moustache. Antennae as described for tigrina, type D (Text-fig.
4), with second segment of style elongate and finely pointed, with longer black bristles in apical
half. Hairs of basal two segments pale. Hairs of basal two segments pale. Palpi and proboscis
black with yellow hairs and bristles. Occiput uniformly covered with white tomentum and pale
yellow hairs.

Thorax. Tomentum of mesonotum extensively grey posteriorly and laterally. Median stripe
very broad anteriorly, indistinctly merging with the two large spots, but ending abruptly level
with bases of wings. Erect hairs rather sparse on dark areas, but dense and very pale on grey
posterior area and on grey scutellum. Pleura with dense ashy grey tomentum, darker brown
round anterior spiracle and margins of mespleuron. MHairs yellow. Postscutellum dark brown
like base of abdomen, and contrasting with grey pleura.

Abdomen whitish on membranous areas of tergites ; sclerotized tergites covered with ashy grey
tomentum, dark brown on first two tergites, and narrowly at bases of others. Hairs pale yellow.
Genitalia as in Text-fig. 13.

Legs. Femora black, tibiae and tarsi orange, other tarsomeres darker. Clothing hairs and
bristles yellow.

Wings dark brown, no paler area. Halteres yellow.

Length of body 6 mm.; of wing 6 mm.

Qsimilar. Eighth segment forming a short ovipositor, but still covered with tomentum.

Holotype 3. KENyA : Gasi-Mombasa, v.1944 (E. Opiko). In B.M. (N.H.).

Paratypes : KENyA: Gasi-Mombasa, v.1944 (E. Opiko), 19; Gasi, xi.1927 (van
Someren), I 3, 12; Diana Beach, vii.1951 (N. L. H. Kraus), 1 9. TANGANYIKA:
Tanga, vi.1932 (Miss A. Mackie & J. Ogilvie), 2 3, 3 &.

Rhipidocephala distincta sp. n.

The thoracic pattern sets this species apart from any other. Besides the usual
large grey area posteriorly, the anterior third of the mesonotum is also entirely
yellowish grey, reducing the dark pattern to a transverse row of three spots.

$ Head. Tomentum yellowish grey: ocellar tubercle, and a transverse band immediately
above antennae, dark brown. Ocellar tubercle with fine yellow hairs, which also appear on
vertex and postvertex, but not on sides of frons. Face with yellowish grey tomentum and a
lunate, bare, shining black area above moustache, which is about half height of face, and has

158 H. OLDROYD

rather scattered yellow bristles. Antennae of type A (Text-fig. 4), black, basal segments with
yellow hairs. Palpiand proboscis black with yellow hairs and bristles. Occiput entirely covered
with white tomentum and pale yellowish hairs.

Thorax. Mesonotum strongly arched, almost semicircular in outline, covered with dense
yellowish grey tomentum, which is extensive anteriorly as well as posteriorly: dull brown
stripes reduced to a transverse row of three spots. Fine, pale yellowish hairs long and erect,
over entire mesonotum and scutellum. Pleura entirely covered with yellowish grey tomentum
and pale yellowish hairs, except for dark brown colour in notopleural area and on posterior
spiracle.

Abdomen dorsally with membranous area indicated, but less extensive than shown in Text-fig.
5. Entire dorsum light brown, laterally with dense tufts of long, fine yellow hairs ; ventrally
with more greyish tomentum and fine yellow hairs. Male genitalia (Text-fig. 14) orange,
exserted, but not very prominent, claspers with a characteristic row of strong bristles.

Legs reddish yellow, fore and middle femora indistinctly darkened, especially anteriorly.
Hairs and bristles entirely yellow, clothing hairs dense and relatively long.

Wings uniformly brown, costal cell broadened.

Length of body 5 mm.; of wing 5 mm.

similar. Ovipositor short, stout, simple.

Holotype 3. MocaMBIQuE: Inhambane, li.1924 (R. F*. Lawrence). In B.M.
(N.H.).
Paratypes: MocaMBIQUE: Inhambane, ii.1924 (R. F. Lawrence), 5 3, 2 9.

Rhipidocephala fulva sp. n.

Distinguished from flavipes and other yellow-legged species by having also a tawny
thorax and abdomen, with abundant long, yellow hairs.

3 Head. Frons and face covered with dark, yellow-brown tomentum, ocellar tubercle and
sides of vertex with rather short, yellow-brown hairs. Face with a large, ovoid, bare spot,
shining brown, and with a rather sparse moustache of well-separated yellow-brown hairs.
Antennae of type D (Text-fig. 4) black, with yellow hairs on first two segments. Proboscis and
palpi with yellow hairs on a black ground. Occiput with yellow-brown tomentum and short
yellow hairs.

Thorax. Mesonotum yellowish grey, with the usual dark brown pattern of median stripe and
two spots, all obscured by dense, long, tawny hairs. Pleura yellowish grey with few yellowish
hairs ; notopleural region and two spiracles dark brown.

Abdomen extensively membranous dorsally, sclerotized areas also clear yellow-brown, with
yellow hairs. Ventersimilar. Male genitalia also yellow, small and inconspicuous (Text-fig. 17).

Legs clear yellow, tarsomeres becoming progressively darker. Hairs and bristles yellow,
bristles of femora and tibiae unusually long and strong.

Wings uniformly brown, costal cell rather expanded. MHalteres yellow.

Length of body 6 mm. ; of wing 6 mm.

Q closely similar. Ovipositor scarcely exists, but lamellae protrude from a shining eighth
segment.

Holotype g. CAPE Province: Kruger National Park, 29.xi.54 (C. H. Andrewes).
In B.M. (N.H.).

Paratypes : CAPE Province : Kruger National Park, 29. xi.1954 (C. H. Andrewes),
MogaMBIQUE : Inhambane, i.1924 (R.F.Lawrence), 1 9. TRANSVAAL: Potgieters-
rust, 6.xii.1953 (F. Zumpt), g. BECHUANALAND: nr, Nata, xii.1954 (F. Zumpt),
IQ. S. RuopeEsta: Sawmills, 31.xii.2z (?collector), r J, 1 .

THE GENUS -RATPIDOCEPHALA 159

Rhipidocephala flavipes Hermann
Rhipidocephala flavipes Hermann, 1926, Verh. zool-bot. Ges. Wien 74 : 177.
Descriptions made from the syntypes in the Naturhistorisches Museum, Vienna.

Conspicuous yellow-brown male genitalia, epandrium divided into two rolled and leaf-like
upper forceps, with a large apical notch. Legs with only femora black, tibiae and some tarsal
segments red.

6 Head. Face and frons covered with dense yellowish white tomentum, leaving bare and
shining black a semicircular area in middle of face to which pale yellow moustache is confined.
Basal antennal segments dull black (Ant. III broken off), with pale yellow hairs. Proboscis and
palpi black with yellow hairs. Beard and occipital hairs all pale yellow.

Thorax. Mesonotum with long, dense, pale yellow hairs. Tomentum dull greyish, stripes
distinct, median stripe broad, complete anteriorly, stopping before reaching scutellum. Pleura
with uniformly greyish tomentum and pale yellowish hairs.

Abdomen. Tergites and sternites covered with dense grey tomentum, only hind margins of
posterior segments a little brownish. Hairs long, pale yellowish. Genitalia yellow-brown,
translucent (Text-figs. 15, 16).

Legs. Coxae black with greyish tomentum ; trochanters shining black. Femora shining
black with orange knees, and a little more orange colour on hind femora. Fore and middle
tibiae dull reddish, black ventrally ; tarsi black ; hind tibiae and basitarsi orange, rest of hind
tarsus black. All legs with yellowish bristles, and covered with recumbent yellow hairs.

Wings rather broad, uniformly dark brown. Halteres pale yellow.

Length of body 5 mm.; of wing 5 mm.

? similar except for rather more orange on all femora. Third antennal segment awl-shaped,
with style and microsegment more than half its total length. Female terminalia also orange,
short, downturned.

Syntypes I g, I 2 in Vienna Museum, lent to me for study. The labelling and
probable origin of these specimens has been discussed earlier in this paper.

Distribution. E.CAPE PROVINCE: Kathberg, 4,000 ft., I-15.1.1933 (R. E. Turner),
2 3, I 2 agree in genitalia, and in other respects except for variations in leg colour.
TRANSVAAL: Skukuze, 23.xi.1959 (F. Zumpt).

Rhipidocephala umbripennis (Loew)

Discocephala umbripennis Loew, 1858, Ofv. Kongl. Vet.-Akad. Forhandl. 14: 351; 1860, Dipt.-
fauna. Siidafr. 1: 97.

The combination of colour-characters given in the key distinguish this species from
any other.

6 Head. Frons and face with greyish white tomentum, only slightly darker on frons ; ocellar
tubercle slightly brownish. Hairs mainly pale yellowish, at sides of frons as well as on ocellar
tubercle, but rather sparse. Moustache of pale yellowish hairs, extending almost up to antennae
with a large, ovoid, bare black spot in its dorsal part. Antennae of type D (Text-fig. 4), basal
segments with pale yellowish hairs. Proboscis and palpi black with pale yellowish hairs.
Occiput whitish with very pale yellowish hairs.

Thorax. Mesonotum yellowish grey with median dark brown stripe extending almost to
scutellum, and with lateral brown areas. Scutellum entirely brownish grey. Hairs erect,
moderately long, pale yellowish. Pleura ashy grey, indistinctly brownish in part, hairs and
bristles pale yellowish.

160 H. OLDROYD

Abdomen dorsally with large, membranous areas as in Text-fig. 5 ; brown in colour; lateral
sclerites ashy grey, with rather long, pale yellow hairs. Venter dark grey with rather shorter
pale yellow hairs. Genitalia bright orange, conspicuous ; upper forceps curled, leaf-like (Text-

fig. 19).

a a CAO LTE 2
- pies ad A 22

Fics. 14-22. Genitalia of Rhipidocephala: 14, distincta 3; 15, 16, flavipes $; 17, fulva
3; 18, umbripennis 9; 19, umbripennis g; 20, tenera 3; 21, tenera 2; 22, zumpti 3.

THE GENUS RHEPIDOCE PH ALA 161

Legs. Coxae like pleura. Trochanters and femora shining black except for knees, which are
red, very narrowly so on fore and middle femora, more broadly so on hind femora. Tibiae dull
reddish, a little darker apically. Tarsi black except for base of basitarsus, which is red, more
extensively so on hind basitarsus. All clothing hairs and bristles yellow.

Wings broad, with broadened costal cell. Dark brown. Halteres orange.

Length of body 6 mm.; of wing 6 mm.

2 similar. Ovipositor short, stout, orange (Text-fig. 18).

Holotype in Riksmuseet, Stockholm.
Distribution. CAPE PROVINCE. NATAL.

Rhipidocephala tenera sp. n.

Closely similar in colour to umbripennis, but distinguished in both sexes by the
very different genitalia (Text-figs. 18-21).

6 Head. Frons and face with dull bronze tomentum, only ocellar tubercle and small areas
near antennae darker brown. Hairs yellow, those on ocellar tubercle and adjoining areas of
vertex rather short. Moustache yellow, extending about half height of face, and with a trans-
verse, shining black area above it. Antennae of type C (text-fig. 4), black, basal segments with
pale hairs. Proboscis and palpi black with yellow hairs. Occiput yellowish grey with pale
yellow hairs.

Thorax. Mesonotum yellowish grey with usual pattern of dark brown stripes. Median stripe
short, lateral stripes distinct. Hairs shorter than in umbripennis, yellow. Pleura grey, meso-
pleuron brown, hairs and bristles yellow.

Abdomen dorsally a large membranous area (Text-fig. 5) is yellow. Sclerites also yellow
anteriorly, especially second segment, but become obscured by greyish tomentum on other seg-
ments. Venter yellowish grey. Hairs short, yellow. Genitalia bright orange, but of a different
form from those of umbripennis (Text-figs. 19, 20).

Legs. Coxae like pleura. Femora shining black, red tips a little more extensive than in
umbripennis. Tibiae and tarsi also more extensively red than in umbripennis, only tips of tarsi
darker. Hairs and bristles orange.

Wings broad, costal cell expanded, uniformly dark brown. Halteres orange.

Length of body 5 mm.; of wing 5 mm.

Q similar. Genitalia orange, shape as in Text-fig. 21.

Holotype 3. TRANSvAaL: Skukuze, 23.xi.1959 (fF. Zumpt). In South African
Institute for Medical Research, Johannesburg.

Paratypes: TRANSVAAL: Skukuze, 23.xi.1959 (F. Zumpt), 3 3, 4 &.

Rhipidocephala zumpti sp. n.

Unique in the genus because of its ashy grey mesonotum, in which the stripes,
though faintly visible, are almost completely obscured. Legs black but moustache
white. The strongly curved male genitalia (Text-fig. 22) are also unique.

6 Head. Face and frons covered with white tomentum, only a little more brownish near
antennal bases and on ocellar tubercle. Hairs of tubercle and sides of vertex short, yellow.
Face with a large, semi-circular bare area, shining black, in its upper half; moustache pale
yellow, thin and straggly but extending over black area as well as over tomented area of epistoma.
Antennae type D (Text-fig. 4), black, basal segments with yellow hairs. Proboscis and palpi
black with yellow hairs. Occiput ashy grey with pale yellow hairs.

162 He. -OLDROYD

Thorax with dense light grey tomentum, and only the merest indication of darker stripes ;
pale yellow bristles dense, long, erect, curved. Pleura grey, mesopleuron more brownish, hairs
pale yellowish.

Abdomen dorsally membranous (Text-fig. 5), dull orange; lateral and ventral sclerites grey-
brown, with short, pale yellowish hairs. Male genitalia orange, large, strongly curved (Text-fig.

22).

Legs black, narrowly red at knees, but on tibiae, not on femora. Clothing hairs and bristles
pale yellow.

Wings moderately broad, a little expanded in costal cell, uniformly dark brown. MHalteres
orange.

Length of body 6 mm. ; of wing 6 mm.
2. Not yet known.

Holotype g. TRANSVAAL: Sabie, i.1952 (Ff. Zumpt). In South African Institute
for Medical Research, Johannesburg.

Rhipidocephala caffra (Macquart)

Discocephala caffra Macquart, 1846, Dipt. exot. Suppl. 1:70; Loew, 1860 Dipt.-fauna. Sudafr.
1s 97:

Rhipidocephala caffra (Loew) [err. pro Macquart] Hermann, 1926, Verh. zool.-bot. Ges. Wien 74:
178.

This is perhaps the most distinctive species of Rhipidocephala, recognized not only
by its dark wings and entirely black legs—which it shares with several other species,
including wmbripennis Loew—but also by the form of the antennae (Text-fig. 48).
The second microsegment is relatively long and hairy, and bears a dorsal spine.
Some authors (e.g. Hull, 1962: 64) have attempted to make this type of antenna
diagnostic of the genus Rhipidocephala, but it is not possible to do so.

$ Head. Tomentum of frons velvety dark brown, with a yellow area on each side of ocellar
tubercle, and a median black line. A tuft of fine yellow hairs on each side of vertex. Ocellar
tubercle black, with numerous long yellowish hairs. Facial tomentum more greyish, with
moustache of black-brown hairs occupying a large, cordiform, bare black area. Proboscis and
palpi black with black-brown hairs. Occiput with white tomentum on eye-margin, more brown
posteriorly, where hairs are yellowish, merging ventrally into a yellowish white beared. Antennae
(Text-fig. 4) of type B, black, with hairs black and yellowish, only first segment perhaps a little
reddish ; second microsegment relatively long, with dorsal notch and spine.

Thorax ashy greyish brown with only indistinct traces of two longitudinal stripes and of dark
areas above wings ; humeri, postalar calli and scutellum of same greyish coloration. Hairs of
mesonotum and scutellum long, fine. Pleura with ashy grey tomentum, and yellow hairs ;
mesopleuron with numerous hairs, both dorsally and posteriorly ; metapleuron with a vertical
band of strong yellow bristles.

Abdomen very incompletely sclerotized on mid-dorsal area ; transverse bridges at base of first
segment and middle of second (Text-fig. 5). Membranous areas brown, sclerotized areas dull
blackish, with narrow yellowish posterior margins on posterior segments. Long yellowish hairs.
Venter similar, but entirely sclerotized. Male genitalia orange, often more or less concealed.
(Text-figs. 23-25).

Legs black, with yellow clothing hairs and bristles.

Wings dark brown to the naked eye, sometimes paler by transmitted light, but only indistinctly
paler apically.

Length of body 7 mm.; of wing 7 mm.

2 closely similar; no ovipositor, only quadrate lamellae visible. Sometimes, especially in
males, the hairs of the head and mesonotum may be black-brown, or even blackish.

THE GENUS RALPIDOGEPHALA 163

Holotype in Muséum National d’Histoire Naturelle, Paris.
Distribution. CAPE PRovINCE. NaTAL. PORTUGUESE East AFRICA.

Rhipidocephala morio Hermann

Rhipidocephala morio Hermann, 1926, Verh. zool.-bot. Ges. Wien 74 : 180.

One of two species with dull, dark brown thorax, and hardly any trace of stripes,
morio is distinguished from caffra by the more strongly convex thorax and the
shorter antennae. R. caffra is a S. African species, and R. mortio is a species of the
southern Sahara, from Abyssinia (type locality) westwards across to the Gambia.
All the western specimens have the bristles paler than those described by Hermann,
but a female from the Katanga has dark bristles.

6 Head. Frons and face strongly contrasting ; frons and ocellar tubercle dark brown, with
yellow hairs on tubercle and on sides of vertex ; face with whitish tomentum, leaving an ellip-
tical, bare, black patch above sparse yellow bristles of moustache. Antennae of type B (Text-fig.
4), black basal segments with yellow hairs [black according to Hermann]; proboscis and palpi
black with yellow hairs. Occiput grey along eye-margins, brown in centre, with rather incon-
spicuous pale yellow hairs.

Thorax. Mesonotum does have usual pattern of dark stripe flanked by two dark spots, but
this is obscured, usually almost completely, by dark brown tomentum ; sometimes pattern
moderately visible ; lateral and posterior margins, including scutellum brown. Hairs dense,
long, erect, yellow-brown. Pleura dark brown, hairs yellow-brown.

Abdomen dorsally membranous (Text-fig. 5), whitish ; sclerotized areas dark chocolate-brown,
with grey hind margins and large grey lateral spots. Venter grey. Hairs pale yellowish, longer
laterally and ventrally. Male genitalia black (Text-fig. 26).

Legs black with yellow clothing hairs and long yellow bristles.

Wings only moderately broad and slightly expanded in costal cell. Uniformly dark. Halteres
orange.

Length of body 5 mm.; of wing 5 mm.

Q quite similar. Ovipositor short, black. (Text-fig. 28).

Holotype ? in Vienna. It was not in the material kindly lent to me by Prof. Max
Beier. Hermann’s description agrees with the specimens described above, except

that he consistently mentions darker, browner hairs.

Distribution. Type locality is AByssINIA: Harar. Specimens seen by me from
S. NIGERIA, GHANA & GAMBIA. A single 2 in the Musée R. de 1’Afrique centrale,
Tervuren is from KATANGA: Mulungivishi, i.1931 (G. F. de Witte), and has dark
hairs as described by Hermann.

Rhipidocephala signata (Hermann)
Holcocephala (Discocephala) signata Hermann, 1907, Z. syst. Hymenopt. Dipterol. 7: 11.
A dark brown species with black legs, distinguished by its inconspicuous yellow
male genitalia, and by the thoracic pattern, with a short, detached, black-brown
median stripe just before scutellum.

6 Head. With dense tomentum, more brownish on frons, more greyish on face. Facial bare
area kidney-shaped, transverse, extended towards mouth margin by a fine line. Face and frons
rather bare of soft hairs; bristles of ocellar tubercle and of moustache stiff, bristly, and of a

164 H. OLDROYD

black-brown, shifting colour, which Hermann calls pechbyaun. Antennae black: second seg-
ment of style long and with a dorsal style (type B, Text-fig. 4). Palpi and proboscis black, with
pechbraun hairs. Occiput white on eye-margins, brown elsewhere, with black hairs.

Thorax. Mesonotum brownish grey with distinct dark brown pattern; a pair of parallel
median stripes terminate just behind suture, and are succeeded by a single, short, median stripe
reaching about to scutellum ; laterally three spots, one near humeral lobe, and other two centred
on transverse suture. Scutellum grey. Hairs dark brown, erect, fine, no strong bristles.
Pleura mainly grey, but triangular area of more yellow-brown tomentum extends from base of
fore coxa to include entire mesopleuron. Hairs and bristles long, yellow.

Abdomen. Dorsum extensively excavated, with the usual transverse bar across the second
segment ; sclerotized areas yellowish grey, with whitish segmentations and yellow hairs and
bristles ; venter similar. Male genitalia yellow, inconspicuous, almost concealed beneath
seventh tergite, which is also yellow.

Legs shining black, covered with yellow hairs and bristles.

Wings blackish, a little paler apically, and with paler centres in a few of the cells. Halteres
yellow.

Length of body 6 mm.; of wing 5 mm.

Q similar. Ovipositor also yellow, short, and mostly concealed beneath a yellow seventh
segment.

Holotype 3 in Naturhistorisches Museum, Vienna, kindly lent to me by Prof. Max
Beier.

Distribution. CAPE PROVINCE: Willowmore (Dr. Brauns), 3 holotype, 1 9
paratype in Vienna. A series of specimens in the British Museum, determined as
signata Hermann by Engel, have quite different, very prominent, male genitalia, as
well as differences in thoracic pattern, and are described elsewhere in this paper as
engelt sp. n.

Rhipidocephala congoiensis sp. n.

A widespread species of the Congo Basin, distinguished from signata Hermann and
scutata sp. n. by the different pattern of thoracic stripes ; characteristic of congotensis
is the extension of the lateral stripes beyond the median stripe.

36 Head. Frons and face with uniform dull golden tomentum, ocellar tubercle brown. Hairs
of tubercle yellow ; no hairs on frons proper. Face with a large, ovoid, bare, shining black spot,
placed dorsally to a sparse yellow moustache. Antennae of type D (Text-fig. 4), but final
segment tapering more, and with longer black hairs; two basal segments with yellow hairs.
Proboscis and palpi black with yellow hairs. Occiput grey with yellow hairs.

Thorax. Mesonotum grey with very distinct brown pattern : median stripe cut short before
reaching scutellum, but lateral stripes unusually long, prolonged by prescutellar spots. Hairs
erect, pale yellow. Pleura ashy grey and brown, brown especially on mesopleuron ; hairs pale
yellow.

Abdomen. Dorsum membranous (Text-fig. 5), orange. Sclerites brown with orange hind
margins, first two more generally orange. Hairs golden yellow. Venter brown with brassy
tomentum and golden hairs. Male genitalia (Text-figs. 29, 30) yellow, fairly prominent.

Legs all black except very narrowly at knees, covered with dense yellow hairs and long yellow
bristles.

Wings only moderately broad, and costal cell not appreciably expanded, uniformly dark brown.
Halteres orange.

Length of body 5 mm.; of wing 5 mm.

Q similar. Genitalia compact, orange.

DHE GENUS RALEPIDOCE PH AEA 165

Holotype 3. Conco: Kasai, 1928 (Dr. Walker) in Musée R. de l'Afrique Centrale,
Tervuren, Belgium.

Paratypes. Conco: Kasai, 1928 (Dr. Walker), 3 3, 3 2; Dolo, xi.1rg12 (F.
Chaltin), 4 3,1 9; Lomami, Katompe, 12.xiii.1923 (Dr. M. Bequaert), 2 9.

Fics. 23-31. Genitalia of Rhipidocephala: 23-25, caffra $; 26, morio $; 27, scutata 3;
28, morio 2; 29, 30, congoiensis g; 31, engeli 3.

166 H. OLDROYD

Rhipidocephala scutata sp. n.

Distinguished from congoiensis sp. n. by the thoracic pattern, in which the stripes
are heavily marked, and coalesced in the anterior half of the thorax ; posterior half
grey, with long yellow hairs. rons and face sharply differentiated in colour.

6 Head. Frons with yellow and brown tomentum, ocellar tubercle dark brown ; hairs short,
yellow, confined to tubercle. Face with brassy yellow tomentum and an oval, bare spot dorsal
to sparse yellow moustache. Antennae of type D (Text-fig. 4), black, basal segments with yellow
hairs. Occiput grey on eye-margins, brown centrally, with yellow hairs.

Thorax. Mesonotum densely covered with tomentum, which is pale yellow laterally and
posteriorly, including scutellum, stripes dark brown, heavily marked, tending to coalesce.
Hairs long, especially posteriorly, yellow. Pleura yellowish grey, with extensive dark brown
patches, especially on sutures and on anterior of mesopleuron. Sparse hairs yellow.

Abdomen. Dorsum extensively membranous (Text-fig. 5), dull orange; sclerites ashy grey
with yellow hairs. Venter blackish, with light yellow hairs. Genitalia black, protruding, but
rather small, clasper with its spines well displayed (Text-fig. 27).

Legs black with yellow hairs and bristles.

Wings dark brown, moderately broadened, costal cell a little expanded.

Length of body 5 mm.; of wing 5 mm.

Q similar. Ovipositor formed from a stout eighth segment, dully shining.

Holotype 3. CONGO: GARAMBA National Park, 3449, II/gd/4, 8.v.1952 (De
Saeger). In Institut des Parcs Nationaux du Congo, Brussels.

Paratypes: Conco: Parc National du Garamba, Mission H. De Saeger, 3440,
II/gd/4, 8.v.1952 (De Saeger), r 3, 1 2; 1887, II/gd/7, 8.vi.1951r (De Saeger),
IQ; 1588, II /hc/4, 20.iv.1961 (Verschuren), 10 3, 49; 1824, II/fd/27, 28.v.1951
(De Saeger), I 3; 4609, I/a/1, 1.v.1950 (Demoulin), 1 2; 3678, Ndelele, 4, 18.vi.52
(De Staeger), I 3, 22; 3323, Pidigala, 23.iv.1952 (De Saeger), 2 3, in Institut des
Parcs Nationaux du Congo, Brussels.

Kivu: Uvira, xi.1922 (Ch. Seydel), 3 3, 4 2; 16-23.11.1953 (P. Basilewsky),
2 3, 3 8, in Coll. Musée de l'Afrique centrale, Tervuren, Belgium.

Rhipidocephala engeli sp. n.

This is a very distinctive species, with characteristic male genitalia (Text-fig. 31).
The series in the British Museum was previously identified by the late Dr. E. O. Engel
as signata Hermann, but the material lent to me from Vienna shows it to be quite
distinct.

A grey species with light brown thoracic markings, concentrated especially pos-
teriorly (including scutellum) ; genitalia in both sexes yellow, especially prominent
in male (Text-fig. 31).

6 Head. Frons and face with grey tomentum and a bar of dark brown tomentum through
bases of antennae. A large, ovoid, shining bare spot underlies most of moustache, which reaches
nearly to antennae. Hairs of frons and of moustache pale yellow. Antennae of type D (Text-
fig. 4), black, basal segments with yellow hairs. Palpi and proboscis black, with pale yellow
hairs. Occiput white with pale yellow hairs.

Thorax. Mesonotum densely tomented, mainly dark brown, including scutellum: in well
preserved specimens three stripes can be seen, and anterior third of mesonotum, as well as lateral

THE GENUS RATPIDOCEPHALA 167

margins, grey: but in other specimens entire mesonotum dull blackish brown. Hairs pale
yellowish, erect, rather long. Pleura grey, mesopleuron partly brownish, hairs and bristles pale
yellowish.

Abdomen dorsally membranous (Text-fig. 5), dark brown, sclerotized areas with grey tomentum
and long, pale hairs; venter similar; both dorsally and ventrally hind margins of segments
indistinctly yellowish. Male genitalia bright orange, relatively huge, about half as long as
abdomen, with prominent lobes on upper forceps (epandrium) (Text-fig. 31).

Legs black with yellow clothing hairs and yellow bristles.

Wings dark brown, expanded, costal cell broadened.

Length of body 5 mm.; of wing 5 mm.

Q similar. Eighth segment stout, yellow, forming short ovipositor.

Holotype g. NatTat: Weenen (H. P. Thomasett). In B.M. (N.H.).
Paratypes: NataL: Weenen (H. P. Thomasett), 6 3, 13 9.

Rhipidocephala fimbriata sp. n.

A distinctive species, with coarse yellow hairs abundant on head and thorax; a
shining black abdomen with very prominent lateral fringes of yellow hairs ; black
femora and orange tibiae ; and genitalia as shown in Text-figs. 32, 33.

6 Head. Frons and face with grey tomentum. On frons it is denser and allows bare, shining
ground colour to show through ; long, yellow hairs abundant laterally and on ocellar tubercle.
Face entirely tomented, without any bare area; moustache of abundant, coarse yellow hairs
extends to bases of antennae. Antennae of type C (Text-fig. 4), basal segments with abundant
yellow bristles. Palpi and proboscis black, with yellow bristles. Occiput with brownish white
tomentum and with abundant, strongly curved, yellow hairs.

Thorax. Mesonotum completely tomented, ashy grey, with usual pattern of black-brown
stripes. Scutellum entirely grey. Mesonotum and scutellum covered with very long, erect,
curved yellow bristles. Pleura black, covered with brassy yellow tomentum and with bright
yellow hairs. Hairs of pronotum also abundantly long, erect, yellow.

Abdomen black, dorsally shining through thin tomentum ; entirely sclerotized, and with no
dorsal membranous areas; dorsally with many short and sparse yellow hairs, but laterally and
ventrally with a broad band of yellowish tomentum, covered with long, golden, bristly hairs.
Male genitalia bright orange, prominent, upper forceps hoodlike, as shown in Text-fig. 33.

Legs. Coxae like pleura. Femora and tarsi shining black with thin yellow tomentum and
yellow hairs and bristles ; tibiae bright orange.

Wings broad, costal cell expanded, entirely dark brown. Halteres orange.

Length of body 8 mm. ; of wing 8 mm.

Qsimilar. Large and shining black eighth segment forms a conspicuous ovipositor (Text-fig.
32).

Holotype 3. CAPE PRovVINCE : Wellington, Witte R., 1922 (Laurence). In the B.M.
(N.H.).

Paratypes : CAPE PROVINCE: 13 m. N.E. of Touws R., 26.x.1938 (R. E. Turner),
29; Ceres, xi.1920 (R. E. Turner), 19; Montagu, x.1919 (R. Tucker), 1 2; Cape
Town, Cape Point, 15-21.xi.1930 (H. W. Simmonds).

Rhipidocephala manicata sp. n.

_ Another species of the group with no facial bare area, and with little or no mem-
branous area dorsally on the abdomen. LEasily distinguished from fimbriata sp. n. by

168 H. OLDROYD

the much less conspicuous lateral fringes of the abdomen, and by the male genitalia
(Text-figs. 33, 34).

6 Head. Frons and face with dense, yellowish grey tomentum; only a small black spot on
frons is bare and shining ; ocellar tubercle with brown tomentum and brown hairs, brown hairs
also laterally on frons. Face without any bare area, and with a moustache of shining brown
hairs extending almost to antennae. Antennae of type D (Text-fig. 4) ; hairs of basal segments
brown. Palpi and proboscis black with brown hairs. Occiput with white tomentum and dense,
silky brown hairs.

Thorax. Mesonotum ashy grey, with rather indistinct brown pattern. Clothing hairs erect,
but only moderately long, brown. Pleura with grey tomentum, brown on mesopleuron ; hairs
silky, brown, dense all over mesopleuron, on metapleuron, and on prothorax.

Abdomen dorsally with some membranous areas, but less so than usual in this genus ; a trans-
verse area on boundary between first and second segments dull red, and obscure areas on other
segments slightly so; sclerotized areas black, shining through very thin brown tomentum, and
with bristly brown clothing hairs covering whole area, longer and stronger laterally, long and
fine ventrally. Male genitalia (Text-fig. 34) red and black, prominent.

Legs. Coxae like pleura. Femora shining black, covered with short yellow bristles ; tibiae
red with black tips ; tarsi black, hairs and bristles concolorous.

Wings broad, costal cell broadened, brown. Halteres yellow.

Length of body 8 mm. ; of wing 8 mm.

Q. Not yet known.

Holotype g. NAMAQUALAND: Springbok (R. Lightfoot). In B.M. (N.H.).

Rhipidocephala speciosa sp. n.

Distinguished from manicata sp. n. by the blackish rather than ashy appearance,
and in particular by having silky yellow hairs, in place of the rather coarse brown
hairs of manicata.

36 Head. Tomentum of frons and ocellar tubercle dark brown, contrasting sharply with dense
yellow tomentum of face. Ocellar tubercle and sides of vertex with yellow hairs ; face entirely
without bare area, and with a rather diffuse moustache covering most of facial area. Antennae
of type C (Text-fig. 4), black, basal segments with yellowish hairs. Occiput with yellowish white
tomentum and dense, curved, yellow hairs.

Thorax. Mesonotum black-brown, with mere traces of longitudinal dark stripes ; humeri,
lateral margins and scutellum covered with yellow tomentum. Pleura with uniformly yellowish
white tomentum. Mesopleuron and metapleuron each with yellow hairs or bristles.

Abdomen almost completely sclerotized, with merest indications of transverse membranous
slit between segments 1 and 2. Black, with very fine yellow hairs which become coarser laterally.
Venter with yellowish tomentum and long, silky yellow hairs. Male genitalia yellow, prominent,
hood-like (Text-fig. 36).

Legs relatively stout for the genus. Femora, tips of tibiae, and tarsi black; tibiae yellow
except for tips. Hairs and bristles yellow, even on tarsi.

Wings broad, with expanded costal cell, uniformly dark brown, scarcely paler at tip. Halteres
yellow.

Length of body 6mm.; of wing 6 mm.

Q. Not yet known.

Holotype g. CAPE PRovINCE: Paarl, 24.x.1954 (C. H. Andrewes). In B.M.
(N.H.).

THE GENUS RATPIDCOCEPHALA 169

Rhipidocephala doornensis sp. n.

Among those species with entirely tomented face, doornensis sp. n. is unusual in
having the dorsum of the abdomen fully membranous as in Text-fig. 5. It differs
from speciosa sp. n. in this, and in having the legs entirely black.

¢ Head. Frons and face entirely tomented ; frons a little darker brown than face, which is
greyish at sides, more brownish in area of moustache, but without any distinct bare area.
Ocellar tubercle and lateral margins of frons with yellow-brown hairs and bristles. Moustache
yellow-brown. Antennae of type C (Text-fig. 4), black, first two segments with yellow hairs.
Proboscis and palpi black with yellow hairs. Occiput grey with white hairs.

Thorax. Mesonotum black-brown, without pattern; scutellum with a narrow grey rim.
Long, erect, pale hairs yellowish. Pleura with grey tomentum and silvery white hairs on meso-
pleuron and before halteres.

Abdomen black-brown. Membranous area fully developed as in Text-fig. 5. Sclerites with
whitish tomentum and pale yellow hairs. Venter similar, hairs dense. Male genitalia orange
(Text-fig. 35).

Legs black with white hairs and yellow bristles.

Wings dark brown, costal cell broadened. MHalteres yellow.

Length of body 5 mm.; of wing 5 mm.

2 not yet known.

Holotype g. CAPE PRovINCE: Doorn River, 3.xi.1931 (Miss A. Mackie). In
B.M. (N.H.).

Rhipidocephala quadrifaria Hermann
Rhipidocephala quadrifaria Hermann, 1926, Verh. zool.-bot. Ges. Wien 74: 179.

Distinguished, as Hermann pointed out, by the contrast between the partly shining
dark crossband of the frons and the entirely tomented face, where darker tomentum
replaces the clear spot that is common in this genus.

Q Head with sharp contrast between frons and upper part of face ; frons partly shining black
through thin dark brown tomentum, especially in a transverse band between ocelli and antennae.
Face entirely tomented, brown on area covered by moustache, white above and laterally ; no
bare spot. Hairs black-brown. Antennae of the form of umbripennis (type D, Text-fig. 4),
black, with black-brown hairs. Proboscis and palpi black with black-brown hairs. Occiput
white on eye-margins, darker elsewhere, with brown hairs.

Thorax. Mesonotum with brown tomentum and only indistinct traces of black-brown
pattern ; scutellum dark brown with grey rim. Hairs black. Pleura more greyish, meso- and
metapleuron yellowish, hairs yellowish, some rather bristly.

Abdomen black : dorsal membranous areas blackish, only a narrow, transverse slit on margin
of segments 1-2 is yellow. Sides with denser grey tomentum and pale yellowish hairs. Ovi-
positor (Text-fig. 38) fairly long, shining black dorsally on seventh and eighth segments.

Legs shining black with yellow hairs and bristles.

Wings uniformly smoky-brown, only a little darker basally. Halteres yellow.

Length of body 5 mm. ; of wing 5 mm.

3. Not yet known.

Holotype in Naturhistorisches Museum, Vienna, lent to me for study.

Distribution. CAPE Province: Algoa Bay, 16.xi.82 (Dr. Brauns).

170 H. OLDROYD

Rhipidocephala angustior sp. n.
Discocephala analis Macquart; Hermann, 1926, Verh. zool.-bot. Ges. Wien 74:177. [mis-
identification]

The following description is based upon the male specimen in the Naturhistorisches
Museum, Wien, which was kindly lent to me by Prof. Max Beier. Hermann wrote of
this specimen that although it had no locality label, and despite certain differences, he
felt convinced that it was correctly identified as being Macquart’s analis. It seems
to be conspecific with a female in the British Museum from S. Africa, which I had
already set aside as an undescribed species.

Fics. 32-38. Genitalia of Rhipidocephala: 32, fimbriata 2; 33, fimbriata 3; 34, manicata 3;
35, doornensis 3; 36, speciosa 3; 37, angustior 3; 38, quadrifaria Q.

THE GENUS RAIPIDOCEPHALA 171

Among the species that have the male epandrium and the female ovipositor
extended horizontally, angustior is distinguished by the leg colour, and by not having
the fore margin of the wing expanded by broadening of the costal cell. In both sexes
it is generally more furry than related species.

6 Head. Frons and face entirely tomented : frons dark dully shining black through brown
tomentum, and with abundant yellow hairs, not only on ocellar tubercle but on sides of frons as well ;
facial tomentum white above and laterally more brown beneath moustache, which is dense and
dark yellow. Antennae black, type C (Text-fig. 4), basal segments with yellow hairs. Palpi
and proboscis black with long yellow hairs [Macquart says: palpes fauves]. Occiput white on
eye-margins, brown elsewhere, with abundant yellow hairs.

Thorax. Mesonotum grey with fairly distinct black-brown stripes, narrow median, tapering
posteriorly, and lateral spots rather restricted in area. Posterior mesonotum and scutellum
grey, latter with dense white rim. Hairs yellow, very long, erect and dense, giving thorax a
furry appearance. Pleura grey, mesopleuron more brown, hairs pale yellowish or white.

Abdomen dorsally shining black, but entirely sclerotized, with no membranous area except for a
narrow, yellow, transverse groove on segments 1-2. Dorsum of abdomen shining black, extreme
sides narrowly margined with yellow tomentum; venter with yellowish tomentum. Hairs
dense, fine, yellow, rather shorter dorsally, longer laterally and ventrally, giving a furry appear-
ance. Male genitalia prominent, orange, epandrium extending posteriorly as two large, pointed
lobes (Text-fig. 37).

Legs. Coxae like pleura. Femora shining black, red only at extreme base; tibiae and tarsi
orange, each segment narrowly black at tip. Hairs and bristles of legs yellow.

Wings pale smoky, darker brown basally, and especially in costal cell, which is not broadened
as it is in several related species.

Length of body 7 mm. ; of wing 6 mm.

? similar, except that hairs are greyish rather than deep yellow. Ovipositor broad, flattened,
bare and shining, eighth tergite being continuous with shining median band of abdomen.

Holotype 9. CAPE PROVINCE: Ceres, ix.1920 (R. E. Turner), In B.M. (N.H.).
I make one of the females holotype because the only male, the one seen by Hermann,
is without even a continental locality.

Paratypes : CAPE PROVINCE: Ceres, I-12.xi.1924 (R. E. Turner), 19; I 3 para-
type, without locality, in the Vienna Museum.

REFERENCES
BRoMLEy, S. W. 1930. New robber flies from Madagascar. Bull. Brooklyn ent. Soc. 25:
283-290.
CaRRERA, M. 1958. Asilidae (Diptera) dacoleg¢4o Seabra. Archos Zool. Est.S. Paulo 11: 147-
170.

HERMANN, F. 1926. Der Verwandschaftskreis des gen. Holcocephala Jaennicke. Verh. zool.-
bot. Ges. Wien 74 : 153-191.

Hui, F.M. 1958. Some genera and species of the family Asilidae (Diptera). Proc. ent. Soc.

Wash. 60 : 251-257.

1962. Robber flies of the world. 2 vols. Washington.

JANSSENS, E. 1951. Un nouveau genre de diptére Asilidae: Owynoton, n.-g. Bull. Inst. r.

Sci. nat. Belg. 27 (54) 4 pp.

1953. Contribution a l’étude des diptéres de l’Urundi. Bull. Inst. r. Sci. nat. Belg.

29 (42), 15 pp.

Macguart, P. J. M. 1846. Diptéves exotiques. suppl. 1: 69-70. Paris.

172 H. OLDROYD

OLpRoyD, H. 1959. Synopsis des Asilidae de Madagascar. Mém. Inst. scient. Madagascar

11 : 247-319.

1963. The tribes and genera of the African Asilidae. Stuttg. Beitr. Naturk. 107, 16 pp.

PRITCHARD, A, E. 1938. Synopsis of North and Central American Holcocephala, with a des-
cription of a new species (Diptera: Asilidae). J. New York ent. Soc. 46: 11-21.

ys

A LIST OF SUPPLEMENTS
TO THE ENTOMOLOGICAL SERIES
OF THE. BULLETIN ‘OF
THE BRITISH MUSEUM (NATURAL HISTORY)

I. MASNER, L. The types of Proctotrupoidea (Hymenoptera) in the British
Museum (Natural History) and in the Hope Department of Entomology, Oxford.
Pp. 143. February, 1965. 5.

2. Nixon, G. E. J. A reclassification of the tribe Microgasterini (Hymenoptera :
Braconidae). Pp. 284 ; 348 Text-figures. August, 1965. 6.

3. Watson, A. A revision of the Ethiopian Drepanidae (Lepidoptera). Pp. 177;
18 plates, 270 Text-figures. August, 1965. £4 4s.

4. SANDS, W. A. A revision of the Termite Subfamily Nasutitermitinae (Isoptera,
Termitidae) from the Ethiopian Region. Pp. 172; 500 Text-figures. October,
1965. {£3 5s.

5. AHMAD, I. The Leptocorisinae (Heteroptera: Alydidae) of the World. Pp. 156;
475 Text figures. November, 1965. {2 I5s.

6. OxapA, T. Diptera from Nepal. Cryptochaetidae, Diastatidae & Drosophilidae.
Pp. 129; 328 Text-figures. 3.

7. GILIOMEE, J. H. Morphology and Taxonomy of Adult Males of the Family
Coccidae (Homoptera: Coccoidea). In press.

8. FLETCHER, D. S. A revision of the Ethiopian species and a check list of the
world species of Cleora (Lepidoptera: Geometridae). In press.

PRINTED IN GREAT BRITAIN BY ADLARD & SON LIMITED, BARTHOLOMEW PRESS, DORKING

A LIST OF THE ODONATA TYPES
DESCRIBED BY F. C. FRASER,
NOW IN THE BRITISH MUSEUM
(NATURAL HISTORY)

D. E. KIMMINS

BULLETIN OF
THE BRITISH MUSEUM (NATURAL HISTORY)

ENTOMOLOGY Vol. 18 No. 6
LONDON: 1966

29 JUN 364)
Robs Or tie DONA TA EYPES DESCRIBED Sy,
BY. © GC. PRASER.NOW. IN THE
BRITISH MUSEUM (NATURAL HISTORY)

BY

D. E. KIMMINS

British Museum (Natural History) BY Ar

)

Pp. 173-227

BULLETIN OF
THE BRITISH MUSEUM (NATURAL HISTORY)

ENTOMOLOGY Vol. 18 No. 6
LONDON : 1966

THE BULLETIN OF THE BRITISH MUSEUM
(NATURAL HISTORY), instituted in 1949, 1s
issued in five series corresponding to the Departments
of the Museum, and an Historical series.

Parts will appear at trregular intervals as they become
veady. Volumes will contain about three or four
hundred pages, and will not necessarily be completed
within one calendar year.

In 1965 a separate supplementary series of longer
papers was instituted, numbered serially for each
Department.

This paper is Vol. 18, No. 6 of the Entomological
series. The abbreviated titles of periodicals cited
follow those of the World List of Scientific Periodicals.

© Trustees of the British Museum (Natural History) 1966

TRUSTEES OF
THE BRITISH MUSEUM (NATURAL HISTORY)

Issued 28 June, 1966 Price £1

me LIST OF THE ODONATA. “ly PES” DESCRIBED
BY FC” FRASER NOW IN THE

BRITISH MUSEUM (NATURAL HISTORY)
By D. E. KIMMINS

CONTENZES
Page
INTRODUCTION . F ‘ : ; j : ‘ : : : 175
List OF TyPEs IN B.M. (N.H.) . ; : ‘ . ; - ‘ 176
REFERENCES . P : < - i F : ‘ 7 219
APPENDIX. LOCATION OF TYPES NOTIN B.M. (N.H.) . ; ‘ : 223
SIN OS'S

The types or syntypes of 379 taxa of the species-group are dealt with, of which the types of
19 taxa, which should have been in the Fraser Collection, have not yet been traced. Neotypes
of a few taxa were marked by Fraser to replace types known to be lost or destroyed. Lectotypes
are designated for 108 taxa.

INTRODUCTION

In 1963, the British Museum (Natural History) received the important collection
of Odonata bequeathed by Lt.-Col. F. C. Fraser. Since this collection contained
the types of nearly two-thirds of the species he had described, apart from those which
he had already presented to the BM (NH), it seemed desirable to list those which are
now there, as well as others which should have been in Fraser’s collection but which
have not as yet been traced.

In 1957 he deposited in the Department of Entomology the greater part of his
Odonata types, on permanent loan during his lifetime, to become the property of
the Museum on his death. This proved to be a fortunate decision, since in the course
of arranging these types certain discrepancies in labelling (locality, date, etc.) came
to notice, and it was then possible to consult Fraser on general details; these dis-
cussions proved valuable in dealing with problems which arose after the remainder
of his bequest reached the Museum.

It may be thought strange that such inaccuracies should have occurred, but
Fraser, as an officer in the Indian Army Medical Service, was naturally subject to
posting from one place to another at intervals, and his entomological work was only
a hobby. In 1939 his collection suffered a more serious disturbance when, as a
wartime precaution, he unmounted and re-papered most of his type-specimens.
After the war, many of these were again relaxed and re-pinned and Fraser agreed
that errors of re-labelling had occurred in a number of cases during that period.
Cases were also found where he had placed type-labels on specimens in his collection,

ENTOM. 18, 6. 14

176 D. E. KIMMINS

although the actual types had already been presented to the BM (NH). Errors in
locality labels were also discovered involving the types already presented to the
BM (NH) by Fraser. Much of this material was in paper triangles when presented
and errors have been due to faulty transcription of data written (not always very
legibly) on the paper triangles. Where it is evident that such errors had occurred,
new labels have been written, based upon the published data.

In common with other workers of his time, Fraser did not always mention or
indicate a type-specimen in his earlier works. In preparing this paper, I have
endeavoured to trace the first reference which can be taken as an indication or
designation of a type-specimen, since the concept of lectotype was then unknown.
For example, a statement such as “ A pair in the British Museum, the male the type ”’,
made subsequent to the presentation of the specimens to the BM (NH), has been
accepted as equivalent to designation of lectotype and the requisite reference has
been quoted after the word Lectotype. Where LECTOTYPE is used, it denotes
present designation, and in the absence of evidence to the contrary, I have accepted
the specimen marked Type by Fraser as lectotype.

A number of Fraser’s types, originally deposited in the collection of the Department
of Agriculture at Pusa, were subsequently given to the then Imperial Bureau of
Entomology and presented by that body to the British Museum (Natural History).
These are indicated as “Ex Pusa Collection ”’.

The recognition of the type-specimens of some taxa has been made difficult by the
fact that, in some of Fraser’s earlier species subsequently placed in synonymy, he
apparently replaced the original determination labels with new ones, omitting any
reference to the original name. Such types have in some cases only been recognizable
by locality data or by some special individuality in structure.

In this paper, the names of the taxa of the species-group are listed alphabetically.
Following the specific name is the original generic name and the reference by year
and page. The status of the type is followed by the data on the various labels. In
the case of manuscript labels, the authorship is indicated by initials as follows
(label F.C.F.], [label D.E.K.] and [label T.B.F.], T. Bainbrigge Fletcher. Following
the label data are any comments concerning these and finally the present status of
the taxon, where it differs from the original combination. The names of taxa for
which no types have been located in the Fraser collection are enclosed in square
brackets.

LIST OF TYPES IN B.M. (N.H.)

abbreviata (ssp. of Rhinocypha biforata), 1928 : 454. Holotype g. Rhinocypha
delimbata abbreviata, Shillong, Assam, 1927, T. B. F. [label F.C.F.]. Now
Heliocypha biforata abbreviata (Fraser).

aculeata (Macromia), 1927 : 68-69. Holotype g. M. aculeata, Maymyo, Upper
Burma, 25.v.25, Col. F. Wall [label F.C.F.].

adami (Ceylonosticta), 1933 : 211, fig. 6. LECTOTYPE g. Type. C. adami,
Madugoda, Ceylon, 5.v.32, F. C. Fraser [label F.C.F.]. Ceylonosticta adami

FRASER ODONATA TYPES IN B.M. (N.H.) 177,

Fras. 3, Lectotype, D. E. Kimmins det. 1964. Of the 7 J and 1 9 syntypes listed,
6 g and r 9 have been traced. Now Drepanosticta adami (Fraser).

adami (Libellago), 1939 : 24, fig. 1. Holotype g. L. adami Fr., Haragama,
Ceylon, 7.v.32, F. C. Fraser coll. [label F.C.F.].

adjuncta (race of Polythore derivata), 1946: 18, fig. 1c. LECTOTYPE J.
Umbria, 19.xi.30. P. derivata race adjuncta [label F.C.F.]. Allotype 2. Same
locality, 23.x1.30.

africanus (Echinopterogomphus), 1926 : 356, figs. I-2, (3); 1928a : 130, fig. 4
(2). Holotype ¢ [imperfect]. West Africa, Sierra Leone, Port Lokko, I.v.1912,
J. J. Simpson. Echinopterogomphus africanus Fraser. Type. [label F.C.F.].
Allotype 2. Lestinogomphus africanus 2 (Fraser), Uganda, Hale Carpenter
(label F.C.F.]. Now Lestinogomphus africanus (Fraser).

albistigma (Ischnura), 1927a : 27-28, fig. 51. Holotype 3. Ischnura albistigma
Fras. g (Type). Coll. P. A. Buxton, Malololelei, Upolu Is., W. Samoa, 7.xi1.25
{label F.C.F.].

albistyla (Gynacantha), 1927 : 75-76. Lectotype ¢ (Fraser, 1936 : 113). Pusa,
5.ix.24, Fletcher coll., in verandah. Gynacantha albistyla Fraser 3 (Type)
{label F.C.F.]. Allotype 9. Pusa, 14.viii.24. Fletcher coll. On ceiling of
bungalow verandah. Gynacantha albistyla Fraser 2 Allotype [label F.C.F.].
Both ex Pusa Collection. Fraser records a single pair in the Pusa Collection, with
date 5.ix.24, but this is surely an error, as the label on which he wrote 2 Allotype
is clearly dated 14.viii.24. Now Acanthagyna albistyla (Fraser).

albofasciata (Lestes), 1943 : 113-115, figs. 1-2. Holotype 3. Lestes albofasciata
Fraser, ¢ Type. Buru. det. F. C. Fraser. [label F.C.F.]. The label bears a
printed date 1957 [one of Fraser’s blank determination labels] which probably is
an indication of the year in which the specimen was relaxed and re-labelled.

ambigua (race of Polythore derivata), 1946 : 19. LECTOTYPE 3. Yumbatos,
Peru, xi.32. P. derivata race ambigua [labels F.C.F.]. Allotype 9. Same
locality, x.32. P. derivata race ambigua [labels F.C.F.]. Specimens labelled
Type and allotype by Fraser.

anascephala (Drepanosticta), 1933c : 114-115, fig. 2. Holotype g. Laos,
Pu Tat. c. 1,200 m., in evergreen forest, 22.iv.1932, Dr. A. Kerr [label D.E.K.]
Drepanosticta anascephala Fraser ¢ Type, D. E. Kimmins det. 1963. The
specimen labelled Type in Fraser’s pinned collection did not agree in either date
or locality with the published description. The true type was located in his
papered collection and now bears my determination label.

[angularis (Lestes), 1929 : 848-849.] The type has not been traced. It is stated to
have been in the Pusa Collection (L. Burma, Minbu, 8.vili.1914), but it was not
listed amongst the Pusa types presented to the BM (NH).

annaimallaicus (race of Lamelligomphus nilgiriensis), 1934 : 279, fig. 83b.
LECTOTYPE g. Type. L. annaimallaicus, Mudis Hills, Coimbatore Dt, v.29,

178 D. E. KIMMINS

F. C. Fraser coll. [label F.C.F.]. Fraser subsequently placed a type label on a 9
from Munar, Travancore, 23.vi.31, S. India, F. C. Fraser [label F.C.F.] and this
may be considered as allotype. It may be noted that the specific name is spelt
“annamallaicus ’’ in the original description but as it is spelt ‘‘ annaimallaicus ”’
in the legend of fig. 83b (two pages earlier), in the index and on the type label,
the spelling over the original description should be considered as a printing error.
Now Onychogomphus nilgiriensis annaimallaicus (Fraser).

annaimallaiensis (Macromia), 1931 : 452-453. Holotype g. Mudis Hills,
12.v.29, S. India, F. C. Fraser [label transcribed by D.E.K.]. M. annaimal-
laiensis ¢ [label F.C.F.]. Allotype 9°. M. annaimallaiensis 2, Mudis Hills,
I2.v.29. S. India, F. C. Fraser [label F.C.F.]._ The holotype lacked a locality
label when received from Fraser, though bearing type and determination labels,
and as only two specimens were taken, the locality label was transcribed from that
of the female.

annandalei (Caconeura), 1921 : 543. Lectotype 3 (Fraser, 1924 : 503). India,
Mahableshwar, 22.iv.1920, 4,500 ft., F. C. Fraser. Caconeura annandalei Fraser,
Type-specimen [label F.C.F.]. Allotype 9°. Same data. Now Prodasineura
verticalis annandalei (Fraser).

anomala (Indolestes), 1946a : 43-44, figs. Ic, 2e. Holotype g. Indolestes
anomala Fraser, Dalat, Siam, 17.x.20, Type [label F.C.F.]. The locality given
in the original description is Doun-moi, without date, but the specimen otherwise
agrees. Now Lestes (Indolestes) anomala (Fraser).

antelopoides (Protosticta), 1931 : 467-468, figs. 5a-c. Holotype g. P. antelo-
poides, Munnar, W. Ghats, Travancore, 30.v.1931, F. C. Fraser [label F.C.F.].
Allotype 2. Data as holotype. Protosticta antelopoides Fras., 9, det. F. C.
Fraser [2 labels D.E.K.]._ This female is referred to as a paratype in the original
description.

apiaensis (Gynacantha), 1927a : 35-36. Holotype g. G. apiaensis, Apia, W.
Samoa, 16.iv.24, J. S. Armstrong [label F.C.F.]. Now Acanthagyna apiaensis
(Fraser).

apicalis (Chloroneura), 1924 : 501-502, fig. 5. LECTOTYPE ¢. S. India,
Coorg, Cauvery River, Fraserpet, 6.v.1923, Maj. F. C. Fraser. C. apicalis Fras.
(label F.C.F.]. Allotype 9. Coorg, 6.xi.23. Fraserpet, Cauvery. C. apicalis 2
(label F.C.F.]. No types were specified in the original description nor in 1933b :
248-250. The above ¢ was sent to the BM (NH) as type. The allotype was so
labelled by Fraser.

apicalis (Gynacantha), 1924a : 83. Holotype 9. Lyallpur, Punjab, October
1921, Dutt coll. Gynacantha apicalis sp. nov. (Type) [label F.C.F.]. Gynacantha
apicalis Fraser. Type 9, Fraser det. 1922 [label F.C.F.]. Ex Pusa Collection.
Now Acanthagyna apicalis (Fraser).

[apicalis (ssp. of Hylaeothemis fruhstorferi), 1924 : 430.] I have been unable to
trace the types of this subspecies, unless the specimens in BM (NH), labelled as

FRASER ODONATA TYPES IN B.M. (N.H.) 179

types, have been given incorrect locality data by Fraser. The status of this
subspecies is uncertain, as Fraser himself omits any reference to it in Fauna of
British India, 1936, but in r946c : 99 he refers to it as ‘“‘race apicalis”’ of
H. indica. See also indica, (Hylaeothemis).

aquicola (Agriogomphus), 19432 : 165. Holotype g. Agriogomphus sylvicola
Selys, Allotype 3, Iquitos, Peru, 10.iii.31, det. F. C. Fraser, 1943 [label F.C.F.].
Agriogomphus aquicolor Fras., g Type, D. E. Kimmins det. 1964. This name
was proposed conditionally by Fraser, and since it is earlier than 1960 (Art. 15 of
Intern. Code), it must be considered as an available name. Fraser (loc. cit.) writes
“Mr. J. E. Roberts, who has seen this supposed male allotype of Agriogomphus
sylvicola from Peru, expresses his opinion that it would not be safe to associate it
with the female A. sylvicola in view of the differences of locality, and suggests that
a name should be given to it, which may be confirmed in the future if the male
should be found to belong to different species ; in adopting his advice, I suggest
the name Agriogomphus aquicola.”” The specific name aqguicola was included in the
Zoological Record for 1943 as a new specific name and I have therefore placed my
determination label on the specimen, so that it shall not be overlooked.

arachnoides (Pseudocopera), 1922 : 56, pl. 7, fig. 4. Holotype g. Margherita,
I4-19.v.1920, Fletcher Coll. Margherita, 15.v.1920, Copera arachnoides Fraser,
Type ¢ [labels T.B.F.]. Ex Pusa Collection. Syn. of Platycnemis annulata
(Selys).

armageddoni (Chlorocypha), 1940 : 552, figs. 1, 3. LECTOTYPE g. East
Africa, Buganda, 1927, G. Hale Carpenter. C. armageddoni, Type 3 [label
F.C.F.]. Allotype 9. Same locality data. C. armageddoni, Allotype 2 [label
F.C.F.]. Syn. of Platycypha lacustris (Foerster).

armstrongi (Amorphostigma), 1925 : 433-434, fig. 3. LECTOTYPE 4, allotype
9. W. Samoa, Apia, 28.x.1923, J. S. Armstrong, g¢ in coitu. Amorphostigma
armstrongi Fraser, Types.

armstrongi (ssp. of Rhyothemis regia), 1956 : 328, fig. 5. This subspecific name
was proposed by Fraser for Rhyothemis regia exul Ris; Fraser, 1927. Fraser
mentions no type specimens in his paper, nor have I found a specimen marked
Type in his collection. I have therefore decided to designate as LECTOTYPE a
$ from the Samoan Is., Tutuila Isl., Pago Pago, 2.xii.1924, P. A. Buxton and
G. H. Hopkins, determined by Fraser as Rhyothemis regia exul. This specimen
also bears Lieftinck’s determination label Rhyothemis regia chalcoptilon (Brauer),
of which species R. 7. armstrongi Fraser has been placed as a synonym by Lieftinck.
Syn. of Rhyothemis regia chalcoptilon (Brauer).

arthuri (Mortonagrion), 1942 : 97-08, figs. 1a, b. Holotype g. Malaya, Butter-
worth, 24.xi.1935, A. Wheeler. Mortonagrion arthuri sp. nov., ¢ Type. F.C. F.

asiatica (Indophlebia), 1935 : 323-324, fig. 2. Holotype 9. Sikkim, Tonglu,
10,000 ft., 29.ix.1924. Indophlebia asiatica Fraser, Female Type.

180 D. E. KIMMINS

[assamica (Enallagma), 1919 : 877-878.| The type is stated by Fraser (1933b :
343) to be in the BM (NH), but there is no record here that it was ever received.
There are no examples in the Fraser collection marked as type, or even bearing
the date of the type series (27.x.1918), but there are examples taken at Shillong
on other dates in October, 1918. This may be another instance where an error
has occurred in the published date of capture. Fraser’s description is not very
detailed and differs somewhat from these examples, and whilst they may have
been part of the type-series, it seems better to record the type as “‘ untraced ”’.
Syn. of Aciagrion tillyardi Laidlaw.

assamica (Indolestes), 1930: 104. Holotype 9. Type of Indolestes assamica
Fras., Shillong, Assam, 27.x.1929. T.B.F. Symp. assamica 9 (Fraser) [labels
FC]:

atrocyana (Agrion), 1935 : 330. LECTOTYPE g. Tonkin. Pseudomatrona
atrocyana Fras., Mss., Type ¢ [label F.C.F.]. Agrion atrocyana Fras. ¢ Lecto-
type, D. E. Kimmins det. 1965. In his account of the species, Fraser mentions
that he had provisionally named the species Pseudomatrona atrocyana, but later
decided that it was not separable from Agvion. Asa name published in synonymy,
Pseudomatrona is not an available name. The remaining examples have been
labelled paratypes.

attenuatum (Aciagrion), 1928a : 126-127. Holotype g. Africa, Nyasaland,
Zomba, Colin Smee. Aciagrion attenuatum ¢ Fraser Type [labels D.E.K.].
Zomba, Nyassaland Protectorate, Coll. Mr. Colin Smee. Aciagrion attenuatum 3
sp. nov. [labels F.C.F.].

atuberculata (Macromia), 1922 : 67-68. Holotype g. Hasimara, Duars, [Ben-
gal], C. M. Inglis coll. 1920. Macromia atuberculata Fraser J, Type [label T.B.F.].
Ex Pusa Collection. Syn. of Macromia flavocolorata Fraser.

aurantiacum (Ceriagrion), 1924 : 492. LECTOTYPE g. Type. Nilgiris, 3,500
ft., Wynaad, 26.vii.1g22. Ceriagrion auranticum Fras., 3. Allotype 9. Nilgiri
Wynaad, Devarashola, 3,000 ft., 13.viii.1922, F. C. Fraser. C. aurantiacum 9
allotype [label F.C.F.]. The specific name on the label of the lectotype is mis-
spelt. Now Ceriagrion olivaceum aurantiacum Fraser.

auranticum (Ceriagrion), 1922a : 236-237. LECTOTYPE g. Bangkok, 20.
viii.rg21. Coll. Gen. E. W. Trotter. Ceriagrion auranticum ¢ [label on paper
triangle, F.C.F.]. Of the three males listed by Fraser, this is the only example
which I have been able to trace ; it was in his papered collection, without any
type-label.

aurea (Tetrathemis), 1924a : 69-70, pl. 9, fig. 1. LECTOTYPE g. [L. Burma],
Mergui, 28.ix.1922, Bott coll. Tetrathemis aurea 3, Type. Fraser det. 1922
(labels F.C.F.]. Ex Pusa Collection. Syn. of Tetrathemis platyptera Selys.

auricolor (Amorphostigma), 1927a : 30-32, figs. 3c—-d, 5ili. LECTOTYPE 2.
W. Samoa, Upolu, Malololelei, 14-30. vii. 1925, J. S. Armstrong. Amorphostigma
auricolor Fras. $ (Type) [label F.C.F.]. The remainder of the type series (1g, 29)
has not been traced.

FRASER ODONATA TYPES IN B.M. (N.H.) 181

auricolor (ssp. of Caconeura dorsalis), 1927 : 90. Holotype g. Upper Burma,
Maymyo, 5.viii.1925, Col. F. Wall. C. dorsalis auricolor Fraser, § Type [label
F.C.F.]. Allotype 2. Same data, Caconeura dorsalis auricolor Fraser 9 [label
D.E.K.]._ Now Prodasineura auricolor (Fraser).

auricolor (Gomphus), 1926c : 482. Holotype 2. Tonkin, Ngaio-Tio, 4,800 ft.,
5.iv.1924, H. Stevens. Gomphus auricolor sp. nov. 9, Type [label F.C.F.].

auricolor (var. of Notiothemis jonesi), 1944 : 40-42, fig. 2a. Holotype d.
Notiothemis jonesi var. aureus Fras., Uganda, 3.xii.1927. Det. F. C. Fraser
[label F.C.F.]. The name aureus on the label may be an error of transcription
from the paper envelope.

autumnalis (Caconeura), 1922 : 43. Holotype 3. Assam, Shillong, 14.x.19g19,
T. B. Fletcher. Caconeura autumnalis, Type [label T.B.F.]. Disparoneura
autumnalis §. T. B. F., Shillong, 14.x.1919 [label F.C.F.]. Now Prodasineura
autumnalis (Fraser).

azureum (Aciagrion), 1922: 51. Holotype g. [Assam], Margherita, 18.v.19g20,
Fletcher coll. Aciagrion azureum sp. nov. [label F.C.F.]. Aciagrion azureum
Fraser, § Type, Fraser det. 1921 [label T.B.F.]. Ex Pusa Collection.

bainbriggei (Gynacantha), 1922 : 75-76. Holotype g. Gauhati, in bamboo
jungle, 18.x.1919, Fletcher coll. Gynacantha bainbriggei Fraser, Type [labels
T.B.F.]. The specimen is rather bleached, confirming that it is the type, which
was a spirit specimen. Now Acanthagyna bainbrigget (Fraser).

bakeri (Ceriagrion), 1941 : 62-63, fig. B, 1. LECTOTYPE 3g. Uganda, Gulu,
Patiko, 23.vii.1929, G. D. Hale Carpenter. Type of C. bakeri sp. n. [label F.C.F.].

beatifica (ssp. of Rhinocypha perforata), 1927 : 86-87, fig. 5. Lectotype 3
(Fraser, 1934 : 45). R. perforata limbata, Mugba [Nungba], Naga Hills, 8.iv.
1924, Dr Annandale. _ R. perforata beatifica Fras. [labels F.C.F.]. Right hand
wings missing. In 1927 : 87, Fraser says “‘ one in my own collection, one in Pusa ”’
and in 1934 he remarks “‘ Type in Fraser collection ’’, which is equivalent to a
designation of lectotype.

beesoni (Rhinocypha), 1922 : 61-63. Holotype g. R. beesoni, Lachiwala, Dehra
Dun Dt, N. India, 18.xi.1920, F. C. Beeson [label F.C.F.]. Now Heliocypha
beesoni (Fraser).

bellicosa (Macromia), 1924 : 453-454, pl. 25, fig. 9. Lectotype ¢ (Fraser, 1936 :
177). India, Coorg, Cannanore Ghat, 28.v.1923, Maj. F. C. Fraser. Macromia
bellicosa $ Type [label F.C.F.].

bicolor (race of Chlorocypha curta), 1941a : 39. The position of this race-name
is confused. Fraser refers to it as a ‘‘ provisional Mss name ”’ for a race of curta,
which he distinguished by the paired black spots on abdominal segments 3 to 5
being confluent with the moderately broad apical black rings. Pinhey, 1962 : 148?

1Pinhey, E. 1962. Publ. cult. Cia Diamant. Angola, 59.

182 D. E. KIMMINS

gives the locality as Bagang (Cameroons), but this is an error, since Fraser implies
that the pair from this locality are typical of curta. As far as I know, Fraser never
gave the locality of his race bicolor. There is in the BM (NH) a ¢ specimen
from Uganda, which is marked by Fraser as Chlorocypha bicolor sp. n.
Neither this nor two paratypes from Uganda agree with his diagnosis, all having the
paired spots on segments 3-5 separated from the apical rings and appear to be
normal curta. I propose therefore that C. bicolor Fraser, 1941 be placed as a
synonym of C. curta (Selys), syn. n.

bicornutus (Gomphus), 1922 : 72-73. Holotype 9. Shillong, 18.vi.1920, Flet-
cher coll. Hovering over stream. Gomphus bicornutus 2 Type. Fraser det.
1g2r [label T.B.F.]. Ex Pusa Collection. Now Megalogomphus bicornutus
(Fraser).

bidentatum (Ceriagrion), 1941 : 64-66, fig. B, 3. Holotype dg, allotype 9.
Uganda, Budama, x.1927, G. D. Hale Carpenter [label D.E.K.]. Ceriagrion
bidentatum Fraser [label F.C.F.]. The locality written on the paper envelopes
by Fraser was Entebbe, but the published locality was Budama.

bidentatus (Leptogomphus), 1930a : 752-753, 2 figs. LECTOTYPE g. Type.
Leptogomphus bidentatus ¢, Shillong, Assam, 21.vii.28, T. B. F. [label F.C.F.].
Allotype 2. Same data. Both specimens are labelled 21 instead of 19 July as
published. Now Dubitogomphus bidentatus (Fraser).

bifenestrata (Rhinocypha), 1922: 63, pl. 8, fig. 2. Holotype 3g. Mangpu
[Darjiling Dt], 3,860 ft., 30.viii.1920, Rhinocypha bifenestrata Fraser, 3 Type
(label T.B.F.]. Ex Pusa Collection.

bifida (Tetrathemis), 1941b : 138-140, figs. 1-3. LECTOTYPE 3. Uganda,
Bwamba Valley, Kidongo, 6.i.1928, G. Hale Carpenter [label D.E.K.]. Tetra-
themis bifida Fras., f [label F.C.F.]._ The published date of capture is 7.1. 1928.

biguttata (Cephalaeschna), 1935 : 321-322. Holotype 3. C. assamica, Shillong,
Assam, 18.v.1924, T. B. Fletcher [label F.C.F.]. This specimen agrees with the
diagnosis of biguttata and I believe that Fraser changed the name of this species
before publication and did not correct the label. I consider it to be the holotype
of C. biguttata.

biharica (Gynacantha), 1927 : 74-75. Holotype g. India, Bihar, Pusa, 7. viii.
1924, T. B. Fletcher. Gynacantha biharica Fraser sp. n. j. Ex Pusa Collection.
Now Acanthagyna biharica (Fraser).

bilineata (Melanoneura), 1922: 55. Holotype 2. Sidapur, Coorg, 25.iv.19g19,
Y. R. Rao coll. Melanoneura bilineata 9? Type, F. C. F. det. r92z. Ex Pusa
Collection. It should be noted that there are a few discrepancies between this
specimen and the description. Firstly, it is a female, as labelled by Fraser, not a
male as in the description. Secondly the abdomen, lacking the last four segments,
measures 30 mm., not 23 mm., and thirdly, Fraser does not mention the transverse
blue band on the face. I consider it to be conspecific with later material placed
by Fraser as M. bilineata. There is in the BM (NH) a further complete ? taken by

FRASER ODONATA TYPES IN B.M. (N.H.) 183

Fraser at Coorg, which has been labelled by him as cotype. This is incorrect,
as the species was based upon a single example.

[binocellata (Agriocnemis), 1922a : 233]. Location of unique type ¢ not stated.
Not traced in Fraser Collection, where types of two of the five species described
in this paper have been located.

binocellata (Macromia), 1924 : 451-452. When checking the Fraser types
deposited at the BM (NH) in 1958, I discovered that the specimens labelled as 3
and 92 types of M. binocellata were in fact the types of Epophthalmia frontalis
malabarensis Fraser, a species synonymous with M. binocellata. In answer to
my query about this, Fraser wrote me (1.ii1.1958) “‘ Inglis [the collector] was
essentially an ornithologist and his only collections were on my behalf, with the
proviso that duplicates and types should be reserved for the [Darjeeling] Museum.
He is dead now and when he left Darjeeling to settle in the Nilgiris in 1936, he sent
me one type in fragmentary condition and said that his assistant had apparently
not looked at the collections for some time as clothes moths and anthrenids had
got at them and they were all in bits and pieces.”

Fraser then suggested that a Malabar specimen of M. frontalis malabarensis
be selected as Neotype of M. binocellata and sent me a printed label to be attached
to the specimen in question. This label was based upon his mistaken belief that
the International Code required that the neotype be deposited in the original
museum (a policy to which he was strongly opposed). Since this is not the case
there was no point in retaining the whole of his label and it has been cut down to
“Type destroyed. Neotype”’ and supplemented by a label in my handwriting.
I take this opportunity of publishing Fraser’s selection of a NEOTYPE for M.
binocellata Fraser.

Epophthalmia frontalis malabarensis Fraser has since 1936 been considered a
synonym of E. frontalis binocellata Fraser and the selection of the type of mala-
barensis as the neotype of binocellata will maintain this usage. A full diagnosis
of the subspecies is given in Fraser, 1936 : 199—20T, fig. 62.

The labels on the NEOTYPE are: E. binocellata 3, Waliyar Gt, S. Malabar,
II.ix.32, F. C. Fraser [label F.C.F.]. Epophthalmia frontalis malabarensis Fras.,
3 Type, det. D. E. Kimmins, 1958. Selected by Fraser, 1958 as NEOTYPE of
Macromia binocellata Fraser [label D.E.K.]. Type destroyed. Neotype. Now
Epophthalmia frontalis binocellata (Fraser).

bispina (Macromia), 1954 : 49-52, fig. 2 (5). Holotype g. Bwamba Forest,
2,400 ft., Fort Portal, Uganda, iv.1951, E. Pinhey. Type of Macromia bispina
Fraser $ [label F.C.F.]. Allotype 9. M. bispina Fras., 9, Sango Bay, Katera,
Uganda, x.1953. E. Pinhey.

borikhanensis (Macrogomphus), 1933c : 136-139, figs. 8a, c. LECTOTYPE 4,
allotype 2. M. borikhanensis, Borikhane, Laos, Siam, 27.iii.32, Coll. A. Kerr
[label F.C.F.].

botti (Caconeura), 1922 : 41-42. Holotype g. King Island, Mergui, Burma,

184 D. E. KIMMINS

4.vi.1g921, Bott coll. Caconeura botti Fras. ¢g Type. Fraser det. 1924 [labels
T.B.F.]. Syn. of Prodasineura collaris (Selys).

burliyarensis (ssp. of Idionyx corona), 1922 : 65; 1924 : 461-462, pl. 26, fig. 6.
LECTOTYPE 9. S. India, Burliyar, 29.vili.1g21, 2,520 ft., Maj. F. C. Fraser.
Idionyx corona race nilgiriensis Fraser 2 Type [label F.C.F.]. Idionyx corona
burliyarensis Fraser 9 Type, det. D. E. Kimmins. Allotype g. I. burliyarensis
3d, Burliyar R., Nilgiris, 2,500 ft., 16.vi.22 [label F.C.F.]. It may be pointed
out that Fraser (1936 : 229) refers to “‘ Type and allotype female in the British
Museum’”’. This is an error, since the type must be a female. (See also nil-
giriensis).

burmanensis (ssp. of Caconeura verticalis), 1933b : 216-218. LECTOTYPE d.
Upper Burma, Maymyo, 11-20.vi.1925, coll. F. Wall. Caconeura verticalis
burmanensis Fras. g [label F.C.F.]. No mention is made of types in original
description. This lectotype agrees with the brief locality data. Now Proda-
sineura verticalis burmanensis (Fraser).

burmicus (Microgomphus), 1925a : 854. LECTOTYPE g. Type. Microgom-
phus burmicus g, Maymyo, Upper Burma, Io.vi.24, F. Wall [label F.C.F.].
Allotype 2. Microgomphus burmicus 2 Maymyo, ro.vi.24, F. Wall [label F.C.F.].
Syn. of Microgomphus loogali Fraser.

butoloensis (Notogomphus), 1952 : 854. Holotype g. N. butoloensis J, sp. n.,
Bwamba Valley, Uganda, 2,500 ft., 8.xii.19g28, A. O. Fisher [label F.C.F.].
Allotype °. Butolo Forest, 30 m. W. of Kampala, May 1951, E. Pinhey. Allo-
type, Notogomphus butoloensis Fras. 2 [label E. Pinhey].

buxtoni (Ischnura), 1927a : 23-24, figs. 3a, b, 5v. LECTOTYPE J, allotype 9,
W. Samoa, Upolu Is., Malololelei, 2,000 ft., 2. vii.1924, J. S. Armstrong. Fraser
appears to have erred in transcribing the locality data. Two females in paper
envelopes in his collection bear data agreeing with that published in the original
description. The g type in BM (NH) was correctly attributed to J. S. Armstrong,
but the date was given as 6.xii.26, not 2.vii.24. The female type also had the
wrong date and the collector’s name was given as P. A. Buxton. I have therefore
removed our museum locality labels (based upon Fraser’s transcriptions) and re-
placed them with labels agreeing with the descriptions. Fraser’s labels have been
retained.

cacharensis (ssp. of Disparoneura campioni), 1933b: 244. Holotype d.
Cachar, Assam, 5.ix.1921, coll. B. Antram. Disparoneura campioni cacharensis
Fras, ¢ Type [labels F.C.F.]. Now Elattoneura campioni cacharensis (Fraser).

cacharicus (Lamellogomphus), 1924a : 81-83. Holotype 9. Dilkoosh, Cachar,
ix.21. Lamellogomphus cacharicus Fraser, 9 Type, Fraser det. 1924. Gomphus
sp. nov. Fraser det. 1922 [labels T.B.F.]._ Ex Pusa Collection. The data given
in the original description had been transcribed as July. Now Onychogomphus
cacharicus (Fraser).

FRASER ODONATA TYPES IN B.M. (N.H.) 185

calcipennis (Microstigma), 1946 : 463, fig. 3f. Holotype 9. Microstigma calci-
pennis Fraser 2 Type. Bolivia, F. C. Fraser det. 1946 [label F.C.F.].

campioni (Disparoneura), 1922 : 43-44. Holotype g. Margherita, 18.v.1920,
Disparoneura campioni Fras. § Type. Fras. det. 1921 [label T.B.F.]. Ex Pusa
Collection. Now Elattoneura campioni (Fraser).

campioni (Orogomphus), 1924 : 467-469. Lectotype g (Fraser, I929a : 149).
S. India, Coorg, 4 mls from Mercara, 2.v.23, coll. F. C. Fraser [label F.C.F.].
Orogomphus campioni Fras. ¢ Type, D. E. Kimmins det. 1964. Allotype 9.
S. India, Coorg, Mercara, Sidapur Rd, 2.v.23, Maj. F. C. Fraser [printed label],
Orogomphus campioni 9, Mercara, Sidapur Rd, Coorg, 2.v.23, coll. F. C. Fraser,
Cotype [label F.C.F.]. The original description does not state the sex of the type
and cotype but Fraser, 1g29a : 149 states ‘‘ Coll. Br. Mus., 1 J (type), 1 2 (cotype)”’.
Now Chlorogomphus campiont (Fraser).

canningi (Caconeura), 1919 : 877. Holotype g (lacking abdominal segments
7-10). Nilgiris, Coonoor, 6,000 ft., Caconeura canningi sp. nov., Type, F. C. F.
(label F.C.F.]. Fraser (1924 : 504) casts doubt on the accuracy of the locality
label of the type and suggests that it may really have been collected in Ceylon,
Dyatalawa, 5,000 ft. Syn. of Prodasineura sita (Kirby), a Ceylanese species.

cardinalis (Oxyagrion), 1946 : 41-42. Holotype 9. Type, Oxyagrion cardinalis
2, Mishuyacu, Peru, 14.iv.30 [label F.C.F., on back of mounting card].

cardinalis (Pseudophaea), 1924 : 512-513. Holotype 3. S. India, Shambanagur,
Madura. P. cardinalis Fraser, the upper specimen the type [label F.C.F., in
pencil]. Fraser omits any details as to the location of the type in the original
description, probably due to an uncorrected printing error. The BM (NH) has
in fact a type and two paratypes, all with locality labels as detailed above, presented
by S. Maulik, as part of a collection of undetermined insects from S. India. Fraser
studied these insects in our collection and made them part of his type-series,
placing the pencil-written determination label below them. This label was
subsequently placed on the upper specimen and the other two labelled as para-
types, either by Campion or myself. In the circumstances I think it is justifiable
to regard this type as holotype rather than lectotype. Now Indophaea cardinalis
(Fraser).

carpenteri (Nilogomphus), 1928a : 133, figs. 5a-d. LECTOTYPE 3. Noto-
gomphus dorsalis g, Bududiri, Mt. Elgon, 4,200 ft., 20.xii.1927, H. Carpenter.
[label F.C.F.]. Uganda, Gulu Dt, W. Nile, viii.ix.1927, G. Hale Carpenter.
Nilogomphus carpenteri, J Type Fraser, D. E. Kimmins det. 1965 [labels D.E.K.].
Allotype 9. Same data. No specimens labelled as Nilogomphus carpentert have
been located either in Fraser’s collection, in the BM (NH) or in the Hope Depart-
ment, University Museum, Oxford. The Fraser collection does however contain
a pair labelled Notogomphus dorsalis (Selys), a species with which Fraser subse-
quently synonymized his Nilogomphus carpentert. These examples differ slightly
in nodal indices from the original description, the locality and date also disagree

186 D. E. KIMMINS

and the $ superior anal appendages are slightly crumpled, compared with the
original figures. I consider this to be another case in which Fraser has re-labelled
his types and in so doing has given them incorrect locality labels. Iam therefore
accepting them as the types and have re-labelled them as above. Now syn. of
Notogomphus dorsalis (Selys).

carpenteri (Oxythemis), 1944a : 85-87, figs. 1a, b. LECTOTYPE g. Type.
Uganda, White Nile, R. Alla, 27.ix.27, G. D. H. Carpenter. Oxythemis carpen-
teri Fraser ¢ [label F.C.F.]. Allotype 9. Same data. Now Aethiothemis car-
penteri (Fraser).

carpenteri (Tetrathemis), 1941b : 140-141, figs. 7, 8. Holotype g. Uganda,
Entebbe, vii.rg2r [label D.E.K.]._ Tetrathemis carpenteri Fras. ¢ Type.
F. C. Fraser det. Syn. of Tetrathemis polleni (Selys).

caudalis (Anisogomphus), 1926a : 423. Holotype g. Assam, 5,000 ft., Shillong,
18 June 1924, Fletcher coll. Anisogomphus caudalis Fraser g Type [label T.B.F.].
Allotype 9. A. caudalis, Barog, 5,000 ft., Simla Hills, 4.viii.29, Capt. Cardew
(label F.C.F.].

cauvericus (Burmagomphus), 1926a: 413. Holotype g. S. India, Coorg,
Bhagmandala Rd, 3.vi.1923, Maj. F. C. Fraser. Burmagomphus cauvericus
Fras. 3 Type [label F.C.F.].

cerinostigma (Protosticta), 1924 : 499-500. There is a discrepancy between the
published locality data and that on the holotype specimen, although it is possible
that the two sets of data may be complementary to one another and refer to the
same locality. The published locality is Nilgiris, Ochterlony Valley, 3. viii. 1922.
The data on the paper envelope (by rubber stamp) are Nilgiris, 3,500 ft., Wynaad,
Pandy R., 2.viii.1922, F. C. F. Protosticta cerinostigma g Fras. Type [label
F.C.F.], with additional comment (in pencil) var. of sanguinostigma. Now var. of
Protosticta sanguinostigma Fraser.

cervus (Heliogomphus), 1942a : 336-337, figs. 1 (3, 6), 2 (14). LECTOTYPE g.
Upper Burma, Maymyo, 4.vii.25, F. Wall coll., Heliogomphus ¢, with on reverse,
H. cervus (Type) [label F.C.F.]. Allotype 9. Same data.

chichibui (Gomphus), 1936b : 148-149, fig. 1. Holotype g. Japan, Nikko,
June °34, Coll. F. C. Fraser. G. chichibui g. Syn. of Trigomphus interruptus
(Selys).

clauseni (Agriocnemis), 1922 : 53-55, pl. 6, fig. 4. Holotype ¢. Shillong,
Fruit Gdn Pond, 2.ix.1919. Fletcher coll. Agriocnemis clauseni g Type,
Fraser det. 1921 [label T.B.F.].

coerulescens (Caconeura), 1932 : 289-291. LECTOTYPE 3. P{eninsular] Siam,
Prachuap Province, Pak Tawan, 2.viii.193r, Dr A. F. G. Kerr [label D.E.K.].
Caconeura dorsalis group, Ab entirely absent, C. coerulescens g Type [label
F.C.F.]. Allotype 9. Same data. Caconeura coerulescens Fraser allotype.
Now Prodasineura coerulescens (Fraser).

FRASER ‘ODONATA TYPES IN B:M. (NH) 187

cordosa (ssp. of Chlorocypha dispar), 1947 : 23. Holotype 3. Gabon. Libel-
lago cyanifrons Selys $ [Selys label]. Chlorocypha rubida Hag., D. E. Kimmins
det. 1948. Chlorocypha dispar cordosa Fraser, 3 Type, D. E. Kimmins det. 1963.
Now syn. of Chlorocypha dispar (P. de B.).

corona (Idionyx), 1921a : 690. Holotype 9. Babuddin Hills, Mysore, r.vi. 1915.
Ramakrishna coll. Idionyx corona 9 Type, Fraser det. May 1921 [label T.B.F.].
Ex Pusa Collection. The type lacks abdominal segments 6-10.

coryndoni (Lokia), 1952a : 249—250, fig. 2d. LECTOTYPE g. Uganda, En-
tebbe, Lake Victoria, x.1952. Lokian. sp. ¢ Q [label F.C.F.].. Lokia coryndoni
Fraser J, D. E. Kimmins det. Allotype 2. Same locality, v.52. Lokia coryn-
doni Fraser 9, D. E. Kimmins det.

cupricincta (Macromia), 1924a:74. Holotype 3. Burma, Thaton distr.,
Mokpalin, 18.v.23, Bott coll. Shady jungle path, hawking low, 11.30 a.m.,
18.v.23 [label T.B.F.]. Macromia cupricincta Fras. ¢g Type [label T.B.F.].
Ex Pusa Collection. Allotype 9. Assam, Nowgong, 2.vi.31, C. Antram. M.
cupricincta.

cupricolor (Hemicordulia), 1927a : 37-39, fig. 4. LECTOTYPE 3. W. Samoa,
Upolu Is., Malololelei, 2.vii.1924, J. S. Armstrong, Hemicordulia cupricolor.
Allotype 2. Same data.

curiosa (Protosticta), 1934a : 134-135, fig. 1. Holotype g. Burma, Mergui,
31.v.32 [Dr A. Kerr]. P. curiosa Fras. ¢ Type [label F.C.F.].

cyaneovittata (Esme), 1922 : 45-46. Lectotype f (Fraser, 1933b : 264). Palnis,
7,000 ft., Kodaikanal, 29.vili.1g21, Fletcher coll. No type was selected by
Fraser. This may be taken as type [label T.B.F.]. Esme cyaneovittata § Fraser
(label T.B.F.].

cyaneofrons (Gomphus), 1923 : 62. This name first appeared in a key to Indian
Dragonflies and there is no indication that it was a new species nor is any precise
locality mentioned. Apart from a similar key later in the same volume, no further
reference is made to this name by Fraser. Laidlaw placed it as a synonym of
Perissogomphus stevenst Laidlaw, 1922.

There are no examples labelled Gomphus cyaneofrons in the Fraser collection,
but there are other instances in which he has removed his original determination
labels when a species has been synonymized and replaced them with new deter-
mination labels. Concerning this first key, the following extract from a letter by
Fraser to Dr F. F. Laidlaw (19.xi.1929) is of interest. He writes ‘‘ I would not
regard ‘ Dragonfly Collecting in India’ too seriously, as it was written rather
hurriedly to order and is what I called ‘ The Young Dragonfly Collector’. When
writing it, I had a number of undescribed species and wishing to make the key as
complete as possible, I named these provisionally and included them. I ought,
of course, to have conserved these names when publishing descriptions later. The
article, however, was never intended as a scientific paper and so I never regarded it
at allseriously.’”’ This statement of Fraser’s does not affect the availability of the

188 D. E. KIMMINS

specific names published in this key (Art. 8 of Int. Code), as there was no published
indication that it was not intended as a scientific publication.

In Fraser’s Gomphidae collection are 1 g and 3 @ of Perissogomphus stevensi
Laidlaw, with dates of capture which would entitle them to consideration as
possible syntypes of Gomphus cyaneofrons Fraser (Bengal, Gopaldhara, 4.vi.1g19,
H. Stevens). I have therefore placed labels on them to that effect, in case at
any time it is desired to designate a lectotype for Gomphus cyaneofrons Fraser.
Now syn. of Perissogomphus stevensi Laidlaw.

[dabreui (Agriocnemis), I91g9a : 454.] The location of the holotype 9 is not given
in the original description. In 1933 : 398, the type is stated to be in BM (NH),
formerly in Pusa Collection. This species is however not listed amongst the types
received from Pusa in 1925 and the type is probably no longer in existence.

davenporti (Ceylonolestes), 1930 : 96-97, pl. 1, fig. 7. Syntypes, 4 J, 2 2, Palni
Hills, 7,000 ft., T. B. Fletcher coll., various dates, 14. vili. 1920—vii.1923. No type
has been selected for this species and it seems better to leave such selection until a
specialist is making a revision. It may be pointed out that in the original des-
cription, the species is attributed to Ris. Now Lestes (Indolestes) davenporti
(Fraser).

davenporti (Protosticta), 1931a : 70-71, pl. figs. 9, 10; 1931 : 467 (2). Holotype
g. S. India, Mudis Hills, 29.v.1929, F. C. Fraser [label D.E.K.]. Protosticta
davenporti Fras., ¢ [label F.C.F.]. Allotype 9. Munnar Ghat, Travancore,
25.v.1932. P. davenporti Fras. 9 allotype, det. F. C. Fraser [label F.C.F.].
The manuscript label on the holotype appears to have been marked Type, but
most of it has been cut off, leaving only the top of the “ T”’.

davina (ssp. of Zygonyx iris), 1926 : 768-769. The holotype 9 has been destroyed
by pests in the Darjeeling Museum. The following specimen, with data “ Z.
davina, Kurseong, Sikkim, vi.39’”’ from Fraser’s collection, was selected by him
as typical and was regarded by him as a neotype, though he did not publish the
designation. As it is desirable that there should be authentic material for this
subspecies, I take this opportunity of designating the above specimen in BM (NH)
as NEOTYPE @ of Zygonyx iris davina Fraser, 1926. See binocellata,
Macromia.

decoloratum (Libyagrion), 1928a:126. LECTOTYPE ¢. Africa, Nyasaland,
Zomba, Colin Smee [label D.E.K.].. Zomba, Nyassaland Protectorate, coll. Colin
Smee. Libyagrion decoloratum Fraser § Typel [abels F.C.F.]. Libyagrion
decoloratum Fraser 3 Type [label D.E.K.]. Allotype 9. Africa, Nyasaland,
Zomba, Colin Smee [label D.E.K.]. Libyagrion decoloratum gen et sp. nov.
(label F.C.F.]. Libyagrion decoloratum Fras. 9 allotype [label D.E.K.]. Now
syn. of Enallagma (Africallagma) subtile Ris.

delineatus (ssp. of Davidius zallorensis), 1926d : 166, pl. 1, fig. 4, text-figs. 3,
iii, viii. LECTOTYPE g. India, Darjeeling Distr., Gangtok, 16.v.25, C. M.
Inglis. Davidius zallorensis delineatus Fras., ¢ Type [label F.C.F}.

FRASER ODONATA TYPES IN B.M. (N.H.) 189

denticauda (Tetrathemis), 1954a : 255-257, figs. Ia, b. Holotype g. Uganda,
Bunyoro distr., Ft. Portal, G. H. Carpenter [label D.E.K.]. Tetrathemis denti-
cauda Fraser ¢ [label F.C.F.].

descendens (ssp. of Acanthagrion apicale), 1946 : 38-40, figs. ga, 10, 4. Holo-
type g. Acanthagrion descendens Fras., 3 Holotype, Mishuyacu, Peru, 2. vii. 30.
From Leeds Mus. Coll. [writing F.C.F., on underside of staging card].

(diminutivus (Onychogomphus), 1924b : 109-110, pl. 1, fig. 1.] The type of
this species has not been traced in the Fraser Collection.

dingavani (Onychogomphus), 1924c : 1005-1006, pl. 2, fig. 7, text-fig. 2, vii.
Holotype g. Onychogomphus dingavani J, [Kalaw] Siam Rd, S. Shan States,
3,000 ft., 20.x.22, Type. Capt. Dingavan coll. [label F.C.F.].

disarmatus (Lestes), 1961 : 11-12, fig. 1. Holotype g. Madi Opei, Acholi,
Uganda, iii.52. T.H.E. Jackson. Lestes disarmata sp. nov. Holotype.

dobsoni (Agriocnemis), 1954b : 147-148, figs. 1a, b, c. LECTOTYPE g. A.
dobsoni Fras., Innisfail, N. Queensland, 22.ix.52, coll. R. Dobson, F. C. Fraser
det. 1953 [label F.C.F.].

dolorosa (Pacificagrion), 1953 : 119-120, figs. 2, 3, 8, 9, 12, 14, 16,17. Holotype
dg. P. dolorosa Fras. g Type. Hellsoi’s Paddock, 1,600 ft., Apia, J. S. Arm-
strong [label F.C.F.].

donaldi (Anaciaeschna), 1922b : 482-483, 699-700. Lectotype ? (Fraser, 1924 :
465). Kodaikanal, Palni Hills, S. India. Coll. F. C. Fraser, May 1908 (Original
specimen) [typewritten label]. Anaciaeschna donaldi Fraser 9. Lectotype 9
Fraser, 1924 [label D.E.K.]. Allotype J. On road side. Anaciaeschna donaldi
Fraser, g. Varatapari, Annaimallai Hills, S. India, 25.iv.1933. det. F.C. Fraser.
vide E. m. M. 76 : 88 [label F.C.F.]._ The specimen in Fraser’s collection labelled
“‘ A. donaldi Fraser 2 Type’”’ was not one of the syntypes, since it was dated
21.v.22. Amongst his examples of A. martini, with which donaldi was once
synonymized, I found a 2 with locality label agreeing with the first example listed
by Fraser (1922b) and marked by him “ Original specimen’’. In 1924 : 465,
Fraser makes the following statements concerning A. donaldi “ described originally
from a female from the Palni Hills ” and “‘ The original female, rather dilapidated,
is in my collection, and was taken by myself in the Palni Hills, May 1908’. I have
accepted these two remarks as being the equivalent of a lectotype designation by
Fraser. His collection also contained two other females (which had been re-
labelled as A. martini by Fraser) which may be part of the type-series of A. donaldi.

donaldi (Indomacromia), 1924 : 515-516, fig. 6. Holotype §. Macromidia
donaldi 3, Sampaji Ghat, Coorg, 7.vi.24, (Type) Coll. F. C. Fraser [label F.C.F.].
Allotype 2. Macromidia donaldi 9. Tamaracherry, S. Malabar, nr Calicut,
15.vii.28. Coll. F. C. Fraser [label F.C.F.]. Now Macromidia donald: (Fraser).

dorothea (ssp. of Agrionoptera insignis), 1927 : 65-66. Lectotype 3 (Fraser,
1936 : 277). India, Bengal, Duars, Hasimara, 1924, H. V. O’Donel [label D.E.K.].
A. insignis dorothea Fraser, J Type, det. F. C. Fraser [label F.C.F.].

ENTOooM. 18, 6. 15

190 D. E. KIMMINS

dorothea (Lestes), 1924 : 484-485, pl. 26, fig. 8. Holotype g. S. India, Coorg,
Mercara, I17.iv.1923, Maj. F. C. Fraser. Lestes praemorsa dorothea ¢ [label
F.C.F.]. Allotype 2. S. India, Coorg, Greenfield, Mercara, 26.iv.1923, Maj. F. C.
Fraser. L. praemorsa dorothea 9 [label F.C.F.]._ The 3 type was presented to the
BM (NH) in 1923 and thus qualifies for the status Holotype by Fraser’s statement
in original description ‘‘ Types in B.M.”’. Another specimen in BM (NH), ex
Pusa Collection, from Kurseong, Lestes dorothea Fraser 3 Type, was incorrectly
labelled by T. Bainbrigge Fletcher and is in fact a paratype of Indolestes helena
Fraser.

drummondi (Lamellogomphus), 1924c : 985-986. Holotype 9. L. risi 9 = (G.
drummondi Fras. Type), Siam Road, Loimwe, S. Shan States, 3,000 ft., 17. v. 1923,
Capt. Drummond [label F.C.F.]. Now syn. of Onychogomphus risi (Fraser).

duarensis (Burmagomphus), 1922c : 421-422, pl. 11, fig. 7. Holotype dg.
' India, Hasimara, Duars, Maj. F. C. Fraser. Burmagomphus duarensis Fras. 3.
Coll. H. V. O’Donel, Hasimara, Duars, Type [label F.C.F.]. The printed label
incorrectly quotes Fraser as the captor. Now syn. of Merogomphus martini
(Fraser).

duaricus (Onychogomphus), 1924c : 1001-1003, pl. 2, fig. 5. Holotype dg.
Onychogomphus duaricus 3, Hasimara, Duars, Bengal, 9.vii.20, H. V. O’D[onel]
(label F.C.F.]. Allotype 9. Same data. Although it is stated in the original
description “‘ Type male and cotype female in the British Museum ”’, these speci-
mens in fact remained in Fraser’s collection until 1957. |

dubia (Porpacithemis), 1954a : 261-262, figs. 2a-d. Holotype 2. Gabon, L.
Asebbe, Fernan-Vaz, 14.i1.1908. Porpacithemis dubia Fraser, D. E. Kimmins
det. 1955.

dundomajoricus (Phyllogomphus), 1957 : 24-28, figs. 2c, g, 4a, 5c, 7c. LECTO-
TYPE ¢g. N. Angola, Lundu Province, Dundo, I1.xii.1947. Phyllogomphus
dundomajoricus ¢ Type, det. F. C. Fraser 1957. Allotype 9. Same locality,
xi.1948. Phyllogomphus dundomajoricus 2 Allotype, det. F. C. Fraser, 1957.

dundominusculus (Phyllogomphus), 1957 : 28-29, figs. 2h, 4b, 5f. Holotype ¢.
N. Angola, Lunda Province, Dundo, Nov. 1948. Phyllogomphus dundominus-
culus ¢ Type, det. F. C. Fraser, 1957.

echinoccipitalis (Onychogomphus), 1922 : 74-75. Holotype 2. Dipping over
roadside stream, Mile 23, Gauhati-Shillong. About 1,000 ft., 2.viii.r9. Fletcher
coll. (Gomphinae). Onychogomphus echinoccipitalis Fraser Type. Fraser det.
1g2I [labels T.B.F.]._ Ex Pusa Collection.

ellisoni (Macromia), 1924 : 457-458, pl. 25, fig. 3. Lectotype ¢ (Fraser, 1936 :
171). M. ellisoni 3, 4,000 ft., Nilgiris, 7.x.21, Mysore Ditch, S. India, F. C. Fraser
(label F.C.F.]. Allotype 9. M. ellisoni 2, Devalashola, Nilgiri Wynaad, 5.xi.
1922, S. India, F. C. Fraser [label D.E.K.]. When received from Fraser, the
allotype bore a label stating that it was a g, from Gudapur.

FRASER ODONATA TYPES IN B.M. (N.H.) I9I

eltoni (Phyllothemis), 1935 : 890-891. This is yet another species in which one
finds discrepancies between the published data and the specimens. Fraser, in
his description, records two males in his collection (King Island, Mergui, Lower
Burma, taken by Mr J. Elton Bott during September), and states that the type
is in his own collection. I have so far traced only one fragmentary ¢ in the
Fraser collection. It was in a paper triangle ; the covering envelope was labelled
Phyllothemis eltoni Type by Fraser but the triangle was labelled Amphithemis
gomphordes.

In 1925 the BM (NH) received, amongst other Odonata types from the Pusa
Collection, a specimen from Mergui, Bott coll., labelled by T. B. Fletcher Amphi-
themis gomphotides 3 Fraser, Type, Fraser det. 1922. I think that the most probable
explanation is that Fraser drew up a description of Amphithemis gomphoides
from Pusa material submitted to him by Fletcher, but did not immediately send
it to press. Years later, when assembling material for the third volume of the
Odonata of the Fauna of British India, he used this description, publishing it under
the name Phyllothemis eltoni. Similar discrepancies between labels and published
names have been found in the Pusa material. Whether there were more than the
two specimens I do not know, but I have discovered only this fragmentary one
in the Fraser collection.

In view of the fragmentary condition of this specimen, and the fact that the
Pusa male has been in the BM (NH) for forty years as the type of a Fraser Mss name,
I designate the Pusa male as the LECTOTYPE of Phyllothemis eltoni Fraser,
and it now bears my determination label to that effect. The fragmentary male
has been labelled by me as a paratype.

ethelae (Dysphaea), 1924 : 480-482. Holotype 3. S. India, Coorg, Napoklu,
Cauvery River, 22.iv.1923, Maj. F. C. Fraser. Pseudophaea etheli ¢ Fras. MS.,
Cauvery River, Coorg, Napoklu, 22.4.23. Type [label F.C.F.]. Allotype 9.
S. India, Coorg, Fraserpet, near Cauvery River, 22.v.23, Maj. F. C. Fraser.
Dysphaea ethelae Cotype 2 [label F.C.F.]. The original description and Fauna of
Brit. India give the specific name as ethela.

euphorbia (ssp. of Argia extranea), 1946 : 457-458, figs. 2e-g. Holotype d.
Type. Argia extranea euphorbia g. Mishuyacu, Peru, Io. viii.30 [label F.C.F.,
on underside of mounting card].

excelsa (Orolestes), 1933d : 181-182, pl. 4, fig. 3, text-fig. 5. Holotype J. Oro-
lestes excelsa, Chaiyabari, N. E. Siam, 3.v.32, Dr A. Kerr [label F.C.F.]. One
pair of wings in Ann Arbor Mus., Michigan. The right hand pair of wings was
presented to Ann Arbor Museum by Fraser.

extraordinata (Atoconeura), 1950 : 56-57, fig. 1. LECTOTYPE g. Uganda,

_ Ruwenzori, Bwamba Valley, 6,500-7,500 ft., 13-15.1.1926, G. D. Hale Carpenter

[label D.E.K.]. Atoconeura extraordinata Fraser g Type (A. eudoxia (Kirby))
F. C. Fraser det. 1951 [label F.C.F.]._ Now syn. of Atoconeura eudoxia Kirby.

flaviceps (Burmargiolestes), 1933a : 120-121. LECTOTYPE g. B. flaviceps

192 D. E. KIMMINS

Fraser § Type. Taweing, Laos, 1,200 m., 6.iv.32, Dr A. Kerr [label F.C.F.].
?Syn. of Burmargiolestes melanothorax (Selys).

flavicolor (Heterogomphus), 1923a : 678, pl. 1, figs. 2, 2a. Holotype 9. Ex
Mackenzie coll. Sine loco. (in box labelled “‘ Saran’’.) Heterogomphus flavicolor
Fras. Fraser det. 1923. Type [labels T.B.F.]. Ex Pusa Collection. Now
Megalogomphus flavicolor (Fraser).

flavifrons (Notogomphus), 1952 : 8-10, figs. Ic, 4a, b, e. LECTOTYPE 4.
Type. Mafuga Rain Forest, Kigezi, Uganda, 7,500-8,000 ft., June 1951, T. H. E.
Jackson. Notogomphus flavifrons Fraser. Mafuga Forest, Kigezi, Uganda,
vi.51. det. F. C. Fraser. Allotype 9. Notogomphus flavifrons Fraser, Mafuga
Forest, Kigezi, Uganda, vi.51. det. F. C. Fraser.

flavocolorata (Macromia), 1922b : 702, fig. 2. Holotype 9. Macromia flavo-
colorata 9, Hasimara, Duars, Bengal, 20.x.21, H. V. O’Donel. Type [label
B.G.E I. Allotype ¢g. M. flavocolorata. Hasimara T. E., Duars, Bengal,
7. vill.31, H. V. O’Donel [label F.C.F.].

flavovittata (Macromia), 1935 : 326-328, figs. 4-5. Holotype g. M. flavovittata
3, Moungpu, Darjeeling distr., 29.v.27 {label F.C.F.].

fletcheri (Disparoneura), 1919 : 876. Holotype 3. Assam, Shillong, 22.ix.1918,
Fletcher coll. D. fletcheri 3, Fras. Type specimen [label F.C.F.]. Allotype 9.
Same locality, 20.ix.1918, Disparoneura fletcheri 2, Fraser, Type specimen [label
F.C.F.].. Now syn. of Elattoneura atkinsoni (Selys).

fletcheri (Gomphidia), 1923a : 669-670, pl. 2, figs. 2, 2a, text-fig. 2a. LECTO-
TYPE g._ S. India, Coorg, Hallery, Mercara, 4,000 ft., 14. vi.23, Maj. F. C. Fraser.
Shot down with dust-shot. F.C. Fraser. Gomphidia fletcheri 3 [labels F.C.F.].

fletcheri (Petaliaeschna), 1927 : 73-74. The holotype 3 was originally in the
Pusa Collection, but there is no record of its having been received at BM (NH)
with the other Pusa types in 1925 and it is now presumed lost. Fraser marked a
replacement specimen in his own collection as neotype, but did not publish its
establishment. I therefore designate this specimen as NEOTYPE ¢. The data
are P. fletcheri, Shillong, Shillong, Assam, 21.v.24. T. Fletcher [label F.C.F].
Type destroyed. Neotype. Petaliaeschna fletcheri g Fraser, det. D. E. Kim-
mins, 1964. NEOTYPE.

floridense (ssp. of Acanthagrion gracile), 1946 : 37-38, figs. 8a, 10 (2, 3). Holo-
type g. Acanthagrion gracile floridense § Type. S. E. Colombia, Florida,
711.1931 [written on mounting card, F.C.F.].

folia (ssp. of Cordulegaster brevistigma), 1929a : 120-121, pl. 12, figs. II-12.
Holotype g. Naini Tal, Kumaon, W. Himalayas, c. 7,000 ft., 5.vi.93 [label
F.C.F.]. Cordulegaster brevistigma folia Fras. g, Type, D. E. Kimmins det.,
1963. Allotype 2. C. brevistigma, Naini Tal, N. W. Himalayas, c. 7,000 ft.,
5.vi.g3 [label F.C.F.]. Cordulegaster brevistigma folia Fras. 2 Allotype. D. E.
Kimmins det. 1963.

BRASE RVODONATA TY PES) DNB eM (New) 193

forficula (ssp. of Argia extranea), 1946 : 456-457, figs. Id-e. LECTOTYPE 4.
Argia extranea forficula g Type. Rio Paranapura, Balsapuerto, 220 m., li. 1933.
N. Peru, F. C. F. det. [written on underside of mounting card, F.C.F.].

fujiacus (Gomphus), 1936b : 147-148, fig. 7. Holotype g. G. fujiacus ¢ Type.
Nikko, Japan, vi.34, coll. F. C. F. [label F.C-.F.].

fujiama (Davidius), 1936b : 150-151, fig. 2. The Fraser collection contained no
example identified as Davidius fujiama. Examination of his series of Davidius
cuniculus Ris however revealed a single ¢ without either locality or determination
labels, which agreed entirely with the description of D. fujiama and it has therefore
been labelled as the LECTOTYPE J, with data as given in the description, (Japan,
Nikko, 3.vi.1934, F. C. Fraser). Fraser’s notes on D. cuniculus list 4 9 from
Nikko ; his collection however contained 5 9, one of which had a label which was
apparently of a later date than the others (the writing being larger and rather
shaky). This 9 appears to be cuniculus Ris, but I believe that it is the 2 originally
associated by Fraser with fujzama and that he later decided it was incorrectly
named and changed the label. I have labelled it as the probable allotype of
D. fujiama.

fulvia (ssp. of Idionyx corona), 1924 : 515-517. Twospecimens, I 4g, I 9, have been
located in the Fraser collection, labelled J. burliyarensis fulvia, Sampaji Ghat,
Coorg, 18.v.1924. Since the description of this subspecies appeared in an appen-
dix published in September 1924, these specimens can be considered as SYNTYPES.
Fraser subsequently placed this subspecies as a synonym of J. burliyarensis and
on the envelope of the 3 the name fulvia is in much more faded ink than the name
I. burliyarensis. Now syn. of Idionyx burliyarensis Fraser.

galeata (Idionyx), 1924 : 517-519, fig. 8a. LECTOTYPE 4. S. India, Coorg,
Khatlkad, 3,800 ft., 30.v.1924, Maj. F. C. Fraser. I. galeata Fras. ¢ [label
F.C.F.]. Allotype 9. India, Katlkad, Coorg, 4.v.23, F. C. Fraser. I. galeata
Fras. Type [label F.C.F.]. The information about the types in the original des-
cription is confused. Fraser writes ‘‘ The types were taken on Katlkad Estate
near Mercara, 17.v.24 by myself. Two more also females were taken on the
following day at the same place ... A fourth female was taken on 25.v.24...
Two males, the first taken by myself, were captured on 30.v.24.”

After much consideration I think that there is an error in the first two sentences,
which I suggest should read “‘ The types were taken by myself on Katlkad Estate,
near Mercara; one female on 17.v.24 and two more, also females, on the following
day at the same place...” This gives four females and two males as syntypes,
all from Katlkad. Coming now to the existing material, there are two males and
one female from Katlkad. One of the males was presented to BM (NH) in 1925
as the g type. A female presented at the same time, marked allotype, is from
Kudremukh, 22.v.24, C. A. Souter, and I consider that Fraser’s statement excludes
it from the syntypes. The only remaining 2 syntype is marked Type and bears the
date 4.v.23, which I believe to be a relabelling error. I consider this to be the
allotype.

194 D. E. KIMMINS

gardneri (Hylaeothemis), 1927 : 66-67. The holotype was originally in the
Museum of the Forest Research Institute, Dehra Dun, N. India and was presented
to the BM (NH) in 1949. The locality label of the type differs from that quoted
in the original description (Kampison, Himalayas), being Naga Hills, 6,000 ft.,
Assam, S. N. Chatterjee, 8.iv.1924. Hylaeothemis gardneri Fras. Major
Fraser det. Type g. The holotype now lacks its head.

gautama (Indagrion), 1922 : 50. Holotype 9. Sadiya, 23 May 1920. Indagrion
gautama Fras. 2 Type. Fraser det. 1921 [label T.B.F.]. Indagrion gen. nov.
gautama sp. nov. 9 [label F.C.F.].. Now Mortonagrion gautama (Fraser).

gigantica (Anotogaster), 1924e : 48-49. Lectotype g (Fraser, 1929a : 94-95).
A. gigantica, Siam Rd, Kalaw, Burma, 8.ix.23. Capt. Drummond [label F.C.F.].
Type. Anotogaster gigantica Fras. Male. Siam Rd, S. Shan States, Burma.
Coll. Capt. Drummond, 8.ix.23 [typewritten label]. Allotype 2. A. giganticum.
Kalaw, Burma, 20.ix.23. Capt. Drummond [label F.C.F.]. Allotype. Anoto-
gaster gigantica Fras. Female. Kalaw, S. Shan States, Burma. Coll. Capt.
Drummond, 20.ix.23 [typewritten label.].

gloriosa (Dysphaea), 1938 : 197-198. Holotype g. D. gloriosa. Pak Tawan,
Prachaup, Siam, 21.viii.31. DrA.Kerr[label F.C.F.]. Allotype?. D. gloriosa,
Muang Ban, Laos, Siam, 28.iv.32. Dr A. Kerr coll. [label D.E.K.].

gracillima (Hemicordulia), 1944a : 87-88, fig. Ic. Holotype 2. Malaya, Kuala
Lumpur, 26.11.32, H. M. Pendlebury. At light. H. gracillima Fras. Holotype
(label F.C.F.].

gynostylus (Cyclogomphus), 1926d : 162-163. Holotype g. Ceylon, Kandy
Lake, 4.ix.24, Col. F. Wall. C. gynostylus [label F.C.F.]. Allotype 9. C.
gynostylus Allotype 9. Kandy, Ceylon, v.1g11 [label F.C.F.].

haemastigma (Ischnura), 1927a : 26-27, figs. 2a, b, 5, vi. Holotype fg. Samoan
Is., Upolu Is., Malololelei, 5.xii.1g25. P.A. Buxton & G. H. Hopkins. Ischnura
haemastigma Fraser g. Malololelei, Upolu Is., W. Samoa, coll. P. A. Buxton,
5-xll.1925. Type [label F.C.F.].

[haematoneura (Sympetrum), 1924a : 70-71.| Type not yet traced. Originally
in Pusa Collection but not listed amongst those presented to BM (NH) in 1925,
and it is therefore presumed lost.

halei (Macromia), 1928a : 137. LECTOTYPE. Uganda, L. Victoria, Entebbe,
xi.1927, G. D. Hale Carpenter. Macromia halei sp. nov. 9 [label F.C.F.]. Syn.
of Macromia nyanzana Grinberg.

hamulata (Argia), 1946 : 447-449, fig. 2d. Holotype g. S. America, Peru,
Yumbatos, xi.1932. Argia hamulata n. sp. § Type [label F.C.F.].

hannyngtoni (Heterogomphus), 1923a : 674-676, pl. 1, figs. I, Ia, text-fig. 4.
Holotype g. S. India, Coorg, Bhagmandala Rd, 5.vi.1923, Maj. F. C. Fraser.
Heterogomphus hannyngtoni Fras. ms. g, Bhagmandala Rd, Coorg. Type

FRASER ODONATA TYPES. IN B.M. (N.H.) 195

[label F.C.F.]. Allotype 9. M. hannyngtoni Fras. 9. Napoklu, Coorg, 2.v.24.
Coll. F. C. Fraser. F. C. Fraser det., 1924 [label F.C.F.]. Now Megalogomphus
hannyngtont (Fraser).

hanumana (Gynacantha), 1922: 76-77. Holotype g. Mangpu, 5,200 ft.,
18.11.1920. C.M. Inglis. Gynacantha hanumana, Fraser, Type [labels T.B.F.].
Ex Pusa Collection. Syn. of Acanthagyna subinterrupta (Rambur).

hasimaricus (Burmagomphus), 1926a : 411. LECTOTYPE 3. Burmagomphus
hasimaricus g, Hasimara, Duars, Bengal, 5.vi.23, coll. H. V. O’Donel, 5.vi.23.
Allotype 9. Burmagomphus hasimaricus 9, H. V. O’Donel, Duars, Bengal
[label F.C.F.]. The reverse of the lectotype’s label is inscribed Burmagomphus
biharicus, H. V. O’Donel, 21.vii.23, Duars, N. I., Type.

hearseyi (Protosticta), 1922d : 5, pl. 1, figs. 3-4. Lectotype ¢ (Fraser, 1933b :
115). Madras, Gudalur, Nilgiri Hills, 26.vi.21, 4,500 ft., Maj. F. C. Fraser.
Protosticta hearseyi $ Fraser [label F.C.F.]. Allotype 2. Same locality, 26.vi.
21. Both the lectotype and allotype bore incorrect date labels, which have been
corrected.

helena (Indolestes), 1922 : 60-61, pl. 7, fig. 1. Holotype g. Kurseong, 6,000 ft.,
27.1x.20. C.M. Inglis coll. Sympecmoides sp. nov. J, Fraser det. 1921. Indo-
lestes helena Fraser § Type. Fraser det. 1921. [labels T.B.F.]. Ex Pusa
Collection. The type now lacks segments g-10 of abdomen. Syn. of Indolestes
cyaneus (Selys).

hemihyalina (ssp. of Rhinocypha quadrimaculata), 1922 : 64, pl. 8, fig. 5.
Holotype g. Khasi Hills, Shillong, 16.x.1919, Fletcher coll. B. H. g, 16.x.19.
Rhinocypha hemihyalina Fraser ms. ¢ Type. Fraser det. 1921 [label T.B.F.].
Penis drawn, J. Cowley, Aug. 1935. Allotype 92. Same locality and date.
Rhinocypha hemihyalina Fraser ms. 9 Cotype, Fraser det. rg21 [label T.B.F.].
Ex Pusa Collection.

hermionae (Allogaster), 1927 : 76-77. Holotype g. India, Bengal, Darjeeling
Distr., [Senchal, 8,000 ft.} 3.vi.26, C. M. Inglis. Allogaster hermione Fraser ¢
Type. The date of capture is given as July in original description. Now
Neallogaster hermionae (Fraser).

hetaerinoides (Leucopteryx), 1933c : 125-128, figs. 5-6. LECTOTYPE 3. Laos,
Muang Cha, 1,100 m., I0.iv.1932, A. Kerr. Leucopteryx hetaerinoides Fras.
(label F.C.F.]. Allotype 9. Archineura hetaerinoides, Muang Cha, Laos, Siam,
10.iv.32, Dr A. Kerr [label F.C.F.]._ Now Archineura hetaerinoides (Fraser).

hilaryae (Rhinocypha), 1927 : 83-86, fig. 4. Lectotype ¢ (Fraser, 1934 : 39).
Burma, Maymyo, vii.1924. Col. F. Wall. Rhinocypha hilaryae Fraser ¢ [label
F.C.F.]. Allotype 9. Burma, Maymyo, vii.1924. Col. F. Wall. Rhinocypha
hilaryae Fras. 9 Allotype [label F.C.F.]. The printed locality labels on type and
allotype were incorrect in date and collector’s name and have been replaced. Now
Aristocypha hilaryae (Fraser).

196 D. E. KIMMINS

hincksi (Mnesarete), 1946 : 26-28, fig. 2. Holotype 3. S. E. Colombia, Umbria,
6.i1.1931. Mnesarete hincksi Fraser f Type. det. F. C. Fraser. 1945. In the
original description the generic name was mis-spelt Mnesarte.

huallaga (Argia), 1946 : 450-451, fig. 2a. LECTOTYPE g. Peru, Huallaga
region, Balsapuerte, x.1932, G. Klug. Argia huallaga n. sp. g Type, det. F. C.
Fraser [label F.C.F.]. Allotype 9. Locality and date as for lectotype 3. Argia
huallaga n. sp. 2 Allotype [label F.C.F.]. The locality given by Fraser is not
correct. The part of the paper triangles upon which he wrote his determinations
gives the locality as Balsapuerte, not Yumbatos, and the labels have been amended
accordingly.

huonensis (Anaciaeschna), 1926c : 473. Holotype g. Germ. New Guinea,
Huon Gulf, W. Potter. Gabmetzung, 20 miles up Markham River, 29.ii.1920.
Anaciaeschna huonensis sp. nov. ¢ Type [label F.C.F.].

icteroptera (Mnais), 1929b : 592-593. Holotype gj. Mnais icteroptera 3, Kalaw,
Chin Hills, Burma. Type [label F.C.F.].

ida (Macromia), 1924 : 449-450, pl. 25, fig. 4. Lectotype § (Fraser, 1936 : 190).
India, Gudalur, Nilgiris, 3,500 ft., 20.1x.1922, F. C. Fraser. Macromia ida Fras.
ms. ¢ Type [label F.C.F.]._ Allotype 2. India, Gudalur, ix.1922, F. C. Fraser.
Macromia ida Fraser 9, det. F. C. Fraser [labels D.E.K.].

imbricata (Idionyx), 1926e : 205-206, pl. 9, fig. 4, pl. 10, fig. 1. Holotype ¢.
Assam, 5,000 ft., Shillong, 8.vi.1924, Fletcher coll. Idionyx imbricata, Fraser g.
Type [label T.B.F.]. Ex Pusa Collection.

immaculata (Zygonyx), 1933c : 134-135. Syntypes, 2 g. Z. immaculata. Pak
Muang, Laos, ca 600 m., 24.iv.32, A. Kerr. Z. immaculata. Hat Kam, Laos,
Wiangchan, 28.iv.1932, A. Kerr [labels F.C.F.]._ Fraser did not mark either of
the two males of the type series as types and they are therefore left as Syntypes,
pending a revision of the genus.

incisura (Gynacantha), 1935 : 325-326, fig. 3. Holotype g. G. incisura.
Loimwe, S. Shan States, 5,600 ft., Burma, 25.ix.23. Type. [label F.C.F.].
Now Acanthagyna incisura (Fraser).

indica (Hylaeothemis), 1946c : 97-100, figs. I-3. LECTOTYPE 9. Hylaeo-
themis fruhstorferi 9 Nilgiris, June 1919, Type, F. C. Fraser [label F.C.F.].
Hylaeothemis indica Fraser 2 Type, D. E. Kimmins det. 1964. Allotype .
Same locality, Hylaeothemis indica Fraser g Allotype, D. E. Kimmins det. 1964.
These two specimens have been in BM (NH) over the label fruhstorferi since 1920,
and at a later date Fraser wrote Type (in red ink) on the label of the female, but
he omitted to put a determination label on either specimen. This was probably
about the time that he described indica. I have therefore labelled the specimens
as types of H. indica Fraser. The situation is still obscure, as in his description
he refers to the ‘‘ form apicalis Fraser”’, which he considered as synonymous
with the new species imdica. H. apicalis Fraser, 1924, however was described as a

FRASER ODONATA TYPES IN BM. (N.H.) 197

subspecies of H. fruhstorferi and the name apicalis therefore has priority over
H. indica, and the species should be quoted as Hylaeothemis apicalis Fraser.
Syn. of Hylaeothemis apicalis Fraser.

indica (Indolestes), 1922 : 58-60. Holotype g. Khasi Hills, Shillong, 20.ix. 1919.
Fletcher coll. BL J, Shillong, 20.x.19. Indolestes indica g Type Fraser, Fraser
det. 1921 [labels T.B.F.]. Ex Pusa Collection. The co-type 2 was not received
from the Pusa Collection.

indica (Macromia), 1924 : 448-449, pl. 25, fig. 5. Holotype g. India, Gudalur,
Nilgiris, 3,500’, I4.ix.1922. Maj. F. C. Fraser. Macromia indica Fras. Ms. ¢
Type [label F.C.F.].

indica (ssp. of Micromerus lineatus), 1928 : 686-687, pl. 1, fig. 5 and pl. 3, fig. 2.
Holotype g. India, Poona, vi.1917, F. C. Fraser. M. lineatus $9, Poona,
June ’17 [label F.C.F.]._Micromerus lineatus indica Fras. ¢ Type, D. E. Kimmins
det. Allotype 9. India, Poona, vi.1917, F. C. Fraser. Micromerus lineatus
indica 2 Allotype, D. E. Kimmins det. Now Libellago lineata indica (Fraser).

indicum (Pseudagrion), 1924 : 495-497, figs. 3i, ii, iv, v. Holotype g. India,
Nilgiri Nadgani, 24.vili.1g22, F. C. Fraser. Pseudagrion indicum Fras. J Type
[label F.C.F.]._ The locality should be Nilgiri Wynaad, Nadgani. Fraser records
types in BM (NH) but only the above specimen was presented before his collection
was received here. Fig. 127a (1933b) gives a poor impression of the bifid superior
appendage of the male.

inflata (Ceylonolestes), 1933b : 67-68, fig. 30. Holotype ¢g. Ceylonolestes
inflata Fras. J, Maymyo, N. Shan Sts, 5.vi.25, F. Wall. Sympycna assamica ¢
Fras. {labels F.C.F.]._ Now Indolestes inflata (Fraser).

infrequentula (Argia), 1946 : 451-452, figs. 2h,7d. LECTOTYPE 3. S. America.
S. E. Colombia: Umbria, 12.xii.1930. Pres. by Leeds City Mus. Argia infre-
quentula Jj Type [label F.C.F.]. Allotype 2. Same data, except date, 25.xi.1930.
The date on the holotype is an error for 12. xi. 1930.

[infumata (var. of Mecistogaster linearis), 1936 : 462.| No examples of this
female variety with ‘‘ evenly and rather deeply enfumed ’’ wings have been traced
in the Fraser collection, nor were any included in the types of new species presented
by the Leeds City Museum. This varietal name was proposed provisionally by
Fraser, but under the International Code (Art. 17, (8), (9)), the name is available
should it at any time be considered desirable to elevate the form to sub-specific
rank.

inglisi (Lamellogomphus), 1924c : 984-985, pl. 1, fig. 2. Lectotype $ (Fraser,
1934 : 275). Lamelligomphus inglisi 3, Mangpu, Darjeeling, C. M. Inglis, 29.ix.23
{label F.C.F.]. Lamellogomphus inglisi Fraser. ¢ Type, D. E. Kimmins det.
1963. Allotype 9. Lamelligomphus risi 2, Riyang R., Mangpu, Darjeeling Dt,
23.1x.23 [label F.C.F.]._ Lamellogomphus inglisi Fraser 2 allotype, D. E. Kimmins
det. 1963. Both locality labels differ in date from that given in the original
description (17.v.23), and the 9 is clearly an example of the re-labelling that

198 D. E. KIMMINS

Fraser occasionally did when a species became asynonym. Syn. of Onychogomphus
vist (Fraser).

inglisi (Stylogomphus), 1922 : 70-71, pl. 7, figs. 3, 3a-b. Holotype g. Teesta
River, 8 June 1920, C. M. Inglis coll. Stylogomphus inglisi Fraser g Type [labels
T.B.F.]. Ex Pusa Collection.

intermedia (Protorthemis), 1936c : 700-701, I fig. Holotype g. Protorthemis
intermedia Fras. § Type, Palawan [on reverse of label, F.C.F.].

interrupta (Agriocnemis), 1927a : 22, figs. Ic, d; 1953 : 125-126. Holotype 3.
W. Samoa, Upolu Is., Malololelei, 2.vii.1924, J. S. Armstrong. Agriocnemis
interrupta Fraser, 3, det. F. C. Fraser. Allotype 2. A. interrupta Fras. 9, A.
Grey’s Reach, Vasigano River, 1,900’, 5.1.1952, J. S. Armstrong [label F.C.F.].

intricata (Idionyx), 1926 : 201-202, pl. 8, fig.6. Lectotype $ (Fraser, 1936 : 237).
Type. I. intricata g, Cherrapunji, Assam, 6,000’, 15.vii—30.viii.23. T. B.
Fletcher. det. F. C. Fraser [label F.C.F.]. Allotype 9. I. intricata 9, Tura,
Garo Hills, Assam, 3,500-4,000’, 15. vii-30.viii.23, T. B. Fletcher. det. F. C.
Fraser [label F.C.F.].

irata (Macromia), 1924 : 454-455, pl. 25, fig. 6. LECTOTYPE g. S. India,
Coorg, Bhagmandala Rd, 3.vi.1923, Maj. F. C. Fraser. Macromia irata j Type
label F.C.F.]. Allotype 9. M. irata 9. Napoklu Rd, Coorg. Shot down.
18.v.1924. S. India. F. C. Fraser [label F.C.F.]. The phrase “‘ Shot down”
indicates that it was captured by shooting with a dust-shot cartridge.

[irma (Megalestes), 1926f : 32-33.) The Fraser collection contained 3 J, I 2 with
data agreeing with the types and to 1 g, I 9 of which he had attached type-labels.
These cannot however be the types, which were placed in the Darjeeling Museum
where, according to a letter from Fraser (see under binocellata, p. 183) they were
destroyed by insect pests. The 4, in addition, has the abdomen 5 mm. shorter
than the measurement given for the type. In my opinion these should all be
considered as paratypes, though eligible for designation as a neotype should such
action be deemed necessary.

isa (ssp. of Zygonyx iris), 1926 : 767-768. Holotype g. Z. isa gj. Maymyo, N.
Shan States, U. Burma, coll. F. Wall. 19.vi.24 [label F.C.F.]. This subspecies
appears to have been based upon a single specimen, which is thus the holotype.

isis (Zygonyx), 1924 : 440-441. Holotype g. S. India, Coorg, Fraserpet, 14.ix.
1923, Maj. F. C. Fraser. Z. isis J, Fras. Type [label F.C.F.]. Now Zygonyx
torrida isis Fraser.

kalarensis (Heliogomphus), 1934 : 329-330, fig. 1orb; 1942a : 340. Holotype dg.
Heliogomphus kallarensis, Kallar, 1,500’, Nilgiris, March 1916, S. India. F. C.
Fraser [label F.C.F.]. Allotype 9. H. kalarensis 9, Kotagiri Ghat, Nilgiris,
S. India, 9. vii. 33, F. C. Fraser coll. [label F.C.F.]. The specific name and locality
are spelt with double “‘ 11” on the holotype label.

kerri (Amphithemis), 1933c : 130-133, fig. 8. LECTOTYPE 3g. Type F. C. F.
Laos, Taweing, 500 m., 4.iv.1932, A. Kerr [label D.E.K.]. Hylaeoth. kerri g.

FRASER ODONATA TYPES IN B.M. (N.H.) 199

[with ‘‘ Amphi” written below in pencil, label F.C.F.]. Allotype 9. Laos,
Muang Baw, 27.iv.1932, A. Kerr [label D.E.K.]. Allotype. A. kerri 2 [labels
F.C.F.]. The specimens were received from Fraser without locality labels and
the above labels have been transcribed from the original description.

kerri (Macrogomphus), 1932 : 286-287, figs. 1a, b. Holotype g. M. kerri ¢.
Pak Tawan, Pran River, Prachaup Province, N. Siam, 1.viii.31, Coll. A. Kerr
[label F.C.F.]. The original description gives the country as P. Siam.

kerri (Onychogomphus), 1933c : 141. LECTOTYPE g. Onychogomphus saun-
dersi kerri ¢. Dawn Ton, N. E. Siam, Dr A. Kerr, 15.v.32, Type d [label F.C.F.].
This species is briefly described as O. kerri from two examples (one of which has
been traced in the Fraser collection), in the discussion of a female of O. saundersi
Selys ?, giving characters by which it may be distinguished from saunderst, but
without locality data.

kerri (race of Rhinocypha iridea), 1933c : 121-122, fig. 4b. Holotype g. R.
iridea kerri, Muang Awm, Laos, ca 900 m., 7.iv.32, (Type), Dr A. Kerr coll.
[label F.C.F.]. Now Aristocypha iridea kerri (Fraser).

kimminsi (Chlorogomphus), 1940a : 55-56, fig. 1. Holotype g. Pulo Nias.
Chlorogomphus kimminsi ¢ Fraser [label D.E.K.]. Allotype 9. Pulo Nias.
Chlorogomphus kimminsi 9 Fraser [label D.E.K.]. Fraser has incorrectly inter-
preted the locality label as Nias Island, Pulo. Pulo is Malay for “island ’’.

kimminsi (Macromia), 1954 : 63-64, figs. 2, 6, 7. Holotype 3. W. Africa,
Sierra Leone, Kaballa, 29.v.1912, J. J. Simpson. M. kimminsi Fraser 3, Type.
F. C. Fraser det. 1953 [label F.C.F.].

klossi (Anotogaster), 1919b : 456-457, pl. 1. Holotype 9. South Annam [label
D.E.K.].. Dran. Eyes green, fore body yellow and black, after body tawny and
black, base of wings deep yellow. Anotogaster klossi Fraser sp. nov. Anotogaster
klossi Fraser 9, sp. nov. [label F.C.F.]. The specimen had at some time lost its
locality label and BM Register label and these have been replaced.

kodaguensis (Gomphidia), 1923a : 671-672. Holotype g. S. India, Coorg,
Bettagiri, 7.v.1924, Maj. F. C. Fraser. G. kodaguensis Fras. J. Type F. C. F.
(labels F.C.F.].

krishna (Aciagrion), 1921 : 543. No specimens labelled krishna were found in
Fraser’s collection but 3 ¢ were located in an envelope labelled “‘ Mahableshwar,
Aciagrion hisopa race’’. The dates fell within the limits of Fraser’s visit to this
locality (20 April-r May 1920) ; it may be mentioned that in the original descrip-
tion the date is given as 23rd May ; this should obviously have been 23rd April.
These three males have been labelled as Syntypes.

kumaonensis (Davidius), 1926d : 170-171, pl. I, fig.6. LECTOTYPE. Assam,
Kumaon, 7,000 ft., v.1923. T. B. Fletcher [label D.E.K.]. Davidius kumaonen-
sis [label F.C.F.]. This specimen has at some time lost its BM Register number
label. I have been unable to trace it in the Departmental Register and it has

200 D. E. KIMMINS

therefore been re-registered as B.M. 1924-536. The type had also been given an
incorrect locality label by Fraser (Kurseong, Sikkim, iv.32 instead of Kumaon,
Assam, v.23) and this has been corrected.

lachrymosa (Pacificagrion), 1926g : 505-507, pl. 49; 1953 : 120-122, fig. I.
Holotype ¢. W. Samoa, Upolu Is., Malololelei, 3.vii.1924, J. S. Armstrong.
Pacificagrion lachrymosa Fras. § Type. Upolu Is., Malololelei, 3.vii.24. F. C.
Fraser det. 1924 [label F.C.F.]. Allotype. P. lachrymosa Fras. 2, Fuluason R.,
above Mooris Dam, 7.x.41, J. S. Armstrong coll. F. C. Fraser det. 1952 [label
F.C.F.]|.

laidlawi (Burmagomphus), 1924 : 475-476. Holotype 9. Burmagomphus laid-
lawi 9, Gudalur, Nilgiris, Pandy R., 8.vii.2r [label F.C.F.]. Allotype 3. B.
laidlawi Fras., Nilgiri Wynaad, 3,500’, 13.viii.22, F. C. Fraser coll. [label F.C.F.].
Originally described as Gomphus sp., Fraser, 1922c : 419.

laidlawi (Idionyx), Holotype 9. Malay Penin.; Pahang, F.M.S., Frasers Hill,
4,200’, 2.vili.1932; on reverse, H. M. Pendlebury, F.M.S. Museum. Idionyx
laidlawi Fr. Type @ [label F.C.F.].

[laidlawi (Pseudagrion), 1922 : 48-50.] Type (originally stated to be in Fraser’s
collection and later (1933b : 296) in BM (NH)) has not yet been traced.

laidlawi (Rhinocypha), 1924 : 482-483, 519-520; 1928 : 457-458, pl. 2, fig. 3.
Lectotype ¢ (Fraser, 1934 : 55). S. India, Coorg-Kanara, 9.xi.1923, Maj. F. C.
Fraser. Rhinocypha laidlawi Type ¢ [label F.C.F.]. Allotype 9. C. laidlawi,
Sampaji Ghat, Coorg, 500 ft., 18.i.24, S. India, F. C. Fraser [label F.C.F.]. The
type ¢ was presented to BM (NH) in 1927. Now Calocypha laidlawi (Fraser).

lankanensis (Ceylonosticta), 1931a : 335, pl. 2, figs. 11-12; 1933 : 220-222, fig. 12.
Holotype 3. Ceylon. Yerbury. Kottawa, 19.iv.g2. Ceylonosticta lankanen-
sis Fraser ¢ Type, det. D. E. Kimmins, 23.xii.1931. Allotype 9. P. lankanensis.
4 miles above Belihul Oya, Ceylon. r1r.v.32. F. C. Fraser [label F.C.F.].
Now Drepanosticta lankanensis (Fraser).

laosica (Calopteryx), 1933c : 128-129. Holotype g. A. laosica. Muang Cha,
Laos, Siam, 1,100 m., 17.iv.32, Dr A. Kerr [label F.C.F.]. Now Agrion laosica
(Fraser).

leoni (Phyllomacromia), 1928a : 138. Holotype 2. P.leoni9. Yegi, Volta R.,
Gold Coast, W. Africa, 13-14.xi.26, H. Hargreaves. M. leoni (Fras.) 9. F. C.
Fraser det. 1953 [labels F.C.F.]. Syn. of Macromia bifasciata (Martin).

{leucostigma (Disparoneura), 1933e : 225-226.] The original description states
“Type and allotype female in Colombo Museum ’’, and a footnote adds “ The
type, in accordance with the usual custom of the Colombo Museum, will be pre-
sented to the British Museum on publication of this paper’. This does not appear
to have been done. The Fraser collection contains a pair from the type series
which should be considered as paratypes. Now Elattoneura leucostigma (Fraser).

FRASER ODONATA TYPES IN B.M. (N.H.) 201

libyana (Gynacantha), 1928a: 136. Holotype g. H. libyana (Fras.), Lake
Victoria, Entebbe, Uganda, G. Hale Carpenter, vii—vili.28 [label F.C.F.]. Now
Heliaeschna libyana (Fraser).

lieftincki (Heliogomphus), 1942a : 338, fig. 1 (6), 3 (13). Holotype 9. Helio-
gomphus 2, Maymyo, Upper Burma, Col. F. Wall. Type. lieftincki Fraser
{labels F.C.F.].

lieftincki (Macromia), 1954 : 65-66, figs. 2, 4. LECTOTYPE ¢. Fernando
Po, Moka, 28.i-3.11.1933, W. H. T. Tams. M. lieftincki Fraser g Type. F. C.
Fraser det. 1953.

lieftincki (Rhodothemis), 1954b : 148-149, figs. Im, n. LECTOTYPE 4.
N. Queensland, Annan River, Cooktown, 30.i1.51, R. Dobson. I have not been
able to trace a female from Annan River or Redlynch in Fraser’s collection and I
have therefore selected a female from Cooktown, N. Q., i.1908, R. J. Tillyard,
BM (NH) coll. as allotype.

lilliputians (Microgomphus), 1923 : 60; 1923b : 329; 1926a : 855-856, pl. 1,
fig. 7. Holotype 9. Microgomphus lilliputians 9, Tenasserim, Burma, 18.v.22,
Type, F. C. Fraser [label F.C.F.]. The first description of this species was in a key
to Indian Dragonflies (1923 : 60); an almost identical description occurs in the
second reference, a key to Indian Gomphidae. Seeunder cyaneofrons (p. 187).
In the third reference a full description is given and the type locality isfcited. In
1934 : 358, the date of the type is given as 18.iv.1923; this must be incorrect,
as the first description was published only two days later!

lindgreni (Onychogomphus), 1923: 65 (in key); 1923b: 332 (Mesogomphus) ;
1924c : 995-996. Holotype 3. Mesogomphus lindgreni, Turzum, Darjeeling Dt,
O. Lindgren [label F.C.F.].. The locality was incorrectly given and the correct
locality has been written on the reverse of the label. Now Paragomphus lindgreni
(Fraser).

lindgreni (Protosticta), 19202: 150. Holotype g. India, Darjeeling, iii.1920,
O. Lindgren [printed label]. Turzum Tea Estate, Darjeeling, ca Mar. 20.
Protosticta sp. nov. P. lindgreni Fras. 3, Type specimen [labels F.C.F.]. Syn. of
Protosticta himalaica Laidlaw.

long fieldae (Enallagma), 1947a : 146-148, figs. Ib, f, 2a. Holotype g. E. Africa,
Uganda, W. shores of L. Victoria, vii-ix.1927, G. D. Hale Carpenter [label D.E.K.,
from pencilled data by F.C.F.]. Enallagma longfieldae g Fraser Type [label
F.C.F.]. Allotype 9. Same data. The left superior anal appendage of holotype
had been glued to inside of paper triangle by Fraser, but it was missing when triangle
was opened for setting (iii. 1965).

longicauda (Bayadera), 1928 : 53. There is some doubt whether the example
from Fraser’s collection is indeed the type. The original description gives no
indication as to the depository of the type but in 1934 : 84, Fraser states that it is
in the Darjeeling Museum. In 1958 Fraser wrote to me (see under binocellata,
p- 183) that the types of Odonata in that museum were almost entirely destroyed.

202 D. E. KIMMINS

The specimen of longicauda from Fraser’s collection has been labelled Type by him,
but it may only be a paratype.

longistigma (Indogomphus), 1922c : 422-424, pl. 11, fig. 8. LECTOTYPE 3.
Merogomphus longistigma, Gudalur, Nilgiris, 3,500’, 14.viii.1g21, F. C. Fraser.
The female of the pair mentioned by Fraser has not been traced in his collection,
although it is possible that a female from the same locality but dated 5.viii.21 may
be the allotype. There are no specimens with their original determination labels
(Indogomphus) and the error in altitude of the lectotype and in the date of the
female may be due to re-labelling. Now Merogomphus longistigma (Fraser).

longistyla (Esme), 1931 : 471, Fig. 6, vii LECTOTYPE g. S. India, S. Kanara,
Peraje, 8.x.1924, F. C. Fraser. Esme longistyla g, Type, Sampaji, Coorg,
(Peraje enclave), 8.x.24, det. & coll. F. C. Fraser [label F.C.F.]. Allotype 9.
S. India, S. Kanara, Peraje, 14.ix.1924, F. C. Fraser. Esme longistyla 2 cotype,
Peraje, S. Canara, 14.ix.24. coll. F. C. Fraser [label F.C.F.].

{loogali (Microgomphus), 1923 : 60; 1923b : 329.] M. loogali was originally des-
cribed in a key to Indian Dragonflies, without indication that it was a new species
or of its locality. (See under cyaneofrons, p. 187). Fraser claims to have the type
of loogali, as well as that of burmicus in his collection but I have not found any
examples labelled Joogali and the specimens of burmicus, which he places as a
synonym of loogali, all have dates later than the publication of loogal. The
type would appear to have been lost.

loringae (Philoganga), 1927 : 79-81. LECTOTYPE 3. Burma, Maymyo, I. viii.
1925, F. Wall. Philoganga loringae 3, det. F. C. Fraser. Type [label F.C.F.].
Allotype 2. Same data. The date on all three of the type-specimens was given as
1g.vi.1924, and has been altered to agree with the published data.

lyttoni (Gynacantha), 1926f : 31-32. Holotype g. Darjeeling Distr., Singla,
April, C. M. Inglis [label D.E.K.]._ Gynacantha lyttoni J sp. nov. (Type) [label
F.C.F.]. Syn. of Acanthagyna bayadera (Selys).

mackwoodi (Caconeura), 1919 : 876-877. Holotype g. Caconeura mackwoodi
sp. nov., Djatalawa, 5,000 ft., Ceylon. Type. F.C. Fraser [label F.C.F.]. Syn.
of Elattoneura centralis (Selys).

magnus (Paragomphus), 1952 : 6-8, figs. 3a-f. Holotype 9. Shimba Hills,
Kenya Col., Dec. 1950, E. Pinhey. Paragomphus n. sp., det. F. C. Fraser, nearest
cognatus (Ramb.). Largest species known [label F.C.F.]. Holotype, Paragom-
phus magnus Fraser 9 [label E. Pinhey].

makoka (Argia), 1946 : 455-456, fig. a. Holotype g. S. America: S. E. Colom-
bia, Umbria, 26.xi.1930. Argia makoka n. sp. $ [with name kokamella], Umbria,
26.xi.30. Argia mokaka gn. sp. Type [labels F.C.F.]. The locality given in the
description is Peru, Mishuyacu, xi.30. The spelling ‘‘ mokaka”’ is obviously a
slip. There is also a 2 (not mentioned in the description) from Umbria, 31. xii. 30,
which is marked Argia makoka n. sp. Allotype in Fraser’s writing. The apex of
the abdomen is missing.

ERASER CDONATA TYPES IN BoM. (NH?) 203

malabarensis (Epophthalmia), 1935 : 328-329. See binocellata (Macromia).

malabarensis (Lamellogomphus), 1924c : 990-991. Holotype 9. Lamellogom-
phus malabarensis 9, Palghat, Malabar, Coll. T. N. Hearsey, 16.vi.2r [label
F.C.F.].. Now Onychogomphus malabarensis (Fraser).

malabarica (Lestes), 1929 : 847-848, pl. 2, fig. 6. Holotype g. Type, L. mala-
barica g, Chevayur, Malabar, 26.vi.28, Fraser [label F.C.F.]. Allotype 9.
~ Same data.

malabaricum (Pseudagrion), 1924 : 494-495. LECTOTYPE 3. Pseudag. mala-
baricum g. Kodaikanal, Palni Hills, 4.ix.22, T. B. Fletcher coll., det. F. C. F.,
Type.

malabarica (ssp. of Zygonyx iris), 1926 : 764-766. Lectotype 3 (Fraser, 1936 :
396). India, Nilgiris, 3-4,500 ft., vi.rg18, F. C. Fraser. Z. iris 3, June ’18,
Nilgiris, 3,000—4,500 ft., [label F.C.F.]. Zygonyx iris malabarica Fras. ¢ Type,
det. D. E. Kimmins, 1962. Allotype 2. India, Nilgiris, 111.1916, 3,500 ft., F. C.
Fraser. Zygonyx iris g, Nilgiris, 3,500 ft., Mar. rg1g, Coll. F. C. Fraser [label
F.C.F.]. Zygonyx iris malabarica Fraser 2 Type, det. D. E. Kimmins, 1962.

malcolmi (Brachydiplax), 1922 : 231. LECTOTYPE g. Siam, Bangkok, ix.
1921, E. W. Trotter. Brachydiplax malcolmi ¢ Type, Coll. E. W. Trotter, Sept. 21,
Bangkok [label F.C.F.]. Allotype 9. Same data.

malloryi (Davidius), 1926d : 167-168, fig. 3iv, pl. 1, fig. 5. LECTOTYPE 4d.
Assam, 5,500 ft., Khasi Hills, Laitlynghat, 2r.iv.1924. Fletcher coll. Davidius
davidi anchorites Fraser Type [labels F.C.F.]. Davidius malloryi Fras., det.
D. E. Kimmins. Ex Pusa Collection. This appears to be another instance of a
specific name being changed by Fraser before publication. I have no doubt that
this specimen is D. malloryi Fraser. There was also a second specimen labelled
Type in Fraser’s collection, but since the Pusa specimen has been marked Type in
B.M. (NH) for about forty years, I see no reason to make a change now.

martini (Davidioides), 1924 : 472-473, fig. 3. Holotype g. Davidioides martini
6, Kunnoth, Malabar, 19.v.23. Type. S. India, F. C. Fraser [label F.C.F.].

martini (Platygomphus), 1922 : 68-69. Holotype g. Shillong, 6.vi.19g20, Flet-
cher coll. Pusa Coll. Platygomphus martini g, Fraser det. rg2r [label T.B.F.].
Ex Pusa Collection. Now Merogomphus martini (Fraser).

mascarenica (Nesocordulia), 1948 : 5-7, fig. 1. Holotype g. N. mascarenica,
Mandritsara, Madagascar (W. Janson), with on reverse, Type F. C. F. [label
F.C.F.]. The allotype @ is said to be in the Paris Museum.

medusa (Protosticta), 1934a : 135-136, fig. 2. Holotype g. P. medusa Fraser,
3 Type, Mergui, Lower Burma, coll. Dr A. Kerr, 31.v.1932, det. F. C. Fraser.

metallica (ssp. of Zygonywx iris), 1931 : 450-451. Holotype 3. Z. metallica Fras.
g- Vayitri Ghat, 32 mile. S. Malabar, 14.iv.29, coll. F. C. F., Type [label
F.C.F.]. Allotype 9. Z. metallica 2, Vayitri Ghat, S. Malabar, 32 mile, 7.iv.29,
F. C. Fraser [label F.C.F.].

204 D. E. KIMMINS

mildredae (Ischnura), 1927 : 87-88. Holotype g. Burma, Maymyo, 7.vi.1925,
Col. F. Wall. [label D.E.K.]. Ischnura mildredae 3, Fraser, Type [label F.C.F.].

mildredae (ssp. of Zygonyx iris), 1926 : 766-767. LECTOTYPE 3. Z. mildredae.
Am Sakan, N. Shan States, Burma, 18.v.25, F. Wall. Allotype 9. Same data.
The allotype lacks its head.

millardi (Gynacantha), 1920a : 147. LECTOTYPE ¢. India, Poona, v.1918,
F. C. Fraser. G. millardi Fraser J, Poona, May ’18, type spec. [label F.C.F.].
Allotype 2. Same data. The dates on the type specimens contradict the des-

cription, which gives x—xi and ii-iii. Now Acanthagyna millardi (Fraser).

[miniata (Macromia), 1924 : 450-451, pl. 25, fig. 7). LECTOTYPE ¢g. Coorg,
Somwarpet, I.vii.23. flavocolorata. [data on envelope, F.C.F.]. Lectotype dg,
Macromia miniata Fraser, det. M. A. Lieftinck. In Leiden Museum. Neither of
the two ¢ syntypes had been traced in Fraser’s collection, but Dr M. A. Lieftinck
recently informed me that he had the specimen listed above, which was presented
to him by Fraser as an example of Macromia flavocolorata Fraser. Fraser had
placed M. miniata as a synonym of M. flavocolorata and, as in similar cases, had
re-labelled it and presumably did not realize that it was one of the types of minzata
when he gave it away. Dr Lieftinck has marked it as the lectotype of M. minzata
Fraser, and since the other syntype of miniata cannot be traced, I take this oppr-
tunity of publishing his designation. Dr Lieftinck also informs me that he con-
siders M. miniata to be distinct from M. flavocolorata.

minima (Idionyx), 1931 : 453-455, fig. 1a. LECTOTYPE g. Type. I. minima,
6.vi.31, Munnar Gt, Travancore, S. India, F. C. Fraser [label F.C.F.]. Allotype 9.
I. minima, Munnar Gt, Travancore, F. C. Fraser [label F.C.F.].

mishuyaca (Argia), 1946 : 453-454, figs. rb, c. LECTOTYPE 3. Type. Argia
mishuyaca 3, Katzenbach (Mishuyacu), 15 km. v. Iquitos, Peru, x.30. via Nagel
[label F.C.F., on reverse of mounting card]. Allotype 9. Argia mishuyaca 9,
Mishuyacu, 4.iv.30 [label F.C.F., on reverse of mounting card].

mollusca (Argia), 1946 : 454-455, figs. 1j, k,l. LECTOTYPE 3. Type. Argia
mollusca 3, Mishuyacu, Peru, 4.iv.30 [label F.C.F., on reverse of card]. Allotype
2. Same locality, 24.vii.30 [label F.C.F., as above].

montana (Coeliccia), 1933c : 116-117. Holotype g. C. montana Fras., Muang
Huang, Laos, Evergreen jungle, A. Kerr, Type, 31.iii.32. det. F. C. Fraser
{label F.C.F.].

montanus (Phyllogomphus), 1957 : 14, figs. 1a, 6b, 7c, 8b. Fraser did not specify
a holotype, but marked a @ in his own collection as Type, probably because it was
in better condition than the mis-identified 2 type of P. aethiops Selys in BM (NH).
I have accepted the Fraser specimen as type and hereby designate it as LECTO-
TYPE 9. The data on the specimen are (1) Phyllogomphus coloratus Kimmins,
Cameroons, 9, det. F. C. Fraser; (2) Phyllogomphus montanus Fraser 2 Type,
det. F. C. Fraser [labels F.C.F.] and (3) Phyllogomphus montanus Fraser 9

FRASER ODONATA TYPES IN B.M. (N.H.) 205

LECTOTYPE, det. D. E. Kimmins, 1965. The mis-identified 2 type of P.
aethiops Selys has been marked as a paratype of P. montanus Fraser.

[montana (ssp. of Vestalis gracilis), 1934 : 128.) The types of this subspecies,
which was a new name for V. gracilis amoena Fraser, have not yet been traced in the
Fraser collection.

mortoni (Protosticta), 1924 : 500-501, pl. 26, fig. 4. There has been some con-
fusion over the labelling of the specimens presented to the BM (NH) in 1923 as
types of this species. The description gives the type locality as Coorg, Sampaji
Ghat, 23.v.23. The specimens bear museum-written labels Mangalore Ghat,
3,000 ft., 13.v.1922. Fraser’s collection contains examples from Sampaji Ghat
with dates 13.v.22, 13.v.24 and 12.v.32. There have obviously been errors in
transcribing dates, etc., both by Fraser and possibly by our setting room staff.
Unfortunately the original locality labels from the paper envelopes were not pre-
served, so there is no way of checking what Fraser actually wrote on these envelopes.
I have therefore decided to give the published locality and date to the types pre-
sented to BM (NH) by Fraser. There are subsequent examples from Sampaji Ghat
dated r.v.24, but I do not consider them as part of the type series.
Lectotype ¢ (Fraser, 1933b : 113). S. India, Coorg, Sampaji Ghat, 23.v.23,
Maj. F. C. Fraser. Protosticta mortoni Fras. ms. (Type), P. mortoni ¢ [label
F.C.F.]. Allotype 9. Locality as above, P. mortoni 2 Cotype [label F.C.F.].

moundi (Phyllogomphus), 1960 : 141-144, figs. a-i. LECTOTYPE 3. Phyllo-
gomphus moundi Fraser $ Type, Ibadan, Nigeria, 1960, leg. L. Mound [label
F.C.F.]. Allotype 9. Phyllogomphus moundi Fraser, @ allotype, Ibadan,
Nigeria, 1960, leg. L. Mound [label F.C.F.].

mudiensis (Esme), 1931 : 472-473. LECTOTYPE 3g. Esme mudiensis 3, Mun-
nar, Travancore, 24.v.29, coll. F. C. Fraser (Type) [label F.C.F.]. Allotype @.
Same locality, 23.v.29. F.C. F. Coll. Allotype (label F.C.F.].

multinervosa (Devadatta), 1933c : 129-130. Holotype ¢ (teneral). Devadatta
multinerve, Pu Tat, Laos, ca 1,200 m., Dr A. Kerr, Evergreen jnl., 22.iv.32.
Devadatta multinervosa Fraser § Type. D. E. Kimmins det. 1965. This type
was located in a store box in Fraser’s collection, without any indication that it was
a type. It agrees in locality and neurational data and I have no doubt that it is
the type. The left fore wing is now missing. The specific name is mis-spelt on
the label.

nadganiensis (Idionyx), 1924: 460. Lectotype 9 (Fraser, 1936 : 238). India,
-Nilgiris, Nadgani Ghat, 3,000 ft., Maj. F. C. Fraser. Idionyx nadganiensis
Fraser 9 Type [label F.C.F.].

naia (Agriocnemis), 1932 : 747; 1933b : 387-389, fig. 166. Lectotype ¢ (Fraser,
1933b : 389). Mergui, Burma, Bott coll. Agriocnemis naia Fraser [labels F.C.F.].
Allotype 9. Labels as above. This species was originally described in a key to
Indian Dragonflies, without locality or indication that it was a new species. See
ENToM. 18, 6. 16

206 D. E. KIMMINS

under cyanofrons (p.187). The type bears a label by Lieftinck “‘ This is the same
species as nana Laidlaw”’. Syn. of Agriocnemis nana Laidlaw.

nietneri (Ceylonosticta), 1931a : 334-335, pl. 2, figs. vi-viii. Lectotype g
(Fraser, 1933 : 218). Type. C. nietneri, Kandy, 2,000’, Col. F. Wall [label
F.C.F.]. The allotype 2 was destroyed but Fraser selected a replacement female
which he suggested should be a neallotype. It is labelled C. nietneri, Belihul Oya,
Ceylon, 1r.v.32, F. C. Fraser. Now Drepanosticta mietneri (Fraser).

nigra (Caconeura), 1922 : 42-43. Holotype 3. King Island, Mergui, Bott coll.
Caconeura nigra Fraser Type 3, Fraser det. 1921 [labels T.B.F.]. Now Proda-
sineura nigra (Fraser).

nigrescens (Vestalis), 1929b : 584-585. LECTOTYPE g. V. nigrescens 3,
Kandy, Ceylon, 1.ix.24, F. Wall coll. det. F. C. F. Type [label F.C.F.].

nigriceps (Lestes), 1924a : 85-86. Lectotype g (Fraser, 1933b: 56). Pusa,
I7.vii.22. Fletcher coll. Lestes nigriceps Fraser, Type ¢ [labels T.B.F.].
Ex Pusa Collection.

nigricolor (Amphithemis), 1922 : 700-701. LECTOTYPE g. Type. Bengal,
Hasimara, Duars, H. V. O’Donel [label D.E.K.]. A. nigricolor ¢ [label F.C.F.].
Amphithemis nigricolor Fraser J, D. E. Kimmins det. 1962. Allotype?. Locality
as above. A. nigricolor 2 [label F.C.F.]. Amphithemis nigricolor Fraser Q,
D. E. Kinmmins det. 1962. Syn. of Amphithemis vacillans Selys.

nigricolor (Hagenius), 1924a : 76-79, pl. Io, figs. I-2, 5-7. Holotype g. Sie-
boldius nigricolor, S. Shan States, 23.ix.23, Capt. Drummond [label F.C.F.].

nigroflavum (Ceriagrion), 1933b : 323-324, figs. 138a—b. Holotype g. Lower
Burma, Ceriagrion nigroflavum Fraser $ Type, D. E. Kimmins det., 1965. This
specimen and a paratype from Siam were found in envelopes in the Fraser papered
collection.

nigrolineatus (Anax), 1935b : 23-25. Holotype g. A. nigrolineatus, Nagri spur,
Darjeeling, 31.v.22, Col. Hirst coll. [label F.C.F.]. Allotype2. A. nigrolineatus
{label F.C.F.].

nilgiriensis (ssp. of Idionyx corona), 1922 : 66-67. See burliyarensis (Idionys),
p. 184. The name of this subspecies was changed to burliyarensis by Fraser, 1924,
when it became a secondary junior homonym, due to the transfer to Idionyx of
Phyllomacromia nilgiriensis Fras., 1918. The latter species was later made the
type-species of Idiophya Fraser, 1934, but although J. nilgiriensis Fraser was
only a secondary homonym, under Art. 59 (c) of the Code the name is not eligible
for restoration, since it was rejected before 1960. Syn. of Idionyx corona burl-
yariensis Fraser.

nilgiriensis (ssp. of Onychogomphus biforceps), 1922c : 425-426, figs. a—c.
LECTOTYPE ¢. India, Nilgiris, Gudalur, 16.x.1g21, T. B. Fletcher [label
D.E.K.]. Lamellogomphus nilgiriensis Fras. ¢ Type [label F.C.F.]. Allotype 9.
India. Nilgiris, 3 miles from Gudalur, 16.x.1921, Maj. F.C. Fraser. Gomphus 9,

FRASER ODONATA TYPES IN B.M. (N.H.) 207

Gudalur, Type; O. biforceps nilgiriensis 2 [label F.C.F.]. The locality labels had
been incorrectly transcribed in our setting room and have been corrected from
the original description.

nilgiriensis (Phyllomacromia), 1918 : 383-384; 1922d : 9. Holotype 9. Phyllo-
macromia? nilgiriensis 9, Nilgiris, 1917. Type Fraser. Now Idiophya nilgiriensis
(Fraser).

nocturnalis (Periaeschna), 1927 : 71-72. Holotype g. Assam 5,000 ft., 14 June
1924, Fletcher coll. Beaten from junglein Fruit Garden. Periaeschna nocturnalis
Fraser g. Type [last two labels T.B.F.]. Ex Pusa Collection.

obscura (Indoneura), 1933b : 257-258, fig. 107. Holotype g. I. ramburi var.
obscura Fras. gf Type. 10.x.1926. Koropur Ghat, Jaipur, 2,000 ft., coll. C. A.
Souter. Type of I. obscura Fras. [label F.C.F.]._ Now Caconeura obscura (Fraser).

occidentalis (Phyllogomphus), 1957 : 18-22, figs. 2a, e, 3c, 6a, 7b. Holotype 9.
Bwake, French Ivory Ct [label F.C.F.]. Phyllogomphus occidentalis Fras. 2
Type, det. Fraser [label D.E.K.].

octomaculata (ssp. of Palpopleura sexmaculata), 1935b : 26-27. LECTOTYPE
g. Assam, Sylhet, 11.viii.1g21 [label D.E.K.]. 11.8.2 [label F.C.F.]. Palpo-
pleura sexmaculata octomaculata Fraser, Sylhet, Assam [typewritten]. Palpo-
pleura sexmaculata octomaculata Fras. J. D. E. Kimmins det. 1962.

odoneli (Caconeura), 1924e : 51-52. LECTOTYPE 3g. Bengal, Hasimara T. E.,
Duars, 8.iv.1923, H. V. O’Donel [label D.E.K.]. Caconeura odoneli, Huldibara
T. E., Duars [label F.C.F.]. Allotype 9. Bengal, Duars, Hasimara T. E.,
I0.iv.1923 [label D.E.K.]. Caconeura odoneli, Silli Toorsa, Duars, 10.4.23
{label F.C.F.]. The discrepancies in locality are probably due to errors in labelling
the papered specimens. The lectotype has the head detached and damaged.
Now Prodasineura odoneli (Fraser).

odoneli (Gomphus), 1922c : 420-421, pl. 11, fig. 5. Holotype 3. G. odoneli 3,
Hasimara, Duars, Bengal, 21.ix.20, Type [label F.C.F.]. Allotype 9. Same
data [label F.C.F.].

odoneli (Gynacantha), 1922b : 700. Holotype g. G. odoneli, Hasimara, Duars,
Bengal, India, H. V. O’Donel [labels F.C.F.]._ Now Acanthagyna odoneli (Fraser).

olympicus (Chlorogomphus), 1933f : 257-258; 1933a : 464-465. Holotype ¢.
Chlorogomphus olympicus 3, Kufri, Simla, 8,400 ft., Capt. Elwes, 9.vi.29 [label
F.C.F.]. The second reference also gives olympicus as a new species but is two
months later.

orientalis (Phyllogomphus), 1957 : 16-18. LECTOTYPE ¢. Phyllogomphus
aethiops 3, Bugalla Is., L. Victoria, r.ix.28. G. Hale Carpenter [label F.C.F.].
Phyllogomphus orientalis Fraser 3, det. F. C. Fraser [label D.E.K.]. Allotype 9.
Same data as g. Although labelled as aethiops by Fraser, these specimens were
placed above a typewritten label orientalis in his collection.

208 D. E. KIMMINS

originata (race of Polythore derivata), 1946 : 18-19, fig. 2b. Holotype ¢.
Katzenbach (Mishuyacu), 15 km. v. Iquitos, Peru, 13.vii.31. Polythore sp. ¢.
via Nagel, 1930. P. derivata race originata g. Allotype 9. Same data as 4,
except date, 22.iii.30.

ornata (Idionyx), 1921a : 688-689; 1922: 6. Holotype. Shillong, 2 June 1920,
Fletcher coll. Shillong, 2.vi.20, in jungle & difficult to see on wing. Idionyx
ornata 2 Type, Fraser det. May 1921 [labels T.B.F.]. Idionyx optata Selys 2
(label F.C.F.].. Ex Pusa Collection. Syn. of Idionyx optata Selys.

Osiris (ssp. of Zygonyx iris), 1936 : 400-401. LECTOTYPE 3. Z. mildredae 4,
Ani Sakan, N. Shan Sts, Burma, 18.v.25, F. Wall [label F.C.F.]. Zygonyx iris
osiris § Fras. Lectotype, D. E. Kimmins det. 1965. The type was not marked
by Fraser, and was not traced until 1965, during the preparation of this paper.
It was recognized by the nodal indices of the wings, complete distal antenodal
nervures and five-celled subtrigone.

[ovulosa (ssp. of Chlorocypha dispar), 1947 : 23.| This subspecies is said to be
based upon an example in the McLachlan collection from Old Calabar. Fraser did
not mark any example in this collection as ovulosa and neither of the two Old
Calabar examples agree at all well with the characters given by Fraser for this
subspecies. There are in fact several inconsistencies in this account of C. dispar
by Fraser, and one can only assume that he made rather hurried notes during one
of his visits to the BM (NH) and later got them mixed. He refers to the type of
dispar (which we do not have) as being in the BM (NH), from Sierra Leone. He
gives a reference to fig. 1f as being from a male from Old Calabar, but according to
his legend it is C. dispar cordosa from Gaboon. In view of these discrepancies,
I find myself unable to recognize a type for the subspecies ovulosa.

pacifica (Hemicordulia), 1925 : 435-436. LECTOTYPE g. Western Samoa,
Apia, 30.xii.1922, J. S. Armstrong. pacifica ¢ [label F.C.F.].

palampurensis (ssp. of Anotogaster basalis), 1929a : 88-89. Holotype gd.
A. basalis palampurensis, Palampur, 4,000 ft., 4.v.25. C. M. Inglis.

pallidum (Ceriagrion), 1933c : 118-119, fig. 3. LECTOTYPE g. Type. Laos,
Muang Cha, 1,100 m., iv.32, A. Kerr [label D.E.K.]. C. pallidum Fras., ¢ Type,
Laos [label F.C.F.]._ There is also a female labelled Allotype by Fraser from the
same locality, which is not mentioned in the original description.

pallida (Macromia), 1924 : 456-457, pl. 15, fig. 8. Lectotype gj (Fraser, 1936 :
175). M. pallida, Hasimara, Duars, Bengal, 22.vi.23, H. V. O’Donel.

pallidistigma (Calilestes), 1926c : 489. Holotype g. Tonkin, Ngaio-Tio, 4,800
ft., 26.iv.1924, H. Stevens. Podolestes pallidistigma sp. n. g (Type) F. C. F.
(label F.C.F.]. The type is teneral and lacks the left fore wing and segments 7—10
of the abdomen.

paludensis (Aciagrion), 1922b : 698-699. LECTOTYPE 4g. India, Nilgiris,
Masnagudi, 27.xi.1g21, F. C. Fraser. Aciagrion paludensis Fraser g Type, det.
F. C. Fraser [labels D.E.K.]. Aciagrion paludensis g¢ Type, Masnagudi, Nilgiris,

FRASER ODONATA TYPES IN B.M. (N.H.) 209

27.xi.1g2I, coll. F. C. Fraser [label F.C.F.]. Allotype 9. Dataas g. Three of
the original type-series of five (those from Masnagudi) were traced in the Fraser
collection, in envelopes labelled Aciagrion occidentalis Laidlaw, with which species
it is now synonymized. A gd and 9 were labelled Type and allotype and the ¢ has
been chosen as the lectotype. Syn. of Aciagrion occidentalis Laidlaw.

papavarina (Hetaerina), 1946 : 33-35, figs. 6a, b, 7a. LECTOTYPE g. H.
papavarina g. Mishuyacu, Peru, 15.viii.30 [label F.C.F.]. Allotype 9. H.
papava [sic], same locality as J, 20.vili.30 [label F.C.F.].

patricia (Lestes), 1924: 486, pl. 26, fig. 7. Holotype g. L. patricia Fras.,
Virajpet, Coorg, 24.vi.23, coll. F. C. F. [label F.C.F.]. This does not appear to
be the original label.

periyashola (Idionyx), 1939a : 93-94. Holotype 9. I. periyashola, Manar,
Travancore, W. Ghats, 17.v.1933, S. India, F. C. Fraser [label F.C.F.].

peruviensis (Protoneura), 1946 : 459-460, figs. 5a,b. Holotype 3. P. peruviense
Fras., Mishuyacu, 2.vii.30 [label F.C.F. on mounting card].

pinheyi (Lestes), 1955: 10, figs. 5-6. LECTOTYPE 3. Rusape, 22.2.48,
E. Pinhey. Lestes sp. n. E. C. G. Pinhey det. Mar. 1948. L. pinheyi Fraser
(label F.C.F.]. Allotype 9. Same data. L. pinheyi Fraser [label F.C.F.].

pinheyi (Nepogomphoides), 1952 : 3-6, figs. Ia, b, 2a-d. LECTOTYPE 4.
Kimboza Forest, Uluguru Mts, Tanga Territory, Oct. 1951, E. Pinhey. Nepo-
gomphoides pinheyi § Type, F. C. Fraser det. 1952. Syn. of Nepogomphoides
stuhlmanni (Karsch).

platyceps (Gomphidia), 1953a : 189-190, figs. n, 0. Holotype g. Gomphidia
platyceps Fraser § Holotype, Bhagmandala, Coorg, 17.vi.24. F.C. Fraser coll.
{label F.C.F.].

platystigma (Ceriagrion), 1941 : 63-64, figs. A (1, 4), B (5). LECTOTYPE 4.
Uganda, N. W. shore of L. Victoria, vii.1927, G. D. Hale Carpenter [label D.E.K.].
Type, Ceriagr. platystictoides [sic] ¢ [label F.C.F.]. Ceriagrion platystigma Fras.
3 Type [label D.E.K.]. Allotype 9. Locality label as above [label D.E.K.],

~N. Western shores of L. Victoria, July-Sept. ’27 [label F.C.F.]. Ceriagrion
platystigma Fras., 2 allotype [label D.E.K.]. Fraser’s determination label on
the lectotype ¢ is “‘ platystictoides ’’ but the published name is “ platystigma ”’.

polychromatica (Drepanosticta), 1931a : 338. Holotype g. Darjeeling distr.,
Gopaldhara, v.vi [label D.E.K.].. D. polychromatica [label F.C.F.]. This is the
only example of the species found in Fraser’s collection and although not marked
by him as the type, I believe it to be the holotype.

poungyi (Coeliccia), 1924e : 5253. Holotype g. Gokteik, [N. Shan States],
1,500 ft., rr. vi.1924, F. Wall. C. poungyisp. nov., Gokteik, 1,500 ft., F. Wall coll.,
det. F. C. F., Type [label F.C.F.]. The description gives the date of capture as
26.x.2I, probably another error of transcription from the original label.

praeclarum (Pseudagrion), 1924 : 493-494, fig. 3, vi; 1931 : 466. Holotype d.
P. praeclarum ¢, Makut, Coorg, 29.xi.23. Coll. F. C. Fraser. Type [labels

210 D. E. KIMMINS

F.C.F.]. India, Coorg, Makut, 29.xi.1923, F. C. Fraser. Pseudagrion prae-
clarum Fras., g Type, det. F. C. Fraser [labels D.E.K.]. Allotype 9. Tamara-
cherry, S. Malabar, 23.vi.28, F. C. Fraser coll. A. mimetes praeclara 9. In his
second reference to this species Fraser refers to the type as being rather teneral,
but neither his original description nor the type suggest this. Syn. of Archibasis
oscillans (Selys).

prateri (Pseudotramea), 1920a: 149. Holotype g. Pseudotramea prateri 3
Type, Turzum, Darjeeling, Bengal, Feb. 1920, F. C. Fraser [label F.C.F.].

preciosus (Orogomphus), 1924a : 75-76, pl. 9, fig. 8; I929a : 154-156. Holotype
Q. Sikkim, Mungpu, 4,000 ft., I0.v.22. Orogomphus preciosus Fras., 2 Type.
Fraser det. 1924 [labels T.B.F.]. Ex Pusa Collection. Allotype g. Chloro-
gomphus preciosus g, Pashok, Darjeeling, C. M. Inglis, 25.v.25 [label F.C.F.].
Now Chlorogomphus preciosus (Fraser).

protostictoides (Protoneura), 1946 : 458-459. Holotype ¢g. P. protostictoides
Fras. Mishuyacu, Peru, 16.vi.30. [label F.C.F., on mounting card].

pruinans (Heliogomphus), 1922c : 416-417, pl. 11, fig. 2. LECTOTYPE 4.
H. promelas Selys, Burliyar, Nilgiris, 3,000 ft., 29.vii.21. F. C. Fraser coll.
(label F.C.F.]._ Heliogomphus pruinans ¢ Fraser. Lectotype, D. E. Kimmins
det. 1963. Allotype?. Samedata. Fraser subsequently synonymized his species
with H. promelas (Selys), removed his original labels and replaced them with new
ones, which make no mention of the fact that the specimens were the types of
H. pruinans. In his re-labelling, the altitude has been changed from 1,500 to
3,000 ft. Syn. of Heliogomphus promelas (Selys).

pruinosa (Metacnemis), 1928a : 129-130, fig. 3. LECTOTYPE g. Uganda,
Budama & Bunyoli, x.1927, G. D. Hale Carpenter. Metacnemis pruinosa sp. nov.,
F. C. Fraser, type [label F.C.F.]. Allotype 2. Uganda, Budama Dist., x.1927,
G. D. Hale Carpenter. Metacnemis pruinosa sp. nov. [label F.C.F.]. Syn. of
Metacnemis singularis (Karsch).

pulcherrima (Ceylonicolestes), 1924 : 487-489, pl. 26, fig. 9. LECTOTYPE @.
S. India, Coorg, Mercara, I.v.1923, Maj. F. C. Fraser. Lestes pulcherrima 3
Fras. ms., Sidapur Rd swamp, Mercara, Coorg, I.v.23 [label F.C.F.]. Allotype
2. S. India, Coorg, Hallery, Mercara, 25.v.1923, Maj. F. C. Fraser. (Cotype)
pulcherrima [label F.C.F.]. Now Lestes (Indolestes) pulcherrima (Fraser).

pulcherrimus (Onychogomphus), 1927 : 78-79. LECTOTYPE 3. Onycho-
gomphus pulcherrimus ¢ Type, Maymyo, Upper Burma, 31.v.25. The specific
name was originally published as “ pulcherrima”’. Now Nihonogomphus pul-
cherrimus (Fraser).

[quadrilateralis (Aeschna), 1927 : 70-71.] The type of this species (now placed
as a synonym of Tetracanthagyna waterhousei McLachlan) has not been traced
in the Fraser Collection, although it was stated to be there (1936: 119). It was
an incomplete g, with the last five abdominal segments missing. Syn. of Tetra-
canthagyna waterhouset McLachlan.

FRASER ODONATA TYPES IN B.M. (N.H.) 211

ramburi (Coenagriocnemis), 1956a : 70-71, fig. r. LECTOTYPE ¢. Type.
Mauritius, Les Mares, 26.xii.1947, J. Vinson, with on reverse, Coenagriocnemis
ramburi n. sp. g, Fraser. C. ramburi Fraser 3 Type, det. F. C. Fraser. Now
Agriocnemis ramburt (Fraser).

ramburi (Indoneura), 1922 : 2-3. LECTOTYPE 4. Madras, Coonoor, 6,000’,
ix.1924. Mjr F. C. Fraser. I. ramburi g Fras., Coonoor, 6,000’, Type [label
F.C.F.]. Allotype 9. Madras, Ootacamund, Nilgiri Hills, 27.vii.rg2r. Mjr
F. C. Fraser. ramburi Fras. Type. Now Caconeura ramburi (Fraser).

reducta (Uracis), 1946 : 468-469, fig. 6a. Holotype g. Peru, Mishuyacu, 31.v.
1931. Uracis reducta sp. nov. <¢ Type, det. F.C. Fraser. Uracis reducta holo-

typus, prap. Dr Buchholz. The last label refers to a preparation of the penis
made by Dr Buchholz.

rhinoceroides (Idionyx), 1934b : 565, fig. 3F. Holotype 9. S. Malabar, Mannar-
ghat, Dhoni, v.1923, F. C. Fraser [label D.E.K.]. I. rhinoceroides 2 Type [label
ECE],

risi (Gomphus), 1922 : 73-74. Holotype 9°. India, Darjeeling Distr., near
Mangpu, 3,400’, 31.viil.1g20, C. M. Inglis [label D.E.K.]. Gomphus risi Fraser
2 Type [label T.B.F.]. Ex Pusa Collection. The holotype of this species was
missing for a number of years in BM (NH), but was discovered in May 1963 in the
type collection, over the same label as the type of Lamelligomphus cacharicus
Fraser. It had apparently been misplaced when the types were removed from the
main collection for safety in 1940. In the meantime, as the type could not be
traced, Fraser had marked a Neotype in his own collection. This specimen now
bears an additional explanatory label. Now Onychogomphus risi (Fraser).

risi (Indoneura), 1931 : 469-471, fig. 6i. LECTOTYPE ¢g. Type. D. risi
(Fraser), Tamaracherry, S. Malabar, 9.x.1928, F. C. Fraser [label F.C.F.]. Now
Caconeura risi (Fraser).

risi (Mesogomphus), 1924c : 996-997. Holotype g. Mesogomphus risi Kalaw,
S. Shan States, coll. C. Dingavan. Now Paragomphus risi (Fraser).

robertsi (Notiothemis), 1944 : 42-43, figs. 1 (4-5), 2a. LECTOTYPE ¢. Notio-

themis robertsi Fraser. L. Victoria, Uganda, vi.27, G. Hale Carpenter [label
rC.F.].

robusta (Protosticta), 1933c : 111-112. Holotype 9. Taweing, Laos, ca 800 m.,
A. Kerr, 6.iv.32. P. robusta Fras. 9 Type, Laos, 6.iv.32, det. F. C. Fraser.

rusticatus (Mesogomphus), 1928a : 135-136. LECTOTYPE 4g. Uganda,
Bugalla, Is. Sesse, L. Victoria, vili.1927, G. Hale Carpenter. Mesogomphus
rusticatus g, sp.nov. Type [label F.C.F.]. Allotype 9. Same locality and date.
Syn. of Paragomphus hageni (Selys).

ruwensoriensis (Tetrathemis), 1941b : 141-142, figs. 5-6. Holotype g. Uganda,

Ruwenzori Mts, 1927, G. Hale Carpenter [label D.E.K.]. Tetrathemis ruwen-
zoriensis Fraser g Type, det. F. C. Fraser.

212 D. E. KIMMINS

saffronata (Idionyx), 1924 : 458-460. Lectotype g (Fraser 1936: 221). S. India,
Coorg, Mercara, 10.v.1923, Maj. F. C. Fraser. saffronata Fraser [label F.C.F.].
Idionyx saffronata Fraser ¢ [label D.E.K.]. Allotype 9. S. India, Coorg, on
Napoklu Rd, 29.iv.1923, Maj. F. C. Fraser. Flying high in jungle, 5.30 p.m.
Idionyx saffronata 2 Fraser [label D.E.K.].

samoensis (Pseudagrion), 1925 : 430-432, fig. 1. LECTOTYPE ¢. Samoa,
Suimu, 27.x.1923, J. S. Armstrong [label D.E.K.]. P.samoensis g. Allotype 9.
Same data. Presented to BM (NH) as holotype and allotype by Fraser, 1953.

sanguinostigma (Ischnura), 1953 : 123-124, figs. 4, 5, 10, 11, 15, 18. Holotype ¢.
I. sanguinostigma Fras. ¢ Type. 1.x.51r, Moa Moa stream, Apia. Segs. 8-10
blue. J. S. Armstrong coll. F. C. Fraser det. 1952. Allotype 2. I. sanguino-
stigma Fras. 9 Allotype. Moa Moa stream, Apia, 26.v.51. J. S. Armstrong.
F. C. Fraser det. pee The date of the allotype is given as 1931 in the original
description. .

sanguinostigma (Protosticta), 1922d : 6-7, pl. 1, figs. 5, 6. LECTOTYPE 4.
Madras, Burliyar, Nilgiri Hills, Mjr F. C. Fraser, with on reverse, 29.vii.2I.
P. sanguinostigma ¢ Fras. [label F.C.F.]. Allotype 9. India, Nilgiris, Burliyar,
7.vili.tg2r, F. C. Fraser [label D.E.K.]. 7.viii.21, Burliyar, P. sanguinostigma
2 Type [labels F.C.F.]. Protosticta sanguinostigma Fraser 9 Allotype [label
D.E.K.]. There has been some confusion over the location and labelling of the
type. The original description states “‘ Type in my own collection, paratypes in
British and Indian Museums.”’ The dates given are 3.viiand 23.vii.1g21. Ihave
not traced any specimens with these dates but 2 ¢ (of the original four) were
presented to BM (NH) early in 1922 in a collection containing “ types of new
species’; one of the males (type) was labelled 7.viii.192r and the other 29. vii.
1921. In addition. a third male has been found in the Fraser papered collection,
also dated 29.vii.rg21. I have therefore assumed that there had been a printing
error in the second date and that the g with a BM Type label (now Lectotype) had
also been given an inaccurate date and this has been changed to 29.vii.192I, in
agreement with the other two males. The fourth male, possibly dated 3.vii.19g21,
presumably went to the Indian Museum.

schmidti (Onychogomphus), 1937 : 163-164, figs. 1a, 2a. LECTOTYPE J.
Onychogomphus [bistrigatus (deleted)] g, Gopaldhara, Assam, H. Stevens,
21.ix.10. [Allo(deleted)]type g. schmidti Fraser [labels F.C.F.]. Onycho-
gomphus schmidti Fraser, g Lectotype, D. E. Kimmins det. 1965. Allotype 9.
Onychogomphus [bistrigatus (deleted)] 9, Simla Sts, 19.vii.28. schmidti Fraser
(labels F.C.F.]. Onychogomphus schmidti Fraser, 2? Allotype, D. E. Kimmins det.
1965. The specimens listed above were the ones which Fraser considered the types
of his new species O. schmidti. In BM (NH) there is also a 3, marked by Fraser as
an allotype of O. bistrigatus Selys, which is in fact a J O. schmidt.

seductus (Macrogomphus), 1926h : 737-738, pl. fig. 2, text-figs. Ia, 2ili-v.
Holotype g. Macrogomphus seductus g. MHasimara, Duars, Bengal. 8-15.
vi.23. H. O’D. [label F.C.F.]. Allotype 9. Same data [label F.C.F.].

FRASER ODONATA TYPES IN B.M. (N.H.) 213

selysi (Chlorogomphus), 1929a : 158-160, fig. 32. LECTOTYPE g. Chloro-
gomphus selysi 3, Mangpu, Darjeeling, 6.v.27, F. C. F. [label F.C.F.]. Chloro-
gomphus selysi Fraser, § Type, Moungpoo, Br. Sikkim, 3,600 ft., coll. F. C. Fraser,
27.v.27 F. C. F., [typewritten]. There is also an allotype 9, Chlorogomphus
selysi 2, Mangpu, Darjeeling, 6.v.27, F. C. F., which is not mentioned in the
original description. Fraser (1936 : 24) might appear to have indicated a lecto-
type (Type in British Museum) but this example remained in his own collection
until 1958.

senchalensis (ssp. of Davidius aberrans), 1926d: 170. Holotype 9. Davidius
aberrans senchalensis, Senchal, Darjeeling Distr., 8,000 ft., 19.v.24, C. Inglis
(label F.C.F.].

shanensis (Macromidia), 1927: 67-68. Lectotype ¢ (Fraser, 1936: 210).
Type. Macromidia shanensis, Maymyo, Upper Burma, 3r.v.25, Col. F. Wall
(label F.C.F.]. Allotype 2. Macromidia shanensis, Maymyo, Upper Burma,
7.Vi.25. :

siamensis (Burmagomphus), 1926a : 411-412. Holotype 9. Burmagomphus
siamensis 2. Don Chai, 5.xi.23, coll. S. Williamson. Type [label F.C.F.].
The original description gives as locality only Bangkok, Siam. Syn. of Burma-
gomphus williamsoni Fraser.

siamensis (ssp. of Onychothemis tonkinensis), 1932 : 284-286. Holotype 3.
O. testacea siamensis. Siam. Dr A. Kerr. [label F.C.F.]. Onychothemis
tonkinensis siamensis Fraser ¢ Type. D. E. Kimmins det. 1965. The unique
type was found in a store box, without indication that it was a type. It was
identifiable from the two rows of cells in the discoidal field and the uncrossed
triangle, both in the right fore wing. Fraser originally described this as a sub-
species of O. tonkinensis, but the present label is obviously a replacement, as he has
now placed it as O. testacea stamensis. It may well prove to be a small example of
O. testacea Laidlaw.

[siamensis (Pseudagrion). 1922a : 235-236, fig. 2a.| The type of this species has
not yet been traced in the Fraser collection.

sikkima (ssp. of Lestes praemorsa), 1929 : 843-844. Holotype 3. N. E. India,
Sikkim, Kurseong, 4,300 ft., O. Lindgren [label D.E.K.]. Lestes, Sikkim, Lind-
gren [label F.C.F.]. Allotype 9. Same data. Additional data obtained from
envelope and original description. Now Paralestes praemorsa sikkima (Fraser).

sinuatus (Burmagomphus), 1933g : 33-35, fig. 6. Holotype 9. Burmagomphus
sinuatus 9, Urugalla, 8.v.32. Type. F.C. Fraser Coll.

smithi (Crocorthetrum), 1921b : 162-163, pl. 3. Holotype g. Crocorthetrum
smithi J Type, Langbiang Prov., Siam-Annam Frontier, 12.19-1.20. F.C. Fraser
{label F.C.F.].

souteri (Disparoneura), 1924 : 521-522. Holotype g. India, Coorg, Sampaji,
18.iv.1924, Maj. F. C. Fraser. D.souteri. [label F.C.F.]. Allotype 9. India,

214 D. E. KIMMINS

Coorg, Sampaji Ghat, 18.iv.1924, Maj. F. C. Fraser, Disparoneura souteri ?
(label F.C.F.].. Now Elattoneura souteri (Fraser).

souteri (ssp. of Microgomphus torquatus), 1924: 474. Holotype g. S. India,
Coorg, Sampaji Ghat, 7.x.1923, Maj. F. C. Fraser. Microgomphus souteri Fras.
dg. Sampaji Ghat, Coorg, 7.x.23, F. C. F. [label F.C.F.]. Allotype 9. M.
souteri Fras., Allotype 9. Hallery, Coorg, 7.vi.23, Fraser [label F.C.F.]. Now
Microgomphus soutert Fraser.

spatula (Lestes), 1946b : 46-48, figs. 1a, b. LECTOTYPE g. L. spatula Fras.
Concordia, R. A., xii.35, coll. Hayward [label F.C.F.]. Allotype 9. Same data.

[spencei (Pseudagrion), 1922 : 47-48.] The holotype 2 from Shillong (which was
in the Pusa collection) has not been traced, and the species was not included in the
list of Pusa types presented to the BM (NH). There is in the Fraser collection a
male marked Type, which is labelled Shillong, 10.v.22. This is not mentioned
in the original description. The label has the appearance of being a fairly recent
one and it is possible that it is the allotype 3, with an incorrect locality label.

[stevensi (Davidius), 1923 : 61; 1923b : 330.] Type not traced. This is another
case in which Fraser has used a new specific name in a key, without any indication
that it is a new species, and apart from the above second reference, has not used it
again. See under cyaneofrons, (p. 187). Itis an available name, and I include it in
this paper to draw attention to it, in the hope that a specialist on Indian Gomphines

may perhaps be able to recognize it and possibly place it in the synonymy of an
earlier species.

stevensi (Idionyx), 1924 : 462-463, pl. 26, fig. r. LECTOTYPE g. Type.
I. stevensi. Gopaldhara, Bengal, 8.vi.21. [label F.C.F.]. Allotype 9. Assam,
Darjeeling Distr., Rungbung Valley, Gopaldhara, H. Stevens [label D.E.K.].
Idionyx stevensi Fras. [label F.C.F.].

stevensi (Protosticta), 1922 : 7-8, pl. I, figs. 1, 2, 7. LECTOTYPE J. S. India,
Nilgiris, Coonoor Rd, 2,500 ft., Burliyar R., 29.vii.rg21, F. C. Fraser [label
D.E.K.]. P. stevensi gf, Burliyar, 2,500 ft., 19.7.21 [label F.C.F.]. Allotype 9.
Same data, P. stevensi @ Fras. [label F.C.F.]. The date in the original description
is given as 24 instead of 29.vii. The lectotype is the g presented to the BM (NH)
as type. Syn. of Protosticta gravelyi Laidlaw.

stevensoni (Gynacantha), 1927a : 36-37. Holotype g. G. stevensoni, Nukua-
lofa, Tonga, 17.ii.1925, G. Hopkins. Now Acanthagyna stevensoni (Fraser).

striatus (Onychogomphus), 1924c : toor. Holotype g. Onychogomphus stria-
tus §, Kalar, Nilgiris, 1,520 ft., v.1917, F. C. F., Type [label F.C.F.]. The label
gives the month of capture as June, not May.

subaequistyla (Copera), 1928a : 127-128. LECTOTYPE g. Uganda, Entebbe,
Shore of L. Victoria, vii—viii.1g27, G. Hale Carpenter. Copera subaequistyla

sp. nov. g. Type [label F.C.F.]. Allotype 9. Same data. Now Platycnemis
subaequistyla (Fraser).

FRASER ODONATA TYPES IN B.M. (N.H.) 215

submontana (Ceylonosticta), 1933 : 214-216, fig. 8. LECTOTYPE g. Type.
C. submontana, Ceylon, Kandy, 1.ix.24, Col. F. Wall. [label F.C.F.]. Allotype
Q. Same data. The ¢ is labelled November instead of September. Now
Drepanosticta submontana (Fraser).

submontana (ssp. of Vestalis apicalis), 1934 : 130. Holotype 3. India, Nilgiris,
Gudalur, 3,000 ft., 20.ix.1922, F. C. Fraser. Vestalis apicalis submontana
Fraser ¢ [labels D.E.K., transcribed from rubber stamped labels]. V.submontana
(type) [label F.C.F.]. Allotype 9. Same locality, date 9.ix.1922. V. sub-
montana allotype @ [label F.C.F.].

subplatystyla (Anisopleura), 1927 : 81-82. LECTOTYPE ¢g. Assam, Shillong,
4.vi.24, T. B. Fletcher. Anisopleura subplatystyla 3 sp. nov., 4.vi.24, Shillong,
Assam, coll. T. B. Fletcher, Type [label F.C.F.]. The type was stated to be in the
Pusa Collection, but it appears to have remained in Fraser’s collection until it
was presented by him to the BM (NH) in 1937.

subtropica (Ceylonosticta), 1933 : 209-211, fig. 5. LECTOTYPE g. Type.
C. subtropica. Pettigalla, Ceylon, 21.vi.26, G. M. Henry. Allotype 9. C. sub-
tropica, Belanagoda, Ceylon, 21.vi.26, G. M. Henry [both labels F.C.F.]. Slips in
re-labelling have occurred in both types, the date being given as “iv ”’ instead of
“‘vi’’, and the collector’s initials are G. M., not C. M. or C. Fraser gives no
indication as to the number of specimens ; so far only the lectotype and allotype
have been found.

superbus (Megalogomphus), 1931 : 460-463, figs. 3, 4iii. LECTOTYPE 4.
S. India, Coimbatore Distr., v.1931, F. C. Fraser. Megalogomphus superbus
Fraser 3, Bolovumpatti Forest, Coimbatore District, F. C. Fraser, May 31 [label
F.C.F.]. Allotype 9. Same data.

superplatypes (Copera), 1927 : 88-89, figs. 6b, 7b. Holotype g. Bengal, Duars,
Hasimara T. E., 6.viii.1923, H. V. O’Donel [label D.E.K.]. Copera super-
platypes Type [label F.C.F.]. This species was based upon a single male. The
Fraser collection also contained a specimen labelled by him as allotype 2 from the
same locality. I have not traced any reference to the description of this specimen.
The locality labels were large, typewritten ones and were not retained after
transcription. Now Platycnemis superplatypes (Fraser).

sylvatica (Macromia), 1954 : 56-58, figs. 4-5. Holotype g. Malaba Forest,
Kabras, Kenya Colony, June 1932, T. E. Jackson. Macromia sylvatica Fras.
Holotype 3 [label F.C.F.]. Allotype 9. Same data.

tamaracherriensis (ssp. of Merogomphus longistigma), 1931 : 460. Holotype
3. Merogomphus tamaracherriensis ¢, Tamaracherri, S. Malabar, 19.ix.28,
Type [label F.C.F.]. Allotype 9. Same locality, date 9.ix.28 [label F.C.F.].
Now Merogomphus tamaracherriensis Fraser.

t-coerulea (Indoneura), 1933b: 256. Holotype g. Type. I. ramburi var
T-coeruleum Fr., Nilgiris, 24.vii.1g21. Coll. F. C. Fraser [label F.C.F.]. Fraser

216 D. E. KIMMINS

infers that he has seen several examples but I have traced only this one specimen
in his collection. Now Caconeura t-coerulea (Fraser).

tenaculatus (Libyogomphus), 1926b : 358-359. Holotype g. Africa, Cameroons
ix.1918. Libyogomphus tenaculatus Fras. Now Tvragogomphus tenaculatus
(Fraser).

terminata (race of Polythore derivata), 1946 : 20-21, fig. 1d. Holotype 2.
San Antonio, Peru, xii.1932. P. derivata race terminata. Allotype 9. Same
locality, 1.1933 [labels F.C.F.]._ The date of the allotype differs from that given
in the original description (xi—xii. 1932).

theebawi (Caconeura), 1922 : 40-41. Holotype g. Mergui, 1921, Bott coll.
Caconeura theebawi Fras., Fraser det. 1921. Type @ [3 labels T.B.F.]. Allotype
Q. King Island, Mergui, Bott coll. Caconeura theebawi Fraser. Cotype Q,
Fraser det. 1921 [2 labels T.B.F.]. Ex Pusa Collection.

tillyardi (ssp. of Diphlebia lestoides), 1956 : 305, figs. 1, 3, 5. LECTOTYPE 32.
D. lestoides tillyardi Fraser g. Cameron Creek, 28.xii.1954, Mt Tambourine,
S. Queensland, R. Dobson coll. det. F. C. Fraser [label F.C.F.]. Allotype 9.
Same locality, date 4.1.1955.

tillyardi (Palaeothemis), 1923 : 33-34, figs. 2, 3. Holotype g. Mergui, 1921,
Bott coll. Palaeothemis tillyardi Fraser, J type, Fraser det. 1922 [labels T.B.F.].
Ex Pusa Collection.

tonkinicus (Onychogomphus), 1926c : 481. Holotype ¢g. Tonkin. Onycho-
gomphus tonkinicus Fraser ¢ Type.

travancorensis (Idionyx), 1931 : 455-456, fig. 1c. Lectotype ¢ (Fraser, 1934:
225). Type. I. travancorensis g, Munnar Ghat, Travancore, 30.v.31. F. C.
Fraser. [label F.C.F.]. Allotype 92. Western Ghat, Travancore, 7.vi.31, F. C.
Fraser [label F.C.F.].

trifoliata (Argia), 1946 : 449-450, figs. th, ii LECTOTYPE 4, allotype 2. S.
America, S. E. Colombia, Umbria, 30.xi.30. Argia trifoliatan. sp. ¢& 9. Type
and allotype, det. F. C. Fraser [label F.C.F.]._ The combined determination label
in Fraser’s writing is on the holotype.

trilobata (Protosticta), 1933 : 112-114, fig. 1. Holotype g. P. trilobata Fraser
§ Type. Laos, coll. Dr Kerr, 28.iv.1932 [label F.C.F.]. Muang Cha, Laos,
ca 1,100 m., coll. A. Kerr, 28.iv.1932 [typed label]. In Fraser’s label, the date
was given as 25.Vi.

trinervulata (Heliaeschna), 1955 : 16-17. Holotype g. Entebbe, Uganda, ’28.
Hale Carpenter. H. trinervulata Fraser, J Type, det. F. C. Fraser [labels F.C.F.].

trituberculata (Macromia), 1921a : 685-686. LECTOTYPE 4. Shillong, 20
Sept. 1920, over stream, Fletcher coll. Macromia trituberculata ¢ Fras., Type
(label T.B.F.]. Ex Pusa Collection. Syn. of Macromia moorei Selys.

[trotteri (Pseudocopera), 1922a : 237—238.| The types of this species have not yet
been traced in the Fraser collection. Probably syn. of Platycnemis annulata
(Selys).

FRASER ODONATA TYPES IN B.M: (N.H.) 27,

umbriaca (Argia), 1946 : 451, fig. 2i,j. LECTOTYPE 4, allotype 2. S. America,
S. E. Colombia, Umbria, 28.xi.1930. Argia umbriaca n. sp. Type & allotype 9,
Umbria, 28.xi.30, det. F. C. Fraser [label F.C.F.]. The allotype now lacks
segments 7—10 of the abdomen.

uncata (Protosticta), 1931a : 75-76. LECTOTYPE g. Upper Burma, N. Shan
States, Gok Teik, 1.vi.24, F. Wall [label D.E.K.]. P. uncatus, Gok Teik, 1.vi.24
{label F.C.F.]. Protosticta uncata Fras. g [label D.E.K.]. Allotype 9. Same
data. The specific name (published as wncatus) was later corrected by Fraser to
uncata. The specimens appear to be rather teneral.

unguiculata (Idionyx), 1926 : 204-205, pl. 8, fig. 3. Lectotype ¢ (Fraser, 1936 :
232). Type. I. unguiculata J, Maymyo, Upper Burma, 16.vi.25, Col. F. Wall,
{label F.C.F.]. Allotype 2. Burma, Maymyo, 31.v.1925, Col. F. Wall. Idionyx
unguiculata sp. nov. Allotype [label F.C.F.]._ The original locality label on the
allotype erroneously gave F. C. Fraser as the collector.

unifasciata (Periaeschna), 1935b : 25-26. Holotype g. P. unifasciata, Moongpu,
Darjeeling Distr., 28.v.27, F. C. Fraser [label F.C.F.]. Allotype 9. Same data.

vallisi (Phasmosticta), 1955a : 227-229, fig. 1. LECTOTYPE 3g. Type. Phas-
mosticta vallisi Fraser J. Rockhampton, N. Q., C. Vallis, iv.1954 [label F.C.F.].
Allotype 2. Same data. The labels on these two specimens give the date of,
capture as ili. 1955.

varralli (Mortonagrion), 1920 : 148. LECTOTYPE g. India, Bombay, Vihar
Lake, Mar. 1919. Mortonagrion varralli Fras. Type [label F.C.F.].

veronica (Indolestes) 1924a: 85. Lectotype 2 (Fraser, 1933b : 77). Kumaon,
7,500 ft., Muktesar, 1 Oct. 1922, Fletcher coll. Indolestes veronica Fraser 2 Type,
Fraser det. 1922 [label T.B.F.]. Indolestes veronica 2 sp. nov. Type [label
F.C.F.]._ Ex Pusa Collection. A nymphal exuvium is associated with the type.
Syn. of Lestes (Indolestes) cyaneus Selys.

versicolor (Telebasis), 1946 : 42-43, fig. 12. Holotype g. Telebasis versicolor
Fras., Umbria, S. E. Colombia. [label F.C.F., on mounting card].

v-flavum (Burmagomphus), 1926a : 414-415. LECTOTYPE 92. Burmagom-
phus V-flavum 2, Maymyo, Burma, F. Wall coll. 15.vi.24. Type [label F.C.F.].

victoria (Agriocnemis), 1928a : 123-125, fig. 1. LECTOTYPE ¢. Uganda,
N. W. shores of Lake Victoria, vii-ix.1927, G. D. Hale Carpenter. Agriocnemis
victoria sp. nov. [label F.C.F.].

vinsoni (Ischnura), 1949 : 140, figs. 1a, b,c. LECTOTYPE 3. Mauritius, Moka,
-12.1.1947, J. Vinson [label D.E.K.]. Ischnura vinsoni Fraser, Type [label F.C.F.].
Allotype 9. Same locality, date 12.xii.1946. Fraser presented to the BM (NH)
in 1949 a J, 9 of this series as type and allotype, but this was an error on his part,
as the specimens were collected later than the type series. The true types were
discovered in his collection after his death.

218 D. E. KIMMINS

violacea (Coenagrion), 1924a : 84-85. Holotype g. Ranchi, 1.ix.22, Fletcher
coll. [label T.B.F.]. Coenagrion violacea sp. nov. (Type) [label F.C.F.]. Ex Pusa
Collection. Wings rather damaged, right pair with only bases remaining.

viridata (Rhinagrion), 1938 : 198. Holotype g. Rhinagrion viridata g, Tenas-
serim—Bachrup boundary, 28.v.32, A. Kerr coll. {label F.C.F.].

viridis (Drepanosticta), 1922 : 39-40. Holotype g. Mergui, Bott coll. Dre-
panosticta viridis ¢ Type [labels T.B.F.]. Ex Pusa Collection.

vitrinella (Rhinocypha), 1935 : 332-333. Holotype g. R. vitrinella g, Cachar,
Assam, Antram coll. [label F.C.F.].

walli (Ceylonosticta), 1931a : 335-336. The holotype could not be traced in
Fraser’s collection. He therefore selected the following 2 specimen, which he
suggested should be considered as NEOTYPE. It bears the following labels:
C. walli 9, Kadugannawa Ghat, Kandy, Ceylon, 14.v.32. F. C. Fraser coll.
[label F.C.F.]. Type 2 untraced. This specimen selected as neotype in consulta-
tion with F. C. Fraser, 1958 [label D.E.K.]. Allotype g. Same locality and date.

walli (Dysphaea), 1927 : 82-83. LECTOTYPE g. D. walli, Gok Teik, Burma,
30.v.24. Col. F. Wall [label F.C.F.]. In the original description the locality is
given as Maymyo; Gok Teik is probably an error in relabelling. The type was
stated to be in the British Museum, but remained in Fraser’s collection until his
death. The other three males listed have not been traced.

walli (Heliogomphus), 1925a : 849-851. Holotype ¢. Heliogomphus walli 9,
Nalande, Ceylon, 10.ix.24, F. Wall coll. [label F.C.F.]. The allotype 3 is in the
Colombo Museum.

walli (Onychogomphus), 1924b : 109. Lectotype 3 (Fraser, 1934 : 288). Nepo-
gomphus walli g. Maymyo, Upper Burma, vi.24. Type ¢ [label F.C.F.]
Allotype 2. Same data. Now Nepogomphus walli (Fraser).

wheeleri (race of Drepanosticta fontinalis), 1942 : 96. Holotype g. Malaya,
Wellesley, Bukit Mertajam, 25.v.1936. D. fontinalis Lieft., race wheeleri [label
F.C.F.]. Syn. of Drepanosticta fontinalis Lieftinck.

wheeleri (Gomphidictinus), 1942 : 100-101, figs. 2-3. Holotype 3. Malaya,
Cameron Highlands Rd, nr Ramaswami Waterfall, 14th mile, 11. viii. 1932, R. L.
Wheeler. Gomphidictinus wheeleri g¢ Type [label F.C.F.]. Syn. of Gomphidia
perakensis Laidlaw.

wilkinsi (Cyclogomphus), 1926d : 161-162. Holotype g. Cyclogomphus wilkinsi
g. Hunse R., Coorg, Mysore, 12.x.24, coll. F. C. Fraser [label F.C.F.]. Allotype
Q. India, Hunse R., Mysore, 12.x.24, F. C. Fraser. Cyclogomphus wilkinsi 9
Fras. [label D.E.K.].

williamsoni (Gomphidia), 1923a : 670-671. Holotype g. India, Bengal, Hasi-
mara, Duars, v—vi.1923, H. V. O’Donel [label D.E.K.]._ Gomphidia williamsoni 3
Type [label F.C.F.]. Allotype 9. Gomphidia williamsoni 9, Hasimara T. E.,
Duars, Bengal, H. V. O’Donel [label F.C.F.]._ The BM (NH) locality label on the
type was inadequate and incorrect, giving no date and Fraser as collector. It

FRASER ODONATA TYPES IN B.M. (N.H.) 219

has therefore been replaced. In Fraser’s collection another j was discovered
incorrectly labelled as type, the type having already been presented to the BM
(NH) in 1923.

williamsoni (Pseudagrion), 1922 : 46-47. Holotype g. Mergui, 1921, Bott coll.
Pseudagrion williamsoni g, Fraser det. 1921. Type [labels T.B.F.]. Ex Pusa
Collection.

{wynaadensis (race of Pseudophaea fraseri), 1922d : 8-g.|_ I have been unable to
trace any specimens referred by Fraser to this race in his collection, and as he
makes no subsequent reference to it, one can only presume that he considered it as
a synonym of frasert. Now Indophaea frasert (Laidlaw).

wynaadicus (Macrogomphus), 1924 : 471-472. Holotype 2. M. wynaadicus Q,
Masnagudi, Nilgiri Hills, 18.viii.22, F. C. Fraser [label F.C.F.].

xanthoptera (Orogomphus), 1919 : 874-875. Holotype 9. Cumbum, Madura
Distr., May—June 1917, S. H. Prater. Orogomphus xanthoptera, F. C. Fraser,
Type. Type ex Bombay Museum. The original description gives the specific
name as xantheptera. Now Chlorogomphus xanthoptera (Fraser).

zeylanica (Macromia), 1927: 69-70. Holotype g. M. zeylanica. Kandy,
Ceylon, 13.1x.24, Col. F. Wall.

REFERENCES TO PAPERS BY F. CG: FRASER

1918. Two new Indian Dragonflies. J. Bombay nat. Hist. Soc. 25 : 383-385.

1919. Notes on some new and other Indian Dragonflies. J. Bombay nat. Hist. Soc. 26: 874-
878.

Ig1ga. Descriptions of four new Indian Odonata. Rec. Indian Mus. 16: 451-455.

I9g19b. Note on a collection of Odonata from South Annam. J. nat. Hist. Soc. Siam, 3: 455-
461, 1 pl., 1 text-fig.

1920. Notes on Indian Dragonflies. Rec. Indian Mus. 19 : 31-33.

1920a. Some new Indian Dragonflies. J. Bombay nat. Hist. Soc. 27 : 147-150.

1921. A list of dragonflies from Mahableshwar. J. Bombay nat. Hist. Soc. 27 : 540-543.

1921a. Indian Dragonflies. Pt X. J. Bombay nat. Hist. Soc. 27 : 673-691, fig.

I921b. Report on a collection of Dragonflies from the Laos Country. J. nat. Hist. Soc. Siam
4 (3) : 161-165, 1 pl.

1922. New and rare Indian Odonata in the Pusa Collection. Mem. Dep. Agric. India ent. Ser.
7 (7) : 39-77, pls. 6-8, 1 fig.

1922a. A collection of dragonflies from Bangkok. J. nat. Hist. Soc. Siam 4: 231-238, 1 pl.

1922b. Notes on new and rare Indian Dragonflies. J. Bombay nat. Hist. Soc. 28 : 698-702,
2 figs.

1922c. in Appendix to Laidlaw, A list of dragonflies recorded from the Indian Empire, with
special reference to the collection of the Indian Museum. Rec. Indian Mus. 24: 415-426,
pl. 11, 1 text-fig.

1922d. New and rare Odonata from the Nilgiri Hills. Rec. Indian Mus. 24: 1-9, pl. 1.

1923. Dragonfly collecting in India. Pts II, IV. J. Bombay nat. Hist. Soc. 29 : 48-69, 2 figs.,
741-756, 3 pls.

1923a. Indian Dragonflies. Pt XVII. J. Bombay nat. Hist. Soc. 29 : 659-680, 2 pls., 4 figs.

1923b. Indian Dragonflies. Pt XVI. J. Bombay nat. Hist. Soc. 29 : 324-333, figs. 1-3.

1923c. A second note on Odonata in the Pusa Collection. Mem. Dep. Agric., India ent. Ser.
8 : 290-34, 3 figs.

220 D. E. KIMMINS

1924. A Survey of the Odonate (Dragonfly) Fauna of Western India, with special remarks on
the genera Macromia and Idionyx and descriptions of thirty new species, with Appendices
I, Il. Rec. Indian Mus. 26: 423-522, pls. 25-27, 8 figs.

1924a. Notes on Indian Odonata in the Pusa Collection. Mem. Dep. Agric. India ent. Ser.
8 (8) : 69-87, pls. 9-10, 6 figs.

1924b. Indian Dragonflies. Pt XIX. J. Bombay nat. Hist. Soc. 30: 106-117, pl. 1, 2 figs.

1924c. Indian Dragonflies. Pt XVIII. J. Bombay nat. Hist. Soc. 29 : 982-1006, 2 pls., 2 figs.

1924d. Zoological results of the Percy Sladen Trust Expedition to Yunnan under the leadership
of Professor J. W. Gregory, F.R.S. (1922) Odonata. J. Asiat. Soc. Beng. 19 (1923):
447-404, 8 figs.

1924e. New additions to the Odonata (Dragonfly) fauna of India. J. Bombay nat. hist. Soc.
30 : 48-53. 1 fig.

1925. A preliminary report on the Dragonflies (Order Odonata) of Samoa. Tvans. ent. Soc.
Lond. 1924 : 429-438, 3 figs.

1925a. Indian Dragonflies. Pt XXII. J. Bombay nat. Hist. Soc. 30: 846-857, 1 pl., 3 figs.

1926. Dragonflies, a revision of the genus Zygonyx Selys. J. Bombay nat. Hist. Soc. 31 : 762-
769, 1 fig.

1926a. Indian Dragonflies. Pt XXIV. J. Bombay nat. Hist. Soc. 31 : 408-426, 2 pls., 5 figs.

1926b. Two new Dragonflies (Order Odonata). Tvans. ent. Soc. Lond. 74: 355-359, 7 figs.

1926c. Notes on a collection of dragonflies (Order Odonata) from the Dutch East Indies and
descriptions of four new species from the neighbouring continent. Tveubia, 8 : 467-494,
7 figs.

1926d. Indian Dragonflies. Pt XXIII. J. Bomba’ mat. Hist. Soc. 31: 158-171, 1 pl., 3 figs.

1926e. <A Revision of the genus Idionyx Selys. Rec. Indian Mus. 28 : 195-207, pls. 8-10.

1926f. Two new species of Dragonfly from the Darjeeling District. J. Darjeeling nat. Hist. Soc.
1926 : 31-33.

1926g. A remarkable new Dragonfly from Samoa. Tvans. ent. Soc. Lond. 73 : 505-507, pl. 49.

1926h. Indian Dragonflies. Pt XXV. J. Bombay nat. Hist. Soc. 31 : 733-741, 1 pl., 5 figs.

1927. Descriptions of twenty new Indian Dragonflies. Rec. Indian Mus. 29 : 63-90, 7 figs.

1927a. Odonata. Insects of Samoa and other Samoan terrestrial Arthropoda. Pt VII, fasc.
1 : 19-44, 5 figs. London.

1928. Indian Dragonflies. Pts XXIX-XXX. J. Bombay nat. Hist. Soc. 32 : 450-459, 2 pls.,
2 figs. ; 32 : 683-691, 3 pls. ; 33: 47-59.

1928a. Odonata of the African Continent. Tvans. ent. Soc. Lond. 76 : 123-138, 6 figs.

1929. Indian Dragonflies. Pt XXXIV. J. Bombay nat. Hist. Soc. 33 : 834-950, 2 pls., 4 figs.

1929a. A revision of the Fissilabioidea (Cordulegasteridae, Petaliidae and Petaluridae) Pt I.
Cordulegasteridae. Mem. Indian Mus. 9 (3) : 69-167, pls. 9-12, 35 text-figs.

1929b. Indian Dragonflies. Pt XXXIII. J. Bombay nat. Hist. Soc. 33 : 576-597, 6 pls., 2 figs.

1930. Indian Dragonflies, Pt XXXV. J. Bombay nat. Hist. Soc. 34: 87-107, 1 pl., 4 figs.

1930a. A new Indian Dragonfly (Order Odonata). J. Bombay nat. Hist. Soc. 34: 752-753,
2 figs.

1931. Additions to the Survey of the Odonata (Dragonfly) Fauna of Western India, with
descriptions of nine new species. ec. Indian Mus. 33 : 443-474, 6 figs.

1931a. Indian Dragonflies. Pts XXXVII-XXXVIII. J. Bombay nat. Hist. Soc. 35 : 66-76,
1 pl., 2 figs. : 325-341, 2 pls., 1 fig.

1932. Notes ona small collection of Siamese Dragonflies, with descriptions of three new species.
J. Siam Soc., nat. Hist. Suppl. 8 (4) : 283-292, 2 figs.

1933. The Platystictas of Ceylon (Order Odonata). Ceylon J. Sci. (B) 17: 201-224, 13 figs.

1933a. Additions to the Dragonfly (Odonate) Fauna of India, with descriptions of new species.
J. Bombay nat. Hist. Soc. 36: 460-468, 4 figs. (April 1933).

1933b. Odonata. Vol. I. Fauna of British India, including Ceylon and Burma. pp. xiii +
423, 180 figs., map. London.

1933c. Dragonflies from the Laos Country. J. Siam Soc., nat. Hist. Suppl. 9: 109-141, 8 figs.

FRASER IODONALTA TYPES EN? BoM: i(NoH.) 221

1933d. A Revision of the Genus Orolesies (Order Odonata). Rec. Indian Mus. 35: 175-182,
pl. 4, text-figs. 1-5.

1933e. A new species of Disparoneura from Ceylon (Order Odonata). Ceylon J. Sci. (B)
17 : 225-226.

1933f. A Revision of the Fissilabioidea (Cordulegasteridae, Petaliidae and Petaluridae).
Part II. Petaliidae and Petaluridae and Appendix to Part I. Mem. Indian Mus. 9:
205-260, 31 text-figs. (Feb. 1933).

1933g. The Gomphines of Ceylon (Order Odonata). Ceylon J. Sci. (B) 18 : 19-36, 6 figs.

1934. Odonata. Vol. II. Fauna of British India, including Ceylon and Burma, pp. xxiii +
39& 4 pls., 120 figs. London.

1934a. New and rare Odonata from the Burma-Siamese Frontier. Stylops, 3 : 134-137, 3 figs.

1934b. Indian Dragonflies. Pt XLII. J. Bombay nat. Hist. Soc. 37 : 553-572, 2 pls., 1 fig.

1935. New oriental Dragonflies (Order Odonata). Rec. Indian Mus. 37 : 321-333, 5 figs.

1935a. A new dragonfly from Lower Burma. J. Bombay nat. Hist. Soc. 37 : 890-891.

1935b. Three new species of Dragonfly from N. India. J. Darjeeling nat. Hist. Soc. 10 : 23-27.

1936. Odonata. Vol. III. Fauna of British India, including Ceylon and Burma, pp. xi + 461,
2 pls., 1 map, 125 figs. London.

1936a. A note on the identification of some obscure genera and species of the family Gomphidae
(Odonata). Proc. R. ent. Soc. Lond. (B) 5: 47-143, 2 figs.

1936b. Odonata collected in Japan, with descriptions of three new species. Tvans. R. ent. Soc.
Lond. 85 : 141-156, 7 figs.

1936c. New Oriental Dragonflies. J. Bombay nat. Hist. Soc. 38 : 700-701, 1 fig.

1937. A note on the confusion of Onychogomphus bistrigatus (Hagen) with a new species of
Onychogomphus (Odonata). Proc. R. ent. Soc. Lond. (B) 6: 161-164, figs. 1-2.

1938. Two new species of oriental Odonata. Proc. R. ent. Soc. Lond. (B) 7: 197-198.

1939. Heliaeschna ikeiets a new species of Dragonfly from Uganda (Order Odonata). Proc.
R. ent. Soc. Lond. (B) 8 : 88-99, 1 fig.

1939a. Additions to the family Corduliidae, including descriptions of two new species and a new
genus. Proc. R. ent. Soc. Lond. (B) 8 : 91-94, fig. 1.

1939b. Libellago adami, a new species of dragonfly from Ceylon (Odonata). Proc. R. ent. Soc.
Lond. (B) 8 : 23-24, 1 fig.

1940. A new species of Chlorocypha (Order Odonata). Ann. ent. Soc. Am. 33 : 552-554, 4 figs.

1940a. A newspecies of Chlorogomphus (Order Odonata). Proc. R. ent. Soc. Lond. (B) 9 : 55-56,
I fig.

1941. New African species of Ceriagrion (Order Odonata). Proc. R. ent. Soc. Lond. (B)
10 : 61-66, 2 figs.

1941a. Results of the Mission P. Lepesme, R. Paulian and A. Villiers in the Cameroons. Order
Odonata. Proc. R. ent. Soc. Lond. (B) 10: 35-42, 2 figs.

1941b. Notes on the genus Tetrathemis Brauer, with descriptions of three new African species

(Odonata). Proc. R. ent. Soc. Lond. (B) 10: 137-142, 9 figs.

1942. Dr. Raymond Wheeler’s collection of Odonata from the Federated Malay States, with
descriptions of a new genus and two new species. Proc. R. ent. Soc. Lond. (B) 11 : 95-104,
4 figs.

1942a. Notes on the genus Heliogomphus Laidlaw, with descriptions of two new species,
Trans. R. ent. Soc. Lond. 92 : 333-334, 3 figs.

1943. Lestes albofasciata : a new species of Odonata from Buru Island. Proc. R. ent. Soc. Lond.
(B) 12 : 113-115, 2 figs.

1943a. Notes on the genus Agriogomphus Selys and Ischnogomphus Williamson, with description
of the male of Agriogomphus sylvicola Selys (Odonata). Proc. R. ent. Soc. Lond. (B) 12:
161-166, 2 figs.

1944. A note on the genus Notiothemis Ris, with the description of a new species. (Order
Odonata). Proc. R. ent. Soc. Lond. (B) 13 : 40-43, 2 figs.

ENTOooM. 18, 6. 17

222 D. E. KIMMINS

1944a. New Ethiopian and Oriental Dragonflies (Order Odonata). Proc. R. ent. Soc. Lond. (B)
13 : 85-88, fig. 1.

1946. Notes on Amazonian Odonata in the Leeds Museum. Tvans. R. ent. Soc. Lond. 96 : 11-
46, 1 pl., 13 figs.; 97: 443-472, 7 figs.

1946a. One rare and two new species of Odonata from Tropical Africa and Asia. Proc. R. ent.
Soc. Lond. (B) 15: 41-45, 2 figs.

1946b. Lestes spatula, a new species of Dragonfly from the Argentine Republic. Proc. R. ent.
Soc. Lond. (B) 15 : 46-48, 1 fig.

1946c. Hylaeothemis indica, a new species of Oriental Libelluline (Odonata). Proc. R. ent.
Soc. Lond. (B) 15 : 97-100, 3 figs.

1947. The Odonata of the Ivory Coast, based upon the Mission of Dr. R. Paulian and P.
Lepesme. Tvans. R. ent. Soc. Lond. 98 : 19-39, 8 figs.

1947a. The synonymy of Enallagma elongatum (Martin) and E. pseudelongatum Longfield, and
their differentiation from a new African species. Proc. R. ent. Soc. Lond. (B) 16 : 143-148,
figs. I-2.

1948. Three new species of Ethiopian Odonata. Proc. R. ent. Soc. Lond. (B) 17 : 5-10, 2 figs.

1949. The Zygoptera of Mauritius (Order Odonata). Tvans. R. ent. Soc. Lond. 100 : 135-146,
3 figs.

1950. Atoconeura extraordinata, a new species of Libelluline from Uganda. Proc. R. ent. Soc.
Lond. (B) 19 : 56-57, 1 fig.

1950a. The Odonata of Mauritius, suborder Anisoptera, and the description of a new species of
Zygoptera. Proc. R. ent. Soc. Lond. (B) 19: 70-77, 1 fig.

1951. Notes on Odonata in the Museum Nationale d’Histoire naturelle. With descriptions of a
new species and a new Zygopterous nymph. Revue fr. Ent. 18 (2) : 93-101, 4 figs.

1952. Notes on African Gomphidae, with descriptions of four new species and a new genus.
Occ. Pap. Coryndon meml Mus. 3 : 3-12, 4 figs.

1952a. New genera and species of Libellulines from the Belgian Congo. Revue Zool. Bot. afr.
48 : 242-256, 4 figs.

1953. Further notes on Samoan Odonata belonging to the Ischnurine complex of species, with
descriptions of two new species and some unknown females. Proc. R. ent. Soc. Lond. (B)
22 : 119-126, 18 figs.

1953a. Notes on the family Gomphidae, with descriptions of a new species and the female of
another. Proc. R. ent. Soc. Lond. (B) 22 : 189-194, figs. a-o.

1954. New species of Macromia from Tropical Africa. Revue Zool. Bot. afr. 49 : 41-76, 4 figs.

1954a. Further notes and descriptions of new species of Libellulidae from Tropical Africa.
Revue Zool. Bot. afr. 50 : 252-268, 2 figs.

1954b. Twonew species of Odonata from Australia. Ann. Mag. nat. Hist. (12) 7: 145-149, 1 fig.

1955. Odonata. Explor. Parc Nat. Upemba, I. Miss. de Witte (1946-1949) fasc. 38 (1) : 1-34,
6 figs. Brussels.

1955a. A new Isostictine Dragonfly from Australia, with some remarks on the subfamily
Isostictinae nov. (Odon. Protoneuridae). Entomologist’s mon. Mag. 91 : 227-230, 1 fig.

1956. A note on the correct habitat of Rhyothemis chalcoptilon chalcoptilon (Brauer) (Odonata,
Libellulidae). Entomologist’s mon. Mag. 92 : 326-328, 10 figs.

1956a. A newspecies of Diphlebia (Odonata, Amphipterygidae) from Australia. Entomologist’s
mon. Mag. 92 : 304-305, figs. I, 5.

1957. A Revision of the Genus Phyllogomphus Selys, with descriptions of five new species.
Revue Zool. Bot. afr. 56 : 9-32, 8 figs.

1960. Phyllogomphus moundi, a new species of Odonata from Ibadan, Nigeria, Tropical Africa.
Revue Zool. Bot. afr. 62 : 141-144, figs. a-i.

1961. in Pinhey, A Survey of the Dragonflies (Order Odonata) of Eastern Africa. vii + 214 pp.,
11 pls., 2 text-figs. London.

FRASER ODONATA TYPES IN B.M. (N.H.) 223
APPENDIX
Location of types not in B.M. (N.H.)

acuminatus, Parogomphus, 1944 Tervuren
adamst, Archaeophya, 1959 Dobson collection
aequatorialis, Oxythemis, 1954 Tervuren
aethiopia, Elattoneura, 1941 Paris
africana, Neodythemis, 1954. Tervuren
albicauda, Nesolestes, 1952 Paris
albicolor, Nesolestes, 1955 Inst. Rech. Sci. Madagascar
anact, Notogomphus, 1955 Tervuren
andamanensis, Caconeura, 1924 Originally in Indian Mus.*
andamanensis, Lestes, 1924 Originally in Indian Mus.
andamanensis, Micromerus, 1924 Originally in Indian Mus.
andamanicum, Pseudagrion, 1924 Originally in Indian Mus.
angelicum, Pseudagrion, 1947 Paris
annandalei, Anotogaster, 1924 Originally in Indian Mus.
annandalei, Drepanosticta, 1924 Originally in Indian Mus.
annulatum, Ceriagrion, 1955 Tervuren
annulata, Protosticta, 1926 Leiden
apicalis, Anectothemis, 1954 Tervuren
apicale (Schmidt Mss.), Pseadagrion, Unknown

1949
arnoultt, Neodythemts, 1955 Inst. Rech. Sci. Madagascar
auripennis, Lestes, 1955 Inst. Rech. Sci. Madagascar
auritum, Ceriagrion, 1951 Inst. Rech. Sci. Madagascar
bali, Gomphidia, 1949 Tervuren
basilewskyi, Aethiothemis, 1954 Tervuren
beesont, Amphiaeschna, 1922 Forest Res. Inst., Dehra Dun, India
berenice, Lokia, 1952 Tervuren
berlandi, Xiphiagrion, 1951 Paris
binocellata, Agriocnemis, 1922 Unknown
bispina, Drepanosticta, 1932 ? Brussels
bispina, Malgassophlebia, 1958 Tervuren
bredot, Trithemts, 1952 Tervuren
calcaris, Argiolestes, 1958 Not stated, ? Dobson coll.
carinata, Idionyx, 1926 Koenigsberg Mus.
carpentiert, Ischnura, 1946 Brussels
cheliferum, Pseudagrion, 1949 Paris
chloroceps, Pseudagrion, 1955 Paris
chromostigma, Ischnura, 1927 Bishop Museum, Honolulu
coerulea, Indolestes, 1924 ? Indian Museum

* The entomological collections were subsequently transferred to Zoological Survey of India, but
enquiry revealed that most of these types could not be traced, possibly lost by flood damage during
wartime storage.

D. E. KIMMINS

comorensis, Thermorthemis, 1959

comorensis, Trithemis, 1959

congoensis, Orthetrum, 1949

congoliath, Anax, 1953

congolica, Macromia, 1955

conspicuum, Pseudagrion, 1947

crenulatipennis, Tatocnemis, 1953

dahli, Chlorocypha, 1956

digitatum (Schmidt Mss), Pseud-
agrion, 1949

diplacoides, Sleuthemis, 1957

dispar (Schmidt Mss), Pseudagrion,
1949

dissimulans, Lestes, 1955

dobsont, Gynacantha, 1951

dorothea, Coenagrion, 1924

drocera, Nesolestes, 1951

dubitalis, Gomphomacromia, 1939

dyeri, Argiocnemis, 1920

elongata, Gynacantha, 1957

emarginatipenms, Tatocnemis, 1960

exclamationis, Himalagrion, 1919

filosa (Schmidt Mss.), Ischnura, 1949

flavipes, Gynacantha, 1956

forficuloides, Lestes, 1950

fulvia, Gynacantha, 1926

geminata, Caconeura, 1926

ghesquieret, Chlorocypha, 1959

goliathus, Anax, 1922

gracillima, Teinobasis, 1926

gravelyt, Argiocnemis, 1920

gregory, Anotogaster, 1923

gregoryt, Mnais, 1924

guttifera, Platycnemis, 1950

hamon, Aciagrion, 1955

hamont, Coenagriou, 1955

hamoni, Pseudagrion, 1955

haywardi, Oligoclada, 1947

henrardi, Elattoneura, 1954

heterosticta, Aciagrion, 1955

hova, Gynacantha, 1956

igniceps, Pseudagrion, 1953

igniceps, Pseudagrion, 1953 (pre-
occupied name, see chloroceps)

Inst. Rech, Sci. Madagascar
Inst. Rech. Sci. Madagascar
Inst. Parcs nat. Congo, Brussels
Tervuren

Tervuren

Paris

Unknown

Copenhagen, Univ. Mus. Zool.
Unknown

Paris
Unknown

Tervuren

Dobson collection

? Indian Mus.

Paris

R. Scottish Mus., Edinburgh
Originally in Indian Museum
Tervuren

Unknown

Originally in Indian Mus.
Unknown

Tervuren

Inst. Rech. Sci. Madagascar
Leiden

Leiden

Tervuren

Forest Res. Inst., Dehra Dun.
Leiden

Originally in Indian Mus.
Originally in Indian Mus.
Originally in Indian Mus.
Paris

Paris

Paris

Paris

Inst. Miguel Lillo, Tucuman
Tervuren

Tervuren

Paris

Paris

FRASER ODONATA TYPES IN B.M. (N.H.) 225

immaculifrons, Gynacantha, 1956

imperfecta, Oligocaemia, 1951

inaequistigma, Mullotgarion, 1953

iufumosa, Umma, 1951

insula, Enallagma, 1920

10, Rhinocypha, 1926

yavica, Gynacantha, 1926

karnyi, Argiolestes, 1926

karnyt, Procordulia, 1926

kirckhoffae (Schmidt Mss), Proto-

estes, 1949

lankanensis, Macrogomphus, 1933

lejeunet-carpentiert, Lithemis, 1951

leonorae (Schmidt Mss), Protolestes,
1949

leopoldi, Neurobasis, 1932

leopoldt, Procordulia, 1932

leucostigma, Disparoneura, 1933

liberiensis, Brachythemis, 1949

linaenea, Cordulia, 1956

longistyla, Congothemts, 1959

longiventris, Paragomphus, 1955

lucast, Leucorrhinia, 1955

lyvatus, Heliogomphus, 1933

maclachlani, Mnats, 1924

madagazureum, Certagrion, 1949

malgassicam, Gynacantha, 1962

malgassica, Nesocordulia, 1956

malgassicum (Schmidt Mss), Pseud-
agrion, 1949

merina (Schmidt Mss), Pseudagrion,

1949
metallica, Parathemis, 1926

muilloti, Orthetrum, 1949

millott, Protolestes, 1949
missionensis, Argia, 1948
molindica, Chlorocypha, 1948
mortont, Chlorogomphus, 1936
multilineata, Diastatomma, 1949
neophytus, Antipodogomphus, 1958
neurothemioides, Hemistigma, 1954
obscura, Neurothemis, 1926

Tervuren

Liége Univ. Mus.

Paris

Paris

Originally in Indian Mus.
Leiden, (¢ holotype)
Leiden, (2 holotype)
Leiden, (3 lectotype)
Leiden (¢ holotype)
Unknown

Colombo Mus.
Liége Univ. Mus
Unknown

? Brussels

? Brussels

Probably Colombo Mus.
IFAN, Dakar

Linnean Soc., London
Tervuren

Tervuren

Hope Dept. of Entomology, Oxford.
Colombo Mus.

Originally in Indian Mus.
Paris

Inst. Rech. Sci., Madagascar
Inst. Rech. Sci. Madagascar
Unknown

Unknown

Originally in Mus. Zool. Bogoriense,
now presumed lost (Lieftinck, i.1.,
1965)

Inst. Rech. Sci. Madagascar

Inst. Rech. Sci. Madagascar

Inst. Miguel Lillo, Tucuman

Tervuren

R. Scottish Mus., Edinburgh

Tervuren

W. Australia Mus.

Tervuren

Leiden (¢ lectotype)

226

D. E. KIMMINS

occidentalis, Macromia, 1954
pallipes, Aeshna, 1947
paludinis, Aethiothemis, 1954
pauliam, Neodythemis, 1952
pauliani, Nesolestes, 1951
pearsom, Gomphidia, 1933
pinheyt, Platycypha, 1950

platystictoides, Paulianagrion, 1941

pretentiosa, Zygonyx, 1957
prodigiosa, Zygonyx, 1958
protostictoides, Platycnemis, 1953
pyriformis, Chlorocypha, 1947
vadama, Gynacantha, 1956
rvanavalonae, Zygonyx, 1949
vasoherinae, Phaon, 1949
renaudt, Pseudagrion, 1953
renet, Crenigomphus, 1936
reticulata, Gynacantha, 1926
reuniense, Coenargiocnemts, 1957
robusta, Drepanosticta, 1926
vodericki, Micromidia, 1959
rubellocerinum, Ceriagrion, 1947
sanguinostigma, Ceriagrion, 1955
schoutedeni, Macromia, 1954
schoutedent, Macrogomphus, 1949
schoutedent, Phyllogomphus, 1957
selysi, Heliogomphus, 1925

selysi, Idionyx, 1926

septemflavum, Onychogomphus, 1955

seydeli, Chlorocypha, 1955
seydeli, Macromia, 1954
sictlicus, Cordulegaster, 1929
steberst, Drepanosticta, 1926
spirillus, Leptogomphus, 1922
splendens, Rhyothemis, 1955
straelent, Chlorocypha, 1949
stygia, Agriocnemis, 1954
subtropicalis, Macromia, 1954
superbum, Pseudagrion, 1956
thenartum, Pseudagrion, 1956
tillyardi, Synlestes, 1948
tinctipennis, Pseudagrion, 1957
trithemoides, Porpacithemis, 1958
tropicana, Trithemis, 1952

Paris

Inst. Miguel Lillo, Tucuman
Tervuren

Paris

Paris

Colombo

Coryndon Mus., Nairobi
Paris

Tervuren

Tervuren

Paris

? Paris

Location not stated
Paris

Paris

Paris

Paris

Leiden (¢ holotype)

Inst. Rech. Sci. Madagascar
Leiden (¢ holotype)
Dobson collection

Paris

Tervuren

Tervuren

Tervuren

Tervuren

Brussels

Brussels

Inst. Parcs nat. Congo, Brussels
Tervuren

Tervuren

Brussels

Leiden (¢ lectotype)
Originally in Indian Mus.
Tervuren

Brussels

Tervuren

Tervuren

Tervuren

Tervuren

Brussels

Inst. Rech. Sci. Madagascar
Tervuren

Tervuren

—a

RAS E ROOD ON AACR Y PES TiN BM es (ONCE) 227

tuberculicollis (Schmidt Mss), Neso-
lestes, 1949

umbrosa, Zygonyx, 1957

uncatus, Cyanogomphus, 1947

ungulatum, Pseudagrion, 1951

untifasciata, Macroma, 1954

villtersi, Ceriagrion, 1951

villverst, Oxythemis, 1951

villierst, Prodasineura, 1948

viridifrons, Gynacanthaeschna, 1922

viridivittata, Aeshna, 1947

vridaght, Elattoneura, 1954

vrydaght, Orthetrum, 1954

wiliamsom, Burmagomphus, 1926

willinkt, Macrothemis, 1947

willinki, Telebasis, 1948

witter, Aeshna, 1955

witter, Chlorocnemis, 1953

wittet, Chlorocypha, 1958

witter, Pseudagrion, 1949

xerophilus, Onychogomphus, 1956

yunnanensis, Brachydiplax, 1924

z-viridulum, Paragomphus, 1955

x

M MYs
<a 1Ys :

29 JUN 1966

Unknown

Tervuren

Inst. Miguel Lillo, Tucuman
Inst. Rech. Sci. Madagascar
Inst. Parcs nat. Congo, Brussels
Paris

IFAN, Dakar

IFAN, Dakar

Forest Res. Inst., Dehra Dun.
Inst. Miguel Lillo, Tucuman
Tervuren

Tervuren

Ann Arbor Mus., Michigan

Inst. Miguel Lillo, Tucuman
Inst. Miguel Lillo

Inst. Parcs nat. Congo, Brussels
Inst. Parcs nat. Congo, Brussels
Inst. Parcs nat. Congo, Brussels
Inst. Parcs nat. Congo, Brussels
? Paris

? Indian Mus.

Inst. Rech. Sci. Madagascar

on

A LIST OF SUPPLEMENTS
TO THE ENTOMOLOGICAL SERIES
OF THE BULLETIN OF
THE BRITISH MUSEUM (NATURAL HISTORY)

I. MAsNER, L. The types of Proctotrupoidea (Hymenoptera) in the British
Museum (Natural History) and in the Hope Department of Entomology, Oxford.
Pp. 143. February, 1965. £5.

2. Nixon, G. E. J. A reclassification of the tribe Microgasterini (Hymenoptera :
Braconidae). Pp. 284; 348 Text-figures. August, 1965. 6.

3. Watson, A. A revision of the Ethiopian Drepanidae (Lepidoptera). Pp. 177;
18 plates, 270 Text-figures. August, 1965. £4 4s.

4. SANDS, W. A. A revision of the Termite Subfamily Nasutitermitinae (Isoptera, —
Termitidae) from the Ethiopian Region. Pp. 172; 500 Text-figures. October, ©

1965. £3 55.
5. AHMAD, I. The Leptocorisinae (Heteroptera : Alydidae) of the World. Pp. 156;
475 Text-figures. November, 1965. {2 15s.

6. OxapA, T. Diptera from Nepal. Cryptochaetidae, Diastatidae & Drosophilidae. —

Pp. 129; 328 Text-figures. £3.

7. GILIOMEE, J. H. Morphology and Taxonomy of Adult Males of the Family

Coccidae (Homoptera : Coccoidea). In press.
8. FLETCHER, D. S. A revision of the Ethiopian species and a check list of the
world species of Cleora (Lepidoptera : Geometridae). In press.

PRINTED IN GREAT BRITAIN BY ADLARD & SON LIMITED, BARTHOLOMEW PRESS, DORKING ‘

;.

REVISION OF THE GENUS OROSIU
DISTANT
(HOMOPTERA : CICADELLOIDEA)

M. S. K. GHAURI

BULLETIN OF
THE BRITISH MUSEUM (NATURAL HISTORY)
ENTOMOLOGY Vol. 18 No. 7
LONDON: 1966

REVISION OF THE GENUS OROSIUS DISTANT
(HOMOPTERA : CICADELLOIDEA)

BY

M. S. K. GHAURI _\,,/

Commonwealth Institute of Entomology, London

Php. 229-252 ; 11 Text-figures

BULLETIN OF
THE BRITISH MUSEUM (NATURAL HISTORY)
ENTOMOLOGY Vol. 18 No. 7
LONDON : 1966

THE BULLETIN OF THE BRITISH MUSEUM
(NATURAL HISTORY), instituted in 1949, 1s
issued in five series corresponding to the Departments
of the Museum, and an Historical serves.

Parts will appear at irregular intervals as they become
ready. Volumes will contain about three or four
hundred pages, and will not necessarily be completed
within one calendar year.

In 1965 @ separate supplementary series of longer
papers was instituted, numbered serially for each
Department.

This paper is Vol. 18, No. 7 of the Entomological
series. The abbreviated titles of periodicals cited
follow those of the World List of Scientific Periodicals.

© Trustees of the British Museum (Natural History) 1966

TRUSTEES OF
THE BRITISH MUSEUM (NATURAL HISTORY)

Issued 12 July, 1966 Price Ten Shillings

REVISION OF THE GENUS OROSIUS DISTANT
(HOMOPTERA : CICADELLOIDEA)

BY
By M.S. K. GHAURI

SYNOPSIS

The genera Orosius Distant and Nesophrosyne Kirkaldy are redefined and their status estab-
lished. Species of Orosius are redescribed and illustrated. A new species of Orosius is described.

MEMBERS of the genus Ovosius Distant (1918) : 85) have been recognized as serious
virus vectors affecting economic plants in many parts of the world ; causing con-
siderable losses to sesamum (Sesamum orientale L.) in India (Vasudeva & Sahambi,
1955 and 1958), and to tomato, tobacco and lucerne in Australia (Helson, 1951a
and Heinze, 1951 : 33). In India alone, the loss to sesamum through Phyllody
has been estimated by Vasudeva & Sahambi (1958) as more than three and a half
million pounds sterling per year. The same authors found several other crops,
winter oilseeds (Brassica spp.), sanhemp (Crotalaria juncea L.), gram (Cicer orientum
L.), berseem (Trifolium alexandrium) and many others, susceptible to the virus
disease carried by the same vector. The importance of Orosius argentatus (Evans,
1938) : 15) was recognized even earlier (vide Oman, 1949a: 11) as virus vector of
tomato big bud and tobacco yellow dwarf.

In spite of their economic importance, the identity of the virus vectors belonging
to this group was not clearly understood. Thus they have been known by different
names, Nesaloha cantonis Oman (19430 : 33), Deltocephalus sp. (Vasudeva & Sahambi,
1958), Orosius argentatus (Evans, 1938b) (Day & McKinnon, 1951) and Nesophrosyne
(Orosius) lotophagorum Kirkaldy (Linnavuori, 1960a:56). Evans (im litt.) has
expressed doubt as to whether differences between these should be regarded as of
specific significance. Other species, such as Ovosius albicinctus Distant (19180 : 85),
Thamnotettix cellulosa Lindberg (1927¢ : 90), Thamnotettix filigranus Haupt (19274:
30), Thamnotettix canberrensis Evans (1938): 15) and Ovosius maculatus Singh-
Pruthi (1930a:67) are also involved. There was, thus, a need for re-describing
various species as well as for establishing their correct generic relationships.

Distant (1918b : 85) described Ovosius, with Ovosius albicinctus Distant as the
type-species by monotypy. Later only two other species were described in Orosius.
These are Ovosius maculatus Pruthi (1930a : 67) and Ovosius santali Pruthi (1934a :
25). Ovosius maculatus Pruthi has been designated the type-species of Pruthiorosius
by the writer (Ghauri, 1964:559). Ovosius santali Pruthi was not available for
the present study. Haupt (1927a:30, T. filigranus) and Lindberg (1927c : 90,
T. cellulosa) described one species each under Thamnotettix and Evans (19380 : 15)
added two more species 7. argentatus and T. canberrensis. Oman (19430 : 33)
described another species Nesaloha cantonis, referable to Orosius. Oman (1949:

ENTOM. 18, 7. 18

232 M. S. K. GHAURI

II, 15) transferred 7. argentatus Evans to Ovosius and said that this species occurs
on many central Pacific islands, the Philippines and Malaya in addition to Australia.
He further thought that it was unlikely that the species was native to Australia,
but was probably introduced there from India and that it might be the same as one
of the species described by Distant or Pruthi from India. Dlabola (19520: 34)
transferred T. fiigranus Haupt from Thamnotettix to Orosius. During 1955 the
Indian Ovosius sp. damaging sesamum was called Deltocephalus sp. (Vasudeva
& Sahambi, 1955: 126). Lindberg (1958: 176) transferred his species T. cellulosa
from Thamnotettix to Nesophrosyne and added Cape Verde Island to the original
distribution (Sudan) of this species. Linnavuori (1960 : 320) synonymized Orvosius
Distant and Nesaloha Oman with Nesophrosyne Kirkaldy (1907/ : 160) as well as
transferring O. argentatus (Evans) from Orosius to Nesophrosyne. He also described
a new variety, Nesophrosyne argentatus var. distans (p. 322). Soon after, Linnavuori
(1960a : 56) considered Ovosius Distant as only a subgenus of Nesophrosyne and
also synonymized O. argentatus (Evans) with Allygus lotophagorum Kirkaldy (19074 :
62), transferring the latter from Allygus to Nesophrosyne. In both of his studies
(1960 : 321 and 1960a: 56) the figures of male genitalia belong to O. lotophagorum
(Kirk.), but the comparative remarks about the external appearance (1960 : 322)
apply to O. argentatus (considered by him as the nominate form) and O. lotophagorum
(Kirk.). The variety N. argentatus distans Linnavuori proved to be O. lotophagorum
(Kirk.). Emeljanov (1962) accepted the synonymy WNesophrosyne Kirkaldy
(= Orosius Distant) and included these with other genera, Hishimonus Ishihara
(19530 : rr and 38), Satswmanus Ishihara (1953a@ : 193) Nesophryne Kirkaldy (1907h :
160), Nicolaus Lindberg (1958 : 176), Navata Linnavuori (1960a@ : 52) and Opsianus
Linnavuori (1960 : 316) in the Subtribe Opsiina Em. of the Tribe Opsiini Em.
However, the definition of the Subtribe Opsiina does not seem to hold good if the
character of the eye is taken into account. The “ hollow” is present in the type of
O. albicinctus Distant whereas it is absent in the type of Nesophrosyne perkinst
Kirkaldy and the type of Hishimonus (= Eutettix) disciguttus Walker (19570 : 172
as Acocephala) ; all three species being the type-species of their respective genera.

This work would not have been accomplished without the help rendered by the
following persons and institutions. The writer takes this opportunity to acknow-
ledge with grateful thanks their assistance for making available material for this
study. Dr. J. W. Evans, Director of the Australian Museum, Sydney, Australia ;
The Director, Zoological Survey of India, Calcutta; Dr. J. P. Kramer and Dr.
J. F. Clarke, U.S. National Museum, Washington, D.C., U.S.A. ; Bernice P. Bishop
Museum, Honolulu, Hawaii; Dr. R. Linnavuori, Finland; Dr. H. J. Muller,
Quedlindburg, East Germany; Dr. E. Swirski, The Hebrew University, Israel.
The writer’s warmest thanks are tendered to Mr. J. P. Doncaster, Keeper of the
Department of Entomology in the British Museum (Natural History), London, for
the facilities provided for studying the types of several species of Cicadellidae in
the collection of the Museum.

The writer suspects that Thamnotettix puellus Melichar (Igt1a : 106) might belong
to Orosius, but as the type could not be traced either at Museum National d’Histoire

REVISION OF GENUS OROSIUS 233

Naturelle, Entomologie, Paris, or at Moravské Museum, Brno, this cannot be
decided definitely. Help given by Dr. A. Villiers of the Museum National, Paris and
Dr. J. Stehlik, of Moravské Museum, Brno, in establishing the fact that the type
of 7. puellus Melichar is untraceable is gratefully acknowledged.

Fic. 1. Nesophrosyne perkinsi (Kirkaldy). Holotype g¢ and paratype @ ;
for explanation of figures, see p. 252.
ENTOM. 18, 7. 18§

234 M, S$. K,'GHAURI

NESOPHROSYNE Kirkaldy
(Text-fig. 1)
Nesophrosyne Kirkaldy, 1907h : 160.
Type-species : Eutettix perkinsi Kirkaldy, 1904a : 178.
Moderately robust leaf-hoppers with pale yellow body colour.

Head : vertex with anterior and posterior margins subparallel, frons proportionately broader,
width equal to about three-fourths of length, lateral margins not sinuate ; clypeus with lateral
margins subparallel, elongate; eyes without “ hollow” next to base of antennae; ocelli on
junction of vertex and frons, visible dorsally as well as ventrally.

Pronotum : as wide as head, with anterior margin convex and posterior margin straight,
lateral margins angled ; scutellum wider at base than median length; tegmen appendiculate,
claval veins branches not well defined, external subapical cell petiolate on both ends, internal
subapical cell narrowed in middle; wing with typical Deltocephaline venation as shown in
Text-fig. 1.

3 genitalia and anal tube: pygofers with fewer (about 10) macrosetae located on posterior
half of lateral lobe, setae nearer margin extended beyond it; subgenital plates elongate
triangular, anteriorly broad, posteriorly gently narrowed, not ending in a finger-like process,
external margin with four or five macrosetae and numerous fine and long hair-like setae ; basal
plate more or less “‘ Y ’’-shaped, apex wide, visibly separated from arms by a narrow stem,
arms almost parallel ; parameres with well developed apodeme, apex produced but not hooked ;
aedeagus with two gonopores, base pear-shaped, posteriorly produced on dorsal side to join
anal tube, two-branched, branches converging, each arising independently from base and
carrying a passage opening through a gonopore beyond which apex of branch is produced as a
narrow, hooked process ; anal tube very long, dorsally not entirely membranous, with a wide,
well sclerotized band, laterally and ventrally sclerotized.

OROSIUS Distant
(Text-fig. 2)
Orosius Distant, 1918b: 85.

Type-species : Orosius albicinctus Distant, 19180: 85.

Small leaf-hoppers, usually with irregular transverse series of small dark brown
spots, markings and striations on head, thorax and tegmen giving a filigreed aspect
to the whole coloration.

Head : vertex, shape variable within certain limits, from a condition with anterior and posterior
margins subparallel to one where anterior margin is conically produced, gently but narrowly
curving into frons ; fyons proportionately narrow, width never more than five-eighths of length,
with sides usually sinuate in middle, moderately convex ; clypeus with subparallel sides, elongate;
eyes with “‘ hollow ” near base of antennae ; ocelli located on border between vertex and frons,
visible on dorsal side, on vertex as well as on ventral side on frons.

Pronotum ; as wide as head, anterior margin convex, basal margin straight or only gently
concave, lateral margins not straight, gently angled; scutellum wider at base than median
length ; tegmen appendiculate, clavus with branched claval vein, external subapical cell never
petiolate on both ends, internal subapical cell narrowed in middle; wing with typical Delto-
cephaline venation as shown in Text-fig. 2.

REVISION OF GENUS OROSIUS 235

Fic. 2. Orosius albicinctus Distant. Holotype 9 and Indian $; A,, B,, C,, D, and L
belong to holotype female, the remaining to Indian male; for explanation of figures,
gee p. 252.

236 M. S.iK.GHAURI

3 genitalia and anal tube : pygofers with several (11-20) moderately long macrosetae located
on posterior half of lateral lobes, setae not arranged in regular rows, apices of those farthest
behind extended beyond posterior margin of pygofers; subgenital plates triangular, basally
broad, posterior half abruptly narrowed and produced in a finger-like process, external margin
with a row of 5-8 macrosetae, limited to only anterior broader part, and numerous fine and
very long hair-like setae covering entire external margin of subgenital plate as well as internal
margin of finger-like process; basal plate more or less ‘‘ Y ’’-shaped with narrow or broad
apex, sometimes very short and in form of a ‘‘U”’ rather than a “ Y’’, arms subparallel or
diverging ; paramere with well developed apodeme, apex more or less hooked ; aedeagus with
two gonopores, base pear-shaped, produced posteriorly on dorsal side where it is joined to anal
tube, two-branched, branches parallel or diverging or slightly converging, each branch arising
independently from base carrying a passage opening through a gonopore beyond which apex
of branch, in some species, is narrowly produced but never hooked; anal tube of moderate
length, basally membranous, except in some specimens where there is a very weakly sclerotized
dorsal narrow stripe, on lateral and ventral sides sclerotized.

Orosius very closely resembles Nesophrosyne, but the structure of anal tube,
shape of subgenital plate and to some extent the shape of frons, paramere and that
of aedeagus separate the two genera. The anal tube in Nesophrosyne is dorsally
partly sclerotized, whereas in Orvosius it is almost entirely membranous, posteriorly
the subgenital plate is not narrowed in form of a finger in Nesophrosyne as it is in
Orostus, the hooked process at the apex of each branch of aedeagus is only present in
Nesophrosyne perkinsi. These differences, apart from some minor ones in venation
and the shape of the paramere, are sufficient for regarding Nesophrosyne and Orosius
as two distinct genera.

Orosius albicinctus Distant

(Text-figs. 2, 3)

Orosius albicinctus Distant, 19180: 85.
Thamnotettix fiigranus Haupt, 1927a: 30. syn. n.

Coloration : body pale ochraceous with dark brown and black mottlings and irregular stria-
tions ; vertex with two spots near anterior margin, two similar spots near posterior margin,
two near inner margin of each eye and a curved line in form of an irregular inverted ‘‘ V ” on
each side of median line, dark brown; eyes dark brown; ocelli yellowish red; frons with
transverse dark striations ; irregular dark markings on other parts of face; pronotum with a
series of dark spots near anterior margin and a network of dark striations on disc ; scutellum
with two dark prominent black spots near base in addition to a few more irregular dark markings ;
remaining parts of thorax above black; on ventral side pale ochraceous with dark spots ;
tegmen greyish white with prominent black spots near costal, apical and posterior margins
and irregular criss-cross dark striation covering whole surface ; legs pale ochraceous with black
spots on all segments ; abdomen above and below black with pale ochraceous margins marked
with dark spots; genital segments in both sexes pale ochraceous with very few faintly dark
brown markings. In g, markings are darker than in 9.

aa

———

REVISION OF GENUS OROSIUS 237

MEASUREMENTS in mm.
Females
Indian Middle Eastern and African

oy NS

a
Holotype of Holotype of From From
O. albicinctus T. filigranus Egypt Uganda

Width across eyes 0:81-0:88 0:86 0:92 0:9!
Width between eyes 0+ 31-0: 37 0°34 0°37 0:38
Median length of vertex 0+27-0:29 0:28 0+30 0+ 32
Length of frons 0°55-0°55 0°59 0°59 0-61
Maximum width of frons 0+ 34-0: 37 0:36 0-41 0°41
Width of pronotum 0:81-0:88 0:88 0:92 0:91
Median length of pronotum 0° 41I-0° 43 0°44 0°47 0°47
Width at base of scutellum 0°47-0°47 0°47 O-51 0°54
Median length of scutellum 0+ 30—-0° 34 0+ 34 0°35 0'37
Length of tegmen 2:05-2:63 2°58 2:66 2:66
Body length 2:63-3:21 3°21 3°29 3°29
Males
Indian Middle Eastern and African
(a — ary
Topotypicof From From
T. fiigranus Egypt Uganda

Width across eyes 0:78-0:79 0:81 0:84 0:81
Width between eyes 0+ 31-0: 34 0:32 0°34 0+ 34
Median length of vertex 0°27-0°25 0°27 0-27 0727
Length of frons 0:*49-0°49 0°54 0°52 0:51
Maximum width of frons 0+ 34-0° 34 0+ 34 0-32 0°34
Width of pronotum 0:78-0:78 0:81 0:84 0:84
Median length of pronotum 0+37-0°40 0°44 0°43 0*42
Width at base of scutellum 0°47-0°42 0°44 0°43 0°45
Median length of scutellum 0+ 30-0: 30 0-32 0:34 0:32
Length of tegmen 2+23-2°23 2°37 237 7 ey
Body length 2+71-2:82 2°90 2*90 2:90

Tegmen with appendix wide. Pygofers with numerous setae. Basal plate with apex
broadly triangular and arms sub-parallel or diverging; aedeagus with diverging branches
whose apices in some specimens turn further outwards ; other parts of male genitalia as figured ;
VII sternum, posterior margin not very constant in outline, but always with a central notch,
laterally more or less sinuate as figured. In a large collection of females from Rajasthan,
India, it was noticed that those females which had not oviposited showed only the median
emargination in the posterior margin of the VII sternum, but others which had started ovipost-
ing, the lateral notches, in addition to the median one, were present.

Material examined. INnp1A: The holotype 2 of O. albicinctus is from Kodaikanal,
S. India, B.M. (N.H.). It has been possible to match other specimens from Delhi,
both 2 and g, with the holotype. Rajasthan, Jaipur, at light, J, 9, 12.vi.65,
Dept. Agric., B.M. (N.H.) ; New Delhi, on Sesamum orientale L., 3, 9, ix-x.1954
(H. S. Sahambi).

PALESTINE: @ holotype of Thamnotettix filigranus Haupt, Ben-Shemen,
7-8.v.1926, det. Haupt, Hft., [on loan from Dr. J. H. Muller, Quedlindburg, E.

238

M.S. K. GHAURI

REVISTON OF GENUS OROSTUS 239

Germany] ; 372 (5), Rahovet, r 9, 28.vii.1957 (Michael) [presented to B.M. (N.H.)
by Dr. E. Swirski, The Hebrew University, Israel]; Jericho, at light, 3, 9,
16.vii.1929 (J. Tapukhi), B.M. (N.H.).

Ecyrpt: Giza, on Cotton, 3g, 2, v-vi. 1961 ; Siwa, J, 2, 8, 15, 28.v.1935 (J. Omer
Cooper), Armstrong College Exped., B.M. (N.H.).

UGANDA: Kampala, on Sim-Sim, I J, I 9, 21.vii.1933 (H. Hargreaves), B.M.
(N.H.).

Recently Linnavuori (1964 : 336, 350) recorded N. (0O.) filigranus (Haupt) from
Egypt and considered it to be of Eremian origin. As is shown here Orosvus filigranus
(Haupt) is a synonym of O. albicinctus Dist. and has a much wider distribution
including the Oriental region.

Although the 2 holotype of O. albicinctus is smaller than the 9 holotype of T.
filigranus, other 9 and g from India are not much smaller than those from the Middle
East and Africa (vide Table, p. 237). In addition to size, there is some variation
in colour also, e.g., both g and 9 from Uganda show darkening of pattern, especially
on the head and the tegmen (vide A,, B,, and C,, Text-fig. 3). Since the ¢ genitalia
do not show appreciable differences, all the specimens from various localities studied
here are considered to belong to O. albicinctus Distant.

Orosius cellulosa (Lindberg)
(Text-fig. 4)

Thamnotettix cellulosa Lindberg, 1927c : go.
Nesophrosyne cellulosa (Lindberg), 1958 : 176.

Coloration : very similar to O. albicinctus Distant ; vertex and pronotum with fewer dark
markings, frons and other parts of face with faint markings ; vertex, face and legs with red spots.

Measurements in mm. of 3 and (2): head, width across eyes 0-79 (2 0:88), width between
eyes 0:29 (0°31), median length of vertex 0-26 (0-31), length of frons 0-51 (0-57), maximum
width of frons 0:37 (0-38); width of pronotum 0-74 (0-88), median length of pronotum 0-41
(0-47), width at base of scutellum 0-46 (0-51), median length of scutellum 0-30 (0-34), length of
tegmen 2:23 (2:50); body length 2-76 (3-02).

Tegmen with wide appendix ; pygofers with about 13 setae; basal plate with rectangularly
narrow apex and slightly diverging long arms; aedeagus with subparallel lateral branches
which curve inwards apically, sinuate in lateral view; other parts of ¢ genitalia as figured ;
9 VII sternum, posterior margin convex with a central depression.

Material examined. BRITISH SUDAN: Paratypes, Khartoum, sucking human
blood, r 3, r 9, 17.ix.24 (Well. T. R. Labs.), in B.M. (N.H.).

Superficially O. cellulosa resembles O. albicinctus, but the shape of basal plate and
the aedeagus are sufficient to differentiate it from the latter. The rectangular apex

Fic. 3. Orosius albicinctus Distant. Geographical variation. Holotype 2? and topotypic
36 of T. filigranus Haupt (a synonym of O. albicinctus), g and 2 from Uganda and ¢
from Egypt of O. albicinctus. A, B, C and L belong to holotype 2; H, I, J and K to
topotypic ¢; A,, B,, C,, H,, I,, J, and K, to g and L, to 2 from Uganda; H, and I,
to g from Egypt. For explanation of figures, see p. 252.

240 M. S. K. GHAURI

of basal plate and the subparallel lateral branches of the aedeagus bring it very close
to the Far Eastern species, O. argentatus (Evans) and others, but the sinuate shape
of lateral branches of the aedeagus of O. cellulosa is different from all others.

Fic. 4. Ovosius cellulosa (Lindberg). Paratype g and 9; for explanation
of figures, see p. 252.

REVISION OF GENUS OROSIUS 241

Orosius lotophagorum (Kirkaldy)
(Text-fig. 5)

Allygus lotophagorum Kirkaldy, 1907d : 62.

Nesophrosyne (Orosius) lotophagorum (Kirkaldy) Linnavuori, 1960a@ : 57.
Thamnotettix argentatus Evans ; Linnavuori, 1960a: 57.

Nesophrosyne argentatus (Evans) ; Linnavuori, 1960 : 320.
Nesophrosyne argentatus var. distans Linnavuori, 1960 : 322. syn. n.

Fic. 5. Ovosius lotophagorum (Kirkaldy). ¢g and 9; for explanation
of figures, see p. 252.

242 M. 8; KX) GHAURI

Coloration : basically of the same pattern as that of O. albicinctus Distant; individuals
showing light to dark and more or less extensive markings ; ocelli red.

Measurements in mm. of male and (2) : head, width across eyes 1-00 (0:98-1:10), width between
eyes 0°36 (0°34-0°41), median length of vertex 0-29—-0°300 (0:32-0:34), length of frons 0°55
(0-57-0'58), maximum width of frons 0-42 (0:41-0:44) ; width of pronotum 0-95 (0:88-0:95),
median length of pronotum 0:42-0:44 (0-44), width at base of scutellum 0-54—0°56 (0°57), median
length of scutellum 0-35—0-37 (0-35—0°36), length of tegmen 2-31 (2:50) ; body length 3-00 (3-10).

Body robust, this difference appreciable from O. albicinctus Distant; tegmen with very
narrow appendix ; pygofers with numerous setae ; basal plate with rectangularly narrow apex
and subparallel arms ; aedeagus with lateral branches subparallel, their apices turned inwards,
in lateral view the shaft wide for most of its length, but abruptly narrowed and drawn out in
form of a long process beyond gonopore ; other parts of ¢ genitalia as figured. ? VII sternum,
hind margin convex with a shallow depression in centre and slight depressions on sides of latter.

Material examined. Fiji: Ovalau 1 4g, 2 9, 19.x.24 (E. H. Bryan Jr.), det.
as Nesophrosyne argentatus distans Linnavuori, Bishop Museum Honolulu, Hawaii.

Wake Is.: Sesuvium, J, 9, 28.vii.23 and 30. vii.23, (E. H. Bryan Jr.), some det.
as Nesophrosyne argentatus distans Linnavuori, Bishop Museum, Honolulu, Hawaii
and from Dr. R. Linnavuori, Finland.

O. lotophagorum Kirkaldy resembles other members of the genus Ovosius super-
ficially and sometimes has been confused with O. argentatus (Evans) but its body,
especially the head and the pronotum, are appreciably robust compared with the
more or less slender form of O. argentatus. The appendix of its tegmen on the
other hand is the narrowest amongst the species of the genus Ovosius. Internally,
the most characteristic feature of O. lotophagorum is the shape of its aedeagus ;
viewed laterally the sudden narrowing of its apex should be sufficient to distinguish
it from all other species.

Orosius argentatus (Evans)
(Text-fig. 6)

Thamnotettix argentatus Evans, 1938b : 15.
Thamnotettix argentatus Evans ; Evans, 1940¢ : 3 and It.
Orvosius argentatus (Evans) Oman, 1949a: II.

Coloration : general colour pattern similar to O. albicinctus Distant, dark brown markings
more pronounced on head, thorax and abdomen. Ocelli reddish in colour.

Measurements in mm. of 3 and (9): head, width across eyes 0-78-0-81 (0-88-0-91), width
between eyes 0:30-0:34 (0:35-0°37), median length of vertex 0:24-0:26 (0-25-0-28), length of
frons 0:47 (0°54), maximum width of frons 0:34. (0:35-0:38) ; width of pronotum 0-78-o-81
(0:86-0:88), median length of pronotum 0:40-0:41 (0-41-0°44), width at base of scutellum
0°42—0:47 (0:43-0:49), median length of scutellum 0-30 (0-30—0-34), length of tegmen 2:10-2:21
(2:37-2:76) ; length of body 2:63-2:74 (2:90-3:29).

O. argentatus is more delicate than its two sympatric species, O. lotophagorum (Kirkaldy) and
O. canberrensis (Evans). Tegmen with wide appendix ; pygofers with numerous setae; basal
plate with rectangularly narrow apex and almost parallel arms; aedeagus with lateral arms
almost parallel to each other, with their apices curving inwards, in lateral view gradually
narrowed, without terminal process, other parts of ¢ genitalia as figured; Q VII sternum,
hind margin more or less similar to that of O. filigranus Haupt holotype 9, as shown in Text-fig.
(ose

REVISION OF GENUS OROSIUS 243

Fic. 6. Orosius argentatus (Evans). and 9; for explanation of
figures, see p. 252.

244 M. S. K. GHAURI

Material examined . AUSTRALIA: Musgrave Ra (S.A.), I g, 2 9, vi.61 (H. Cogger) ;
Broken Hill, N.S.W., 2 g, 2 9, ii.1924 (F. W. Shepherd), Australian Museum,
Sydney ; Burnley, Vic., on weeds, 1 g, 3 9, x.30 (J. Evans); Blundale, F. C. T.,
Ig, I 9, 7.1.30 (J. Evans) ; Lockyer, 1 3, 1 9, ex lucerne; 2§: vii. 39 (D. 0. Aa;
B.M. (N.H.) ; Canberra, 9, 11.vi.46 (A. V. Hill) ; Cronula, N.S. Wales, 9, xii. 1924
(H. Petersen), det. as Nesaloha cantonis Oman = Th. argentatus Evans ?, U.S. Nat.
Mus. Washington.

Fic. 7. O. argentatus (Evans) from Java. ¢g and 9; for explanation
of figures, see p. 252.

REVISION OF GENUS OROSIUS 245

NoRFOLK ISLAND : flying over large patch of Ipomoea batatas Lam., 1 3, 10.v.1939
(I. McComish), B.M. (N.H.).

Fij1: Viti Levu, Suva, 1 g, ili.1951 (N. L. H. Krauss), det. as lotophagorum
Kirk. (= argentatus Ev.), Dr. R. Linnavuori Coll., Finland.

W. CAROLINE Is. : Paia I., 2 9, 28-29.iv.1954 (J. W. Beardsley), det. Nesophrosyne
(Orosius) argentatus (Evans), Bishop Museum, Honolulu and Dr. R. Linnavuori
Coll., Finland.

DANGER IsLAND : Motu Katava, 1 9, 1.11.24 (E. H. Bryan Jr.), det. Nesophrosyne
argentatus (Evans), Bishop Museum, Honolulu.

O. argentatus (Evans) has been confused with O. lotophagorum (Kirkaldy) in the
past, but can be readily separated from the latter by its wider appendix and by the
shape of its aedeagus, whose branches narrow gradually.

O. argentatus (Evans) from Java
(Text-fig. 7)

Coloration : very similar to O. argentatus from Australia, but with reddish markings on face,
tegmen and legs.

Measurements in mm. of 3 and (2) : head, width across eyes 0:78-0:79 (0:86), width between
eyes 0°3I1—0°32 (0°34), median length of vertex 0:24-0:27 (0-29), length of frons 0-49—-0°51 (0:51—
0°54), maximum width of frons 0-34 (0-37) ; width of pronotum o-78-—0-82 (0-88), median length
of pronotum 0:41-0:43 (0:43—-0-44), width at base of scutellum 0-41-0-47 (0-45), median length
of scutellum 0-27—0-34 (0°34-0-42), length of tegmen 2-37 (2:53-2°58) ; body length 2-90 (3-16-
9:21).

Specimens from Java are very similar to the nominate form from Australia, only
differing in minor details—basal plate, with rectangularly narrow apex slightly
more developed, aedeagus with lateral branches slightly diverging.

Material examined. WEstT JAVA: Bogor, on Arachis hypogea, 3, 9, vii.1954
(B. H. H. Bergman), B.M. (N.H.).

Orosius argentatus novaebritanniae ssp. n.
(Text-fig. 8)

Coloration : very dark, darker than most of the specimens of the nominate form from Australia,
black markings, especially on frons, much more extensive and deeper in shade.

Measurements in mm. of 3 and (9) : head, width across eyes 0-78 (0:81—0°84), width between
eyes 0:29 (0-32—0-33), median length of vertex 0-24 (0-25-0-27), length of frons 0-47 (0:51),
maximum width of frons 0-34 (0-34); width of pronotum o-78 (0:81-0:82), median length of
pronotum 0-41 (0-41), width at base of scutellum 0-54 (0°54), median length of scutellum 0-34
(0-37-0-41), length of tegmen 2:23 (2:23) ; body length, 2-76 (2:82-2:90).

Tegmen with wide appendix; pygofers with comparatively fewer setae; basal plate with
slightly longer, rectangularly narrow apex and subparallel arms ; aedeagus with lateral branches
slightly diverging, but otherwise similar to those of O. argentatus ; other parts of ¢ genitalia
as figured.

Holotype g. New Britain: Rabaul, Kerarat, 31.viii.1959, ex Crotalaria
goreensis (A. J. von Velsen), B.M. (N.H.).

246 M. S. K. GHAURI

Paratypes: 2 9, with same data as holotype 3.

Although the material examined is not adequate, the new subspecies can be
distinguished from the nominate form by the shape of its basal plate, aedeagus and
the colour of its body which is very dark.

Fic. 8. O. argentatus novaebritanniae ssp.n. dg and 9; for explanation
of figures, see p. 252.

REVISION OF GENUS OROSIUS 247

Orosius cantonis (Oman)
(Text-fig. 9)

Nesaloha cantonis Oman, 19430 : 33.
Orosius cantonensis (Oman) Linnavuori, 1960a : 56.

Coloration : general pattern of coloration similar to that of O. albicinctus Distant, but with
fewer lighter markings.

Measurements in mm. of 3 and (9): head, width across eyes 0:76-0:78 (0:84—0-86), width
between eyes 0-32 (0-34—0-36), median length of vertex 0-20 (0-24), length of frons 0-44 (0-49-
0-51), maximum width of frons 0-32—0-34 (0:36—-0°37) ; width of pronotum 0:74 (0:79-0:81),
median length of pronotum 0-37 (0-41), width at base of scutellum 0-44—0-47 (0-44—0-51), median
length of scutellum 0-30 (0-28—0-34), length of tegmen 2-10 (2-10) ; body length 2-63 (2:63-2:74).

Small and delicate leaf-hoppers ; vertex rounded anteriorly, with anterior and posterior
margins almost parallel, median length and length next to eye almost equal ; tegmen with wide
appendix ; pygofers with numerous setae ; basal plate with apex squarish and anterior arms
slightly diverging; aedeagus with lateral branches slightly diverging, their apices curved
inwards, in lateral view gradually narrowed and turned upwards like a ‘“‘ beak’; other parts
of $ genitalia as figured. 9 VII sternum, hind margin truncate in middle, with broad emargina-
tion on lateral angles, as shown in Text-fig. 9, L.

Material examined. Canton Is.: Ex Boerhaavia diffusa L., 2 3, I 2 paratype,
I.vili.4o (Van Zwaluwenberg) ; ex Boerhaavia tetranda Forst., I 2, 26.ix.1940
(R. Damer). ENIwETOK ATOLL: Eniwetok, Portulaca quadrifida, 2 3, 13.v.46
(Townes 61) ; Aomon, swept grass, I J, 16.v.46 (Townes 218) ; Eniwetok, Japtan,
ex grass, I g, 17.v.46 (Oakley 112). Fassari I.: Uliathiat, r J, 10.vii.46 (H. K.
Townes, No. 1052). All material in U.S. National Museum, Washington, D.C.

Although the shape of the various parts of ¢ genitalia are broadly similar to
those of O. argentatus (Evans), the small size of the body and the shape of its head
with parallel margins of vertex separate O. cantonis (Oman) from all other species of
the genus Orosius.

Recently Ishihara (1963 : 121-123) redescribed and figured Eutettix orientalis
Matsumura (Ig14: 192) as Nesophrosyne orientalis (Matsumura). From the figures
it appears that the specimen belongs to O. cantonis (Oman). As the above-mentioned
study by Ishihara is based on a Formosan material and not on the type or topotypic
specimen, his conclusion cannot be accepted as final. As the type of Eutettix
orientalis is not available at the moment, no further comments can be made.

Orosius canberrensis (Evans) comb. n.
(Text-fig. 10)
Thamnotettix canberrensis Evans, 19380 : 15.

Coloration : pale ochraceous, bright, almost immaculate; eyes greyish; ocelli bright red.

Measurements in mm. of 3 and (2) : head, width across eyes 0-88 (o-98—1-00), width of vertex
between eyes 0-35 (0:42-0:44), median length of vertex 0-25 (0:27), length of frons 0:54 (0-63),
maximum width of frons 0:37 (0-45-0-47) ; width of pronotum 0-88 (0-98—I-0o), median length
of pronotum 0-47 (0:47-0:49), width at base of scutellum 0-51 (0:54-0°57), median length of
scutellum 0-37 (0-41), length of tegmen 2-71 (2:76) ; body length 3:29 (3:42-3:45).

248 M. S. K. GHAURI

Fic. 9. Ovosius cantonis (Oman). Paratype g and 9; for explanation
of figures, see p. 252.

REVISION OF GENUS OROSIUS 249

Fic, 10. Ovosius canberrensis (Evans). 3 and 9; for explanation of
figures, see p. 252.

250 M.S. K. GHAURI

Ovosius canberrensis (Evans) is the largest species of the genus Ovosius, with robust body,
very wide head and the vertex with anterior and posterior margins almost parallel; pronotum
short ; pygofers with numerous setae; basal plate with rectangularly narrow and long apex
and parallel anterior arms; aedeagus with lateral branches comparatively short and widely
parallel, without their apices much produced beyond gonopore as seen in lateral view; other
parts of ¢ genitalia as figured. @ VII sternum, hind margin bulging slightly in middle.

Material examined. AUSTRALIA: Canberra, F.C.T., 1 9 paratype, 1 g, Mar., 1930
(J. Evans), B.M. (N.H.); Broken Hill, N.S.W., 1 9, iti.1924 (F. W. Shepherd),
Australian Museum, Sydney.

The large size, immaculate coloration, parallel-sided vertex, very long apex of
basal plate and the short lateral branches of its aedeagus readily separate O. can-
berrensis (Evans) from other species of Ovosius.

Orosius aegypticus sp. n.
(Text-fig. 11)

Coloration :; general body colour similar to that of O. albicinctus Dist. ; vertex and pronotum
with fewer dark markings, frons and other parts of face with scanty light brown markings,
ocelli reddish yellow ; ovipositor reddish.

Measurements in mm. of 3 and (9): head, width across eyes 0-88 (0-91), width between
eyes 0°37 (0°35), median length of vertex 0-32 (0-34), length of frons 0-61 (0-62), maximum
width of frons 0-37; width of pronotum o-g1 (0-95), median length of pronotum 0-42 (0-42),
width at base of scutellum 0-53 (0°54), median length of scutellum 0-37 (0-40), length of tegmen
2°55 (2°63); body length 3-10 (3:21).

Vertex somewhat flat, wider and longer than most other species ; tegmen with wide appendix ;
pygofers with fewer setae. Basal plate very small compared to other species of Ovosius, stem
reduced, giving it the shape of a “‘U”’ rather than that of a “‘ Y”’; aedeagus with parallel
branches in lateral view showing their sudden narrowing from the globular base ; other parts
of the ¢ genitalia as figured; 9 VII sternum with a shallow, wide emargination on the hind
margin.

Holotype g. Ecypt: Siwa, 12.v.1935, (J. Omer-Cooper), Armstrong College
Expedition B.M. 1935-354, in B.M. (N.H.). Holotype g bearing a determination
label Ovosius cellulosus Ldbg. by W. Wagner, 1955.

Paratypes. Several 9, same data as holotype.

This species is characterised by its flat vertex, smaller basal plate, form of aedeagus
and style, by which it differs from all other species of Ovosius.

The status of the new species should be considered, however, as tentative. Since
only a single 3 was available, the shape of the genitalia and that of the vertex could
not be confirmed by reference to a second g. The holotype $ might be a diseased
specimen although no signs were found to suggest this.

KEY TO THE SPECIES OF OROSIUS

1 Anterior and posterior margins of vertex parallel : . : ‘ ; , 2
— Anterior and posterior margins of vertex not parallel . ; : : ‘
2 Body immaculate . : ; ; ; ‘ ‘ canberrensis (Evans) (p. 247)

HAG: Iz.

REVISION OF GENUS OROSIUS 251

Orosius aegypticus sp. n. Holotype g and paratype 9; A, and L belong to
female, the others to holotype male ; for explanation of figures, see p. 252.

252 M.S. K. GHAURI

— Body with dark markings t . cantonis (Oman) (p. 247)

3 Body robust, appendix of tegmen ; narrow, apex of aedeagus beyond gonopore
abruptly produced as a narrow process . i lotophagorum (Kirkaldy) (p. 241)

— Body of moderate size, appendix of tegmen broad, apex of aedeagus variously
narrowed, but never abruptly so . ; : ; z ‘ ‘ 3 4 4

4 Body of basal plate reduced to a transverse ‘“‘ bar’’, (stem of ‘‘ Y”’ absent)
aegypticus sp. n. (p. 250)
— Body of basal plate developed (stem of “‘ Y ”’ triangular or rectangular) ‘ é 5
5 Body of basal plate triangular, branches of aedeagus widely divergent
albicinctus Distant (p. 236)
Body of basal plate rectangular, branches of aedeagus parallel or sub-parallel : 6
Shaft of aedeagus sinuate. ; 7, ; : : cellulosa ae (p. 239)
Shaft of aedeagus not sinuate : 7
Brown markings on body not extensive, normal, like most of the other species of
Orosius : ‘ argentatus (Evans) (p. 242)
Brown markings very exianeiye, body dark brown
argentatus novaebritanniae ssp. n. (p. 245)

S300 1 ON al

LETTERING USED IN TEXT-FIGURES

Aand A, dorsal view of head and thorax G dorsal view of anal tube
Band B, ventral view of head and thorax H dorsal view of basal plate
CandC, tegmen I lateral view of aedeagus
DandD, wing J dorsal view of aedeagus

E lateral view of pygofers K dorsal view of style

F ventral view of subgenital plate L ventral view of 2 VII sternum

REFERENCES

References not listed herein will be found in the Bibliography of the Cicadelloidea (Homoptera :
Auchenorhyncha) Fascicle VI, March, 1964, by Z. P. Metcalf.

EMELJANOV, A. F. 1962. New Tribes of Leaf-Hoppers of the subfamily Euscelinae (Aucheno-
rhyncha, Cicadellidae). Ent. Obozr. (2) 21 : 388-397.

IsHIHARA, T. 1963. Some Genera, especially ‘ Eutettix ’’ of Japan and Formosa (Hemiptera :
Deltocephalidae). Tvans. Shikoku ent. Soc. (4) 7: 119-124.

GuaurRI, M.S. K. 1964. A new Genus based on Ovosius maculatus Pruthi and re-assignment
of some of Pruthi’s species of Cicadelloidea (Auchenorhyncha : Homoptera). Ann. Mag.
nat. Hist. (13) 6: 559-563.

Heinze, K. 1951. Die Ubertrager pflanzlicher Viruskrankheiten. Mitt. biol. Reichsanst.
Ld-u. Forstw. 71 : 1-127.

LINDBERG, H. 1958. Hemiptera Insularum Caboverdensium. Commentat. biol. 19 : 1-246.

LinnAvuorlI, R. 1960. Insects of Micronesia, Homoptera : Cicadellidae. Insects Micronesia

(5) 6 : 231-344.

1960a. Cicadellidae (Homoptera, Auchenorhyncha) of Fiji. Acta ent. fenn. 15: 1-71.

1964. Hemiptera of Egypt, with remarks on some species of the adjacent Eremian region.

Suomal. eléin-ja kasvit. Seur. van. kasvit. Julk. 1 : 306-356.

VasupvEva, R. S. & SaHAmBI, H. S. 1955. Phyllody in Sesamum (Sesamum orientale L.).
Indian Path. 8 : 124-129.

1958. Interrelationship and perpetuation of phyllody disease of some oilseed and

other common plants. Pl. Path. Ent., Nishigahara 1: 1-8

A MY
Ce
S Z

13 JUL 1966
a es

A LIST OF SUPPLEMENTS
TO THE ENTOMOLOGICAL SERIES
OF THE BULLETIN OF
THE BRITISH MUSEUM (NATURAL HISTORY)

i. Masner, L. The types of Proctotrupoidea (Hymenoptera) in the British
Museum (Natural History) and in the Hope Department of Entomology, Oxford.
Pp. 143. February, 1965. £5.

2. Nixon, G. E. J. A reclassification of the tribe Microgasterini (Hymenoptera: —
Braconidae). Pp. 284; 348 Text-figures. August, 1965. 6.

3. Watson, A. A revision of the Ethiopian Drepanidae (Lepidoptera). Pp. 177;
18 plates, 270 Text-figures. August, 1965. £4 4s.

4. Sanps, W. A. A revision of the Termite Subfamily Nasutitermitinae (Isoptera,
Termitidae) from the Ethiopian Region. Pp. 172; 500 Text-figures. October,
1965. {£3 5s. | |

5. AHMAD, I. The Leptocorisinae (Heteroptera : Alydidae) of the World. Pp. 156; —
475 Text-figures. November, 1965. {£2 I5s.

6. Oxapa, T. Diptera from Nepal. Cryptochaetidae, Diastatidae & Drosophilidae.
Pp. 129; 328 Text-figures. £3.

7. GILIOMEE, J. H. Morphology and Taxonomy of Adult Males of the Family

Coccidae (Homoptera : Coccoidea). In press. 4
8. FLEtcHER, D. S. A revision of the Ethiopian species and a check list of the —
world species of Cleora (Lepidoptera : Geometridae). In press.

D
PRINTED IN GREAT BRITAIN BY ADLARD & SON LIMITED BARTHOLOMEW PRESS, DORKING

A REVISION OF THE GENUS ANTHEN
FROM THE ORIENTAL REGION
(LEPIDOPTERA : LYCAENIDAE)

Gy. EB. THTE

BULLETIN OF
THE BRITISH MUSEUM (NATURAL HISTORY)
ENTOMOLOGY Vol. 18 No. 8
LONDON : 1966

A REVISION OF THE GENUS ANTHENE
FROM THE ORIENTAL REGION
(LEPIDOPTERA : LYCAENIDAE)

BY

Gob. EEE }
XU" °
British Museum (Natural History)

Php. 253-275; 2 Plates; 28 Text-figures

BULLETIN OF
THE BRITISH MUSEUM (NATURAL HISTORY)
ENTOMOLOGY Vol. 18 No. 8

LONDON: 1966

THE BULLETIN OF THE BRITISH MUSEUM
(NATURAL HISTORY), instituted im 1949, 1s
issued in five series corresponding to the Departments
of the Museum, and an Historical series.

Parts will appear at irregular intervals as they become
ready. Volumes will contain about three or four
hundred pages, and will not necessarily be completed
within one calendar ‘year.

In 1965 a@ separate supplementary series of longer
papers was instituted, numbered serially for each
Department.

This paper is Vol. 18, No. 8 of the Entomological
series. The abbreviated titles of periodicals cited
follow those of the World List of Scientific Periodicals.

© Trustees of the British Museum (Natural History) 1966

TRUSTEES OF
THE BRITISH MUSEUM (NATURAL HISTORY)

Issued 21 September 1966 Price Twelve Shillings

A REVISION OF THE GENUS ANTHENE
FROM THE ORIENTAL REGION
(LEPIDOPTERA : LYCAENIDAE)

By G. E. TITE

CONTENTS
; Page
INTRODUCTION ; ‘ : : : ; F : : F 255
Anthene Doubleday . : ' : : : : ‘ ; 256
Pseudodipsas Felder ; : : : ; ‘ : : : 272
REFERENCES . : E : : : : ; : . - 272
INDEX . ‘ : F , - ‘ : 3 : ; : 274

SYNOPSIS

As a result of the study of the male genitalia the grouping of the species and subspecies in
the genus is revised. Seven new taxa are described, and the species described as Lycaenesthes
vufimargo Rothschild is removed to the genus Pseudodipsas.

INTRODUCTION

THE rearrangement of the material in the British Museum (Natural History) has
elicited the fact that Fruhstorfer’s interpretation (1916 and 1923) of the relation-
ships of the species and races contained in the genus Anthene, then known as
Lycaenesthes, is much in need of revision. That author divided the genus into only
two species which he gave as (a) Lycaenesthes emolus and (b) Lycaenesthes lycaenina ;
rather inconsistently he went on to say that Lycaenesthes philo must be separated
from (a) on genitalic grounds. In the main, his species (a) and (b) do correspond
with (a) and (b) below, but his placing of /ycaenoides as a subspecies of emolus is
quite wrong. Corbet & Pendlebury (1956) in dealing with the Malayan species
did make some modifications, but a more general survey dealing with the named
forms from the whole of the oriental region is now desirable. Toxopeus visited
Tring Museum in the 1930s, and evidently intended to publish a revision of the
genus ; he selected types, placing name labels on the specimens; unfortunately
the intended work never materialized. In the present survey, these names have
been adopted where deemed necessary ; they are indicated by the words “ Toxopeus
MS.” in acknowledgement of their originator, and unless otherwise stated, the
actual specimens selected by Toxopeus are designated as types.

All the forms here dealt with are very similar in outward structure and appearance.
Study of the ¢ genitalia does however reveal that there are two main groups:
(a) the emolus Godart group, which includes seltuttus Rober, paraffinis Fruhstorfer,
and philo Hopffer; and (b) the lycaenoides Felder group, which includes licates
Hewitson, villosa Snellen, and lycaenina Felder.

ENTOM. 18, 8, 19

256 G. E. TITE

Many subspecific names founded on inconstant or doubtful characters are to be
found in the literature. In this work an attempt is made to assign such names to
their correct species, although the inclusion of a name does not necessarily imply
agreement that it does in fact represent a well differentiated subspecies. It should
be noted that the localities given in the text are those represented in the B.M. (N.H.)
collection ; they do not in every case portray the full extent of the area inhabited
by the species or subspecies concerned. The word (Type !) after a reference indicates
that the type is in the B.M. (N.H.), and that it has been examined.

ACKNOWLEDGEMENTS

The author wishes to express his thanks and appreciation to Sir Keith Cantlie,
C.I.E. for his kindly aid in elucidating the status of lycaenoides and lycaenina.

ANTHENE Doubleday, 1847

Type-species : Papilio larydas Cramer, 1780.

Anthene Doubleday, 1847 : 27.

Lycaenesthes Moore, 1865 : 773.

Lycaenesthes Moore ; Bethune-Baker, 1910: 14.

Lycaenesthes Moore ; Fruhstorfer, 1916: 96.

Lycaenesthes Moore ; Fruhstorfer, 1923 : 896.

Lycaenesthes Moore ; Aurivillius, 1924 : 435.

Anthene Doubleday ; Hemming, 1935 : 435 [designation of type].

Diagnoses of the genus have been given by Bethune-Baker, Fruhstorfer and
Aurivillius (as Lycaenesthes) and by Corbet as Anthene.

KEY TO SPECIES

1 Underside fore wing, the median band is unbroken. @ forewing with a large white

area in the disc on both surfaces. : . lycaenoides (p. 270)
— Underside forewing, the median band is broken at either vein 2, 3,0r 4. Q fore wing

never with a clear white area on both surfaces . , : : : : ; 2
2 Underside fore wing, the median band is broken at vein3 . ‘ ‘ ‘ 3

— Underside fore wing, the median band is broken at vein 2, or at 2 and 3 ; , 5

3 Underside fore wing without markings inwards of or below the discoidal spot . ‘
lycaenina (p. 269)

— Underside fore wing with irregular white markings inwards of or below the discoidal
Spot + : 4

4 Underside fore wing, the white tharkings inwards of the discoidal spot talie the form

of a short stripe in the middle of the cell, and a long excurved stripe below it in
areal . : , . licates (p. 270)

— Underside fore wing, the white markings below the discoidal spot take the form of a

line starting at the lower portion of the discoidal spot, and proceeding towards the

hind margin and the base in a wide curve; in some ce this line is still further
extended to form a complete circle . : . . villosa (p. 271)

5 Underside fore wing with transverse band very ‘irregular in areas 3, 4, 5 and 6.
Fringes all wings above, whitish , ; : , ; , ‘ : ' 6

REVISION OF ORIENTAL ANTHENE 257

—- Underside fore wing with transverse band fairly regular in areas 3, 4, 5 and 6.

Fringes all wings pale fuscous . : ‘ , ; ; 3 q 7
6 Upperside all wings, intense shining violaecous. “Underside, all markings large and
well defined . : : Dhilo (p. 266)

— Upperside all wings, dull violaceous. "Underside pale, ‘all markings weakly defined
lycaenolus (p. 267)
7 Underside hind wing, the median band of spots is evenly curved, and scarcely nearer

the margin in areas 4 and 5 than it is in area 2 : ; . emolus (p. 257)
— Underside hind wing, the median band of spots bulges outward so that it is consider-

ably nearer the margin in areas 4 and 5 than it is in area 2 : 8
8 Upperside all wings with a black marginal band of up to 1 mm. in width, Hind wing,

with a submarginal series of internervular spots ; , seltuttus (p. 259)
— Upperside all wings with a scarcely perceptible hair-like black marginal band. Hind

wing, with submarginal spots in areas 2 and 3 only . : ; paraffinis (p. 263)

Anthene emolus (Godart)

The oldest name for this species would seem to be Lampides balliston Hibner
(1823: 11). Previously, authors have given the date 1819 or 1823 as the date of
publication for emolus Godart, but according to Sherborn and Woodward (1899 : 595),
(1906 : 578) the correct date is 1824; thus emolus is the junior name. As the name
balliston has not been used as a primary synonym for more than 50 years, and as
the species under consideration has for that period been known to entomologists
as emolus Godart, it would appear that this is a case for application to the Inter-
national Commission on Zoological Nomenclature—under Article 23 (B) (i) of the
I.C.Z.N.—for the suppression of the name balliston Hiibner as a nomen oblitum.
This action is being taken, and it is hoped that the synonomy used in this paper
will be confirmed in due course.

Subspecific names have been given as shown below ; they do not always represent
very distinct races, and so are listed without comment. Continental dry season
forms are considerably paler in colour on both surfaces.

(1) A. emolus emolus (Godart)

(Textfigs. 7, 21 and 22)

Lampides balliston Hiibner, 1823 : 11, figs. 229-230, ‘‘aus Georgien in Florida”, [nomen
oblitum].

Polyommatus emolus Godart, 1824 : 656, Bengal.

Lycaenesthes bengalensis Moore, 1865 : 773, pl. 41, fig. 9.

Lycaenesthes balliston (Hiibner) Semper, 1879 : 165.

Plebeius balliston (Hiibner) Kheil, 1884 : 28.

Nacaduba klanga Corbet, 1938 : 143.

Anthene emolus (Godart) Corbet, 1956 : 457.

In B.M. (N.H.) from South India, NW. India, Bhutan, Sikkim, Assam, Burma,
and Tenasserim.
ENTOoM. 18, 8. 19§

258

G.E. TITE

9 10 11 iz

Fics. 1-12. 1-6. gunderside fore wing (diagrammatic) : 1. Anthene lycaenoides ; 2. Pseudo-
dipsas rufimargo; 3. Anthene lycaenina; 4. A. licates; 5. A. villosa; 6. A. philo.
7. fore and hind wing, A.emolus. 8. g hind wing, A. seltuttus. 9 and 11. clasper:
9. Pseudodipsas rufimargo; 11. P. digglesii. toand 12. aedeagus: 10. P. rufimargo ;
12. P. digglesit.

REVISION OF ORIENTAL ANTHENE 259

(2) A. emolus andamanicus (Fruhstorfer)

Lycaenesthes emolus andamanicus Fruhstorfer, 1916 : 97, Andamans.
Anthene emolus andamanicus (Fruhstorfer) Cantlie, 1963 : 67.

In B.M. (N.H.) from Andamans.

(3) A. emolus goberus (Fruhstorfer)

Lycaenesthes emolus goberus Fruhstorfer, 1916: 97, North Borneo.
Anthene emolus goberus (Fruhstorfer) Corbet, 1956 : 291.

In B.M. (N.H.) from Siam, Malaya, Sumatra, Borneo, and Hainan.

(4) A. emolus modesta (Staudinger) comb. n.
Pseudodipsas modesta Staudinger, 1889 : 104, Palawan.

In B.M. (N.H.) 2 3, 1 9, Palawan.

(5) A. emolus minor (v. Eecke) comb. n.

Lycaenesthes emolus minor v. Eecke, 1918 : 94, Pulo Babi.

(6) A. emolus javanus (Fruhstorfer) comb. n.

Lycaenesthes emolus javanus Fruhstorfer, 1916: 97, E. Java.

In B.M. (N.H.) from Java and Sumbawa.

Anthene seltuttus (Rober)

The races grouped here under seltuttus and paraffinis have long been treated as
conspecific with emolus. Examination of the male genitalia shows that while
seltuttus and paraffinis are indistinguishable, emolus differs in the size and shape of
the clasper, rendering its specific separation necessary. According to the data
labels on the material examined, both the other species occur together in certain
localities, yet over parts of the range, each occurs as a well differentiated geographic
entity. They can be distinguished in the male by the presence on the upperside
of seltuttus of a dusky wide margin of up to I mm. on all wings, and of a usually
complete series of black submarginal spots on the hind wing; in paraffinis, on the
upperside by the brighter violaceous ground colour, the hair-like appearance of the
wing margins, and by the presence on the hind wings of only two black submarginal
spots situated in areas 2 and 3. The undersides of the two species in both sexes
are ‘very similar, but careful comparison reveals the following points: seltuttus, on
the fore wing the spots forming the median band tend to coalesce, those in areas
5-6 though irregularly placed form a fairly straight line, the hachures enclosing the
submarginal spots are sharply acute, especially those on the hind wing ; paraffinis :
the spots of the fore wing median band are smaller, rounder, and those in areas
3-7 form a rough arc, the submarginal hachures tend to be less acute, those on the
fore wing often being almost straight.

G. EB. STITE

260

‘CHUN) ‘IW’ 04} Ut Terezeur oy} Aq poourysur se uoTsayy uendeg 94} ur s1uufpsvd *p pure snynyas auayjup JO UOT{NQLIYAsIp 9YL

Diasou

tu |
tuossiu DAIDIOIA

solyyjow “: & SYN} yas
‘yuy mG | $
siuijyouod . sisuakay

la Mild

S SPDISINO] ay

<

saprojowa siuijjyoiod SIsuaUIOqWD sN{injjas b/

Q 44D] POOMS RB
Vas s| aS
arnt Mares

| sa1dwog
"| uD>|

saga|a>

° fe Ve
‘| Auonbs «4 =
SDIYLDW ‘3S > oe

pslayoujoy

REVISION OF ORIENTAL ANTHENE 261

(x) A. seltuttus amboinensis (Butler) comb. n.
Lycaenesthes amboinensis Butler, 1899 : 343, Amboina. (Type!).

The material listed below is insufficient to provide a sure guide to the subspeciation
obtaining in the Moluccas. As far as can be judged, the males from Halmaheira
and Obi do not differ from Butler’s type from Amboina. The female that that
author described and compared with L. turnert (= lycaenoides godeffroyi Semper)
is almost certainly an example of A. lycaenoides lycaenoides (Felder), as he states that
it was in the Hewitson collection, and the only specimen from that collection in
the B.M. is certainly of that species. The solitary amboinensis 9 in the B.M. is from
the island of Obi; it is described below, but in view of the fact that it is not known
whether it is identical with females from Amboina, it is not proposed to select this
insectas the neallotype.

Fore wing upperside, fuscous with a large but ill defined patch of lavender, which covers the
basal two-thirds of area 1, the basal halves of areas 2 and 3, and extends into areas 4 and 5,
and the cell in the neighbourhood of the discoidal.

Hind wing upperside, fuscous with a sparse scattering of lavender scales, a series of six inter-
neural black submarginal spots, each spot being encircled by an obscure lighter ring and bounded
outwardly by a white stripe.

The underside is similar to that of the ¢; it is not more varied with white as stated by Butler.

A @° from Batchian is larger and darker, the lavender patch on the fore wing being
represented only by a very few scattered lavender scales disposed mainly in area I.
More material is needed in order to ascertain if the Batchian examples represent a
new subspecies.

In B.M. (N.H.) from Motuccas: Amboina, 1 g, 1910 (Rey); Amboina, I g
(holotype) (Coll. Walker) ; Halmaheira, 8g; Obi, 11g, 1 2; Ternate, 1 f (Doherty) ;
Batchian, 2 9, 1 9.

(2) A. seltuttus keyensis ssp. n.

Lycaenesthes emolus keyensis Toxopeus MS.

This subspecies is very similar to s. amboinensis, only differing from it as follows : upperside 6,
the dusky marginal line is distinctly narrower ; upperside 9, the lavender scaling is much less
intense, appearing as a pale cloud on the dusky ground of the forewing; underside $9, the
ground colour is distinctly paler than that of either amboinensis or seltuttus.

As Toxopeus selected an abnormal specimen as holotype, an example that serves
better to characterize the race has been designated below.

Holotype g. Key Istanps: Little Key, 19.iii.1897 (H. Kuhn), B.M. Type
No. Rh. 18345.

Allotype 2. As holotype, 21.ii.1897, B.M. Type No. Rh. 18346.

Other examples in B.M., Key IsLanps ; Little Key, 10g, 1 9; Kissoei, Watoebela,
I 3 (Kuhn).

262 G: E. TITE
(3) A. seltuttus seltuttus (Réber) comb. n.

(Pl. x, fig. 37; Pl. 2, fig. 49. Text-figs. 8, 25, and 26)

Plebeius seltuttus Rober, 1886: 67, Pl. 5, fig. 24, Aru.
Lycaenesthes aruana Butler, 1899 : 344, Aru.

The 3 is not to be distinguished in any way from that sex of s. amboinensis.

Rober figures and describes the 2 as having a white patch on the fore wing. Butler
also describes the white patch, and mentions two such examples in the B.M. These
white-patched insects are still in the collection together with six other females
from Aru which exhibit no indication of a white patch; this suggests that in the
majority of females the fore wing is completely fuscous, as are all seltuttus females
from New Guinea.

In B.M. (N.H.) from Aru: 21g, 8 9, various collectors. NEw GUINEA: Waigeu,
I ¢ (Waterstradt) ; Salwatti, 1 g (Wallace); Kapaur, 3 g, 2 2 (Doherty); I. of
Amberfron, 1 g, 1909 (C. & F. Pratt); Ansus, Jobi, 1 9, 1892 (Doherty); Dorei
Bay, i 2, 1909 (C. & F. Pratt) ; Humbolt Bay, 2 3, 1 2 (Swinhoe Coll.) ; Humbolt
Bay, 23, 6 9, 1892 (Doherty) ; Astrolabe Bay, 7 3, 2 2 (C. Wahnes) ; Stephansort,
14; Berlinhafen, 1 g (Fruhstorfer) ; Woodlark I., 1 3, 4 9, iv.1897 (A. S. Meek) ;
Fergusson I., 7 3, 3 9, 1894-5 (Meek) ; Kiriwini, Trobriand Is., ro g, 5 9, 1895
(Meek).

(4) A. seltuttus violacea (Butler) comb. n.

Lycaenesthes violacea Butler, 1899 : 343, St. Aignan.

Slightly smaller, but very similar to the nominate subspecies, only differing as follows.

Upperside g. The colour is clearer less dingy violaceous, and the blackish submarginal
spots on the hind wing, and the margins of all wings are more definitely defined and neater in
appearance.

Upperside 9. A very slight greyish lightening is present on the disc of the fore wing.

Underside g9. Very like that of the nominate race.

In B.M. (N.H.) from LouisiabDEs : St. Aignan, 3 ¢ (including type), 3 2; Rossell
Island, 2 J, 2 2; Sudest Island, 6 J, 1 (all collected by A. S. Meek).

(5) A. seltuttus affinis (Waterhouse & Turner) comb. n.

Lycaenesthes modestus Waterhouse, 1903 : 198 [nec modesta Staudinger].
Lycaenesthes emolus affinis Waterhouse & Turner, 1905 : 801.

Differs from all other subspecies by the almost lilac violaceous colour of the male
above, and by the pale grey-fawn ground of the underside in both sexes.

In B.M. (N.H.) from AustRALiA: Port Darwin; Cape York; Thursday I. ;
Groote Eylandt, Northern Territory ; Kuranda; Cooktown; Cedar Bay ; Bowen ;
Mackay.

REVISION OF ORIENTAL ANTHENE 263
Anthene paraffinis (Fruhstorfer)

(1) A. paraffinis emoloides ssp. n.
(Pl. 1, figs. 35-36; Pl. 2, figs. 47-48)
Lycaenesthes emoloides emoloides Toxopeus MS.

3g. The upperside on all wings is deep violaceous, and the dark margins are extremely fine.
Dusky submarginal spots, each edged distally by a whitish line, are present in areas 1 and 2
of the hind wings.

9. The upperside on all wings is dingy fuscous, having on the fore wing a diffusely margined
whitish grey-blue area of individually variable extent. This area usually covers the basal
half of areas 1 and 2, sometimes extending into the lower part of the cell and the adjoining parts
of areas 3 and 4. On the hind wing, there is a series of dusky triangular submarginal spots,
each spot margined outwardly with a whitish stripe.

Holotype 3. British NEw GuINEA: Hydrographer Mountains, 2500 ft., i. 1g18
(Eichhorn Bros.), B.M. Type No. Rh. 18347.

Allotype 9. As holotype, i-ii.r918, B.M. Type No. Rh. 18348.

Other examples in B.M. (N.H.). British NEw GUINEA: as holotype, 12 4,
i-li.rg18 ; Kumusi R., 5 g, 1907 (A. S. Meek); Aroa R., 4 g (Meek); Welsh R.,
2 3 (Wetske) ; Milne Bay, 2g, 1 2, 1899 (Meek) ; Port Moresby, 2g; St. Joseph R.,
Ig (Weiske) ; Astrolabe Range, 2 3, 2 9, 1917 (Dodd) ; Haidana, Collingwood Bay,
I 9,iv.1907 (Meek); YuleI., 1 f, 1875 (L. M. d’Albertis). MANDATED NEW GUINEA:
Vulcan I., 7 3, 4 9, 1913-4 (Meek). DutcH NEw GuINnEA: Nr. Oetakwa R., Snow
Mts., 3 5, 1910 (Meek) ; Base Camp, Utakwa R., Sea Level, 3 3, 1912-3 (A. F. R.
Wollaston) ; Upper Setekwa R., Snow Mts., 2—3000 ft., 4 J, 1910 (Meek) ; Eilanden
R., 2 g, xil.1910 (Meek); Mt. Goliath, r Q, ili.rgrr (Meek); Sabron, Cyclops
Mts., 2 J, v.1936 (L. E. Cheesman) ; Njau Limon, S. of Mt. Bougainville, 300 ft.,
I g, 1936 (Cheesman) ; Sorong, 2 g, 1876 (A. J. Bruijn); Humbolt Bay Area
(variously labelled) 43,1 9: Kapaur, 64,1 2 (Doherty) ; Jobil., 53,8 9 (Doherty) ;
Mefor I., r 3, 1 2 (Doherty); Roon I1., 5 g, 1 2 (Pratt); Roon I., 1 g (Doherty) ;
Amberfron I., 2 g, r 2 (Pratt); Biak, Schouten Is., 2 2 (Pratt); Wandesi, 8 3
(Doherty) ; Wangaar R., 15 miles from the Coast, 600 ft., 1 g, 1921 (Pratt) ; Waigeu,
Ir g,5 2; Salwatti,2g¢. Motuccas: Misol,5¢,1 2; Buru, 1,3 9; Sula Is., 2 9.

(2) A. paraffinis paraffinis (Fruhstorfer) comb. n.
(Text-figs. 27-28)
Lycaenesthes emolus paraffinis Fruhstorfer, 1916: 99, Neu Lauenberg (Duke of York I.).

The ¢ is not to be distinguished from that of p. emoloides, but the absence of the whitish
grey-blue patch on the fore wing of the 9 renders the separation of the two races necessary.
The 2 occurs in two forms.

(a) All wings are fuscous brown above. There is on the hind wing a series of 5 triangular
submarginal black spots, each edged inwardly by a white hachure, and outwardly by a white
stripe ; on the basal portions of all wings is a scattering of bright blue scales,

264 Ge. Bs TPE

(b) Similar to above, but the blue scaling is extended and intensified so that it covers about
two-thirds of the wing area, leaving only the costa, the apical area, and the distal margin of
the fore wing fuscous. In some examples on the hind wing the blue reaches almost to the
distal margin.

In the B.M. (N.H.) from Bismarcxs: New Britain; FeniI.; Witu (French I.) ;
New Ireland; Duke of York I. Sotomons: Bougainville; Vella Lavella; Guizo
I.; Choiseul I.; Alu, nr. Shortland I.; Rendova; Treasury.

(3) A. paraffinis nissani ssp. n.
(Pl. x, figs. 31-32; Pl. 2, figs. 43-44)

g. On the upperside, only differs from nominate race by the brighter, more intense purple
lustre.

Q. The upperside is similar to that of the blue form of p. paraffinis, the blue colour being
even denser and more extended with a resulting reduction in width of the fuscous discal margin
of the fore wing, and causing the black submarginal spots on the hind wing to stand out clearly
on the blue field.

In both sexes, the underside is characterized on all wings by a chalk-like whitish suffusion
of the ground colour, which, though present in all parts of the wings is most intense between
the median band and the submarginal markings. All the wing markings are dun coloured and
contrast strongly with the whitish ground, thus forming a parallel development to that of
Catopyrops ancyra distincta Tite (1963: 106) from the same habitat.

Holotype g. SoLtomons: Nissan I., 1924 (A. F. Eichhorn), B.M. Type No.
Rh. 18349.

Allotype 9, as holotype, B.M. Type No. Rh. 18350.
Other examples in B.M. (N.H.), 3 g, I @.

(4) A. paraffinis matthias ssp. n.
(Pl. 1, figs. 29-30; Pl. 2, figs. 41-42)

g. The upperside ground colour is of a lighter more reddish violaceous tint than is that of
the otherwise very similar p. paraffinis.

9. Onthe upperside like that of the nominate race, the colour varying individually from fuscous
brown with a small basal admixture of blue scales to examples in which the blue colour extends
over most of the wings.

Underside in both sexes. The ground colour is pale grey-brown, much colder in tone than
that of p. paraffinis ; a distinct difference in colour, which is immediately observable on arranging
a series of each subspecies side by side. The pattern of the spotting exhibits no distinctive
character, but there is no trace of the white suffusion so evident in p. nissant.

In size it is consistently smaller than the nominate subspecies.

Holotype g. St. MatruiAs Group: St. Matthias I., vi.r9q23 (A. F. Eichhorn),
B.M. Type No. Rh. 18351.

Allotype 9. As holotype, vi-vii.1923, B.M. Type No. Rh. 18352.

Other examples in B.M. (N.H.) from St. Matrut1as Group: as holotype, 34,6 9;
Squally I., 7 g, r 9, 1923 (A. F, Eichhorn).

.
y
:
:
.
.
:
.
.

moe,

265

REVISION OF ORIENTAL ANTHENE

Ot eneceens eee”

i7

Fics. 13-20. Odd numbers clasper, even numbers aedeagus : 13-14. Anthene lycaenoides ;
15-16. A, licates ; 17-18, A. lycaenina ; 19-20, A, villosa,

266 G2 Es REE E

(5) A. paraffinis nereia ssp. n.
(Pl. 1, figs. 33-34; Pl. 2, figs. 45-46)

Lycaenesthes emoloides nereia Toxopeus MS.

Both sexes are very similar to p. paraffinis, but they differ in the fore wing by having a slightly
more obtuse apex and a more convex distal edge.

On the underside, the white edges of the spots forming the transverse median band are wider
and more noticeable ; the spots themselves are larger, and because of this appear to fuse with
one another, giving the appearance of a solid band rather than a chain of separate spots.

9. The blue scaling on the upperside is present in all examples seen, but it is of a subdued,
more violaceous hue than that found in females of p. paraffinis.

As Toxopeus labelled an example of #. paraffinis from Guizo Island as allotype,
a female as listed below has been designated.

Holotype g. SoLtomons: Guadalcanar, v.1891 (A. S. Meek), B.M. Type No.
Rh. 18353.

Allotype 9. As holotype, iv.1891, B.M. Type No. Rh. 18354.

Other examples in B.M. (N.H.) from SoLomons: as holotype, 1g; Guadalcanar,
Ig, 1 2 (Woodford) ; Gela (= Guadalcanar) or Guadalcanar without further details,
936; Floridal., 2g; Tulagi, 1g, 1 9; Ugi, 19,1 9 (G. F. Mathew) ; Uru Bay,
Malaita, 2 3 (Woodford).

Anthene philo (Hopffer)
(1) A. philo philo (Hopffer) comb. n.
(Text-figs. 6, 23, 24)
Lycaena philo Hopffer, 1874 : 27, 3.
Lycaenesthes leocrates Hewitson, 1878 : 220, pl. go, figs. 5 and 10, Macassar, (Type!), 9.
Lycaenesthes philo f. praeclava, Fruhstorfer, 1923 : 897, Kalawara.

It is most probable that emolus, philo, and seltuttus are of common stock, and
each so far as is known inhabits separate territory. In spite of this, they do each
exhibit constant differences in both pattern of the markings on the underside, and
in the formation of the valves. These facts serve to indicate that the divergence
has reached a stage that renders treatment as separate species desirable.

This species and /ycaenolus can be instantly recognized from other members of the genus by
the arrangement of the spots forming the median band on the underside of the fore wing ;
spots 1, 2, and 3 are in echelon, spot 3 being nearest the distal margin ; spot 4 is placed inwards
towards the base of the wing ; spot 5 is placed outwards ; spot 6 is placed well inwards. This
arrangement gives the whole band a most distinctive sinuous appearance.

Fruhstorfer described the form praeclara from Kalawara to the south of Palu
(Paloe), stating that it is larger, lighter above, and with more rounded hind wings.
The only example from that locality in the B.M. is not noticeably lighter in colour,
but it is certainly rather large, and has rounded hind wings.

In B.M. (N.H.) from CELEBES: Bangkei, 1 g, 1885 (H. Kuhn); Saleyer, I g
(Fruhstorfer) ; South and East Celebes, 16 3, 4 2 (including type of Jeocrates from
Makassar). SuLa IsLaAnDs: Sula Mangoli, 6 g, x.1897 (Doherty) ; Ufola, Mangoli,
3 3g, xi.1897 (Crowley Bequest). ToEKAN BeEsI; Binongka, 1 3 (Kuhn).

REVISION OF ORIENTAL ANTHENE 267

(2) A. philo scintillans ssp. n.
(PICT dps 38> Plog. he. 50)
Lycaenesthes emolus scintillans, Toxopeus, MS.

6. The upper surface scarcely differs from that of the nominate subspecies, examples in
the B.M. only showing a more vivid shining purple colour. On the under surface, on all wings
the general tone is slightly deeper brown, and the submarginal spots and lunules are more
clearly marked. All the spots forming the median band are much larger, and exhibit a greater
tendency to coalesce ; the double spot in area 1 of the fore wing is extended inwards towards
the base, forming a rough representation of a heart. All the markings are clearly emphasized
by white rings. The tornal spot on the hind wing is narrowly margined inwardly by a dingy
orange lunule.

Q. On the upper surface, the ground colour is rich nigger-brown, deeper in tint than that of
Celebes females. The under surface is similar to that of the males.

Allotype 2. As holotype, B.M. Type No. Rh. 18356.
Other examples in B.M. from TaLauTt: as holotype, 5 g. SANGIR: 64,1 8,

Anthene lycaenolus sp. n.
(Plea ig. 303 (Pl 2 fis..5 7)
Lycaenesthes lycaenolus Toxopeus MS.

From the material available, it is not possible to be sure of the true status and
affinities of this insect. The male genitalia are not to be differentiated from those
of philo. The exterior, although exhibiting most of the characters of philo, is
distinctive enough to suggest a subspecific separation, were it not for the fact that
the species in question is apparently already represented in Talaut by philo scintillans.
Possibly lycaenolus is just a form occurring together with sczntillans, but this seems
rather unlikely as no similar forms are to be found among the series from Sangir,
Celebes, or the Sulu Islands. The labelling of all specimens concerned is not very
detailed, and may possibly conceal the fact that lycaenolus actually does occur in
isolation ; in view of these considerations, it is deemed advisable to treat it as a
species until material with much more detailed data is available.

g. The upperside is dull mauve, without any indication of the iridescence so characteristic
of p. scintillans. On the hind wing, the submarginal black spots in areas 1 and 2 are vestigial
or absent. The underside is much paler in tone ; all the median markings are in lesser contrast
with the ground colour, being also less distinctly margined with white. The spots forming the
median series are much smaller, rounder, and usually distinctly separate. In contrast, the
submarginal spots on the hind wing are black, and stand out clearly, each on a clear near-white
area which is bounded outwardly by the dark marginal jine, and inwardly by the submarginal
lunule ; except for the tornal spot in area 2, which is inwardly accompanied by a large orange
red lunule of quite three times the extent of the similarly placed lunule in any of the philo
subspecies.

Q. The upperside is brown with a slight tawny tint, on which the submarginal spots of the
hind wing are more clearly defined than in that sex of scintillans, The fringes are dingy white,
and the underside is like that of the g.

268 G. E. TITE

24

25 26 27 28

Fics, 21-28. Odd numbers clasper, even numbers aedeagus: 21-22. Anthene emolus ;
23-24. A. philo; 25-26. A. seltuttus ; 27-28. A. paraffinis,

REVISION OF ORIENTAL ANTHENE 269

Holotype g. Tataut: (Doherty), B.M. Type No. Rh. 18357.

Other examples in B.M. (N.H.) from TaLauT: as holotype, 6 J, 2 9.

Anthene lycaenina (Felder)

This species can be recognized by the rather acute apex of the fore wing, the
definitely lavender hue of the upper surface in the male, and by the spot at the base
of area 7 on the under surface of the hind wing being in the great majority of indi-
viduals black and punctiform ; the last named character must be used with caution,
as in occasional examples this spot shows a transitional tendency towards the colour,
size, and shape of that obtaining in other species of the genus.

Corbet (1938a@: 250) in dealing with the subspecies mzya, states that lycaenina
is the Ceylon subspecies of /ycaenoides. Examination of the genitalia of the type
of lycaenoides proves beyond doubt that the two are not conspecific, and the name
lycaenina is the oldest name for the species under discussion.

(1) A. lycaenina lycaenina (Felder)
(Text-figs. 3, 17 and 18)

Lycaenesthes lycaenina Felder, 1868 : 281, Ceylon.
Lycaenesthes orissica Moore, 1884 : 23, Orissa.
Anthene lycaenina (Felder) Cantlie, 1963 : 67.

In B.M. (N.H.) from Ceylon and South India.

(2) A. lycaenina lycambes (Hewitson)

Lycaenesthes lycambes Hewitson, 1878 : 220, pl. 90, figs. 11-12, North India (Type !).
Anthene lycaenina lycambes (Hewitson) Cantlie, 1963 : 67.

In B.M. (N.H.) from NW. India, Sikkim, Assam, Burma, Laos, and Hainan.

(3) A. lycaenina miya (Fruhstorfer)

Lycaenesthes lycaenina miya Fruhstorfer, 1916 : 101, Borneo ; NE. Sumatra.
Anthene lycaenoides miya (Fruhstorfer) Corbet, 19384 : 249.
Anthene lycaenoides miya (Fruhstorfer) ; Corbet, 1956: 291.

In B.M. (N.H.) from Andamans, Siam, Malaya, Sumatra, and Borneo.

(4) A. lycaenina togata (Fruhstorfer)

Lycaenesthes lycaenina togata Fruhstorfer, 1916 : 101, Lombok.
Lycaenesthes lycaenoides bogorensis Toxopeus, 1929: 219, E. Java. syn. n.
Anthene lycaenina togata (Fruhstorfer) Corbet, 1938a: 250.

In B.M. (N.H.) from Java, Lombok and Sumbawa.

270 Gs; 3B; Ta TE

(5) A. lycaenina villosina (Fruhstorger) comb. n.
Lycaenesthes lycaenina villosina Fruhstorfer, 1923 : 898, Luzon.

Represented in the B.M. (N.H.) by three very ancient examples, all of which on
the underside of the hind wing have the spot in the base of area 7 lozenge-shaped and
coloured as are the other spots.

In B.M. (N.H.) from Luzon: 1 ¢ (ex Semper).

Anthene lycaenoides (Felder)

Both sexes can be readily distinguished by the unbroken median band on the
underside of the fore wing. The ¢ is similar in shape and colour to that sex of
lycaenina, but the @ differs by the presence of a white discal patch on both upper
and under side of the fore wing.

(1) A. lycaenoides lycaenoides (Felder)
(Text-fig. 1)

Dipsas lycaenoides Felder, 1860 : 454, Amboina, (Type !).
Anthene lycaenoides (Felder) Corbet, 1956: 457.

In B.M. (N.H.) from Amboina, Ceram, and Ceram Laut.

(2) A. lycaenoides pegobates (Holland) comb. n.
Lycaenesthes pegobates Holland, 1900: 71, Buru.
In B.M. (N.H.) from Buru, Halmaheira, Ternate and Obi.

(3) A. lycaenoides sutrana (Fruhstorfer) comb. n.

Lycaenesthes lycaenina sutrana Fruhstorfer, 1916: 102, Snow Mountains, Dutch New Guinea,
(Type !).
In B.M. (N.H.) from Key Islands, and New Guinea.

(4) A. lycaenoides godeffroyi (Semper) comb. n.
(Text-figs. 13-14)

Lycaenesthes godeffroyi Semper, 1879 : 165, Bowen.
Lycaenesthes turneri Miskin, 1890 : 39, Cape York.
Lycaenesthes turneri Miskin; Waterhouse, 1903: 199 [=L. godeffroyi Semper].

In B.M. (N.H.) from Queensland.

Anthene licates (Hewitson)

The formation of the aedeagus and claspers together with the superficial characters
given in the key serve to illustrate the specific separation of lycaenoides, licates and
villosa.

REVISION OF ORIENTAL ANTHENE 271

(1) A. licates licates (Hewitson)
(Text-figs. 4, 15 and 16)
Lycaenesthes licates Hewitson, 1874 : 350, Makassar, (Type !).
Lycaenesthes lycaenina licates Hewitson ; Fruhstorfer, 1916 : 101, (part.).
Anthene licates (Hewitson) Corbet, 1938a: 249.
In B.M. (N.H.) from CELEBES: Makasser, 1 g (Hewitson Coll.), (Type) ; Mine-
hassa, I g, 1897 (Semper Coll.).

(2) A. licates dusuntua Corbet
Anthene licates dusuntua Corbet, 1940: 40, Selangor, (Type!).

In B.M. (N.H.) from MaLaya: Dusuntua, Selangor, 2 3 (including type) (W. H.
Evans) ; Kuala Tahan, 1 ¢ (Evans) ; Ginting Sempak, I 4, v.1933 (Ex. F. M. S.
Mus.); Malakka, 1 3, 1904 (J. Waterstradt). SUMATRA: Lebong Tandai, W.
Sumatra, 4 g, 1923 (C. J. Brookes) ; NE. Sumatra, 2 3 (Dr. Martin).

(3) A. licates addenda (Fruhstorfer)
- Lycaenesthes lycaenina addenda Fruhstorfer, 1916 : to1, Palawan.
Anthene licates addenda (Fruhstorfer) Corbet, 1938a: 250.
There are no specimens of this subspecies in the B.M., but from the description
there can be little doubt that it is correctly placed here.

(4) A. licates philetas (Fruhstorfer), comb. n.
Lycaenesthes lycaenina philetas Fruhstorfer, 1916 : 102, Batjan.

Fruhstorfer’s mention of the basal white stripes of the fore wing underside suggests
that he was dealing with a race of Jicates.

In B.M. (N.H.) from Mo uccas: Laiwui, Obi, 1 ¢, ix.1897 (Doherty). NEw
GUINEA: Momi Coast (= Wariab), Arfak Pen., r g, 1928 (Dr. E. Mayr) ; Kapaur,
low c., I g, 1896-7 (Doherty).

Anthene villosa (Snellen) comb. n.
(Text-figs. 5, Ig and 20)
Pseudodipsas villosa Snellen, 1878 : 24, pl. 1, fig. 6.
Lycaenesthes lycaenina licates Hewitson ; Fruhstorfer, 1916: 101, (part.).

Fruhstorfer assumed from Snellen’s description and figure that villosa was a
synonym of licates. Comparison of Hewitson’s type and the Celebes material in
the B.M. makes it clear that vallosa is a distinct species, differing not only as stated
in the key, but showing also distinctive characters in the shape of the claspers and
of the aedeagus.

In B.M. (N.H.) from CELEBEAN REGION: Sangir, I 9, 1892 (Doherty) ; Talaut,
I §, 1892 (Doherty) ; Siao, r 3, 1896 (Doherty) ; South Celebes, 16 3, I Q (various) ;
Tawaya, N. of Palos Bay, 1 ¢ (Doherty).

As the next species was described as a Lycaenesthes it is included here.

272 G. E. TITE
Pseudodipsas rufimargo (Rothschild), comb. n.
(Pl. 1, fig. 40; Pl. 2, fig. 52. Text-figs. 2, 9 and Io)
Lycaenesthes rufimargo Rothschild, 1915 : 390, Vulcan Island, (Type !).

Examination of the unique holotype reveals that the species cannot be retained
in the genus Anthene, and that in fact it is closely allied to Pseudodipsas digglesit
Hewitson.! The characters substantiating this are: the eyes are naked; the palpi
are covered with compact scales giving a smooth effect, whereas those of Anthene
are hair-like and give a shaggy appearance; the terminal joints of the palpi are
shorter than those of Anthene; on the hind wing, veins 2, 3, and 5 are shorter ;
there are no indications of the tornal hair tufts so characteristic of Anthene.

The inclusion of this species in Pseudodipsas may be only a temporary expedient ;
both rufimargo and digglesii exhibit certain characters in common, differing from
those of the type species of Pseudodipsas (eone Felder) to a degree which might be
considered sufficient to justify generic separation therefrom.

Some difficulty has been encountered in attempting to reconcile the original
description with the characters exhibited by the type; therefore a more detailed
description of the underside is given below, omitting the points already mentioned.

Underside ¢. On the fore wing, the ground colour is pale grey-brown, having a slight satin-
like texture, quite unlike that of Anthene. The submarginal markings are almost identical
with those of digglesii, consisting of a series of fine dark lunules, the lunule in area 1 being
expanded into an irregularly shaped patch. The pattern in the disc consists of rounded spots,
of a deeper shade of the ground colour, each enclosed by a dingy white ring. These spots are
arranged in three transverse series as follows: there are two spots placed one above the other
in the cell; the discoidal spot is large, above it being two small obliquely placed elongate
spots in areas 7 and g respectively, and below it a tiny spot in the base of area 3, followed by a
larger one in the base of area 2; beyond these, is the median band consisting of 5 spots placed
in areas 2-6, those in areas 2 and 3 being nearer to the margin than are those in areas 4-6.

The hind wing is coloured like the fore wing; its margins are very like those of digglesit,
having dark submarginal lunules, each accompanied distally by an orange spot; these orange
spots are each edged outwardly with metallic silvery blue. The ends of veins 1 and 2 each
bear a black spot. In the disc, there are four series of rounded spots arranged thus: a basal
series, consisting of a spot near the costa, two conjoined spots in the cell, and a fourth below the
median vein; next, a curved series of irregularly shaped and partly coherent spots running
from below vein 1, through the cell, and terminating with an oval spot in area 7; then, a series
of three spots, the first at the base of area 2, the next at the discoidal, and the third in area 7
near the costa; lastly a series of 8 spots, forming a band and broken at veins 2 and 7.

The formation of the genitalia is very similar to that of digglesii, the main difference being
the absence of a pronounced saccus.

In B.M. (N.H.) from New Guinea: Vulcan Island, 1 g (holotype), 1913-4 (Meek).

REFERENCES

AvrIvILLius, C. 1908-25. In Seitz, The Macrolepidoptera of the World, 13, 613 pp., 80 pls.,
Stuttgart.

BETHUNE-BAKER, G. T. 1910. A Revision of the African species of the Lycaenesthes group
of the Lycaenidae. Tvans. ent. Soc. Lond. 1910: 1-84, pls. I-13.

1 P, digglesii (Text-figs. 11-12).

— ° ee ee ee ess

OE

REVISION OF ORTENTAL? ANTHE NE 273

Butter, A. G. 1899. Descriptions of new Species of the Genus Lycaenesthes. Ann. Mag.
nat. Hist. (7) 4: 341-344.

CANTLIE, SIR K. 1963. The Lycaenidae Portion (except the Arhopala Group) of Brigadier
Evans’ Identification of Indian Butterflies (revised). 160 pp., 6 pls. Bombay.

CorBET, A. S. 1938. <A Revision of the Malayan Species of the Nacaduba Group of Genera

(Lep. Lyc.). Tvans. ent. Soc. Lond. 87 : 125-146, 1 pl., 26 text-figs.

1938a. Observations on certain Species of Rhopalocera from the Malay Peninsula.

J. fed. Malay St. Mus. 18 : 242-262, 6 text-figs.

1940. Revisional Notes on Malayan Rhopalocera. Entomologist 73 : 39-43.

CorBET, A.S. (& PENDLEBURY, H.M.). 1956. The Butterflies of the Malay Peninsula. Edn. 2, pp.
I-537, pls. 1-55. Edinburgh.

DouBLEDAY, E. 1847. Lust of the Specimens of Lepidopterous Insects in the Collection of the
British Museum. Pt. II, pp. 1-57. London.

FELDER, C. 1860. Lepidopterorum Amboinensium. Sber. Akad. Wiss. Wien 40: 448-462.

FELDER, R. 1868. Diagnosen Neuer von E. Baron v. Ransonnet in Vorder-Indien gesam-
melter Lepidopteren. Verh. zool.-bot. Ges. Wien. 18 : 281-286.

FRUHSTORFER, H. 1908-27. In Seitz, The Macrolepidoptera of the World 9, 1197 pp., 177 pls.

Stuttgart.

1916. Uebersicht der indo-australischen Lycaenesthes-formen. Zodl. Meded. 2 : 96-102,

4 text-figs.

GoparT, J. B. 1824. Encyclopédie Méthodique 9: Pt. 2, pp. 329-828. Paris.

HEMMING, F. 1935. The Rhopalocera of Abyssinia by G. D. Hale Carpenter. Appendix 1.
Trans. R. ent. Soc. Lond. 1935 : 435-4306.

HEwIitson, W. C._ 1863-78. Illustrations of Diurnal Lepidoptera 275 pp., 107 pls. (includ.

Supplt.). London.

1874. Descriptions of new species of Lycaenidae, from his own collection. Tyvans. ent.

Soc. Lond. 1874 : 343-355.

Ho.iannp, W. J. 1900. The Lepidoptera of Buru. Pt. 1, Rhopalocera. Novit. Zool.
7: 54-85.

HopFFeErR, C. 1874. Beitrag zur Lepidopteren-Fauna von Celebes. Stettin ent. Ztg. 35:
17-47-

HUBNER, J. 1818-37. Zutrége Sammlung exotischer Schmettlinge [sic]. Augsburg.

KuHEIL, N. M. 1884. Die Rhopalocera der Insel Nias. 38 pp., 5 pls. Berlin.

Miskin, W. H. 1890. Descriptions of hitherto undescribed Australian Lepidoptera (Rhopalo-
cera). Proc. Linn. Soc. N.S. W. 5 : 29-43.

Moor_E, F. 1865. On the Lepidopterous Insects of Bengal. Proc. zool. Soc. Lond. 1865:

755-822, 3 pls.

1884. Descriptions of some new Asiatic Lepidoptera ; chiefly from specimens contained

in the Indian Museum, Calcutta. J. Asiat. Soc. Beng. 53: 16-52.

RGBER, J. 1886. Neue Tagschmetterlinge der Indo-Australischen Fauna. Dt. ent. Z. Iris
1: 45-72.

ROTHSCHILD, LorD. 1915. On the Lepidoptera in the Tring Museum sent by Mr. A. S. Meek
from the Admiralty Islands, Dampier, and Vulcan Islands (continued). Novit. Zool.
22 : 387-402.

SEMPER, G. 1879. Beitrag zur Rhopalocerenfauna von Australien. J. Mus. Godeffroy.
5 : 138-194, 2 pls.

SHERBORN, C. D., & Woopwarp, B. B. 1899. On the Dates of the ‘“‘ Encyclopédie Métho-

dique ”’: Additional Note. Proc. zool. Soc. Lond. 1899 : 595.

1906. On the Dates of Publication of the Natural History Portions of the “‘ Encyclopédie
Méthodique’”’. Ann. Mag. nat. Hist. (7) 17: 577-582.

SNELLEN, P. C. T. 1878. Opgave van en Aanteekeningen over Lepidoptera in Zud-West
Celebes Verzameld. Tijdschr. Ent. 21: 1-43, pl. 1.

STAUDINGER, O, 1889. Lepidopteren der Insel Palawan. Det. ent. Z., Iris 2: 3-180,

274 GG, 4 TELE

TitE, G.E. 1963. ASynonymic List of the genus Nacaduba and allied Genera (Lepidoptera:
Lycaenidae). Bull. Brit. Mus. (nat. Hist.), Ent., 13 (4): 67-119, Pls 1-2, 91 text-figs.

ToxorEus, L. J. 1929. De Riodinidae en Lycaenidae van het eiland Java. (Lycaenidae
Australasiae vi.). Tijdschr. Ent. 72 : 215-244.

vAN ErEckE, R. 1918. Studies on Indo-Australian Lepidoptera—III. Some Rhopalocera
and Netrocera (sic.) from Simalur, Pulu Lasia, Pulu Babi and Sumatra. Zodl. Meded.

4: 70-101, 2 pls., 3 text-figures.

WATERHOUSE, G. A. 1903. Notes on Australian Rhopalocera: Lycaenidae, Pt. iii. Proc,

Linn. Soc. N.S. W. 28 : 132-275, pls. 2-3.
WATERHOUSE, G. A., & TURNER, R. E. 1905.

Notes on Australian Rhopalocera : Lycaenidae,

pt. iv. Proc. Soc. N.S. W. 29 (1904) : 798-807.
WILEMAN, A. E. 1908. New and unrecorded species of Rhopalocera from Formosa. Anno-

tatnes zool. Jap. 6: 307-335.

INDEX

(Synonyms are shown in tfalics)

addenda (Fruhstorfer), Anthene licates ssp.,
271

affinis (Waterhouse & Turner), Anthene
seltuttus ssp., 262

amboinensis (Butler), Anthene seltuttus
ssp., 261

andamanicus (Fruhstorfer), Anthene emolus
Ssp., 259

ANTHENE Doubleday, 256

aruana Butler, Lycaenesthes, Anthene sel-
tuttus ssp., 262

balliston Hiibner, Lampides, Anthene emolus
syn., 257

bengalensis Moore, Lycaenesthes, Anthene
emolus syn., 257

bogorensis Toxopeus, Anthene lycaenina
togata syn., 269

dusuntia Corbet, Anthene licates ssp., 271

emoloides ssp. n., Anthene paraffinis ssp., 263
emolus (Godart), Anthene, 257

goberus (Fruhstorfer), Anthene emolus ssp.,
259

godeffroyi (Semper), Anthene lycaenoides
ssp., 270

javanus (Fruhstorfer), Anthene emolus ssp.,
259

keyensis ssp. n., Anthene seltuttus ssp., 261
klanga (Corbet), Nacaduba, Anthene emolus

syn., 257

leocrates Hewitson, Lycaenesthes, Anthene
philo syn., 266

licates (Hewitson), Anthene, 270, 271

LYCAENESTHES Moore, 256

lycaenina (Felder), Anthene, 269

lycaenoides (Felder), Anthene, 270

lycaenolus sp. n., Anthene, 267

lycambes (Hewitson), Anthene lycaenina
ssp., 269

matthias ssp. n., Anthene paraffinis ssp., 264

minor (van Eecke), Anthene emolus ssp., 259

miya (Fruhstorfer), Anthene lycaenina ssp.,
269

modesta (Staudinger), Anthene emolus ssp.,
259

modestus (Waterhouse), Lycaenesthes, Anth-
ene seltuttus, syn., 262

nereia ssp. n., Anthene paraffinis ssp., 266
nissani ssp. n., Anthene paraffinis ssp., 264

orvissica Moore, Lycaenesthes, Anthene lycae-
nina syn., 269

paraffinis (Fruhstorfer), Anthene, 263

pegobates (Holland), Anthene lycaenoides
ssp., 270

philetas (Fruhstorfer), Anthene licates ssp.,
271

philo (Hopffer), Anthene, 266

praeclara (Fruhstorfer), Anthene philo f., 266

PSEUDODIPSAS, 272

INDEX 275

rufimargo (Rothschild), Pseudodipsas, 272

scintillans ssp. n., Anthene philo ssp., 267

seltuttus (R6ber), Anthene, 259, 262

sutrana (Fruhstorfer), Anthene lycaenoides
ssp., 270

togata (Fruhstorfer), Anthene lycaenina ssp.,
269

turneri Miskin, Lycaenesthes, Anthene lycae-
noides godeffroyi syn., 270

villosa (Snellen), Anthene, 271

villosina (Fruhstorfer), Anthene lycaenina
ssp., 270

violacea (Butler), Anthene seltuttus ssp., 262

Fic.
Fia.
Fic.
Fic.
Fic.
FIG.
Fic.
Fic.
Fic.
FIG.
Fic.
Fic.

PLATE 1

(Uppersides)
Anthene paraffinis matthias, 3, B.M. Neg. No. 41099.
. paraffinis matthias, 2, B.M. Neg. No. 41101.
. paraffinis nissani, 3, B.M. Neg. No. 41105.
. paraffinis nissan, 9, B.M. Neg. No. 41103.
. pavaffinis nereia, 3, B.M. Neg. No. 41111.
. paraffinis nereia, 9, B.M. Neg. No. 41109.
. paraffinis emoloides, 3, B.M. Neg. No. 41113.
. paraffinis emoloides, 2, B.M. Neg. No. 41118.
. seltuttus seltuttus, 3, B.M. Neg. No. 41107.
. philo scintillans, 3, B.M. Neg. No. 41095.
Anthene lycaenolus, 3, B.M. Neg. No. 41097.
Pseudodipsas rufimargo, 3, B.M. Neg. No. 41093.

As AS AR AR AS AA Aa

Bull. B.M. (N.H.) Entom. 18, 8 PLATE 1

[EITIPITTTPTTT TTT TTT

‘wil 2

VIG, 44
VIG, 44.
VIG, 44
VIG, 4,
Vig, 47] /
WiG, 49,
VIG, 4
iG,
VIG, 1,
VIG, 94,

, PLAVH 4
(Undernden)

A, pavafinhs nea, 4, VOM, Sey, Mo, Atta,
SO tf BM. Nee, No, A480,

i
A, hyeMenitns, 4, HM, s / Na, Moy,
Pavudodipans vuprmurya, 4, Van, Seg, Me, MOa4,

2

PLATE

Bull. B.M. (N.H.) Entom, 18, 8

¥
°

Fl of ae i tes *
2 ee, ee ae eb a ee, La

Le

A LIST OF SUPPLEMENTS
TO THE ENTOMOLOGICAL SERIES
OF THE BULLETIN OF
THE BRITISH MUSEUM (NATURAL HISTORY)

MasneER, L. The types of Proctotrupoidea (Hymenoptera) in the British
Museum (Natural History) and in the Hope Department of Entomology, Oxford.

Pp. 143. February, 1965. 5.

. Nixon, G. E. J. A reclassification of the tribe Microgasterini (Hymenoptera :

Braconidae). Pp. 284; 348 Text-figures. August, 1965. {6.

. Watson, A. A revision of the Ethiopian Drepanidae (J-epidoptera). Pp. 177;

18 plates, 270 Text-figures. August, 1965. £4 4s.

. Sanps, W. A. A revision of the Termite Subfamily Nasutitermitinae (Isoptera,

Termitidae) from the Ethiopian Region. Pp. 172; 500 Text-figures. October,
1965. £3 5s.

. AHMAD, I. The Leptocorisinae (Heteroptera : Alydidae) of the World. Pp. 156;

475 Text-figures. November, 1965. {2 15s.

. OKADA, T. Diptera from Nepal. Cryptochaetidae, Diastatidae & Drosophilidae.

Pp. 129; 328 Text-figures. £3.

. GILIOMEE, J. H. Morphology and Taxonomy of Adult Males of the Family

Coccidae (Homoptera : Coccoidea). In press.

. FLETCHER, D. S. A revision of the Ethiopian species and a check list of the

world species of Cleora (Lepidoptera : Geometridae). In press.

PRINTED IN GREAT BRITAIN BY ADLARD & SON LIMITED, BARTHOLOMEW PRESS, DORKING

1 %
8 FEB 196,

% $
Spay Wo

REVISIONAL NOTES ON AFRICAN
CHARAXES
(LEPIDOPTERA : NYMPHALIDAE)
PART IV

V. G. L. van SOMEREN

BULLETIN OF
THE BRITISH MUSEUM (NATURAL HISTORY)
ENTOMOLOGY Vol. 18 No. 9
LONDON: 1967

ad

Ae
8 FEB I%

Z <
4 A
Cra Ww

REVISIONAL NOTES ON AFRICAN CHARAXES
(LEPIDOPTERA : NYMPHALIDAE)
PACT TV

BY.
V. G. L. van SOMEREN_ ,

4

P.O. Box 24947, Karen, Kenya

Pp. 277-316 ; 4 Maps, 9 Plates

BULLETIN OF
THE BRITISH MUSEUM (NATURAL HISTORY)
ENTOMOLOGY Vol. 18 No. 9
LONDON : 1967

THE BULLETIN OF THE BRITISH MUSEUM
(NATURAL HISTORY), tmstituted in 1949, 1s
issued in five series corresponding to the Departments
of the Museum, and an Historical serves.

Parts will appear at irregular intervals as they become
veady. Volumes will contain about three or four
hundred pages, and will not necessarily be completed
within one calendar ‘year.

In 1965 @ separate supplementary series of longer
papers was instituted, numbered serially for each
Department.

This paper is Vol. 18, No. 9 of the Entomological
series. The abbreviated titles of periodicals cited
follow those of the World List of Scientific Periodicals.

World List abbreviation: Bull. Br. Mus. nat.
Hist. Entomology.

© Trustees of the British Museum (Natural History) 1967

TRUSTEES OF
THE BRITISH MUSEUM (NATURAL HISTORY)

Issued 7 February, 1967 Price £2 2s.

REVISIONAL NOTES ON AFRICAN CHARAXES
(LEPIDOPTERA : NYMPHALIDAE) PART IV

By V. G. L. van SOMEREN

CONTENTS
Page
SYNOPSIS . : : : : i 27)
1. Charaxes ethalion BorsDuvaL AND ITS SUBSPECIES . : : 4. 270
Descriptions and Notes . , . : 5 : ; : 281
Systematic list : : : : . - 299
2. Charaxes ansorgei ROTHSCHILD AND ITS SUBSPECIES . : : a S08
Descriptions and Notes . : ‘ ; ‘ ; , om, 1303
Systematic list ; ; : : ‘ : .. 310
3. Charaxes pollux CRAMER AND ITS SUBSPECIES ; : ; ; eta 5
Descriptions and Notes . : ; : ‘ , , sp we SrT
Systematic list ; ; , ; , ; ; : a AES
ACKNOWLEDGEMENTS . : : : : ‘ : : ; re 8
REFERENCES. . : ' : : ; ; ‘ : : 316
APPENDIX . : ; ; : ; ; ‘ F : : r-. -316
SYNOPSIS

Three species of the genus Charaxes are dealt with and eleven new subspecies and five new
forms are described.

1. CHARAXES ETHALION BolsDUVAL AND ITS SUBSPECIES

WHEN we revised the Charaxes etheocles-viola-ethalion complex, (van Someren &
Jackson, 1952:8, 1957: 23) we referred very briefly to the species ethalion
Boisduval ; this was due to the paucity of material from many areas of its range.
This deficiency has, in part, been made good in recent years by the use of the modern
methods of “ trapping ’’ adopted by most collectors in South, Central and East
Africa. <A great deal of additional material has come to hand and has been made
available for study.

The species occurs from the eastern Cape Province north through Natal, Rhodesia,
and Zambia to Malawi, then through parts of Tanzania to south-west Uganda in the
west, and eastward up the east coast of Kenya almost reaching the southern border
of Somalia, then northward to the southern border of Ethiopia at Mt. Marsabit
to the east of Lake Rudolf. There are however some apparent breaks in the distri-
bution, but to what extent this is due to natural ecological barriers, or to lack of
collecting in these blank areas, it is difficult to determine. (Vide Map 1.)

The original description is based on a female from Natal; there is no mention of
the male sex. Rothschild & Jordan (1900), cite a male mentioned by Butler
(1896 : 360) as the male of ethalion. Butler (1895) however, associated this male
with his Charaxes etheocles rosae. Rothschild & Jordan (1900: 487) also treated

ENTOM. 18, 9. 20

280 V.G.L. VAN SOMEREN

vosae as a female of etheocles. We were fully satisfied on geographical grounds and
by an examination of the type, that it is a female of ethalion. This has been proved
by breeding: a female of form ethalion produced offspring of forms ethalion and
rosae and typical male ethalion.

Rothschild & Jordan (1900 : 479) give a full synonymy and refer to the variations
in the female f. e¢halion and figure two examples, pl. 12, fig. 4 being a broad-banded
variation and fig. 6 a blue-spotted and barred form to which Rothschild does not
give a form name; this specimen came from Taveta in Kenya. It resembles
swynnertont Poulton (1919: 79) from Rhodesia. These three authors failed to
recognize vosae as a form of ethalion.

van Someren & Jackson (1952: 260) enumerated the female forms which were
then recognized and indicated their distributions according to the material available.

The following alterations in status were made :

I. vosae Butler, 1895, originally described as a species but subsequently placed
to etheocles by Rothschild & Jordan, was reduced to a female form of ethalion ;
2. ethalionoides Carpenter, 1945, described as a form of etheocles was transferred
to a form of ethalion ;
3. the name vosa van Someren, 1932 became a synonym of vosae Butler since this
is now considered an ethalion (vide I ante).
van Someren & Jackson, 1952 : 260 added the female f. howard to the ethalion
group of the Chepalungu Forest, Sotik, Kenya. We also described ssp. alpinus
from the higher altitudes of the Vumba Mts., Rhodesia.

The material in the British Museum (Natural History), including the Joicey and
Rothschild Bequests, was rearranged according to this regrouping in 1956. With
additional material available, especially from Zambia, Malawi and Tanzania, it is
now possible to study the species in more detail.

If the specimens are arranged geographically, it is at once apparent that the males
differ hardly at all on the upperside throughout the range of distribution, that the
number of female forms varies in each area, and that the basic female forms are :
f. ethalion, f. swynnertont, and f. rosae, each somewhat modified within the distribu-
tion of the species from south to north.

A close study of the material seems to indicate that the species ethalion has not
evolved very far along the lines of subspeciation, though there would appear to be
some degree of limitation and segregation of female forms. There is evidence to
show that environment has played a part in modifying both size and pattern, thus
we note that ethalion of the dry thorn bush country is generally small and is thus an
ecological strain, not a subspecies in the true sense, for savanna merges into forest
either abruptly or gradually. Conversely, ethalion of the higher forests with con-
siderable precipitation tend to be larger and more brightly and intensely coloured.
However, complete isolation of the insect on Mt. Marsabit in the Northern Frontier
District of Kenya has resulted in the development of a race with only one known
female form with strong characteristics.

There are however other regional groups of ethalion which exhibit considerable
divergence from the nominate form yet are basically the same, and others which,

REVISIONAL NOTES ON AFRICAN CHARAXES 281

though having the three basic female forms, have in addition extreme modifications
of them. Some of these regions are well separated and ethalion within them consti-
tute a “‘ breeding community ’’, and one must assume that there is genetical control
over the female forms in each. However, this complexity, and often similarity of
forms, makes it difficult to divide the species into subspecies or geographical races
in the orthodox manner. The alternative is to recognize each “ breeding com-
munity ’’ asa genetical group. To facilitate the comparative study of this complex,
I have divided the species into geographical groups by regions, and each region into
areas. Each region has, to a greater or lesser degree, some isolation and a differing
ecological environment.

Emphasis has been placed on the fact that throughout the greater part of the
range of this species, the three basic forms of females, f. ethalion, f. rosae and
f. swynnertont, occur in part or whole, either similar to the typical form or somewhat
modified. To appreciate the variation in these forms to the best advantage, it is
advisable to compare each separately.

In this revision, the insect designated by van Someren & Jackson (1957 : 42) as
ethalion alpinus, from the Vumba Mts., Rhodesia, is omitted. I now treat it as a
species with a race inhabiting the Nyika Plateau, Malawi, and these will be dealt
with in another paper.

COMPARATIVE DESCRIPTIONS OF MEMBERS OF EACH GROUP

GROUP 1
Charaxes ethalion ethalion Boisduval
(Pl. x, figs. 1-2, Map 1)

Charaxes ethalion Boisduval, 1847 : 593 [9].
Charaxes vosae Butler, 1896 : 360 [g partim].

Mate. General shape of wings short and broad ; fore wing length 30 mm. (average) ; outer
border hardly at all incurved, in some specimens almost straight. Upperside. Forewing,
ground colour velvety black without sheen ; either immaculate or with a trace of a bluish spot
subcostal beyond end of cell, and with one or two obscure upper subapical spots ; no glaucous
border to the wing, but edge with a distinct white fringe hardly broken by the dark ends of the
veins, this white line very obvious in side lighting. Hind wing almost immaculate except for
traces of a submarginal row of small white dots, often limited to area above upper tails, usually
double at anal angle where the spots may be lilac with white central dot. Border of wing may
be slightly maroon above upper tail, or maroon mixed with olive, then olive to anal angle, where
the olive scaling is expanded ; extreme edge black with slight indication of white fringe. Tails
usually straight and slender, of about equal length, upper 5, lower6mm. Underside. Ground
colour dark ashy-grey-brown with strong satiny sheen varying with direction of light but
mostly in the disc of the fore wing and upper apical area ; a quadrate dark mark separates the
apical satiny area from the discal; the black marks on the basal half of the wing narrow but
clear ; the lower part of the discal line outwardly strongly black ; the tornal double black spot
well marked but not large, the black scaling extending half-way up the submarginal series of
spots. Hind wings ground colour as fore, with the discal satiny bar strongly marked; the
postdiscal zigzag line usually well marked, olive and maroon in colour, inwardly defined by a
black line ending above anal angle in a double or single crescentic black mark ; border reddish

282 V.G.L. VAN. SOMEREN

above upper tail then olive to anal angle which carries two black dots. It should be noted that
the ground colour alters slightly with age, fresh specimens are dark while older ones tend to be
more brownish (PI. 1, fig. 1).

REGION I. EASTERN SoutH AFRICA, NATAL, ZULULAND, MOZAMBIQUE, RHODESIA.

2 form ethalion Boisduval (PI. 1, fig. 2)

Fore wing length 40-42 mm., size comparatively large with outer margin almost straight.
Upperside. Fore wing, ground colour brown-black, darker in fresh specimens, with a strong
blue sheen in side light. No spot in the cell, as a rule ; discal spots beyond end of cell, upper
large, lower small or vestigal ; no spot in 4, but those from 3 to hind margin increasing in size
so that the bar is pyramidal, but inner edge almost straight, white in colour with slight bluish
scalingin 1b. The postdiscal spots white, arranged as follows : three subapical in line, spot in 4
set in, remaining spots in line with outer edge of wing, but spots tend to fade out in 2—1b.
Hind wing discal bar comparatively wide, extending from the costa to above anal angle where it
cuts across the lower part of the inner fold at right angles, disc of bar white or with slight bluish
tinge on borders. Submarginal line white or bluish, interrupted by dark veins ; marginal border
reddish above upper tails, then olive to anal angle; anal angle greenish with double black
dots ; extreme edge of wing blackish. Underside. Ground colour greyish brown with rufous
tinge at base and distal portion of fore wing; the light spots of upperside well represented,
strongly in the disc but less marked in postdiscal line, but both rows accentuated by black
proximally, though faint in postdiscal series; tornal double spots black with more diffuse
blackish spots in spaces above. Hind wing slightly more rufous in ground colour, but discal
whitish bar strongly marked, widest at costa then tapering to 1c and represented above anal
angle by a quadrate or triangular buffish mark ; postdiscal lunate marks accentuated in black
proximally, buffish in upper half then more rufous, terminating above anal angle in a double
black line with greenish in between ; submarginal line buffish to whitish, slightly outlined in
black above upper tail then represented by black triangular marks and a double dot at anal
angle ; admarginal line reddish above upper tails then olive to anal angle outlined distally with
narrow white lines; margin reddish brown. Tails long, upper one often spatulate at end,
of almost equal length, 7 mm.

Range : General in Region 1, and almost the only form found south of the Tugela
River (teste Pennington in litt.)

Variation 1. Differs from the nominate form by the larger white spots in fore wing from
hind margin to 2 in discal line ; the bolder postdiscal spots which are contiguous with, or merge
with the discal marks in 1b-2, the upper discal and postdiscal spots are often connected by
diffuse whitish rays. Hind wing discal patch considerably wider throughout, often white,
or with a slight bluish tinge on the borders. Underside drab greyish brown with slightly more
brownish at bases and on the distal portion of the fore wing. The white marks of upperside of
fore wing and the discal bar of hind wing distinct.

Range: Occurs sporadically in Region r.

Variation 2a. Represents a very slight modification of the nominate form in that the
upper spots of the fore wing are slightly tinged with ochre. The hind wing discal bar may be
slightly widened in area of the cell, being angled on the outer border.

Range : General in region I.

REVISIONAL NOTES ON AFRICAN CHARAXES 283

2b. @ form aurantimacula forma n. Similar to above but fore wing spots orange-ochre in
upper half but whitish in lower portion of rb and long mark in ra. Underside rather generally
rufous with upperside fore wing marks showing through and suffused over and indistinct, hind
wing bar completely obscured. Size rather small, fore wing length 34-36 mm.

Range: Most specimens come from the Natal area and have been taken in the
“dry season’’. This form has been bred, but intergrades toward variation 2a
and the nominate form are present in the same family.

2c. Similar to f. aurantimacula, equally small, fore wing length 35-36 mm. Upperside. The
fore wing discal bar very narrow to rb and only very slightly wider in 1a, the upper spots ochre,
the rest white. Postdiscal spots limited to three subapical, remainder to 1b vestigial. Hind
wing bar narrower than usual. Underside more rufous especially on distal portion of fore wing
and over the whole of the hind wing, so that the discal bar is hardly visible.

Range: Natal area, mostly in “ dry season ”’.

These three variants should not be confused with ethalion of the Port St. Johns
area which constitutes a distinct subspecies and is dealt with later.

2 form swynnertoni Poulton
(Plo 2. Fig. 70)

Charaxes ethalion 2 {. swynnertont Poulton, 1918: p. 79.

Attention was first drawn to this form by Rothschild & Jordan, rg00, when they
figured a specimen taken at Taveta, Kenya. The type selected by Poulton came
from Charinda, Melsetter, Rhodesia.

Size rather variable, majority with fore wing length 40-42 mm. Outer margin of fore wing
straight or only very slightly incurved. Upperside. Ground colour blue-black, rather browner
at bases of wings. Fore wing discal spots as follows : one medium and one very small beyond
end of cell, spots from 3-1b not expanding rapidly but streak in 1a long, extending distad.
Postdiscal series all of nearly equal size, three subapical in line, spot in 4 set well in, forming an
angle with one above, remainder following the curve of the wing margin; double spots in 1b
free or just touching the discal spot in this area. Subapical spots may be white, rest blue.
Hind wing discal band usually narrow at costa but expanding over cell area, then tapering to
above anal angle where it cuts across the inner fold at right angles, colour dull bluish, slightly
paler in the disc. Submarginal linear marks white or pale violet outlined distally in black ;
marginal border reddish above upper tail then olive to anal angle, extreme edge black. Tails
long and fine, of almost equal length, 6-7 mm. Underside. Grey-brown with a slight satiny
band in the disc of fore wing, varying with the angle of light, richer rufous on the outer margin
in some specimens. Hind wing with more rufous suffusion over whole area. The markings
of upperside not very apparent but postdiscal series show up slightly. The tornal double black
spot limited to 1b. In the hind wing the submarginal pale line shows up well and the zigzag
postdiscal line is usually strong, especially in the lower half.

It should be noted that in newly emerged specimens, the upper side ground colour is darker.

Range: This form occurs sparingly in upper Natal, rare south of the Tugela,
but more frequent in Rhodesia.

284 V.G.L. VAN SOMEREN

2 form rosae Butler
(Pl. 1, figs. 3, 4 and 7)
Charaxes rosae Butler, 1895 : 255.

A comparatively large form with fore wing length 38-40 mm. but size variable. Upperside.
Ground colour brownish black, more brownish in old specimens, with slight violet-blue sheen in
side light. Disc of fore wing with a curved white band formed as follows : a triangular mark at
end of cell, beyond three whitish streaks and some white scaling on costa ; a triangular mark at
base of 3 followed by elongate marks in 2-1b, rounded at distal end, concave on proximal end
followed by a narrow elongate mark in 1a on hind margin ; two oval white spots beyond end of
cell, lower one sometimes small or vestigial. Postdiscal series : three subapical in line, spot in
4 set in slightly, followed by a spot in line 3. Hind wing base and distal border brownish
black. Disc of wing crossed by a whitish band extending up to costa as a solid block and
reaching to above the anal angle and crossing the inner fold at right angles. The band is usually
white with a slight greenish blue flush over the cell area on inner border, and slightly on outer.
Underside. Ground colour ashy-grey-brown, rather more rusty on distal part of fore wing and
over the disc of the hind wing, but the pattern of upperside is reproduced in varying strength
and may be obscured in the hind wing. The submarginal pale line well marked ; marginal red
and olive border strong.

Range: This form is scarce south of the Tugela River, more plentiful in northern
Natal to Beira and Rhodesia. It is subject to some variation within the range of
the nominate race as follows :

Variation1. Natal. Differs from the nominate form by a slight reduction in the width of the
fore wing bar in 3, the absence of the second spot beyond the cell. In the hind wing the discal
patch does not reach the costa and the patch is suffused all over with violet-blue.

Variation 2. Melsetter area. Large specimens with upperside fore wing curved band very
wide even at the costa, the mark in 2 13 mm. long, those in ta—1b equally long with very slight
blue scaling proximad. Hind wing bar complete from costa to above anal angle, widest in 3-4,
white throughout or with some blue scaling on distal border. Underside. With the pattern of
upperside of fore and hind wings well represented.

Variation 3. Mt. Selinda. A small specimen with upperside fore wing curved bar tapering
rapidly toward the costa; only one spot beyond the cell; three subapical spots white, small
but distinct ; rest of postdiscal series only just visible. Hind wing band represented at costa
by a bluish discal spot and two postdiscal, the bar comparatively narrow and strongly flushed
over with lilac-blue. Underside. Only the fore wing curved bar represented ; the hind wing
discal patch not showing in the almost uniform rufous grey ground.

Variation 4. Vumba. Small specimen very similar to Mt. Selinda example but upperside
hind wing patch stops short at 4 and rather angled on outer aspect and represented at subcosta
by a detached spot in the discal line, the whole patch suffused with lilac, but slightly paler
in the upper proximal corner. Underside. Pattern of fore wing upperside reproduced clearly
with an extension of the white scaling to include the two white spots beyond the cellend. In
the hind wing the greyish brown ground colour hardly interrupted in the discal zone (PI. 1,

fig. 3).

REVISIONAL NOTES ON AFRICAN CHARAXES 285

GROUP 2.

PONDOLAND, Port St. JOHNS, BASHEE RIVER.

Charaxes ethalion pondoensis ssp. n.
(Pl. 1, figs. 5, 6, 8 and g, Map 1)

My attention was first directed to this race of ethalion in 1956, when it was noted
that the males differed considerably from those of the nominate race in shape and
underside coloration and the females exhibited characters which supported the
suggestion that the Port St. Johns insects represented a distinct subspecies.

The material then available was very limited, but I have now received a full
series of males and females from Messrs W. Teare and C. MacMaster which sub-
stantiate the validity of this supposition.

MaLeE. Fore wing length 32-33 mm., outer margin more incurved than nominate ethalion,
thus apex definitely acuminate. Upperside. Ground colour velvety black ; fore wing without
any blue marks or with small obscure subcostal mark beyond cell and one or two vestigial sub-
apical dots; margin without glaucous but white fringe strong. Hind wing with very small
bluish white submarginal dots, double at anal angle ; margin maroon above upper tail, then
olive to anal lobe. Tails short, often robust, of about equal length, lower slightly up-curved,
5mm. Underside. Has a strongly variegated appearance, the ground colour being more
greyish, slightly rusty tinged at base and outer border of fore wing, the satiny bars through
apex of cell, the discal line and postdiscal zone set off by intervening darker bars; the black
lines in basal half clear ; the tornal double black marks strong with black marks in rb strong.
Hind wing black lines in basal half not strong ; the discal satiny bar offset by a dark brownish
band proximal to the postdiscal zigzag line of olive and maroon, which is bordered by the
submarginal greyish line; border reddish above upper tail then more golden olive to anal
angle.

FEMALE. Fore wing length 35-36 mm., outer margin slightly incurved at 3-4. Upperside.
Ground colour black ; fore wing pattern strong, the spots being orange-ochre to orange-tawny ;
the discal band rather narrow made up as follows: one large and one small subcostal in 5-6,
a small spot set well in base 4 (sometimes absent), spots from 3-1b in increasing size followed by
long streak in ta. The marks in ta and rb slightly paler ochre. The postdiscal spots well
developed and complete from costa to 1b, orange-tawny in colour and completely separate
from the discal marks. Hind wing ground colour black, discal band rather restricted, extending
from the costa to short of the anal angle where it fades out, centre of band white, inner border
slightly blue, outer border strongly blue or mauvish blue. Submarginal line of spots not very
strong, mauve with central white dots; marginal border strongly reddish above upper tail,
then olive to anal angle, edged black. Tails comparatively short, 6 and 5mm. Underside.
General colour greyish brown, more brownish at base and darker on outer border of fore wing ;
the discal bar of upperside represented clearly in buffish white, the postdiscal spots in grey,
both accentuated proximally by black lines; tornal spots well marked, with dark marks in
areas above fading out. Hind wing ground colour slightly more brownish, but discal band
strongly indicated and bordered by brownish zones; postdiscal zigzag lunate marks greenish
with maroon distally and black line internally, strongly marked ; submarginal whitish line
strong, accentuated distally in black, with triangular blue-black marks toward the anal angle.
Marginal border reddish above upper tail then olive to anal angle.

Holotype male. Soutu Arrica: Pondoland, Port St. Johns (Transvaal Museum,
Pretoria).

286 V.G.L. VAN SOMEREN

Allotype female. SoutH ArricA: Pondoland, Port St. Johns, Bashee River
(Transvaal Museum, Pretoria).

Variation: Female similar to the nominate form but fore wing discal marks more restricted,
the upper spots pale ochre, the lower ones to 1b white ; the postdiscal spots fading out in 2 to
hind angle.

It is interesting to note that this strongly orange-spotted female is the only form
of female associated with the distinctive male in the Port St. Johns area of Pondo-
land. It bears some resemblance to the ochre-spotted variation of nominate
female ethalion referred to as aurantimacula on p. 283.

This subspecies flies in association with Charaxes karkloof Van Someren &
Jackson in the Port St. Johns area and the females have a superficial resemblance
to one another.

GROUP 3.

Charaxes ethalion nyasana ssp. n.
(Pl. 2, figs. r1-18, Pl. 3, 19-30, Map 1)

The species ethalion is represented in Malawi by a subspecies which, while having
females very similar to the three basic forms found in the nominate race, has in
addition several forms not occurring in any other race.

MALE: Fore wing length 30-31 mm., outer margin almost straight, thus similar to nominate
male ethalion. Upperside. Ground colour of both wings deep velvety black; fore wing
immaculate or with very obscure bluish subcostal spot either beyond end of cell or in upper sub-
apical area. Fringe narrowly white. Hind wing with submarginal spots obscured or apparent
in region of tails, bluish to lilac with white central dots ; border dull maroon above upper tail
then olive to anal angle. Tails short, 4 and 5 mm. long, black with median green line. Under-
side. Drab greyish brown, rather browner on hind wing. Fore wing satiny sheen rather
diffuse and not in distinct bars, but outer border darker brownish from tornal black spots,
which are not very strong, to just below the apex; black lines in basal half clear but fine.
Hind wing black lines in upper half fine; postdiscal zigzag olive and maroon line not strong
except in lower half ; border above upper tail red, then olive to anal angle.

Holotype male. MALAwi: Nkata Bay (J. D. Handman). In B.M. (N.H.).

FEMALES. The ethalion pattern forms in Malawi are, in the main, very similar to those of
Group I, but they are brighter; the ground colour is stronger and blacker and the amount of
blue-green scaling at the base of the wing more in evidence.

Region 1. MALAWwI.

FEMALES OF ETHALION PATTERN.

I. 9 form nyasana

A large form, fore wing length 40-42 mm. Upperside. Ground colour of fore wing blackish
with a strong greenish sheen at base. Pattern generally similar to nominate ethalion but
bolder ; discal upper spots larger ; there is a large diffuse whitish spot at end of cell and a spot

REVISIONAL NOTES ON AFRICAN CHARAXES 287

at base of 4; white marks in 1a—2 wider. Postdiscal spots bold and complete from the sub-
apex to 1b, where the mark merges with the discal mark. Hind wing discal bar wider, especially
in mid position, the white limited to an oblique central bar widest at costa, the borders suffused
with greenish blue, particularly basad. Submarginal line of whitish lilac linear marks rather
diffuse as a rule, double anal marks lilac. Marginal border dull reddish above upper tail,
then olive-green to anal angle. Underside. Greyish brown, satiny in disc but suffused with
brownish on distal part of fore wing; discal and postdiscal white marks of above strongly
represented in fore wing ; hind wing suffused with rufous over the whole, so that the discal bar
may be obscured or suppressed. Tails long and slender 7-6 mm. (PI. 2, fig. 15.)

Variation a. Upperside. Pattern basically similar to the preceding, but the discal and
postdiscal spots of the fore wing show an indication of being joined by rays.

Variation b. Somewhat similar to var. a. but fore wing spots other than those of 1a and 1b
(which are white) are strongly tinged with ochre.

2. 2 form

Upperside. Fore wing pattern as in ethalion ethalion but discal marks in 1ta—1b wider and
strongly blue-scaled at ends. Hind wing discal area strongly suffused all over with blue scaling
except toward the costa, thus obscuring the usual white bar in mid area. Underside. Whitish
marks other than those of 1a—1b suffused over with buffish grey, and obscuring the pattern.
This variation appears to indicate some transition toward f. swynnertoni. Some specimens may
have the fore wing discal and postdiscal spots in upper half tinged with ochre.

FEMALES OF SWYNNERTONI PATTERN.

The swynnertoni-like forms in Malawi divide up into four distinct forms, one of
which does not appear to occur elsewhere ; there are also some interesting variations
apparently restricted to Malawi.

1. 2 form cithaeronoides forma n.
(PL. 3 eS 25 620, 20,220).

On an average smaller than nominate swynnertoni, fore wing length 37 mm. (average).
Upperside. Ground colour is a deeper black with strong dark blue or greenish blue sheen in side
light. Fore wing pattern much as in the nominate swynnertoni, but the discal blue spots are
brighter blue ; there are two spots beyond the cell end, the upper large, lower smaller, both blue,
spots in 3-2 well separated, the spot in rb triangular so that the lower part of the bar expands
rapidly, widest in 1a. The spots in upper part of 1b, 2-3 smaller than in swynnertoni. In the
postdiscal row, the three upper subapical spots are conspicuously white and in line, spots in 3-4
set in at an angle to subapical, spots in 2-1b slightly larger and free of the discal marks ; these
spots are blue. Hind wing discal area bright blue, represented by costa by a spot in discal line,
central whitish bar just visible. Submarginal elongate marks well developed white or bluish
white ; marginal border reddish above upper tail then olive to anal angle ; extreme edge black.
Because of the brighter blue contrasting with the black surround, the patch shows up brilliantly ;
the insect thus bears a strong resemblance to a small fresh specimen of Charaxes cithaeron
nyasae. Underside. Ground colour grey-brown with a rusty flush on the distal part of fore
wing, and over the disc of the hind wing. The subapical spots of fore wing show up, but the
remainder of the upperside pattern is obscured ; the tornal black marks are very distinct and
accentuated by a paler surround. Hind wing black lines hardly visible, discal bar obscured,
but the submarginal pale line is strongly marked and entire; marginal border reddish above
upper tail, then olive to anal angle, with triangular black marks and a double spot at anal angle.

(Pl. 3, figs. 26, 29.)

288 V.G.L. VAN SOMEREN

Holotype female. MAtAaw1: Mlosa Stream, Lauderdale, 23.ix.62 (J. D. Hand-
man).

Paratypes : 3, same data as holotype.

2. 9 form

Upperside. Very similar to f. cithaeronoides but discal bar in fore wing more strongly de-
veloped, wider and whitish blue; spots in 2-3 larger. Postdiscal spots whiter throughout.
Hind wing discal patch whiter, less bright blue especially in upper half ; submarginal line not so
distinct, more greenish ; marginal border with red scaling extending to lower tail, but olive at
anal angle. Underside. Ground colour greyish brown, the discal marks of fore wing show
up buff or whitish, more especially in 1-3. Hind wing discal patch obscured. (Pl. 3, figs.
25:28.)

These specimens show an approach to nominate swynnertoni but at the same time suggest
some influence of an ethalion-like strain in genetical make up.

3. 9 form suppressa forma n.
(Pl. 3, figs. 20, 32)

Upperside. Resembling somewhat f. cithaeronoides, but base of fore wing with strong deep
blue sheen. The discal bar completely suppressed except for a blue spot beyond end of cell,
the long streak in 1a and a diffuse blue mark in 1b. The postdiscal spots are however very
distinct, white from subapex to 2, blue in 1b. Hind wing discal band broad, whitish, but with
strong blue sheen over all, or on distal border. Submarginal linear white marks well developed ;
marginal border reddish to upper tail, then olive green to anal angle. Underside. As in
cithaeronoides or with even darker brownish coloration. (PI. 3, figs. 20, 23.)

A very distinct form represented by several specimens.

Holotype female. Marawi: Port Herald, Mlosa stream, 26.vii.62 (J. D.
Handman).

4. 2° form demaculata forma n.
(Pl. 3, figs. 21, 24)

This boldly marked variation is the converse of form 3 as regards the pattern
of the fore wing ; it appears to be an unusual form.

Upperside. Ground colour as in form 1, but with a greenish blue sheen over the base of
the fore wing. Postdiscal spots limited to three subapical, rather indistinct, remainder com-
pletely obscured. The discal bar however is very pronounced and made up of two white spots
beyond the cell, an arrow-shaped spot in 3, a larger spot in 2 rather angular, mark in 1b broader
and at base as wide as streak in 1a, all pure white, but last two with strong greenish blue ends.
Hind wing discal band comparatively narrow, whitish in upper half toward costa, strongly
greenish blue on borders and over lower end, not extending through to inner fold but represented
above anal angle by a free bluish mark. Submarginal line white ; marginal border broadly red
above upper tail, then olive to anal angle. Underside. Ground colour strongly rufous grey,
more rusty on distal portion of fore wing; uniformly rufous shaded on hind wing. The fore
wing discal bar strongly represented in 1a—3 in buffish white; the hind wing band hardly
visible ; the postdiscal zigzag line obscured in upper half but strongly maroon opposite the
tails. Submarginal line present but most marked above tail area, where there are three black
marks ; marginal border reddish above upper tail, then olive to analangle. (PI. 3, figs. 21, 24.)

Holotype female. Matawr: Lugeri Stream, Mlanje, 2.ix.62 (J. D. Handman).

REVISIONAL NOTES ON AFRICAN CHARAXES 289

FEMALES OF ROSAE PATTERN.

There are two main vosae-like forms in Malawi, 1., a large form in which the fore
and hind wing light areas are not greatly in contrast, and 2., one in which the ground
colour is intensely blue-black with some blue sheen over the basal area of fore wing
the fore wing bar intensely white and in strong contrast to the strong blue of the
hind wing patch. There is also an interesting variation which I describe below.

rt. © form near rosae.

In size and general facies this form is reminiscent of rosae of the nominate race. Upperside.
Fore wing pattern is similar, the curved white bar extends into the upper part of the end of the
cell, the inner curve of the bar is regular, the outer more irregular, two white spots, the second
variable in size midway between the cell mark and the subapical spots of the postdiscal series
which consists of 5 white spots, the lower two forming an angle with the subapical three in line.
The long marks in 1a and 1b tinged with blue. The hind wing bar is fully represented at the
costa as a block, it varies in width, usually fairly even on the outer border but irregular on the
inner, widest in the region of the cell, may be whitish suffused with blue, with the oblique white
showing up clearly or the white line may be obscured ; the lower part usually stops short of the
inner fold above anal angle but is represented on the fold by a pale spot. The submarginal
broken line of linear marks and the marginal border as in the nominate vosae. Underside.
Rufous grey with slightly more rusty tinge on outer border of fore wing and over the disc of the
hind wing. The base of the fore wing with rather strong black lines, and the curved bar of above
strongly represented ; the hind wing patch is strong, while the postdiscal zigzag line of greenish
maroon is especially strong in lower half ; marginal border as usual.

Comparative fore wing lengths are: Chintechi 40 mm., Mlaye 40 mm., Monkey Bay 38 mm.
(P1..3; figs. 27, 30.)

Variation a. Upperside. Similar to the foregoing in the pattern of the fore wing but the
marks in 1b—2 also broader than usual ; the costal portion of bar broad ; the long streak in ta
blue with this colour extending up the inner edge to 1b, the postdiscal spots elongate and larger
than usual. The hind wing patch uniform pale blue and represented toward costa by two
separate spots. The submarginal line conspicuous, the marginal red and green border wider
than usual. A very ornate example, from Mlanje.

Variation b. Upperside. Fore wing curved bar almost uniform throughout, the white mark
in 2—-1b shortened at distal end, so that the postdiscal spots are increased to six, the lowest in 2
being just free of the discal mark. The hind wing patch does not extend to the costa and
besides being narrower than usual, is pale blue, without signs of an oblique white discal line.
The submarginal crescentic linear marks are almost contiguous; the marginal border well-
developed. Mlosa Stream, Mlanje.

Variation c. Upperside. Ground colour fore wing blacker, the curved white bar narrower
than usual, the mark in 2 being half its usual length so that the postdiscal spot in this area is
free. The two spots beyond end of cell and all the postdiscal series are strongly orange-ochre.
The hind wing band goes through to the costa and at the tapered anal end touches the mark
on the inner fold ; the band is suffused with mauvish blue but the oblique white discal bar is
clearly visible toward the costa. Submarginal line of somewhat triangular marks mauve, with
white central dots; marginal border conspicuously red above upper tail, then olive to anal
angle. Underside. Light brownish at bases of wings, black lines fine ; curved bar of above
distinct, and beyond this the wing is strongly satiny brownish grey but the margin of wing
strongly brownish ; the tornal spot large and black, with some brownish diffuse marks in spaces
above distal to the postdiscal spots, which are slightly indicated.

| This unusual specimen was taken on Kubula Hill, Blantyre (H. McKay).

290 V.G. L. VAN SOMEREN

2. 2 form imitans forma n.

Pattern of fore and hind wing conforming more or less to f. vosae but considerably
modified, the colour of fore and hind wings being in strong contrast.

Upperside. Fore wing length 36-38 mm. average 36mm. Ground colour when fresh, a
deep blue-black, with brighter blue sheen at base of fore wing in side light, distal portion black.
Discal curved bar more solidly pure white, less broken by black veins on outer border; the
costal portion of the bar is wide, since the subcostal streaks are extended distad to touch or
merge with the two white spots beyond the cell end, as in the type specimen ; the bar is con-
stricted in 4 and is widest in 2, the streak in Ia is not however longer than the mark in 1b.
In five specimens the postdiscal white spots are limited to the three subapical, the rest are
vestigial or absent ; (in only one specimen is the series complete to 3). The hind wing patch
is rather narrow, represented at the costa by a blue mark in discal line, the outer border is thus
curved, the inner margin is defined at the upper half, and diffuse in the lower, where it meets the
inner fold; the whole patch is strongly suffused with blue scaling, especially strong on the
margins, with high light in disc in side illumination but there is no clear white discal line.
Submarginal linear white marks small but clear, greeny tinged at anal angle ; marginal border
maroon and olive green. Tails rather long and slender, 7 mm., upper one slightly spatulate at
end, lower pointed. Underside. Ground colour darker than vosae, with a large satiny area in
disc of fore wing, offset by a darker brownish distal border. Fore wing discal curved bar
strongly marked but hind wing without indication of discal patch. Submarginal whitish to
greyish line strong. Tornal black mark in fore wing strong. (Pl. 2, figs. 16-18; Pl. 3, figs.
19; 22:)

Holotype female. Matawi1: Lauderdale Est., Mlaye, Mlosa Stream, 5.ix.62
(J. D. Handman).

Paratypes. 4, Lugeri Stream, Mlanje; Mlosa Stream, Mlanje; Hynde Dam,
Limbe ; Swazi Est., Mlanje (Handman & Martin).

Variation a. A small specimen with the costal portion of fore wing curved bar less wide, so
that whitish spot beyond the cell is free. The postdiscal spots are all more or less obscured.
In the hind wing the discal blue patch is very restricted to the mid and lower portions only,
and there is an obscure bluish spot on costa entirely free. Submarginal line of spots obscured
above upper tail but distinct in lower half ; marginal border as usual. Malawi, Mlosa Stream
(Handman).

Variation b. The fore wing curved bar is very much reduced and tapers toward the costa,
the two spots beyond the end of the cell entirely free, the distal portions of white marks in 1a—2
with black scaling obscuring the ends, the streak in 1a strongly blue. The hind wing discal blue
patch reduced in upper half and represented by a blue spot in discal line at costa and in space
below, the outer border is thus very angled. Chitakali Est., Mlanje (Martin).

The form imitans bears a resemblance to females of Charaxes cithaeron nyasae
van Someren.

REGION 2. ZAMBIA, MUMBWA-LUSAKA.
FEMALES OF NYASANA PATTERN.
i. Q form nyasana.

A small form, fore wing length 34-36 mm. In general appearance very similar to form 2
of Malawi, but smaller. Upperside. Discal and postdiscal spots in upper half of fore wing,

REVISIONAL NOTES ON AFRICAN CHARAXES 291

white, but those in 1a—1b suffused with blue scaling. Hind wing discal band rather wide for
size of insect, strongly suffused with blue on borders and toward the anal angle, but white central
bar usually visible. Submarginal line of spots not strongly indicated, but marginal border as
usual. Underside. Ground colour rusty greyish, slightly more rufous over hind wing. Fore
wing light marks apparent only in areas I1a—1b, and in the hind wing the discal bar is almost
obscured. (PI. 2, figs. 13, 14.)

This is a small savanna form, with a distinctly seasonal appearance.

2. 2 form.
A single example from this region is identical with the nyasana 2 form near
cithaeronoides of the Malawi series.

° form near imitans.

A small form bearing a close resemblance to the type form described above. Upperside. The
outer border of the curved bar in fore wing more irregular and with a strong blue suffusion on
lower marks in 1b—Ia especially distad. Hind wing band wider and whitish at costa but other-
wise blue. Submarginal linear mauvish white marks strong; border as usual. Underside.
Ground colour greyish brown, more rusty on border of fore wing ; discal curved bar fore wing
strongly represented, but hind wing band hardly indicated ; submarginal linear marks very
pronounced ; tornal black marks on fore wing strong.

REGION 3, SOUTHERN TANZANIA, NEWALA, NORTH OF RuvUMA R. AND NJOMBE.
FEMALES OF NYASANA PATTERN.
° form near imitans

Very similar to form 2 of the Malawi series with nyasana pattern, with upper part of upper-
side fore wing curved bar wide and touching the spot beyond end of cell, but lower portion
narrower, streak in 1a dull bluish. Hind wing patch not extending to the costa, but stop-
ping at 6, uniform bluish.

Underside. As in Malawi form 2, but upper part of fore wing bar goes through to the spots
beyond the end of cell. Basal black lines and tornal spot strong. Hind wing band not re-
presented below, but submarginal pale line strong.

Variation : An attractive variation has the postdiscal spots and those beyond the end of the
cell in the fore wing orange-ochreous. Tanzania: Rutumba Forest, Lindi area, 27.v.1965
(McCleery).

FEMALES OF SWYNNERTONI PATTERN.

rt. 2 form near cithaeronordes.

Upperside. Fore wing, lower blue spots in discal row obscured with blue scaling ; hind wing
band wide.

2. 9 form near swynnertont.

Upperside. Fore wing discal and postdiscal marks strong and well separated ; hind wing
band narrow, greenish blue. The associated male is very similar to that of ethalion nyasana.

3. 2 form near citheronotdes.

Upperside. Agreeing with f. cithaeronoides of the Malawi area but fore wing discal band less
abruptly expanding in areas 1a—1b and the streak in 1a not so long, the postdiscal double mark

292 V.G.L. VAN SOMEREN

in tb is free and not merging with discal mark. The postdiscal spots are distinct, the four
upper white, the rest bluish white. Hind wing discal patch wider and represented at the costa
by a quadrate off-white mark, borders strongly blue-scaled, the white oblique bar rather diffuse.
Submarginal white linear marks well separated and distinct in the majority of specimens ;
marginal border reddish above upper tail, olive-green to anal angle.

Underside. Generally less rufescent, more greyish than Malawi form 2; satiny areas in fore
wing wider with a reduction of the rufous shading on the distal margin. The black lines on the
basal half of the wing more clear but the upper side pattern is only slightly reproduced below.

Tanzania, Newala, north of Ruvuma River.

GROUP 4.

Charaxes ethalion nyanzae ssp. n.

(Pl. 4, figs. 31-35, Pl. 5, figs. 42, 45-48, Map 1)

Mate. Fore wing length 31-32 mm.; shape, short and broad; outer border of fore wing
hardly at all incurved. Upperside. Ground colour black, white fringe on edge rather con-
spicuous in side light ; no blue spots on subcosta or an obscure one just beyond end of cell anda
minute dot in subapex at subcosta and sometimes a trace of a second subapical spot, otherwise
wing immaculate. Hind wing black, almost immaculate except for a trace of submarginal
off-white spots in tail area ; marginal border with obscured reddish above upper tail and trace of
greenish to anal angle. Tails short, upper 4mm, lower 4mm. Underside. Drab grey-brown
with a satiny quadrate mark in cell, a satiny bar through disc and on postdiscal area and upper
part of apex, with intervening areas slightly darker brown. Basal black lines fine; post-
discal black line fairly distinct ; tornal black spot and submarginal dark marks not strong.
Hind wing ground colour as fore wing, black lines fine, disc with a slight satiny bar, postdiscal
zigzag line of olive and maroon lunules finely outlined proximally in black, fairly strong in
lower half; submarginal pale line faint ; marginal border narrow, reddish above upper tail,
then olive to anal angle with small black dots proximad.

Holotype male. S. W. UGanpDA: Kikagati, Kagera River. In B.M.(N.H.).

REGION I. KENYA, WEST OF THE RIFT, CHEPALUNGU-MARA AREA, EAST SIDE OF
LAKE VICTORIA.

Q ethalion pattern.

The only specimen now available for comment is a large one, fore wing length 38 mm,
Upperside. Fore wing, discal bar strong but rather narrow in lower portion, the mark in tb
is separated from the streak in 1a, and rounded on its distal end so that the postdiscal spot in
this area is free. Hind wing band not very broad, represented at the costa by an elongate free
mark ; the band widens over the area of the cell, then tapers to inner fold ; a discal white bar
is visible in the upper half but the band is suffused with blue in the lower portion. Submarginal
line rather indistinct except at anal angle; marginal border reddish then olive and well
developed.

This specimen resembles the ethalion pattern from the Kagera River in Region 2.

Lake Victoria, south and eastern side : Suna-Chepalungu-Sotik.

REVISIONAL NOTES ON AFRICAN CHARAXES 293

2 rosae pattern
2 form howard: van Someren & Jackson

Charaxes ethalion 2 f. howardi van Someren & Jackson, 1952 : 260.

This unique form has the vosae pattern, but all marks including the fore wing
curved band strong ochre-yellow ; hind wing band also ochre but paler toward the
costa (Pia, tgs. 32,35):

It will be noted that in this group, not only is the female of the ethalion pattern
wide-barred, but the form like vosae is also very wide-barred, as is also howardi, in
which the pattern is ocheous in colour, not white.

Kenya, west of Rift, Chepalungu.

REGION 2. S. W. UGANDA, KAGERA RIVER AND AREA SOUTH.

9 ethalion pattern.

A large form, fore wing length 37-41 mm. Upperside. Closely resembling some specimens
from northern Malawi, but fore wing spots in discal line in 2 and 3 more rounded and entirely
free ; mark in 1b quadrate, streak in 1a longer. Postdiscal spots bold, entirely free to 1b where
the spot touches or merges into the discal mark. Base of fore wing with a strong greenish sheen
in side light, the greenish blue extending to proximal ends of white marks in 1b and 1a. Hind
wing discal patch widest over cell area, generally broad and reaching costa, the discal white
bar visible but suffused over with bluish scales, especially over the lower portion. Submarginal
linear marks distinct, whitish or bluish, but may be rather obscured in upper portion ; marginal
border may be reddish above upper tail or suffused with greenish, greenish to anal angle.
Underside. Ground colour ashy-grey-brown, rather darker rusty to distal portion of fore wing ;
generally brownish on hind wing. Fore wing pattern of upperside not strongly represented,
except on 1b. Hind wing discal patch suffused over and indistinct though apparent below
costa. Black tornal mark fore wing variable in size and density.

Variation a. In one specimen there is a large white spot at base of 4 in fore wing ; otherwise
similar to above.

Variation b. The discal mark in rb is short so that the postdiscal spot in this area is quite
free. The hind wing band is narrower than usual; the submarginal line is obscured in the
upper half ; the marginal border normal, reddish above upper tail, then olive to anal angle.

Males associated with females of this group do not exhibit any outstanding characters by
which to distinguish them from nominate ethalion males. Fore wing length 31 mm., outer
border slightly concave. Ground colour velvety black ; two obscure subapical blue spots and a
trace of one beyond cell. Hind wing without submarginal dots; border barely indicated,
anal angle with double green spot. Underside. As in nominate race.

Kagera River, S. W. Uganda.

REGION 3. TANZANIA, EAST SIDE OF L. TANGANYIKA, KIGOMA AND KUNGWE AREAS.

2 ethalion pattern.

A single specimen has the following characters: Upperside. Fore wing pattern bold, in the
discal line the white mark in 3 is elongate, those in 1a—1b broad, suffused proximally with blue.
There is an additional spot at base of 4 in postdiscal row which goes to 3 and is here contiguous
with the discal spot insame area. In the hind wing the discal band is represented at the costa
by a single quadrate mark, but the patch widens rapidly and is very broad in the cell area, the
extension being basad ; the outer border is irregular, the whole is suffused with greenish blue

ENTOM. 18, 9. 21

294 vV.G.L. VAN SOMEREN

scales but mostly on the inner border. Submarginal line a series of strong white marks ; mar-
ginal border reddish and olive as usual.

The underside is unfortunately rather rubbed, the white pattern of upperside is well represent-
ed on both wings; the tornal spot of the fore wing is strong. (Pl. 4, figs. 34, 35.)

Eastern side Lake Tanganyika. Mpanda-Kungwe.

2 rosae pattern.

A very interesting specimen of this form was taken at Kungwe. Its characters are:
Upperside. Fore wing discal curved band very wide, the subcostal white scaling extending to
and contiguous with the two white spots beyond the cell end, the costa above also white, the
mark in 2 is extended distad and that in rb only a little less, while the streak in 1a is extended
proximad. The hind wing discal band, mostly white, extends up to the costa, uniformly wide
though irregularly dentate and sharply defined in mid area, then tapers gradually to above anal
angle where it crosses the inner fold. Submarginal white linear marks well developed ; marginal
border strongly marked reddish and green. Underside. Ground colour pale greyish brown, the
pattern of upperside well represented ; basal black marks strong, especially in sub-base in 1b ;
tornal black spot fore wing large, rounded and strong. (PI. 4, fig. 31.)

Lake Tanganyika, east side Kungwe.

REGION 4. E. KATANGA, LAKE MWERU AREA.

2 ethalion pattern.

A single specimen from near Lake Mweru seems to belong to this group. It exhibits interest-
ing features: the upper discal and postdiscal spots are large and there is an extra spot at base
of 4; the lower discal marks form a triangle, very broad at base and conjoined up to 3 in a solid
block, the postdiscal spot in 1b part of the block. The hind wing band is comparatively wide,
commencing at the costa and merging into the inner fold, mainly white with only a slight tinge
of blue on the borders. Fore wing length 35 mm. The specimen is unfortunately very worn.

Lake Mweru, Katanga-Zambian Border.

A small series recently received from Major Iain Grahame, from the Kigoma
area of Tanzania are illustrated on pl. 5, figs. 42, 45-48. The specimens are large
and some of the ethalion and rosae 2 forms are decidedly creamy in their light mark-
ings, thus showing a tendency toward the 2 form howard:.

Note however that no example of the vosae pattern has so far been recorded from
the region of the Kagera River in S.W. Uganda.

GROUP 5
Charaxes ethalion littoralis ssp. n.

(Pl. 4, figs. 37-39, Map 1)

Mate. Fore wing 33-34mm.; shape short, broad, outer margin fore wing very slightly
concave at 2-3. Upperside. Fore wing, ground colour deep velvety black, almost immaculate
but for a subcostal blue spot subapex and a trace of one in space below and sometimes an in-
dication of a spot beyond end of cell; fringe white, broken by dark veins. Hind wing velvety
black, greyer on inner fold. Submarginal spots very obscured, bluish in region of tails ; border

REVISIONAL NOTES ON AFRICAN CHARAXES 295

obscured, very faintly reddish above upper tail, dully greenish to analangle. Underside. Drab
greyish brown, strongly satiny with a large quadrate darker brownish mark in subapex, followed
by diffuse brownish marks in postdiscal row, submargin with row of dark contiguous marks,
but tornal mark not accentuated. Black lines distinct but fine. Hind wing ground colour as
fore wing, with a satiny bar in the discal line, bordered by a darker zone carrying the zigzag
series of olive and maroon lunules outlined in black proximally ; submarginal row of greyish
lilac marks triangular in region of tails, accentuated with black distally ; marginal border
maroon to upper tail, then olive to anal angle. Black lines in base of wing very thin. Tails
short, upper 4 mm., lower 5 mm.

The above description is based on specimens from Shimba Hills. Examples from Amani,
Usambara, are slightly larger, slightly more incurved on outer margin of fore wing, with the
apex thus more pointed. The fore wing may be immaculate or with only one large blue spot
beyond end of cell. Hind wing as in Shimba examples; tails of about equal length, 5 mm.
Males from the savanna bush country of Kibwezi-Voi-Taveta are smaller.

Holotype male. KErNyA: Shimba Hills, Kwale Dist., x. 1960 (van Someren).
In B.M.(N.H.).

REGION I. KENYA COAST BELT AND HINTERLAND TO TEITA HILLS, KILIMANJARO,
DAR-ES-SALAAM, AMANI.

Q ethalion pattern.

A large insect, fore wing length 40-41 mm., resembling somewhat the ethalion-like female
of the Malawi series. Upperside. There is often a spot or an indication of one in the fore wing
cell; there is usually a small white spot at base of 4. The discal marks are large, even larger
than in the Malawi form, the mark in 3 is rounded or elongate and entirely free, but the other
marks, increasing in size, are contiguous, longest at the streak in 1a on hind margin. The upper
spots are white, but those in the lower portion are bluish white or strongly blue. The postdiscal
spots are well developed and free to 2, then touching or merging with mark in 1b of discal row.
The upper discal and postdiscal spots are rarely connected by rays. The hind wing discal patch
is broad, as wide as or wider than the streak in fore wing 1a, at the costa, and in 6 but tapering
toward anal angle and crossing the fold as a quadrate mark ; a white discal bar is well marked
but the patch is shaded on either border with strong blue. The submarginal line is obscured in
the upper half though indicated by some white scaling, in the lower half violet, accentuated by
black dots. Tails long, slender, 5-6 mm. long. Underside. Ground colour dark greyish with
just a slight brownish tinge basad, and on the discal line but mof with the silvery discal area
seen in Chavaxes contrarius Weymer, which this form resembles. Light spots and bars of
upperside strongly represented, the postdiscal lunate marks on hind wing greenish and maroon,
especially strongly marked in lower half of zigzag line.

2 rosae pattern.

Form a. Size rather variable, fore wing length average 36mm. Upferside. Very similar
to vosae of nominate race but differing as follows: the curved white bar in fore wing tapers
rapidly toward the costa but does not reach it ; the two spots beyond the cell are elongate and
entirely free ; the postdiscal spots are elongate and extend as far as 3; at this point the discal
marks are of equal width to the hind margin, the outer border dentate as the marks are angled on
the outer ends. Hind wing band rather narrow, but goes through to the costa, and is mostly
white with slight blue scaling along inner border. Submarginal linear marks hardly visible ;
marginal border normal. Underside. Ground colour pale drab greyish brown; the fore and
hind wing patterns of above clearly indicated. Tornal black mark of fore wing small.

Form b. Fore wing curved bar narrower, of almost uniform width throughout; lower
discal marks slightly blue on proximal ends, otherwise white ; postdiscal smaller and rounder,

296 V.G.L. VAN SOMEREN

reaching to 2 but here joined to discal mark by some white scaling. Only one spot beyond the
cell end. Hind wing patch not reaching the costa, though here represented by a dull bluish
mark ; band comparatively wide, dull blue all over with an indication of darker obscure marks
in discal line ; submarginal white linear marks strong; borderasnormal. Underside. Slightly
more brownish than Form a, especially on hind wing. Fore wing white curved bar well
represented but postdiscal spots obscured ; hind wing without indication of discal band, but
postdiscal zigzag olive-red line strong. Tornal black spot on fore wing small.

In spite of intensive trapping in the coastal region of Kenya, no examples of form
swynnertont have ever been taken.

REGION 2. KIBWEZI-VOI-TAVETA BUSH COUNTRY.

2 ethalion pattern.

Specimens from the dry thorn bush country are placed to the coastal race, though they are in
fact a small ecological form. The fore wing length varies from 33-34 mm. They are character-
ized on the upperside by the more restricted discal marks in the fore wing, less developed post-
discal spots and usually narrower hind wing band, which is less brightly blue on the borders.
The underside is less greyish and more shaded with light rufous. The upperside pattern of
fore wing clearly reproduced, but the hind wing discal band tapers out toward the anal angle,
seldom crossing the inner fold.

2 swynnertont pattern.

The only record of this form that I know, from the thorn-bush area is that men-
tioned and figured by Rothschild & Jordan, 1900, (pl. 12, fig. 6.)
Other forms in this region are f. nr ethalion and f. nr rosae.

REGION 3. COAST REGION OF TANZANIA, DAR-ES-SALAAM, USAMBARA (AMANI)
INLAND : MoroGoro, ARUSHA.

2 swynnertont pattern.

Specimens from this intermediate area are somewhat like those from Newala (Group 3,
Region 3), but are larger. Upperside. The discal bar of the fore wing generally wider, the
upper marks white, while the lower are bluish, or all marks may be blue ; marks in 1ta—1b are
expanded in both directions. The postdiscal series of spots are white in the upper half, blue
in the lower, with the double spot in 1b free or contiguous with the discal spot. Hind wing
patch slightly narrower than the Newala specimens, but variable ; marginal and submarginal
lines similar. Underside. Ground colour more rufescent drab, the pattern of upperside
obscured as a rule, but two specimens from Dar-es-Salaam exhibit this pattern clearly.

These Tanzanian specimens do not really belong to the Kenya Coast aggregate
but are in fact a cline toward the Malawi group, but material from the intervening
country is not available for study.

GROUP 6
Charaxes ethalion kikuyuensis ssp. n.

(Pl. 4, fig. 36)

Mate. Fore wing length 32-33 mm.; shape short and broad. Fore wing outer margin
almost straight or very slightly incurved in 3. Upperside. Ground colour velvety black ;

REVISIONAL NOTES ON AFRICAN CHARAXES 297

fore wing immaculate or with one subapical subcostal spot ; marginal white fringe very distinct
in side view. Hind wing ground colour black, immaculate except for submarginal row of lilac,
white centred spots, faint in upper part but very distinct in region of tails and double at anal
angle. Marginal border narrow but distinct, reddish above upper tail, then olive to anal angle.
Underside. Generally dark brownish, relieved by a slight satiny bar in discal line and in
postdiscal zone to upper part of apex, bordered internally by fine black line expanding in 1b,
and externally by a submarginal line of dark spots terminating in the large black tornal spot.
Hind wing ground colour as fore, with a suggestion of a satiny discal bar ; the postdiscal zigzag
line of olive and maroon lunules, accentuated in black proximally, very distinct from costa to
above anal angle, where there is a pale quadrate mark above on inner fold and a black mark
below ; submarginal pale line accentuated in black distally complete and clear ; marginal border
reddish above upper tail, olive to anal angle.

Holotype male. KrENyA: Nairobi district, Karura Forest, i. 1949 (E. Pinhey).
In B.M.(N.H.).
Range: Kenya Highlands, Nairobi, Meru, Mt. Kenya, Njombeni.

2 ethalion pattern.

Slightly larger than the savanna form, fore wing length 35-37 mm. Upperside. Rather vari-
able, specially in regard to the hind wing patch. Fore wing discal and postdiscal spots white
and well developed ; a small spot often present at base of 4 ; discal spots in 2—3 well separated,
but marks on rb-tra increasingly large, the streak in 1a often extending proximad so that the
inner border of the discal bar is concave. The postdiscal spots are entirely free up to 1b,
though the spot here may merge with the discal mark. Hind wing discal patch wide, extending
through to the costa, widening slightly in the cell area and tapering gradually to the inner fold,
which is crossed above the analangle. This patch is often pure white or with just a slight bluish
tinge on its borders. Submarginal linear marks white and well developed ;_ marginal border
dull red and olive, strong. Tails slender, upper one slightly spatulate at end, lower pointed,
7-6mm. long. Underside. Greyish brown, slightly more rusty on hind wing ; light markings
of upperside strongly represented, those of hind wing divided into discal and postdiscal bars ;
submarginal whitish line strong, accentuated in region of tails by black marks.

Variation a. Differs only in that the discal bar in rb is less wide, so that the postdiscal spot
in this area is free. The hind wing band is narrower and more bluish on borders, so that the
oblique white discal zone is clearly visible.

Variation b. Slightly larger, with fore wing discal and postdiscal spots connected by diffuse
rays. Hind wing band narrow from costa to cell area, shaded blue only on the proximal side
where it reaches the inner fold.

2 swynnertont pattern.

The swynnertoni-like form occurs sparingly in the Nairobi area and upper Kikuyu
and Meru, but does not appear to go west of the Rift Valley.

Upperside. Ground colour black, slightly brownish at base of fore wing, which has a slight
blue sheen in side light. Fore wing blue discal spots comparatively large in upper half, though
second spot beyond the cell may be vestigial ; spots in 3—2 large and free, the spot in 1b not
much larger, may be free or fused with longer mark in 1a, the lower part of the discal bar is thus
not expanded to any extent. The postdiscal spots are whitish in the subapex, the rest bluish,
the double spot in rb is free. Hind wing patch not reaching to the costa as a rule, but there
may be a diffuse subcostal mark in the discal line ; the patch is relatively wide, whitish in the
disc but strongly violet-bluish on the borders. Submarginal linear marks whitish or mauve ;
the marginal border as usual, reddish above upper tail then olive to anal angle. Underside.

298 V.G.L. VAN SOMEREN

Ground colour greyish brown, slightly stronger brown to outer margin of fore wing. The
upperside pattern clearly indicated on the fore wing, tornal black spot well marked but discal
band on hind wing less apparent, submarginal pale line strong ; marginal border clear.

Q rosae pattern.

This form is represented in the Kenya Highlands by rather large specimens, fore wing length
38-40 mm. Upperside. Fore wing ground colour rather brownish at the base but blacker on
distal half. The curved white band is uniformly wide, the costal and subcostal narrow white
marks clear, the two marks beyond the end of the cell large and oval; discal mark in 2 longest,
with obliquely cut ends ; marks in 1b—1a of equal length, rectangular, 10 mm. wide. Postdiscal
white spots clear, those in subapex very white, that in 3 rather obscured. Little or no blue
scaling on lower discal marks. Hind wing discal band well developed and extending through to
the costa by a rectangular mark ; the band widens in the cell area, then tapers toward the inner
fold and crosses it above the anal angle ; band generally white with some bluish violet scaling
on proximal border. Submarginal linear marks obscured and greenish ; marginal border well
marked. Underside. Drab greyish brown on fore wing, browner on hind wing. The curved
discal bar well represented below, but postdiscal marks obscured. Tornal black spots strong.

On the hind wing the discal band is moderately distinct. The postdiscal zigzag line not very
strong. The submarginal line present but more strongly marked above in the region of the tails,
accentuated distally by black spots and crescentic greenish marks above anal angle.

Marginal border reddish and olive as usual.

GROUP 7

Charaxes ethalion marsabitensis ssp. n.

(Pl. 5, figs. 40, 41, 43, 44, Map 1)

Matz, Fore wing length 35 mm. outer margin slightly incurved. Upperside. Fore wing,
ground colour velvety black, almost immaculate but for one minute dot subcostal beyond cell,
and one or two subapical blue dots. Marginal white fringe conspicuous when viewed from side.
Hind wing velvety black, dull black at inner fold. Submarginal row of lilac spots obscured in
upper half but clear in region of tails ; marginal border dull red to between tails, then green to
analangle. Tails 5-6 mm. long, lower slightly up-curved, upper with reddish central line, lower
slightly reddish or ochreous. Underside. Very similar to the Kenya Highland race kikuyuensis,
brownish on fore wing with satiny patches beyond the cell and upper part of apex with darker
brown patches between and on the border; black lines in basal half of wing fine, postdiscal
lines faint except for black mark in 1b, submarginal dark marks indistinct but strong in tornus.
Hind wing ground colour more rusty brownish but with satiny flush in side light in discal line ;
postdiscal zigzag line not bold except in region of tails ; submarginal linear lilac marks obscured
in upper half, stronger at tails and accentuated with black angles. Marginal border red, in-
cluding upper tail, then mixed red and olive to anal angle.

Holotype male. KENyA: Northern Frontier Province, at Mt. Marsabit, iii. 1946.
In B.M.(N.H.).
Range: Mt. Marsabit, Northern Frontier Province, Kenya.

2 swynnertont pattern

FEMALE. The only female form taken in the Marsabit area, in spite of intensive collecting,
has been of the swynnertoni pattern. Fore wing length in a majority of a long series, 42 mm.
(a few smaller examples may occur in drought periods). Upperside. Fore wing, ground
colour brownish black, slightly darker in the disc. Fresh specimens blacker. Discal and post-

REVISIONAL NOTES ON AFRICAN CHARAXES 299

discal spots white, above area 1b large and bold ; discal spots in 2-3 rounded and usually free ;
postdiscal spots in these areas angular, double spot in 1b free or may be slightly joined to discal
mark by some violet-blue scaling, the mark in 1a—1b are widest and violet-blue in colour.
Hind wing discal patch broad, carried up to the costa as two separate spots in discal and post-
discal lines, the intervening area blackish. The patch is generally suffused all over with
violet-blue scaling but there is an indication of a paler central zone, especially toward the inner
fold ; the outer border of the patch irregular. Submarginal linear mauvish white line, blackish
edged distally, is strongly indicated ; marginal border, strongly reddish above upper tail or
between tails, is olive to the anal angle. Tails long, upper one rounded at end, lower pointed,
10-9 mm. long, the upper tail with red central line, lower with pale ochre line. Underside.
Ground colour greyish brown, greyer in the disc of fore wing but strong rufous on distal border ;
the upperside pattern clearly indicated in discal line and upper part of postdiscal series. Hind
wing generally suffused with rusty bloom, the discal band only slightly indicated ; submarginal
line strongly marked ; marginal border strongly reddish then green to anal angle.

BIOLOGICAL NOTE.

The following races of Charaxes ethalion have been bred:

ethalion ethalion, which lays on Albizzia sassa (fastigiata Oliv.) and Albizzia charin-
densis (teste Platt.) (Mimosaceae).

ethalion littoralis, which lays on Parkia filicoidea, Albizzia adianthifolia, A. sassa,
A. coriaria, Newtonta (Piptadenia) buchanan, Tamarindus indicus (Mimosaceae),
and Scutia myrtina (Rhamnaceae).

ethalion kikwyuensis, which lays on Albizzia sassa, A. coriaria, A. gummtfera
(Mimosaceae) and Scutia myrtina (Rhamnaceae).

SYSTEMATIC LIST
Charaxes ethalion Boisduval

With an indication of the female forms within Groups and Regions.

Group I
Charaxes ethalion ethalion Boisduval, 1847. Type locality: “ Natal”’.
2 f. ethalion Boisduval, 1847. Type locality: Natal.

et vars.

2 f. swynnertoni Poulton, 1919. Type locality : Mt. Selinda, Rhodesia.
et vars.

2 f. rosae Butler, 1845. Type locality : Natal.
et vars.

2 f. aurantimacula forma n.
Range: From Durban area north to Beira, extending inland to
Rhodesia.
Group 2
Charaxes ethalion pondoensis ssp.n. Type locality: Port St. Johns.

9 ethalion pattern.
Range: Pondoland, Port St. Johns. Transkei, Bashee River,
Embuyti Forest.

300 V.G.L. VAN SOMEREN

Group 3
Charaxes ethalion nyasana ssp.n. Type locality: Malawi and Nkata Bay.

2 f. ethalion pattern = nyasana forma n.
2 f. swynnertoni pattern.
° f. cithaeronoides forma n.
2 f. suppressa forma n.
2 f. demaculata forma n.
2 f. rosae pattern imitans forma n.
et vars.
Region 1, Malawi.
Region 2, Zambia, western block.
Region 3, Tanzania, southern highlands and Ruvuma River area.

Group 4.
Charaxes ethalion nyanzae ssp.n. Type locality: S. W. Uganda, Kikagati and
Kagera River.
Region 1, 2 ethalion pattern.
2 rosae pattern = howardi van Someren & Jackson. Kenya,
west of Rift, Chepalungu-Mara area, east side Lake
Victoria.
Region 2, 2 ethalion pattern.
Uganda, S. W. Kagera River and area south.
Region 3, 2 ethalion pattern.
2 rosae pattern.
Tanzania: eastern side L. Tanganyika, Kigoma and
Kungwe areas.
Region 4, 2 ethalion pattern.
E. Katanga, Lake Mweru area.
Range: S. W. Uganda, Kagera River; Kenya, E. side L. Victoria ;
Tanzania, Kigoma, Kungwe.

Group 5
Charaxes ethalion littoralis ssp.n. Type locality: Shimba Hills.
Region I, 2 ethalion pattern, et vars. ethalionoides Carpenter.
2 rosae pattern, et vars.
Kenya Coast belt and adjacent hinterland ; Tanga and
? Zanzibar.
Region 2. 2 ethalion pattern.
2 rosae pattern.

2 swynnertont pattern.
Kenya, Dry bush country: Voi, Kibwezi and southern

Ukambani.

REVISIONAL NOTES ON AFRICAN CHARAXES 301

Intergrading cline between Groups 3 and 5

Q: ethalion pattern, swynnertoni pattern and _ rosae
pattern.
The area embracing the central Tanzania Highlands,
Morogoro, Dodoma; north to Usambara, western
Kilimanjaro, Mt. Meru and Arusha.

Group 6
Charaxes ethalion kikuyuensis ssp.n. Type locality : Nairobi.

2 ethalion pattern, swynnertont pattern and rosae pattern.
Range: Kenya Highlands east of the Rift : Ngong, Nairobi, Kikuyu,
to Nyeri and Meru Mt. Kenya.

Group 7
Charaxes ethalion marsabitensis ssp.n. Type locality: Mt. Marsabit.

2, swynnertont pattern only.
Range: Mt. Marsabit, Northern Frontier district Kenya.

CHARAXES ANSORGEI RoruscuI_p,
AND ITS SUBSPECIES

The species Charaxes ansorget can be divided into two groups, based on the
character of the hind wing band: (a) a northern group in which this band is en-
tirely white with slight blue scaling, and (b) a southern group in which the band is
divided into two sections, tawny orange in the upper half and bluish white in the
lower.

The species appears to have a somewhat restricted distribution (Maps 2, 3) and is
found on the high ground in the regions of the Eastern Rift and the Western Rift
in Eastern Africa. It is limited to montane evergreen forest at elevations of
6000-9000 feet and is closely associated with its food plant Bersama abyssinica
et ssp. (Melanthaceae) which may occur in the forest, on the forest margins or as
saplings and shoots from roots left in de-afforested land. Females may thus be
noted outside the forest when ovipositing on these young plants, which they seem to
prefer to the old trees.

Because of its restricted habitat, the species exhibits a strong tendency to split
up into well defined localized races or subspecies, most of which are readily recog-
nizable. Poulton (1933) reviewed the species and races known up to that time,
and gave a very full comparative description of the subspecies and their respective
distributions. One or two aggregates were however “left in the air’’ and undecided
as to their relationship with adjoining defined races. Vide also Poulton (1932 : 6-9).

A large amount of additional material has come to hand of recent years, some from
outside the range of the species as cited by Poulton and three new subspecies are
described. I have also taken the opportunity to re-examine the doubtful specimens
mentioned by Poulton in 1932.

302 vV.G.L. VAN SOMEREN

SUDAN i
‘
Th ote : BIDINGA = == Cie Se ETHIOPIA <*
x , a’ ¢
en bats ~)- ACHOLI 7 ass ae)
AS eat ~\ 7 ,
. eal ss \ Sen F a
IMATONG MTS. :aiaces 2
\ i
UGANDA \ \
\ {
L. ALBERT \ KENYA '
oe se | sOM-
>
MT. ELGON Mig 2% ALIA
é @. me ie \
| ee 3
CONGO / G) RUWENZORI x ® wf I
7 ‘re, i Se \
i @:
L. EDWARD ? z (
2 >
@ te \
7 sa

5 ‘ 7 asa is x2 2
One ‘ee.
_ i sg
ANDA - ree
L. KIVU ass ans i ore
Veyee ;, 3 ))
ae Were | » Z *
< Gh usamaura 1, oon
& a 7 MT. MERU
& URUNDI Pag Oo 6 : MOMBASA
/ Z ig, \
if > "tg \
4 = roe 5 ~
TANZANIA is vil
TANGANYIKA) 3 “%,
L. TANGANYIKA ( a | wns $ PEMBA
a hn, o “4, € USAMBARA
Hee has

Vv

7
4
WV xuncwe
c

L. RUKWA pia,

%,

2

we
i
s,
att
Me ven

yy \S43
wed eS
nnn gs US Cette,
ae 'é - ak wine e
L.MWERU > eee ie
~ ~ * 3 : # “
3 s “ows et *
' aren oF _NYIKA
RY = £F PLATEAU
{ z AS ne ee
t eto Ae
Pie
KATANGA |
3S ZAMBIA , L. NYASA
$= “SL ps (NORTHERN RHODESIA) ,
\ ~ : a MOZAMBIQUE
a ' (
\
eae) \ MALAWI
j_ (NYASALAND) KEY
"6, ‘
ee
cea gate * CHARAXES ANSORGEI
a7 an
ai att
ery ‘7 ANSORGE! ANSORGEI
‘ Lp te
2 x ANSORGEI JACKSONI
t

ANSORGEI JACKSON! ATYPICAL

ANSORGEI RUANDANA

+45 eS c@

%,
AG
RHODESIA fy ANSORGEI RUANDANA ATYPICAL
(SOUTHERN. RHODESIA) ‘ones
VUMBA MTS ANSORGEI KUNGWENSIS
of
Fd ANSORGEI LEVICKI
7
at ANSORGEI KILMANJARICA
PAN
OO sansoncet RYDONI
é
a > ANSORGEI ?S8. SP.

-

Map 2. Distribution of Charaxes ansorgei Rothschild and subspecies.

REVISIONAL NOTES ON AFRICAN CHARAXES 303

There is still some doubt concerning the status of those aggregates which inhabit
the forests of the several mountain ranges and more isolated volcanoes to the west and
east of Lake Kivu and north west of Lake Tanganyika ; these are dealt with in the
appropriate place.

Charaxes ansorgei ansorgei Rothschild

(Pl. 6, figs. 49, 53)

Charaxes ansorget Rothschild, 1897 : 181.
Charaxes ansorget Rothschild ; Rothschild & Jordan, 1898, pl. 5, f. 2.
Charaxes ansorgei ansorgei Rothschild ; Poulton, 1933: 12.

Mate. Fore wing length 44 mm. (average), outer border slightly incurved at 2-4; margin
bluntly dentate, hind wing margin serrate. Upperside. Ground colour fore wing base to discal
line rufous, with some quadrate black marks, one in cell, one elongate transverse at end of cell, a
mark at base of 4, triangular marks sub-basal in 3—2 and a black dot below in rb, a quadrate
mark beyond cell and a triangular black patch between the V formed by the discal and post-
discal orange spots ; the remaining discal and postdiscal orange marks fused together, but with
an indication of black scales at line of fusion in 2 and 3; the lower part of the band in 1b
6-7 mm. wide. Border of wing black, with small orange spots on the margin, often double in rb.
Hind wing basal area more brownish, paling to greyish on the inner fold ; disc of wing crossed
by a white band, narrow and white at costa, slightly constricted in 6, then widening over cell
area and tapering to inner fold above anal angle, the borders of lower half suffused over with
blue scaling. Border black, with orange admarginal spots most distinct above upper tail ;
edge black with slight white fringe between dentate projections ; tails rather short, tapering
quickly to a point, 4-3 mm. long ; anal angle with double bluish white to lilac dots. Underside.
Pattern rather complex, the ground colour of the cell and bases of 2-7 chestnut, crossed by black
lines heavily edged in silvery white; the lower discal bar pinkish, shading to orange distally, with
a series of black dots down the mid line to 2; a submarginal row of black spots increasing in size
to tornus, those from 3 to subapex accentuated with white ; border of wing more greyish with
small orange marginal spots with black between. Hind wing basal ground colour chestnut,
crossed by an intricate pattern of black lines strongly outlined in white, bordered distally by
irregular discal bar, widest at costa then tapering to 2 but extending through the inner fold
above the anal angle as a triangular patch ; beyond the bar, the border of the wing is olive-
brownish, with a series of black ocelli outlined in white, the ocelli at anal angle and in space above
with lilac spots ; on the admargin a line of white and black linear marks, bordered by a marginal
band of tawny-olive ; edge black with slight white line internally running into mid line of tails ;
fringe slightly white.

FEMALE. Somewhat male-like but much larger, fore wing length 48mm. Upferside.
Base of fore wing slightly darker rufous chestnut; the black marks similar, but the discal
and postdiscal spots paler orange-ochre ; distal border of wing not so blackish, thus the marginal
inter-nervular orange spots and white fringe slightly less distinct. Hind wing darker brownish
at extreme base and brownish black on border ; the discal band rather narrow for size of insect,
white at costa, more shaded with bluish on borders of lower half, especially in the cell ; border
with orange admarginal spots duller and limited to the three or four above upper tail, those
at upper angle distinct, the rest fading out ; a purplish spot at anal angle ; margin black, with
white fringe especially strong at anal angle; margin strongly serrate; tails relatively short,
6-7 mm. Underside. General pattern as in the male but bolder, and distal borders of wings
more olive.

The female lays on Bersama abyssinica. For full account of life history vide
van Someren & Rogers, 1927-28.

304 V.G.L. VAN SOMEREN

Range: The nominate race of Charaxes ansorgei is confined to the high country
west of the Rift Valley and is found in the outlying hills and forests of W. Mt. Elgon,
on S. Elgon also on the east and in the higher riverine forests of the Trans Nzoia ;
on the Nandi Hills, Lumbwa, Mau, Cherangani and Elgeyu. Specimens taken by
Harrison, labelled “‘ Nairobi ’’ belong here.

Charaxes ansorgei jacksoni Poulton
(Pl. 6, figs. 50, 54, Map 2)

Charaxes ansorgei Rothschild ; Butler, 1900 : 915.
Charaxes ansorget Rothschild ; van Someren & Rogers, 1928 : 31-32, 153-156.
Charaxes ansorget jacksont Poulton, 1933 : 13.

MALE. Fore wing length 38-40 mm. ; shape as nominate race. Upperside. Ground colour
fore wing slightly darker rufous chestnut, shading into the black marks which extend to 1b on
the proximal side of the discal bar, this bar narrower than in nominate ansorgei ; the postdiscal
spots darker than the discal and the dark spots indicating the line of junction of the two series
more apparent ; the bar not more than 4 mm. wide at 1b, the distal black border thus slightly
wider from tornus to 3. Marginal orange spots slightly stronger than in ansorgei ansorgei.
Hind wing ground colour as in nominate race, the discal white patch, though narrow at costa,
widens out more basad over cell area and is slightly tinged with blue ; the admarginal orange
spots are bolder and often extend from upper angle to a point mid way between the tails ;
the serrate margin is black with distinct white fringe; the tails longer, 7-5 mm. Underside.
Much as in nominate race but the chestnut areas of fore wing rather larger, as also are the black
bars in bases of 1b-3 ; the pinkish ochreous discal band, shading into the postdiscal line, is
narrower in Ia~—2, the intervening dark spots reaching to 1b; the submarginal black dots with
the distal lunate white stronger and extending to the tornal black spots. Hind wing shows less
difference, but the postdiscal black and white ocelli larger and more conspicuous in the olive
border of the wing ; the marginal orange-olive border more defined.

FEMALE. Fore wing length 47-48 mm. Upperside. Ground colour darker rufous chestnut
at base than in nominate ansorgei, so that the black spots show up less, but the spots of the discal
series are white, slightly ochre tinged from 2 to subcosta, the spot in 4 is set well in and there are
two tawny streaks between it and the costa; the postdiscal spots are tawny-orange, distinct in
the subapex and less strong to 3, so that the whole wing bar is narrow,6mm.in1b. Hind wing
much as in the nominate race but the discal white patch is wider, extending more proximad over
the cell area; the admarginal orange spots are strong from upper angle to upper tail, then
obscured ; a blue and green mark present at the anal angle. Margin black, with white fringe
which outlines the strong serrations and the tails, which are 7-6 mm. long. Underside. There
is a similar difference on the underside of this race when compared with nominate ansorgei, as
noted in the males. In the fore wing, the discal bar is whiter and narrower, in the hind wing the
discal bar is narrower and whitish ; the postdiscal line of irregular ocelli is stronger and the
ground colour is tinged with chestnut; the marginal border stronger greenish orange, then
olive green toward the tails, this colour running down the mid line of the tails, which are black
bordered with white.

Range : This subspecies is found east of the Rift Valley in the high forests on the
eastern side of the Aberdare Mountains and also at the south end in the upper
Kikuyu country at Uplands and the Kirita forest on the Katamayu River.

A slightly differentiated aggregate is found on the eastern side of Mt. Kenya in the
upper Meru forests and also on the Njombeni Range to the north east of Mt. Kenya.
Too few examples are available for study to judge the constancy of the characters,

REVISIONAL NOTES ON AFRICAN CHARAXES 305

which are: female, fore wing length 50 mm. ; discal band entirely white and wider ;
hind wing discal patch larger. Ground colour darker. Poulton (1932: 800)
records a female taken by Maj. Moysey on the high country between the Imatong
and Acholi Hills in Southern Sudan, which may belong to a distinct race.

Charaxes ansorgei ruandana Talbot

(Pl. 6, figs. 51, 52, 55, 56, Map 2)

Charaxes ansorgei ruandana Talbot, 1938: 289. [9].
Charaxes ansorgei ruandana Talbot; Poulton, 1933: [3].

It is rather unfortunate that the name ruandana should have been selected by
Talbot for this subspecies since the type locality is not within the country of Rwanda,
but in Burundi; it is also unfortunate that the type male should be one from
further north-west, of Lake Tanganyika; furthermore, the female figured by
Poulton (1933, pl. 2, fig. 2) is one from the Congo-Uganda border. Poulton’s
quotation from LeCerf’s communication indicates the differences to be found amongst
the several aggregates of this very complex country. Although this area, taken as a
whole, can be looked upon as an ecological entity within certain limits, the broken
nature of the terrain due to mountain ranges and isolated ancient volcanic forested
mountains, which originated during and after the period of rifting of the Western
Rift, we find that certain montane species tend to break up into localized aggregates
with no very decided characters to distinguish them. This is definitely so in the case
of males of ansorgei, and the females are hardly at all known. However, until much
more material becomes available, especially females, the subspecific name ruandana
can be applied to the whole. I shall indicate the differences in the aggregates
where possible.

Mate. a. (Topotypical, N.W. Lake Tanganyika, Grauer Coll., 5 specimens.) Fore wing
length 37-38 mm. Upperside. Rufous chestnut at base of wing, darker than in nominate
ansorgei and resembling more that of jacksoni ; pattern of black spots adjoining the discal pale
bar heavier than nominate race but not so heavy as in jacksoni ; the upper discal-postdiscal
orange spots forming the V clearer than in jacksoni, the conjoined spots from 3 to hind margin
broader, but less broad than in nominate ansorgei, with the dark spots at line of fusion extending
to 1b. Marginal spots smaller and paler. Hind wing discal white band very constricted at the
upper end at costa at 6-7, the whole band narrower than in jacksoni ; admarginal orange spots
small and rather indistinct. Underside. Fore wing pattern very similar to nominate race but
bolder, the discal bar wider than in jacksoni; distal border more olive and postdiscal ocelli
bolder.

This agrees with type specimen figured by Poulton (1933, pl. 2, fig. r).

b. (S.W. Uganda, Kigezi; Ruhiza, Mafuga, Kanaba Gap). Compared with topotypical
ruandana, the fore wing basal chestnut brighter, more rufous ; the discal-postdiscal orange bar
wider at 1a—2; the hind wing discal white band wider and much less restricted at costal end ;
admarginal orange spots stronger. Underside very similar to form (a) but hind wing white
bar wider, especially at costal end.

c. (Mountains west of Lake Kivu). 3specimens. Very similar to the S.W. Uganda form (b)
in fore wing pattern ; hind wing white band as wide or wider at costal end. Underside similar.

306 V.G.L. VAN SOMEREN

FEMALE: a. No females are available.

b. (S.W. Uganda, Kigezi, Mafuga Forest.) Fore wing length 45-47 mm. Upperside. Base
of wing chestnut ; black marks: one in upper part of cell, a line beyond end of cell, quadrate
marks contiguous with the discal white spots from costa to 2 ; discal spots white or very slightly
tinged ochre to 1b, streak in 1a white; postdiscal spots: two upper ones white or slightly
tinged ochre, a faint indication of other spot 3-2. Hind portion of band narrow, seldom
more than 6 mm. wide at 1b. Distal portion of wing black ; marginal spots small. Hind wing
brownish at base ; black border wide, admarginal spots obscured ; discal band slightly narrow
at costa, then widening over disc and fading out in greyish inner fold, but narrower than in
jacksoni. Cf. description of female and figure, Poulton (1933, pl. 2, figs. 2) from the Uganda—
Congo border.

c. No females available from west of Lake Kivu.

The female lays on Bersama abyssinica var. Specimens were raised from larvae
found on this plant at edge of Mafuga Forest, Kigezi, Uganda.

Range: Type female recorded from N.E. end Lake Tanganyika, 12 miles north
of Usambura, (Urundi). Type male from N.W. Lake Tanganyika. Specimens
placed to ruandana from forests west of Lake Kivu; a long series from the high
forests of Kigezi, S. W. Uganda: Mafuga, Ruhiza, Kanaba Gap. Also one specimen
from Namwamba Valley, east side of Ruwenzori.

Ch. ansorget ssp. nr. ruandana Poulton, 1933. Mention is made of a series of six
males taken in the Kitembo area on the west side of Lake Kivu in 1931-32, at
about 7—9,000 feet. The characters are: the fore wing band is much broader and
more heavily marked with black than in rwandana; the postdiscal spots rather
larger and continued posteriorly as far back as area ra ; the fore wing marginal spots
more developed ; the hind wing band much broader and with no constriction toward
costa. (Paris Museum).

Charaxes ansorgei kungwensis ssp. n.
(Pl. 8, figs. 65, 66, 69, 70, Map 2)

Mate. Fore wing length 4omm. Upperside. Fore wing, general colour closely resembling
nominate ansorgei, the basal areas of the wing being a bright rufous chestnut, but the black
marks bolder, the discal and postdiscal orange spots paler, more orange ; the lower part of the
discal bar wide, 7-8 mm. in 1b; the marginal orange spots large and distinct. Hind wing,
basal area brownish black, forming a rather strong angle with the greyish brown of the inner
fold and accommodating the discal white patch, which has a rather strong black spot where the
band is constricted below the costa, then widens out over the disc ; the band is almost as wide as
in jacksoni ; and as in that race, the admarginal orange spots are large from upper angle to just
above the upper tail, then more obscured to anal angle, which has a violet spot and a green
admarginal line. Border as in other races; tails short, sharply pointed, 3-4 mm. long.
Underside. Basal chestnut of ground colour slightly more obvious, due to narrowing of black
and white bars ; discal-postdiscal band wider and pinkish ochreous ; dark spots as in nominate
race. Hind wing very like nominate, but fore part of discal bar wider and pinker.

FEMALE. Unfortunately, the only two females available for study are both rather battered
but they are sufficiently good to show that the females of kungwensis differ from ruandana by
having a considerably wider fore wing discal bar, the upper discal spots larger ; the two upper
postdiscal spots are white, the remainder very obscured. On the hind wing the discal white

REVISIONAL NOTES ON AFRICAN CHARAXES 307

bar is wider than in ruandana. The underside is too worn to show up details, but the most
obvious character is the increased width of both fore and hind wing discal bands.

Holotype male. TANZANIA: East shore, Lake Tanganyika, in Mahali forest on
the Kungwe Peninsula, vii. 1954 (T. H. E. Jackson). In B.M.(N.H.).

Allotype female. Same data as holotype.

A series of six males and two females, not considered paratypes.
Range: Tanzania, East shore, Lake Tanganyika, Mahali forest, Kungwe
Peninsula.

Charaxes ansorgei levicki Poulton
(Pl. 7, figs. 60, 64, Map 2)
Charaxes ansorget levickt Poulton, 1933: 15.

MALe. Fore wing length 38-39 mm. Upperside. Fore wing, general tone of ground colour
and pattern very similar to male of ansorgei jacksoni, the discal-postdiscal bar equally narrow ;
the black marks more distinct, and the orange marginal spots stronger, almost quadrate in some
examples. Hind wing basal area blacker, the discal bar almost divided into two blocks in form
and colour: the two upper costal spots long ovoid, the next small, all orange, the rest of the
band a long “ droplet ’’ with the apex toward the anal angle, pale bluish with a whitish bar
crossing it obliquely ; admarginal orange spots large and distinct from upper angle to mid point
between tails; border black with very thin white fringe; tails short, black, 4-5 mm. long.
Underside. Fore wing, a basic pattern similar to other races but chestnut area strong in costal
area ; the dark bars at basal area of wing, especially in 1b—2, broadly black ; the discal bar
ochreous orange with a central pale line, the margin with orange and black spots. Hind wing
discal band creamy with silvery sheen; the border ground colour chestnut with the row of
ocelli strong ; the margin reddish ochre, then more olive to anal angle ; marginal line white and
black, the white going up the middle of the tails.

FEMALE. Fore wing length 45-47 mm. Upperside. Fore wing, ground colour at base dark
chestnut, shading to almost blackish toward the discal line, with the usual black marks in cell
and beyond ; upper discal spots are a streak in subcosta, followed by an elongate then quadrate
spot, spot in 4 set well in, spots in 3 to hind margin increasing in size, the spot in 1b 6-7 mm.
All spots in discal line yellowish ochre, slightly more whitish in 1a. Postdiscal spots in almost
a straight line, those in 5-6 very slightly off-set, all spots bold and orange-ochreous in colour,
spot in 2 may touch the lower point of discal mark. Border of wing deep brownish black ;
margin with distinct though not large orange marks. Hind wing basal triangle black, shading to
greyish brown on inner fold ; disc crossed by a wide cream-white band with slight greenish blue
on inner border, greenish tinged on outer border. Border black, admarginal orange spots clear
to upper tail, then more obscure to lower tail ; margin black with thin white fringe ; tails black,
robust but tapering rapidly. Underside. Pattern basically similar to that of male, chestnut
areas bold ; black marks in mid wing strong ; discal bar creamy white with ochre tinge proxim-
ally ; border of wing orange-olive with submarginal arrow-head marks strong. Hind wing
boldly marked, the discal bar creamy; the border of wing light chestnut to olive, with sub-
marginal ocelli strong, that at anal angle and space above olive, black ringed ; border orange-
olive, black edged and outlined in white proximally, fringe white.

This race has been bred in some numbers on Bersama abyssinica at Njombe,
S. Tanzania Highlands, by Dr. C. H. McCleery.

Range: The Southern Highlands of Tanzania, the Rungwe and Poroto Mts.
Manow, Njombe, and Iringa ; and on the Nyika Plateau in Malawi.

V.G.L. VAN SOMEREN

308
“a =e ?
eT gg \ ee ETHIOPIA ae ee i: ie
/
Ne lglg /
/
4
\ L. RUDOLF (
‘e l
} KENYA
- H
RUWENZORI oe l
NJOMBENI l SOMALIA
t
{
0 I 0
ie root si ==
t
2 %
ar
CONGO 3 NAIROBI
- 3
crane
= =%
MERU re KILIMANJARO
A, ge \
6, * a A -€
5 . Grae. 5
TANZANIA i hee st PEMBA
(TANGANYIKA) +. Ps §,,_» USAMBARA $
oye ant x
@ vuncwe MOROGORO i ZANZIBAR
ola,
SOUTH HIGHLANDS
sida x : F
KATANGA \ NJOMBE
| 10
LIVINGSTONE MTS.
a
f
3 \
mn
3%
. ; ; L. NYASA
ston MALAWI
~ s
(NYASALAND) }, MOZAMBIQUE
ZAMBIA i. x
(NORTHERN RHODESIA) me EN atte x
\
es fee
{
bash
yen,
ges
31%
71%
sare nt
. =
Map. 3. Montane Forest areas, Eastern Africa (adapted from Vegetation Map, UNESCO,

1958), in relation to distribution of Charaxes ansorgei Rothschild and subspecies.

REVISIONAL NOTES ON AFRICAN CHARAXES 309

Charaxes ansorgei kilimanjarica ssp. n.
(Pl. 7, figs. 57, 58, 61, 62, Map 2)

Mate. Fore wing length 38mm. Upperside. Fore wing, general colour and pattern very
similar to Jevicki and as dark, the black marks proximal to the discal line slightly heavier ; the
discal and postdiscal spots conjoined from 3—1b, the postdiscal series slightly darker rufescent
than the discal ; border of wing black, with orange spots on margin well developed. Hind wing
as in levicki but orange marks in upper part near costa more defined as three streaks, the lower
part of band a long ovoid, bluer and smaller than in /evicki. The admarginal orange spots large
and extending to just above lower tail. Tails slightly longer and thinner, 5mm. Underside.
Basic pattern as in levicki, fore wing white lines more silvery ; the discal bar narrower. Hind
wing white lines and the discal bar silvery ; submarginal marks bolder, the mark opposite the
upper tail is a strong silvery rosette, ocelli below dark olive ringed in black, that at anal angle
olive, ringed in black and with two blue dots.

FEMALE. Fore wing length 45mm. Upperside. Fore wing, basal area slightly darker
chestnut than /evicki, thus black marks more obscured. Discal and postdiscal spots as in
levicki but spots in 2-3 touch, all spots orange-tawny ; marginal spots small and may be partially
obscured. Hind wing discal band narrower than Jevickt, first two marks at costal end square-cut
on inner ends and forming a defined angle with the expanded patch over the disc, the costal
marks ochreous, the rest white with a blue suffusion on the borders. Admarginal spots strong.
Hind wing discal bar narrow; the ground colour of the border lighter olive-chestnut ; the
submarginal marks bold, the spot in 6 strongly silvery, that in 4 a silvery rosette outlined in
black, the lower spots bold and olive, ringed in black, that at anal angle with two lilac spots ;
border orange with olive tinge edged black and white as in the male.

Holotype male. TANZANIA: Western foot hills, 6~7o00 ft., Kilimanjaro.
ii. 1964 (A. Brown & J. G. Williams). In B.M.(N.H.).

Allotype female. Same data as holotype. In B.M.(N.H.).
Range: Known only from the western side of Mt. Kilimanjaro, Tanzania.

Charaxes ansorgei rydoni ssp. n.
(Pl. 7, figs. 59, 63, Map 2)

Two male specimens, unfortunately very damaged, were taken recently on the
Usambara Range, Tanzania. They belong to the southern group of ansorget in
that the hind wing band is bicolored as in Jevickt. One specimen, though damaged,
is reasonably fresh.

MALE. Fore wing length 4omm. Upperside. Fore wing, colour brighter chestnut at
base than in either Jevicki or kilimanjarica, more like rwandana, so that the black marks stand
out strongly to 1b. The discal and postdiscal series of orange-tawny spots well developed,
conjoined from 3 to hind margin 1b, the band relatively narrow, but wider than kilimanjarica
and paler; marginal spots very bold. Hind wing discal band in two sections, the upper part
at costal end orange-tawny, the third mark not strongly defined, the bluish white portion of the
band larger and paler than kilimanjarica, even larger than in levickt and extending basad in
cell area; marginal orange spots large and bold and extending to lower tail; there is also a
short blue line in the black border basad to the tails. Underside. Less dark than in either
levicki or kilimanjarica, the chestnut basal areas paler and conspicuous, but the black marks of
sub-base in 1b large and linked up with black mark sub-basad in 2 ; the discal bar comparatively
natrow, mostly orange, slightly paler on proximal side ; marginal spots strong. On the hind

ENTOM. 18, 9. 22

310 V.G.L. VAN SOMEREN

wing the chestnut patches are conspicuous ; the discal bar narrow ; the submarginal irregular
marks strong in the olive-chestnut ground, which fades to olive-ochre toward the anal angle ;
the marginal border orange, then orange-olive to anal angle ; edge narrowly white and black.

Holotype male. TANZANIA: Usambara Range, Magamba Forest, nr. Lushoto,
8.x.1962 (A. H. B. Rydon). In B.M.(N.H.).

Female unknown.
Range : Known only from the higher forests of the Usambara Range, Tanzania.

SYSTEMATIC LIST
Charaxes ansorgei Rothschild

Charaxes ansorget ansorget Rothschild, 1897. Type male. Type locality, Patsho,
Nandi; in B.M. (N.H.). Type female. Type locality, Lumbwa,
W. Mau in University Museum, Oxford.

Range : Uganda, W. & S.W. slopes of Mt. Elgon ; Bufumbo, Bulago,
Butandiga, Mt. Kokamjero. N.W. Kenya, eastern slopes of Mt. Elgon,
Trans Nzoia, Kitale, Turbo, Nandi forests, Lumbwa, Kericho, Mau,
Elgeyo, Cherangani. All west of the Rift valley.

Charaxes ansorget jacksont Poulton, 1933. Typemale. Type locality. Katamayo
Riv. Uplands, Upper Kikuyu; in University Museum, Oxford.
Allotype female, Roromo, Kikuyu ; in B.M. (N.H.).

Range: Kenya, east of the Rift Valley, in high forest of the Kikuyu
country : Uplands, Katamayo ; eastern Aberdares, Mweiga, Moyo.

Specimens from Mt. Kenya at Embu, Meru and Njombeni Range,
slightly different.

Charaxes ansorget ssp. (?) Poulton, 1932, records a female taken by Maj. Moysey
on the high country between the Imatong and Acholi Hills in Southern
Sudan.

Charaxes ansorget ruandana Talbot, 1932. Typefemale. Typelocality. 12 miles
north of Usambura, north end of Lake Tanganyika, Kibera Forest,
Burundi. In B.M. (N.H.).

Allotype male. Type locality. N. W. Lake Tanganyika. In
B.M. (N.H.).

Range: (Nominate form) Mountains N.E. and N.W. of Lake Tan-
ganyika. (Atypical form) : Mountains east of Lake Kivu; Uganda :
In Kigezi, S.W.: Mafuga, Kanaba Gap, Ruhiza ; Ruwenzori eastern
side, Nawamba Valley. These are slightly different and agree with
specimens from west side Lake Kivu, in Congo: Lushasha, 2-22°S,
38-51°E ; Bobandana, 1°42°S, 29°E. ; Katana, 50°S, 22-9°E.
Charaxes ansorget kungwensts ssp. n.

Range: Tanzania: Mt. Kungwe, Mpanda district east side Lake
Tanganyika. Types in B.M. (N.H.).

REVISIONAL NOTES ON AFRICAN CHARAXES 311

Charaxes ansorget levickt Poulton, 1933. Type male. Type locality. N. Lake
Nyasa (no exact loc.) Allotype female. Manow, north of Lake
Nyasa in S. Tanzania, west of New Langenberg. In B.M. (N.H.)
Range: The southern Tanzania Highlands: Porto, Rungwe, Manow,
Njombe and Owindi ; extending to the Nykka Plateau in N. Malawi.

Charaxes ansorget kilimanjarica ssp. n.
Range: Known only from the western slopes of Mt. Kilimanjaro at
Wasendo, 6000 ft. Types in B.M. (N.H.).

Charaxes ansorget rydont ssp. n. Kilimanjaro at Wasendo, 6000 ft. Type ¢.
Range: Tanzania, Usambara Range, high forests at Lushoto. In
B.M. (N.H.).

CHARAXES POLLUX CRAMER AND ITS SUBSPECIES

This is a common species with an almost continuous distribution from the West
Coast of Africa to Kenya on the east, extending southward on the western side to
North Angola and Katanga, and on the eastern side to Mozambique and Rhodesia
but not extending into Transvaal. (Map 4.)

It has been divided into two main aggregates, a northern group and an eastern
group with some evidence of transitionals in the south-east of Katanga Province
of the Congo and in Zambia. The subspecies will be discussed in detail hereafter.

It is mainly a forest species, but is also found in gallery forest and heavy woodland
associated with its food plants, which in Eastern Africa are: Bersama abyssinica
et ssp., (Melanthaceae), Fluggea microcarpa (Euphorbiaceae), Deinbollia kilimanjarica
and D. burbonica (Sapindaceae). I cannot find any records of the species having
been bred on the West Coast.

A full synonymy is given by Rothschild & Jordan (1898-1900) and need not be
repeated here, but will be modified, as necessary, when dealing with the geographical
races.

The manuscript name euronotus Le Cerf appears under the series of Charaxes
pollux from Mozambique and Rhodesia, in both the British Museum (Natural
History) and the Paris Museum and in the latter, one specimen bears a type label.
Extensive search through relevant literature has failed to provide evidence that this
name was ever published.

Charaxes pollux pollux Cramer

(Pl. 9, figs. 75, 76, 79, 80, Map 4)

Charaxes pollux Cramer, 1775: 61.
Charaxes pollux Cramer; Rothschild & Jordan, 1900: 427.
Charaxes pollux pollux Cramer ; van Someren & Rogers, 1927-28 : 146-149. [Life history.]

Mate. Fore wing length 36-40 mm.; Upperside. Fore wing, outer border very slightly
concave at 3-4; basal area rufous tawny, followed by a broad, orange-tawny, disco-postdiscal
band, broken in the upper half by black marks, one at upper angle of cell, a black quadrate
spot at end of cell, sometimes joined to black mark basal in 3, a black mark beyond continuous
with black mark in 4 and forming a triangle, a further triangular black mark separating the two

ENTOM. 18, 9. 22§

312 Vv. G. L. VAN SOMEREN

arms of the V of orange marks of the upper part of disco-postdiscal band, which is slightly in-
curved at 3-4. Border broadly black, almost uniform in width, sometimes with very small
tawny internervular spots on margin. Hind wing basal triangle rufous tawny, slightly paler
toward costa, merging into tawny greyish on inner fold, the outer edge almost straight ;
discal band orange-tawny, paler at costal end, narrowing toward inner fold above anal angle ;
outer border almost straight, leaving a broad black border to mm. wide at 4-6, without any
tawny spots on admargin ; two bluish spots in anal angle often tinged violet ; fringe narrowly
white between veins ; margin strongly serrate ; tail moderately long 5—4 mm., sharply pointed.
Underside. Basal half of fore wing rusty chestnut, with heavy black bars outlined in silvery
white, the black bars in sub-bases of 1b-3 heavier and edged in white; the disco-postdiscal
band whitish ochre, shading to orange distally ; the postdiscal line with a series of black linear
marks from subcosta to 2; border of wing tawny rufous with triangular black marks outlined
in white to bluish in submarginal zone, blacker at tornus; border of wing more rusty with
black marks at ends of veins. Hind wing, basal half up to discal bar deep chestnut, divided up
into segments by silvery lines, the costal and subcostal ones double, the lines in the centre of
the disc forming a somewhat diamond-shaped mark ; black and white marks on inner fold form
elongate triangles, with apices toward base of wing. Discal bar comparatively narrow, silvery
white to buffish, clear cut on inner edge but more diffuse on outer, where it merges into the
rusty tawny olive of the wide border; in the mid line of the latter there is a chestnut zone
bordered in the postdiscal line by rather irregular black triangles edged with white ; and in the
submarginal line by linear black marks; anal angle with a large greyish ocellus outlined in
black and with two lilac central spots ; border rufous tawny shading to greenish olive on anal
angle ; edge black.

FEMALE. Fore wing length 45 mm., (average). Very similar to the male, but larger and
usually paler. Shape as male. Upperside. Fore wing pattern like male, basal colour similar
or slightly paler; black spots basad in 2—3 may be less strong. Black border wide, margin
with very small tawny spots, usually complete. Hind wing pattern similar to that of male,
but discal band may be paler ; black border wide, without any yellow spots or just the slightest
indication of a spot at upper angle.

Anal angle with or without blue spots. Margin serrate, with slight white fringe. Tails long,
upper 8mm., lower 5mm. Underside. General pattern as in the male but lower portion
of discal bar fore wing wider, and the series of linear black marks from costa to 2 in disco-
postdiscal line stronger; the submarginal triangular spots stronger; the marginal border
interrupted by black streaks through 5-6, the black marks shorter. Hind wing pattern as in
the male but bolder ; the postdiscal black and silvery marks stronger, especially that in 4,
which is more silvery ; submarginal lunate black marks strong; ocellus at anal angle very
bold ; admargin rusty brown with olive tinge, more olive toward anal angle ; margin black with
slight white fringe. In older specimens the chestnut bases and border tend to be paler and
brighter.

Variation : a. Very similar to nominate form but subcostal and sub-basal black marks 3-4
of fore wing greatly reduced or obsolete. Sex not stated ; but afemale character. Var. ongeus
Stoneham (1932: 2).

b. A strongly marked colour variety in which the basal areas are darker than usual and the
black borders more intense blue-black. Sex notstated. ... Var. bungense Stoneham(1932 : 2)
Both are trivial variations.

c. Original description: ‘‘ g Upperside: Fore wings with the whole of the black subcostal
and discal spots fused together with the marginal border, so that rather more than the apical
third of the wing is solidly black ; all the yellow discal spots above vein 3 obliterated. Under-
sidé : The ground colour of the whole of the costal area of fore wings is white, without any trace
whatever of the red patches of pollux; black spots normal, the post-discal spots between
vein 3 and the costa more broadly bordered with silvery grey. Hind wings normal, but the
white discal band a little broader than usual.”

Sierra Leone. ... ab. subalbescens Hall (1930 : 279).

REVISIONAL NOTES ON AFRICAN CHARAXES 313

Range: The nominate subspecies, with slight variations not related to
environment, ranges from the West African countries of Guinea, Sierra Leone,
Ghana, Nigeria, Cameroon, Gabon, Congo Republic, Central African Republic,
Katanga and north Angola to parts of Zambia bordering on Katanga, thence to
Rwanda, Burundi, Uganda and southern Sudan, to N.W. Kenya, Elgon, Mamdi,
Sotik, Mau then across the Rift Valley to the Aberdares, Mt Kenya, Meru and
Kikuyu to the Nairobi District.

Charaxes pollux geminus Rothschild
(Pl. 8, figs. 67, 68, 71, 72, Map 4)

Charaxes pollux geminus Rothschild in Rothschild & Jordan, 1900: 497.
Charaxes pollux geminus Rothschild ; van Someren & Rogers, 1927-28 : 149.
Charaxes pollux zingense Stoneham, 1932: 2, syn. n.

Rothschild was in error when he referred to the specimens of Charaxes pollux
Trimen (nec Cramer) as belonging to geminus. (See under gazanus ssp. n.)

The subspecies geminus Rothschild, is a mixed aggregate, as indicated in the
original description. It consists of three forms which occur together, in greater or
lesser degree, within the type area and throughout its distribution ; the forms are:
(a) typical geminus (as type); (b) a transitional form and (c) which is very like
nominate pollux but with narrow black borders to both wings.

Mates. a. Nominate geminus from Lauderdale, Malawi. Similar to nominate pollux
in general appearance but upperside of fore wing disco-postdiscal band wider from 3 to hind
margin and slightly paler, thus reducing the width of the black border, which has rather strong
tawny spots on the margin ; hind wing discal band wider, the black border reduced in width in
consequence ; the admargin with conspicuous orange tawny spots, distinct above upper tail,
less so at tails, the spots well divided by black ; anal lobe with one blue spot. Tails rather thin
and sharply pointed. Border of wing serrate. The underside exhibits a broader discal band ;
the basal pattern of fore wing as in nominate pollux but arrow marks in submarginal border
rather more distinct from sub apex to tornus. Hind wing pattern basically similar to nominate
race but discal band very slightly wider ; the ground colour of the border slightly paler so that
the postdiscal and submarginal dark marks show up more distinctly, the anal ocellus clear.
(Pl. 8, figs. 67, 71, type.)

b. Topotypical: in this form the fore wing marginal tawny spots are as well developed as
in the type geminus, but the hind wing admarginals are clear above upper tail but obscured to
anal angle. PI. 8, fig. 68.

c. Topotypical. Fore wing marginal spots minute, those of the hind wing admargin limited
to one or two at upper angle ; black borders narrow. PI. 8, figs. 72.

FEMALES. This sex exhibits variations similar to those in the males. Since all three forms
are found in and around the type locality and throughout Malawi extending northward to
Tanzania, the Kenya coastal belt and adjacent hinterland, the name geminus must apply to all
forms within this aggregate. It should be noted however that the nominate geminus form is
found almost to the exclusion of the others along the Kenya coastal belt and inland to the
Teita Hills and the Taveta Forest, on the Usambara Range and on the foothills of East
Kilimanjaro.

The name zingense Stoneham, proposed for specimens from the Kenya Coast just north of
Mombasa is a synonym of nominate geminus.

ENTOM. 18, 9. 228§

314 V.G.L. VAN SOMEREN

Range: From Malawi and adjacent parts of Zambia east of the Loangwa Valley,
north through southern Tanzania to Usambara, the coastal belt of Kenya and
hinterland, Teita Hills, Taveta to East Kilimanjaro.

When this paper was in galley proof, a new subspecies of C. pollux was recog-
nised. For its description, see Appendix, p. 316.

Charaxes pollux gazanus ssp. n.

(Pl. 9, figs. 73, 74, 77, 78, Map 4)

Charaxes pollux Trimen, 1894 : 41, nec Cramer, 1775.
Charaxes geminus Rothschild ; Pinhey, 1948 : 284.

When Trimen reported on the F.C. Selous collection of butterflies from Manicaland
in 1894, he made the following observation: “‘ A female from Christmas Pass and
male from Mineni Valley . . . are both distinguished from West African examples...
by possessing considerably larger ochre-yellow spots ... on the margin of the
fore wing ... and a complete and conspicuous series of ochre-yellow lunules along
the entire (outer) margin of the hind wings; ... they further both lack the...
lowermost black spots between 2nd and rst median nervules...’’ Trimen adds

further that he believed these specimens to be the furthest south-east records of the
species.

Rothschild (1900 : 497) quoted this reference as applicable to his subspecies
geminus but he selected as the type of his new subspecies a specimen from Lauder-
dale, Malawi which belongs to a different ecological race, and he mentions other
specimens from Zomba, also in Malawi. The low Zambesi Valley is the ecological
barrier between the Malawi and the Manicaland-Gazaland aggregates.

Mace. Fore wing length 38-40 mm. but on an average smaller than the nominate Pollux.
Upperside. Fore wing, general pattern similar to other subspecies but basal and discal areas
wider and paler, the disco-postdiscal band wider, the bars being fused up to area 6; the sub-
costal black areas reduced, the black spots in 3—2 reduced, often missing in 2; the marginal
black border thus reduced in width and on its margin are conspicuous ochre-yellow spots
complete from apex totornus. Hind wing basal discal band paler and wider with a consequent re-
duction in the width of the black border which carries a conspicuous and complete row of
admarginal ochre-yellow lunules almost contiguous and only narrowly separated by black veins.
Underside. Fore wing generally paler than in other races, the white edging to the basal black
lines slightly more conspicuous ; the disco-postdiscal band wider; ground colour of border
paler so that the submarginal dark arrow marks and the marginal internervular marks are more
conspicuous. Hind wing generally paler, the discal pale band usually wider, and the marginal
lunules more conspicuous. A few males may be more strongly orange-tawny at bases of
wings, but the general paler pattern is maintained.

FEMALE. Fore wing length 43-45mm. Upperside. Fore wing, pattern very similar to that
of the male, exhibiting a general paleness compared with females of other races; the
disco-postdiscal band considerably wider and the black marks in 3-4 reduced in size; the
black border narrower but the marginal orange-ochre spots large. In the hind wing, there is a
corresponding narrowing of the black border but the admarginal ochre-yellow lunules are large
and very conspicuous and continuous to lower tail. Underside. This exhibits a corresponding
paleness due to the increase in the width of the disco-postdiscal bands ; the hind wing with more
conspicuous marginal lunules.

REVISIONAL NOTES ON AFRICAN CHARXAES 315

Holotype male. S. MOZAMBIQUE: Amatongas, v.1962 (G. van Son).

Allotype female. Same data as holotype. Types g, 2 in Transvaal Museum.

Paratypes. Same data, in British Museum (Natural History) presented by
Transvaal Museum. Other paratypes in the National Museums of Rhodesia,
Bulawayo and in the National Museum (Coryndon Museum), Kenya, from Vumba
Mountains.

Range : This southern race inhabits the area of Manicaland-Gazaland to Amaton-
gas in S. Mozambique and extends to the Vumba Range to as far as Mt. Selinda,
Rhodesia. It does not appear to go north of the Zambesi River nor south to the
Transvaal.

SYSTEMIC LIST
Charaxes pollux Cramer

Charaxes pollux pollux Cramer, 1775, Type locality : ‘‘ Guinea ”’.
Range: West Africa, Guinea, Sierra Leone, Ghana, Nigeria,
Cameroon, Gabon, N. Angola; Congo Republic and Katanga and
adjacent area of Zambia west of the Loangwa Valley, Rwanda,
Burundi, S. Sudan. Uganda to N.W. Kenya, Elgon, Nandi, Sotik
and Mau then crosses the Rift Valley to the Aberdares, Kikuyu to
Mt. Kenya and Meru and in the Nairobi district.

Charaxes pollux geminus Rothschild, 1900. Type locality : Lauderdale, Zomba,
Malawi.
Range: Malawi and adjacent Zambia east of the Loangwa Valley ;
southern and eastern Tanzania to Kenya coastal belt and hinterland to
Teita and East Kilimanjaro.

Charaxes pollux maua ssp.n. Type locality: western foothills, Mt. Kiliman-
jaro, Maua. (See Appendix).
Range: western foothills, Mt. Kilimanjaro and Mt. Mweru.

Charaxes pollux gazanus ssp.n. Type locality : Amatongas.
Range: Southern Mozambique, Manicaland and Gazaland, eastern
Rhodesia, Vumba Mts. to Mt. Selinda.

ACKNOWLEDGEMENTS

I am indebted to the following for assistance and the loan of essential material and
record my thanks: The Keeper and Staff of the Entomological Department of the
British Museum, (Natural History), particularly Mr. T. G. Howarth; Mr. Ernest
Taylor, Hope Department of Entomology, University Museum, Oxford; Dr. G. van
Son, Transvaal Museum, Pretoria ; Dr. Elliot Pinhey, National Museum of Rhodesia,
Bulawayo, Rhodesia ; Dr. C. B. Cottrell, University College, Salisbury, Rhodesia ;
Mr. K. M. Pennington, Balgowan, Natal; Mr. J. D. Handman of Limbe, Malawi,
whose contributions were of outstanding value; Mrs. J. Wedekind, Mumbwa,
Zambia; Mr. D. C. Plowes, Umtali, Rhodesia; Mr. Duke, Salisbury, Rhodesia ;
Mr. J. Lawson, Durban Museum, Natal; Mr. C. R. Barrett of Johannesburg ;
Mr. W. Teare, Benoni, Transvaal; Mr. C. McMaster of Stutterheim, South Africa ;
Dr. C. McCleery, Njombe, Tanzania ; Mr. Ionides, Newala, Tanzania ; Dr. A. Rydon,

316 Vv. G. L. VAN SOMEREN

Arusha, Tanzania; Major Iain Grahame, Jinja, Uganda; Mr. J. H. Bailey,
Menengai, Kenya; Mr. Barton Eckett, Hoeys Bridge, Kenya; Mr. T. H. E.
Jackson, Kitale, Kenya ; Mr. R. H. Carcasson, National Museum (Coryndon Museum),
Nairobi.

REFERENCES

Full details of references will be found in earlier parts of this issue.

VAN SOMEREN, V.G.L. 1963. Revisional Notes on African Charaxes (Lepidoptera : Nympha-
lidae). PartI. Bull. Br. Mus. nat. Hist. (Ent.) 13(7) : 195-242, 19 pls, 5 text-figs.

1964. Part II. Ibid., 15 (7) : 181-235, 23 pls, 4 maps.

—— 1966. Part III. Ibid., 18 (3) : 45-100, 16 pls, 5 maps.

APPENDIX
Charaxes pollux maua ssp. n.

Specimens of pollux from the foothills of Western Kilimanjaro differ appreciably
from ~. geminus Rothschild of Malawi and the Tanzania coastal belt, the Kenya
coast to the Teita Hills of the hinterland. Whilst retaining some of the characters
of geminus, i.e. the orange spotting on the submargin of the hind wing, the West
Kilimanjaro insects are generally darker above and below, in both sexes.

MALE. Fore wing length 39-40 mm. Upperside. Basal triangle rufous chestnut, darker than
in geminus and in greater contrast to the orange disco-postdiscal bar ; subcostal black marks
bolder and stronger; the marginal black border wider from 4 to apex; marginal orange spots
distinct. Hind wing basal area darker so that the subcostal portion of the discal band is
noticeably creamy but stronger orange toward the inner fold above the anal angle ; black
border wider, with large orange spots 4-5 in number, decreasing in size from upper angle to upper
tail. Underside. Pattern strong and darker than in geminus.

FEMALE. Upperside. follows the general darker tone and wider black border, noted in the
male. Underside also darker and more boldly marked.

Holotype male. TANZANIA: western foothills, Mt. Kilimanjaro, Maua, ix. 1966
(African collector for Major Grahame).

Allotype female. Same data. Types will be deposited in B.M. (N.H.).

A long series of both sexes, taken over a period of three monthsin 1966and through-
out all months during previous years, exhibits a constancy in characteristics through-
out the year.

Range: In the forested areas of the frthills of western Mt. Kilimanjaro and
Mt. Meru, 6~-9,000 ft. Arusha, Ngurdoto Crater, 1963 (A. H. Rydon); Maua
Estate, West Kilimanjaro (A. F. Brown) ; Lyamungu Moshi, Marangu.

On map 4 the sign for pollux geminus west of Mt. Kilimanjaro refer to this new
subspecies.

VIdOIHL3

x vv’
7" ie 6
ine _—_—-— ~

- ¥
\
?
’

NVONs

Pisach et
Charaxes ethalion Boisduval

Fics. 1 and 2. ethalion Boisd., 3 (Portuguese East Africa) and 9 (Vumba District, Bomponi), upper and
undersides.

Fic. 3. ethalion 9 form rosae Butler, 9 (Rhodesia : Umtali), hind wing discal bar much reduced in upper half.

Fics. 4 and 7. ethalion 2 form rosae Butler, Type 2 (Natal), upper and underside. Photos B.M. (N.H.),
Nos. 34015 and 34016.

Fics. 5 and 8,6and 9. pondoensis ssp. n., Holotype ¢ (Pondoland, Port St. Johns) and Allotype 2 (Bashee
River), Bridges leg., upper and undersides.

Approx. 2/3 natural size.

Bull. Br. Mus. nat. Hist. Entomology, 18, 9. PLATE. z

PLATE 2
Charaxes ethalion Boisduval

Fic. 10. ethalion 2 form swynnertoni Poulton, 9 (Portuguese East Africa: Amatongas), upper and underside.

Fics. 11and12. myasana ssp. n., Holotype $ (Lake Nyasa, Nkata Bay), Handman leg., upper and underside.

Fics. 13 and 14. myasana ? form nyasana forma n., Allotype 2 (Malawi [Nyasaland]: Mlaye), fore wing spots
white, no spot in cell or 4 of distal row ; hind wing bar narrower, strongly blue on lower portion, upper and under-
side.

Fic. 15. mnyasana ssp. n., 9 variety (Lake Nyasa, Nkata Bay), base of wings with green sheen ; spot in cell and
base 4; all spots large and white ; hind wing discal band white extending proximad, upper and underside.

Fics. 16 and 17. mnyasana 9 form imitans forma n., Type 9 (Malawi [Nyasaland]: Mlaye, Mlosa stream),
fore wing white bar wide at costa ; hind wing band strongly blue, upper and underside.

Fic. 18. mnyasana 9 form imitans forma n., 9 (Malawi [Nyasaland] : Mlaye), hind wing band very narrow, not
reaching costa ; strongly blue, upper and underside.

Approx. 2/3 natural size.

Bull. Br. Mus. nat. Hist. Entomology, 18, 9 PLATE o

23§

PLATE 3
Charaxes ethalion nyasana ssp. 0.

Fics. 19 and 22. Variety of 9 form imitans forma n., (Mlanje, Chitakali), fore wing discal curved bar very
narrow and broken; spot beyond cell large, postdiscal spots, except in subapex, missing ; hind wing band re-
duced, strongly blue, upper and underside.

Fics. 20 and 23. Q form suppressa forma n., Holotype 2 (Malawi [Nyasaland]; Port Herald, Mlosa stream),
fore wing discal bar almost entirely suppressed ; postdiscal spots strong and complete ; hind wing band strongly
blue, upper and underside.

Fics. 21 and 24. form demaculata forma n., Holotype ? (Malawi [Nyasaland]: Lugeristream), fore wing discal
bar strong but postdiscal spots almost absent ; hind wing band goes to costa but is narrow, white in discal line,
strongly blue on borders, upper and underside.

Fics. 25 and 28. @ form cithaeronoides approaching form swynnertoni (Malawi [Nyasaland]: Port Herald),
fore wing spots large, mauvy blue ; hind wing band white with mauvy blue suffusion, upper and underside.

Fics. 26 and 29. Q form cithaeronoides forma n., Type 9 (Matawi [Nyasaland] : Mlaye, Mlosi stream), fore wing
spots greeny blue ; hind wing band strongly greeny blue upper and underside.

Fics. 27 and 30. @ form near rosae (Lake Nyasa, Monkey Bay), fore wing bar and spots beyond well developed ;
hind wing band broad, reaching costa ; mostly white especially in disc, faintly blue on borders, underside strongly
marked, upper and underside.

Approx. 2/3 natural size.

Bull. Br. Mus. nat. Hist. Entomology, 18, 9

PLATE 4
Chavraxes ethalion Boisduval

FIG. 31. myanzae ssp. n., 9 vosae pattern (E. side Lake Tanganyika, Kungwe).

Fics. 32 and 33. myanzae 2 form howardi van Someren & Jackson, Type Q (Kenya: Chepalungu), rosae
pattern but fore wing curved band very wide in lower portion, upper spots large; all pale marks in fore and
hind wings ochreous, upper and underside. Photos B.M. (N.H.), Nos. 35017 and 35018.

Fics. 34 and 35. myanzae ssp. n., 9 ethalion pattern (E. side of L. Tanganyika, Mpanda-Kungwe), fore wing
discal spots large, an extra spot in 4; upper spots white, lower strongly blue ; hind wing band wide, narrowly
white in discal line, strongly blue on borders, upper and underside.

Fic. 36. kikuyuensis ssp. n., Q ethalion pattern (Kenya: Nairobi District), fore wing discal and postdiscal
spots white, slightly bluish over 1a—1b; hind wing band very broad, white in disc, borders slightly bluish, upper
and underside.

Fic. 37. Jittoralis ssp. n., 2 ethalion pattern (Kenya Coast : Sekoke-Arabuku forest), fore wing discal and post-
discal spots conjoined in posterior area, rayed in upper, spots large, discal bar wide; hind wing band very broad
even toward costa, white with pale blue on borders, upper and underside.

Fic. 38. Jittoralis ssp. n., 2 (Kenya Coast: Shimba Hills, Kwale), fore wing discal bar narrow, no spot in
cell or base 4, white with slight bluish on hind margin in 1a; hind wing band going through to costa, rather
narrow, white in disc, strongly blue on borders, upper and underside.

Fic. 39. Jittoralis ssp. n., 29 (Tanzania [Tanganyika]: Usambara Range, Amani), fore wing discal bar reduced
in width, postdiscal spots small but clear ; hind wing band white, extending to costa, strongly white in discal line,
bluish on borders, upper and underside.

Approx. 2/3 natural size.

Bull. Br. Mus. nat. Hist. Entomology, 18, 9 PIAS. 4

4
Bi Gios

} ae, .

PLATE 5
Chavaxes ethalion Boisduval

Fics. 40 and 43, 41 and 44. marsabitensis ssp. n., Holotype ¢ and Allotype 2 (Kenya: Northern Frontier
Province, Mt. Marsabit), fore wing of 2 with upper discal and postdiscal spots white, though spots in 1a—1b are
mauvy blue ; hind wing band mauvy blue; submarginal linear marks strong, upper and underside.

Fics. 42 and 45-48. myanzae ssp. n., § and 49, (North-eastern side Lake Tanganyika, Kungwe and Kigoma
region at Mukuyu), Group 4 Region 4; 45 and 46, 2 ethalion pattern variety ; 47, 2 swynnertoni pattern; 48,
2 vosae pattern ; these ? forms are characterized by very bold, enlarged patterns. Cf. Plates 2 and 3, 2 of myasana,
upper and undersides. Specimens in Coll. I. Grahame.

Approx. 2/3 natural size.

Bull. Br. Mus. nat. Hist. Entomology, 18, 9 PEALE s

PLATE 6
Charaxes ansorgei Rothschild

Fies. 49 and 53. amnsorget Rothschild, § and 9 (topotypical Kenya: Mt. Elgon and Nandi
Hills), upper and undersides.

Fics. 50 and 54. jacksoni Poulton, g and 9 (topotypical Kenya: Upper Kikuyu, Katamayu
forest), upper and undersides.

Fics. 51 and 55. Near ruandana Talbot, g and 9, (Uganda: Rutenga District, Mafuga
forest, Kigezi), upper and undersides.

Fics. 52 and 56. ruandana Talbot, $ (topotypical N.W. of Lake Tanganyika), upper and
underside.

Approx. 2/3 natural size.

Bull. Br. Mus. nat. Hist. Entomology, 18, 9

PLATE 7
Charaxes ansorgei Rothschild

Figs. 57 and 61, 58 and 62. hkilimanjarica ssp. n., Type g and Type 2 (Tanzania [Tan-
ganyika]: West Kilimanjaro, 6—7,000 ft.), A. Brown and J. G. Williams, upper and undersides.
Types in B.M. (N.H.).

Fics. 59 and 63, rydoni ssp. n., Type $ (Tanzania [Tanganyika] : Usambara Range, Magamba
forest, Lushoto), A. Rydon leg., upper and underside . Type in B.M. (N.H.).

Fics. 60 and 64, levicki Poulton, g and 2 (Tanzania [Tanganyika]: Southern Highlands,
Njombe). C. H. McCleery leg., upper and undersides.

Approx. 2/3 natural size.

PLATE 7

Bull. Br. Mus. nat. Hist. Entomology, 18, 9

PLATE <8
Charaxes

Fics. 65 and 69, 66 and 70. ansorgei kungwensis ssp. n., Holotype g¢ and Allotype 92
(Tanzania [Tanganyika]: Eastern side of Lake Tanganyika, Kigoma District, Buholo area,
Kungwe Peninsular, Mahali Mountains), T. H. E. Jackson, leg., upper and undersides. Types
in B.M. (N.H.).

Fics. 67 and 71. pollux geminus Rothschild, Type g (Malawi [Nyasaland]: Lauderdale),
upper and underside.

Fics. 68 and 72. pollux geminus Rothschild, g and 2 topotypical form b (Malawi [Nyasaland] :
Zomba), upper and undersides.

Approx. 2/3 natural size.

Bull. Br. Mus. nat. Hist. Entomology, 18, 9

PLATE 9
Charaxes pollux Cramer

Fics. 73 and 77, 74 and 78. gazanus ssp. n., Holotype g and Allotype 2 (Portuguese East
Africa: Amatongas), upper and undersides.

Fics. 75 and 79, 76 and 80. pollux Cramer, § and 2 (W. Uganda: Bwamba Valley), upper
and undersides.

Approx. 2/3 natural size.

PLATE 9

Bull. Br. Mus. nat. Hist. Entomology, 18, 9

INDEX TO VOLUME XVIII

New taxonomic names are in bold type

abyssinica, Antanartia . : : cee SE
abyssinica abyssinica, Antanartia 31, 40 (fig.),

42 (fig.), 43 (fig.) Pl. 3
abyssinica jacksoni, Antanartia . 31, Pl. 3
abyssinica vansomereni, Antanartia 31, Pl. 3

aculeata, Macromea ; : - <0 176
adami, Ceylonosticta i 2 : = 1176
adami, Libellago . ; , : eee 7
adjuncti, Polythore : : é oe177
aegypticus, Orosius . : ; 28250
africanus, Echinopterogomphus : a 3h77
albimacula albimacula, Horaga : err
albimacula albistigmata, Horaga 2129
albimacula anara, Horaga . ; 4 Watrye)
albimacula anytus, Horaga. : Sat29
albimacula bilineata, Horaga ; . 130
albimacula chalcedonyx, Horaga kzo
albimacula malaya, Horaga . : 128
albimacula taweya, Horaga T2900) a
albimacula triumphalis, Horaga. Ant An ea te
albimacula viola, Horaga é : e127
albimacula violetta, Horaga : 2128
albistigma, Ischnura : : : eee 177
albistyla, Gynacantha . : : of LT,
albofasciata, Lestes ; . ; sai,
alpinus, Charaxes . : 98
alpinus alpinus, Charaxes 84-8 Bh PL. 12
alpinus nyikensis, Charaxes 85-86, Pl. 12
amethysta amethysta, Horaga : Ts
amethysta isna, Horaga . ney sPlea 3
amethysta overdijkinki, Horaga é qeeekS2
amethysta purpurescens, Horaga . Asc
amor, Rathinda I0g-I10, Pl. 1, 2
amphidon, Ancistrotermes . : : 9
angustior, Rhipidocephala . 170-171
anascephala, Drepanosticta . : E77,
angularis, Lestes_ . A : . e077,
annaimallaiensis, Macromia_ . : 2 7d
annandalei, Caconeura . : : ae Wik)
anomala, Indolestes ; : , athe!
ansorgei ansorgei, Charaxes 303; F126
ansorgei jacksoni, Charaxes . 304, Pl. 6

ansorgei kilimanjarica, Charaxes 309, Pl. 7
ansorgei kungwensis, Charaxes 306-307, Pl. 6

ansorgei levicki, Charaxes 207. la7
ansorgei ruanduna, Charaxes . 205; bls 6
ansorgei rydoni, Charaxes . 309, Pl. 7
antelopoides, Protosticta , i 178
apiaensis, Gynacantha . : : 7S
apicalis, Chloroneura : : i mL S
apicalis, Gynacantha : ‘ F eai7S
aquicola, Agriogomphus . : é wal79
arachnoides, Pseudocopera : , a 79

argentatus, Nesophrosyne . ' eat

argentatus, Orosius
argentatus, Orosius

argentatus novaebritanniae, Orosius . 245

armageddoni, Chlorocypha
armstrongi, Amorphostigma
arthuri, Mortonagrion
asiatica, Indophlebia
assamica, Enallagma
assamica, Indolestes
atrocyana, Agrion .
attenuatum, Aciagrion
atuberculata, Macromia .
aubyni aubyni, Charaxes
aubyni australis, Charaxes
aubyni ecketti, Charaxes
aurantiacum, Ceriagrion .
auranticum, Ceriagrion
aurea, Tetrathemis.
auricolor, Amorphostigma
auricolor, Gomphus
autumnalis, Caconeura
azureum, Aciagrion

bainbriggei, Gynacantha.
bakeri, Ceriagrion .
beesoni, Rhinocypha
bellicosa, Macromia

bellula albimacula, Horaga

berkeleyi, Charaxes
bicornutus, Gomphus
bidentatum, Ceriagrion
bidentatus, Leptogomphus
bifenestrata, Rhinocypha
bifida, Tetrathemis
biguttata, Cephalaeschna
biharica, Gynacantha
bilineata, Melanoneura
binocellata, Agriocnemis
binocellata, Macromia
bispina, Macromia .
borbonica, Antanartia

borbonica mauritiana, Antanartia
borikhanensis, Macrogomphus .

botti, Caconeura
burmicus, Microgomphus
butoloensis, Notogomphus
buxtoni, Ischnura .

cacharicus, Lamellogomphus
cacuthis, Charaxes .

caffra, Rhipidocephala
calcipennis, Microstigma.
campioni, Disparoneura 3

242
245

179
179
179
179
180
180
180
180
180
7778, PL 10
79-80, Pl. 10
ia PISO
180
180
180
180
181
181
181

181
181
181
181
128
74-71 ‘PL
182
182
182
182
182
182
182
182
183
183
183

P 30
80;,Ele3
183
183
184
184
184

184

‘ 70
162-163

: 185
‘ ~ 185

318 INDEX

campioni, Orogomphus . : : » 285
canberrensis, Orosius . ‘ 5 Deny
canningi, Caconeura : z 5 . 185
cantonis, Orosius . ; ; f nt 247
cardinalis, Pseudophaea . : - . 185
carpenteri, Nilogomphus. : : . 185
carpenteri, Oxythemis . : ; 050
carpenteri, Tetrathemis . < : . 186
caudalis, Anisogomphus . : : . 186
cauvericus, Burmagomphus_ . : 186
cavithorax, Ancistrotermes 8 (figs.), Q-II
cellulosa, Orosius . : ; ; 230
cerinostigma, Protosticta ; F . 186
cervus, Heliogomphus_ . : : « 186
chichibui, Gomphus : : : . 186
ciniata ciniata, Horaga 132-134
clauseni, Agriocnemis . 3 : Aue &=10)
coerulescens, Caconeura . : . 186
congoiensis, Rhipidocephala 164-165
corniculum, Horaga . : ; - “120
corona, Idionyx . ; ‘ : a LO7
coryndoni, Lokia . : : : 087
crucifer, Ancistrotermes . 7 (fig.), 8 (fig.), 11-13
cupricincta, Macromia_ . ; : aad 57)
cupricolor, Hemicordulia : : 137,
curiosa, Protosticta : ; : pe tO?
cuzneri, Rathinda : : ; 2 nS
cyaneofrons, Gomphus . : : Se ees
cyaneovittata, Esme : ; : eae itesy,
dabreui, Agriocnemis ri 4 5 . 188
davenporti, Ceylonolestes ; ; . 188
davenporti, Protosticta . : ; . 188
decoloratum, Libyagrion. : ; 188
delius delius, Antanartia. 26, 40 (fig.), 41 (fig.),
43 (fig.) Pl. 1

delius guineensis, Antanartia 27, 40 (fig.),
2d ear

delius f. kamitugensis, Antanartia : 26
denticauda, Tetrathemis . . - . 189
diminutivus, Onychogomphus . : = 3239

dimorphica aethiopica, Antanartia 35, Pl. 5
dimorphica comoroica, Antanartia 35, Pl. 5
dimorphica dimorphica, Antanartia

33, 40 (fig.), 42 (fig.), 43 (fig.)

dimorphica mortoni, Antanartia. 36, Pl. 5
dingavani, Onychogomphus_. : - 189
disarmatus, Lestes . , ‘ : : 189
distincta, Rhipidocephala 157-158
divestita, Rhipidocephala_ . , 156
dobsoni, Agriocnemis : s . 189
dolorosa, Pacificagrion . , : . 189
donaldi, Anaciaeschna . : F «= SO
donaldi, Indomacromia . : . « 189
doornensis, Rhipidocephala : » 160
dorothea, Lestes . : , ; 4 190
drummondi, Lamellogomphus . p . 190
duarensis, Burmagomphus ‘ . «, / E90
duaricus, Onychogomphus : : . 190
dubia, Porpacithemis F ; : . 190

dundomajoricus, Phyllogomphus_ . == 5QO
dundominusculus, Phyllogomphus . . 190
echinoccipitalis, Onychogomphus_. - 190
ellisoni, Macromia . : : ; . 190
eltoni, Phyllothemis ; : : SAGE
emolus, Anthene . ‘ ; ; i 259
emolus javanus, Anthene z : - 259
emolus minor, Anthene : : - 259
emolus modesta, Anthene . : 259
engeli, Rhipidocephala 166-1 #
equatorius, Ancistrotermes ., 9 ig, 2

etesipe abyssinicus, Charaxes . 61-62, Pl. :
etesipe etesipe, Charaxes. 59-61, Pls. 4-5
etesipe gordoni, Charaxes 63, PI..6

etesipe patrizii, Charaxes : : a Oe

etesipe pemba, Charaxes 64, Pl. 6
etesipe tavetensis, Charaxes 62, Pl. 6
ethalion ethalion, Charaxes Zoi, Eveck

ethalion ethalion fm. aurantimaeula,
Charaxes . 283
ethalion ethalion fin, ethalian. Charades
282. Pl.2%
ethalion ethalion fm. rosae, Charaxes. 284, Pl. 1
ethalion ethalion fm. swynnertoni, Charaxes

2835 PL 2

ethalion kikuyuensis, Charaxes 296-297, Pl. 4

ethalion littoralis, Charaxes ; . 294
ethalion marsabitensis, Charaxes

298-299, Pl. 5

ethalion nyanzae, Charaxes. +, 202

ethalion nyasana, Charaxes. 286, Pt 2.24

ethalion nyasana fm. cithaeronoides,

Charaxes . - 287, Pl. 3
ethalion nyasana fm. demaculata,
Charaxes . 288, Pl. 3

ethalion nyasana fin. imitans, Charaxes

290, Pl. 2
ethalion nyasana fm. nyasana, Charaxes . 286
ethalion nyasana fm. suppressa,

Charaxes . ‘ 288, Pl. 3
ethalion pondoensis, Charaxes 285, Pl. 1
ethelae, Dysphaea . é : , <, TOL
excelsa, Orolestes . : ; “ «, Lon
extraordinata, Atoconeura ; : . %Or
filigranus, Thamnotettix . : 7 230
fimbriata, Rhipidocephala . 7 . 6X67
flaviceps, Burmargiolestes : , - Toe
flavicolor, Heterogomphus : : « i0e
flavifrons, Notogomphus. : : eee
flavipes, Rhipidocephala . : : . 159
flavocolorata, Macromia . A ; +. £62
flavovittata, Macromia . ; ; . TOr
fletcheri, Gomphidia ; be : :, £a2
fletcheri, Petaliaeschna . é ; ». 7192
fujiacus, Gomphus . ; 5 : . ees
fujiama, Davidius . f 193

fulgurata, Charaxes 92-84, 98, Pl. 15-16
fulva, Rhipidocephala ’ . 158

INDEX 319

galeata, Idionyx . , . 193
gallagheri, Charaxes - 94-96, ai Piers
gardneri, Hylaeothemis . : 194
gautama, Indagrion : ‘ ; . 194
gigantica, Anotogaster . ; : Seeiey
gloriosa, Dysphaea . : : : . 194
gracillima, Hemicordulia. : ; 194
guineensis, Ancistrotermes 8 (figs. ‘ 14-15
gynostylus, Cyclogomphus 3 ‘ » 194
haemastigma, Ischnura . : ; Pea Gey
halba, Horaga _. : : : 2.120
halei, Macromia_. : : : . 194
hamulata, Argia. F : . 194
hannyngtoni, Uicterovomuhor ; : ey £5LO4
hanumana, Gynacantha . : : - 195
hasimaricus, Burmagomphus . A = LOS
hearseyi, Protosticta 4 ‘ : . 195
helena, Indolestes . : ; ; SOs
hermionae, Allogaster . : : . 195
hetaerinoides, Leucopteryx é . 195
hilaryae, Rhinocypha_ : : . 195
hincksi, Mnesarete . ? : ; 24) 15QO
hippomene, Antanartia . : ase

hippomene hippomene, Antanartia 32, 40 (fig.),

42 (fig.), 43 (fig.)
hippomene madegassorum, Antanartia 33, Pl. 4

holothura, Horaga : : : + 120
Horaga . F ; F ‘ I10—-137
huallaga, Argia z : ‘ j . 196
huonensis, Anaciaeschna. ; : 5 2196
icteroptera, Mnais . A : ? 1 a9:
ida, Macromia ; - P 3 . 196
imbricata, Idionyx . : , ‘ . 196
immaculata, Zygonyx . , , . 196
incisura, Gynacantha . : LOO
inconspicua, Rhipidocephala ; ily
indica, Hylaeothemis . : . 196
indica, Indolestes . : ‘ 2 = FLO7,
indica, Macromia . ‘ : Eko”
indicum, Pseudagrion . ; 2 . -197
inflata, Ceylonolestes ; : F 197,
infrequentula, Argia ; : : SeeCLO?,
inglisi, Lamellogomphus . E : e197,
inglisi, Stylogomphus_ : : . 198
intermedia, Protorthemis : ; los
interrupta, Agriocnemis . ‘ é 3 LO8
intricata, Idionyx . : . ‘ . 198
irata, Macromia . : 5 : A Ke}
irma, Megalestes_ . ; : : . 198
isis, Zygonyx. : : : : « 198
kalarensis, Heliogomphus : : . 198
karkloof, Charaxes . : 98
karkloof capensis, Charaxes 83-84, PL Tr
karkloof karkloof, Charaxes 81-83, Pl. 11
kerri, Amphithemis ; : z a 198
kerri, Macrogomphus : . 199
kerri, Onychogomphus . : : mae ele)

kimminsi, Chlorogomphus ; : » 199

kimminsi, Macromia , ; F 3 299
klossi, Anotogaster. : é : 2 LOO
kodaguensis, Gomphidia . : , “7 uLO9
krishna, Aciagrion . : : : 5 £99,
kumaonensis, Davidius . ; ; te LOO
lachrymosa, Pacificagrion é : . 200
laidlawi, Burmagomphus : : 9200)
laidlawi, Idionyx . ; : , . 200
laidlawi, Pseudagrion . ; : 7 200
laidlawi, Rhinocypha_ . ; : =» 200
lankanensis, Ceylonosticta : 4 3.200
laosica, Calopteryx. : ; : . 200
latinotus, Ancistrotermes 8 (figs.), 15-17
lefebvrei lefebvrei, Horaga : ; 2 135
lefebvrei osma, Horaga . . ; won
lefebvrei osmana, Horaga . (2130. Pirai.3
lembomboensis, Ancistrotermes. : 15
leoni, Phyllomacromia . : : en 200
leucostigma, Disparoneura : : +) 1200
libyana, Gynacantha ; : : iL ecOr
licates, Anthene . ; : ez 7O
licates addenda, Anthene : : Peat
licates dusuntua, Anthene : : Bei Bi
licates licates, Anthene . ; ; eet
licates philetas, Anthene. : Sy api
lieftincki, Heliogomphus . : ; e201
lieftincki, Macromia - 3 ‘ ar 20r
lieftincki, Rhodothemis . ; ; 2200
lilliputians, Microgomphus : : 201
lindgreni, Onychogomphus : ‘ 2 Zor
lindgreni, Protosticta . : ; oe e201
longfieldae, Enallagma . F : 59 "201
longicauda, Bayadera. : : 7 201
longistigma, Indogomphus : ; 1, 1202
longistyla, Esme . 2 ‘ ; we 202
loogali, Microgomphus_ . : ; = 202
loringae, Philoganga : 4 ; 21202
lotophagorum, Orosius . : : me 2A
lycaenina, Anthene ; : : 200
lycaenina lycaenina, Anthene : . 269
lycaenina lycambes, Anthene . 3 . 269
lycaenina miya, Anthene : : . 269
lycaenina togata, Anthene ; ‘ . 269
lycaenina villosina, Anthene . ‘ 2270
lycaenoides, Anthene 527 O
lycaenoides bogorensis, Lycaenesthes e209
lycaenoides lycaenoides, Anthene . = 270
lycaenoides godeffroyi, Anthene . . 270
lycaenoides pegobates, Anthene . RiPerie:
lycaenoides sutrana, Anthene : eo
lycaenolus, Anthene . : ; 2 e267
lyttoni, Gynacantha : : . . 202
mackwoodi, Caconeura . : : = 202
maenala syrinx, Horaga . 120-121, Pl. 1
magnus, Paragomphus . : : a E202
makoka, Argia é : : e202
malabarensis, Epophthalmia : : 203
malabarensis, Lamellogomphus : sy, ZOOS
malabarica, Lestes : : : i203

320

malabaricum, Pseudagrion
malcolmi, Brachydiplax .
malloryi, Davidius .
manica, Charaxes . :
manicata, i aa
Margaritola

marieps, Charaxes .
martini, Charaxes
martini, Davidioides
martini, Platygomphus
mascarenica, Nesocordulia
medusa, Protosticta
melera, Horaga

microdens, Anciatrotermes :

mildredae, Ischnura
millardi, Gynacantha
miniata, Macromia
minima, Idionyx .
mirabilis, Mar¢garitola
mirabilis, Rhipidocephala
mishuyaca, Argia
mollusca, Argia
montana, Coeliccia.
montanus, Phyllogomphus
morio, Rhipidocephala
mortoni, Protosticta
moundi, Phyllogomphus .
mudiensis, Esme
multinervosa, Devadatta

nadganiensis, Idionyx
naia, Agriocnemis .
nietneri, Ceylonosticta
nigra, Caconeura
nigrescens, Vestalis
nigriceps, Lestes
nigricolor, Hagenius
nigroflavum, Ceriagrion .
nigrolineatus, Anax
nilgiriensis, Phyllomacromia
nocturnalis, Periaeschna .

obscura, Indoneura

obscurata, Rhipidocephala ;

occidentalis, Phyllogomphus
odoneli, Caconeura .
odoneli, Gomphus .
odoneli, Gynacantha
olympicus, Chlorogomphus
onyx akronyx, Horaga
onyx cingalensis, Horaga
onyx fruhstorferi, Horaga
onyx moltrechti, Horaga
onyx onyx, Horaga

onyx rana, Horaga

onyx sardonyx, Horaga .
onyx zuniga, Horaga
onyxitis, Horaga . .
orientalis, ola a ge
ornata, Idionyx

F 80-81, 98, Pl. 11

INDEX

203
203
203

86-91, 98, Pl. 13-14

167-168
149

96-07, Pl..16
‘ 203
203

203

203

. i. E35
8 (fig.), 17-18
204

204

204

204

153

153-4

«- 204

204

204

204

163

205

205

205

205

205
205
206
206
206
206
206
206
206
207
207

207
154

207

207

207

207

207

123, Pht

F II4-115
117-118, Pl. 1
: <- PES
125, 0Phr

116

II6-117

116

128-129

207

208

pacifica, Hemicordulia
pallida, Macromia . :
pallidistigma, Calilestes .
pallidum, Ceriagrion
paludensis, Aciagrion
papavarina, Hetaerina
paradoxa, Charaxes

paraffinis emoloides, Anthene
paraffinis matthias, Anthene

paraffinis nereia, Anthene
paraffinis nissani, Anthene .

paraffinis paraffinis, Anthene

Paroxynoton

patricia, Lestes

pembanus, Charaxes
penricei dealbata, Charaxes
penricei penricei, Charaxes

penricei tanganyikae, Charaxes .
8 (fig.), 18-20

periphrasis, Ancistrotermes
periyashola, Idionyx
peruviensis, Protoneura .
philo philo, Anthene .
philo scintillans, Anthene
pinheyi, Lestes

pinheyi, Nepogomphoides
platyceps, Gomphidia
platystigma, Ceriagrion .
pollux gazanus, Charaxes
pollux geminus, Charaxes
pollux maua, Charaxes
pollux pollux, Charaxes .

pollux zingense, Charaxes .

polychromatica, Drepenosticta
poungyi, Coeliccia .
praeclarum, Pseudagrion
prateri, Pseudotramea
preciosus, Orogomphus
protostictoides, Protoneura
pruinans, Heliogomphus .
pruinosa, Metacnemis
pulcherrima, Ceylonicolestes
pulcherrimus, Onychogomphus
punctulata, Holcocephala

quadrifaria, Rhipidocephala
quadrilateralis, Aeschna .

ramburi, Crenagriocnemis
ramburi, Indoneura
rarasana, Horaga

reducta, Uracis
rhinoceroides, Idionyx

risi, Gomphus

risi, Indoneura

robertsi, Notiothemis
robusta, Protosticta
rufimargo, Pseudodipsas
rusticatus, Mesogomphus
ruwensoriensis, Tetrathemis

208
208
208
208
208
209

67-68, 70, Pl. 8

263

$204

266, Pl. 1-2
264

263

149

209

70-72, Hl, :
. 66, Pl. 7, 8

65, Pl. 7
67, Pl. 8

209
209
266
267
209
209
209
209
314, Pl. 9
313, Pl. 8
316
311
313
209
209
209
210
210
210
210
210
210
210
152

169

210

211
211
136 Pl. Dye
211
211
211
211
211
211
272
211
211

saffronata, Idionyx
samoensis, Pseudagrion .
sanguinostigma, Ischnura
sanguinostigma, Protosticta
schaenia diluta, Antanartia

schaeneia dubia, Antanartia
schaeneia schaeneia, Antanartia

INDEX

Zi2
212

212

212

29, Pl. 3
20 bl 2G
28, 40 (fig.),

41 (Gg.), 43 (8), Pl x

schmidti, Onychogomphus 212
scutata, Rhipidocephala 166
seductus, Macrogomphus : ip
selina, Horaga 134, bl. 1,23
seltuttus, Anthene . j . 259
seltuttus affinis, Anthene : of 202
seltuttus amboinensis, Anthene . : 261
seltuttus keyensis, Anthene 201
seltuttus seltuttus, Anthene 262
seltuttus violacea, Anthene . 262
selysi, Chlorogomphus : 213
semitestacea, Rhipidocephala . I 5a 53
shanensis, Macromidia 213
siamensis, Burmagomphus 213
siamensis, Pseudagrion 213
signata, Rhipidocephala . 163-164
sinuatus, Burmagomphus 213
smithi, Crocorthetrum 213
souteri, Disparoneura 213
spatula, Lestes ; 214
speciosa, Rhipidocephala 168
spencei, Pseudagrion eae |
stevensi, Davidius . 214
stevensi, Idionyx 214
stevensi, Protosticta 214
stevensoni, Gynacantha . 214
striatus, Onychogomphus 214
subaequistyla, Copera 214
submontana, Ceylonosticta 215
subplatystyla, Anisopleura 215
subtropica, Ceylonosticta 215
superbus, Megalogomphus 215
superplatypes, Copera 215
sylvatica, Macromia 215
syrinx artontes, Horaga . Host 20, Gor I
syrinx camiguina, Horaga . 123
syrinx decolor, Horaga 122
syrinx joloana, Horaga 122
syrinx moulmeina, Horaga 119
syrinx onychina, Horaga 121
syrinx paulla, Horaga. 122-123
syrinx permagna, Horaga : 124
syrinx pririgna, Horaga ‘ yey:
syrinx samoena, Horaga 124-125
syrinx schoutensis, Horaga. 126, Pl. 1
syrinx sikkima, Horaga . 119g
syrinx syrinx, Horaga 125

t-coerulea, Indoneura
tenaculatus, Libyogomphus
tenera, Rhipidocephala
theebawi, Caconeura 3
tigrina, Rhipidocephala .
tigrinum, Paroxynoton
tillyardi, Palaeothemis .
tonkinicus, Onychogomphus
travancorensis, Idionyx .
trifoliata, Argia

trilobata, Protosticta
trinervulata, Heliaeschna
trituberculata, Macromia
trotteri, Pseudocopera

umbriaca, Argia ‘
umbripennis, Rhipidocephala :
uncata, Protostrata
unguiculata, Idionyx
unifasciata, Periaeschna .
usambarae, Charaxes .

vallisi, Phasmosticta
varralli, Mortonagrion
veronica, Indolestes
versicolor, Telebasis
v-flavum, Burmagomphus
victoria, Agriocnemis
villosa, Anthene .
vinsoni, Ischnura

violetta, Charaxes .
violetta maritima, Charaxes
violetta meru, Charaxes .
violetta violetta, Charaxes
violacea, Coenagrion
viridata, Rhenagrion
viridis, Drepanosticta
vitrinella, Rhinocypha

walli, Ceylonosticta

walli, Dysphaea

walli, Heliogomphus

walli, Onychogomphus
wasmanni, Ancistrotermes
wheeleri, Gomphidictinus
wilkinsi, Cyclogomphus .
williamsoni, Gomphidia .
williamsoni, Pseudagrion
wynaadicus, Macrogomphus

xanthoptera, Orogomphus

zallorensis delineatus, Davidius
zeylanica, Macromia :
zumpti, Rhipidocephala .

321

215
216
161
216
156
156-157
216
216
216
216
216
216
216
216

217
159-161
217

217

2 ZL,
"73-74, Pl. 9

217
217

217

217,

217;

217

27%

217

Pe ele
Lyrae SA}
53, Pl. 4
50, Pls. 1-3
218

218

218

218

218
218
218
218

20
218
218
218
219
219

219

188
219
161-162

4 meh ad
eee 3a

CU aks qlee

ed (4 Cnty Ban

’ Dette ale al

feel
/
1 | eT |
'
‘
r
;
+
cu a
: i:

aa by J ai “
nee gr “7c
i :

es 7w*ae

t Pot
s »
wR
age
7

eS

A LIST OF SUPPLEMENTS
‘TO THE ENTOMOLOGICAL SERIES
OF THE BULLETIN OF
THE BRITISH MUSEUM (NATURAL HISTORY)

1. MASNER, L. The types of Proctotrupoidea (Hymenoptera) in the British
Museum (Natural History) and in the Hope Department of Entomology, © -ford.
Pp. 143. February, 1965. £5.

2. Nixon, G. E. J. A reclassification of the tribe Microgasterini (Hymenoptera :
Braconidae). Pp. 284; 348 Text-figures. August, 1965. 6. |

3. Watson, A. A revision of the Ethiopian Drepanidae (Lepidoptera). Pp. 177;
18 plates, 270 Text-figures. August, 1965. £4 4s.

4. SANDS, W. A. A revision of the Termite Subfamily Nasutitermitinae (Isoptera,
Termitidae) from the Ethiopian Region. Pp. 172; 500 Text-figures. October,
1965. £3 5s.

5. AHMAD, I. The Leptocorisinae (Heteroptera : Alydidae) of the World. Pp. 156;
475 Text-figures. November, 1965. {2 I5s.

6. OKADA, T. Diptera from Nepal. Cryptochaetidae, Diastatidae & Drosophilidae.
Pp. 129; 328 Text-figures. £3.

7. GILIOMEE, J. H. Morphology and Taxonomy of Adult Males of the Family
Coccidae (Homoptera : Coccoidea). In press.

8. FLETCHER, D. S. A revision of the Ethiopian species and a check list of the
world species of Cleora (Lepidoptera : Geometridae). In press.

g. HemMminG, A. F. The Generic Names of the Butterflies and their type-species
(Lepidoptera: Rhopalocera). In press.

Io. STEMPFFER, H. The Genera of the African Lycaenidae (Leipdoptera : Rhopa-

locera). In press.

PRINTED IN GREAT BRITAIN BY ADLARD & SON LIMITED, BARTHOLOMEW PRESS, DORKING

rte! ;
i beets i r M
Shiba ih
} 3 $ ; sits ; f t :
; ath # ‘
7: + tii . 3 i) :
it : } at
i t : i { Hite itt
tH settes ridteta tity +3 it
sitetiets Siiteaitistes Catia att
i itis } tie ? He ait ne itt $3 Hi HatEsttS : sash ;
ry . Bras 4 : esihstee titre inal
4 : $38! it sisetitet ist ‘ Steas 4 itp Bh Hits ais >
iH Bate tterttt ents $s Hes eeatietatsetasesrtities Spat ereteia gto
t tt sairineaptrertsetan ais ttt Min ssseattt tf $ i st ttet
* +t + iE: soh4 St i + it a tf +4 33 iY t iS $ i sesh
Http aanaaeat eagiRaT ERR nat CTR EE ertretiestatestt ; { PT
HH tteastserst rs } t tt sf : ph etot
st . * wat +4 tetas > it ete ts it + tht ta si $4 e st ah efseeees
: : RUEEH IEE HU InSIE ante tite bess tacatguirsenstatseuataetiteteieietata tet : eitreaetateth
Hid ttettis Hn Hina sit Rt eertatetaiith mpseapestaenees itt tee
Ht astaagtt bosLSSH Sti tisest set tiesea itersirisemtetitsetetetatt i esteraet itt
aah naa tata neue Hine guint tia Hema
Testes . $4 piesa. cl tt *
i SURES EG aah Sea a a Saaeoaents
netgtates: heer satatitssesetesties! és i +t sith : st
eicteity He therhitstt sets! Heat HE i f $i sit $ fitheistttt erert petegetee seat
N t it; 4 sat tH eet et bt ; eke $ prise ipestarstitotie ietattese $4 Sree (oie
pieienert STR a Ren aut Gt Gataieurtr re tstutaL isthe teeth
ope cae 7 > os hTiStec se l@iseesses oe >
STE bass tates iii iewesettii fetid pariteetttiagasit felgrarseastitinisrstenertitate : tet : ey
titi HES 7 sgeacratsateetagraegt tes rtiteiatte seiesitst ist f i iitratiratias peters
48 4 t ¢: Hse ey +4 th +i} ttt ue t ? s Saish , ite 32 :
fete Het Hater i aL ester tesetauaes stig isecsstartesitevcatgnites esesetecatiteeentstat Hite ay
; it +t Halaitntstitt ates 3 Hitstisit oeit deseseesaierses ts fester tite ots
Hi apvet ersten at titsithn i titiesten bataterttteetettatet ett spurianettrtts it
Sitges setanttetiat Ht ; ; Shige 2 : pete ets, ‘
Hit i seispstsectatecgssesey # th se .
netiatetntaeststatttt Riftidisiesnig siti titt ; {
EHH ERU HUANG TE pati aan
ibis tet ar Hine Ht ut it ; i +t Streit ie see $3
figs teteeiecptarstacseeitaasnigs 3 eas i
ty tt HSU RSSiateeathessiteatast ites $i} stustesarapeccetaeas:
Ht Mitoetustttsseeeueite cr stitatp tat
stipite itt aigeesiatecstitasesse taag iets $esi33 + piartse rte, =
HSE i Telitpaititatise re iefate Hasisasiss Beatie
i sifee leastel : tbrerbsrebestetstereegtaiees tgs
Hititettets an sts iH ii ; patients Pirstemtasteiocs
Bins iit stat tin paetecacmst
ati: tte Hepat rt xt tyerehesat setts
seteitt $ betr ee Sait} Sratstre
satartitssisistetatat tarts it a tore
ce i: . :
saatitises saaatsesees ; prtiesste
rreetaetigiepiessi ts 3 ibastaiss HS Ht yak
Heaateittstissteepereieeteprale : t
Bs tet sees
33 *s aiaee
: ts SUriets
: ae =e
at phan 6! hes shyt:
i ess
i i it eseet ts
Hie i settee Etats 7 psteetama ae
a tn Settettritetst peel iatertedtatstabterstitescasetyteerechedeetetapeatity oF
ats Setanta Saretentecas ae aR
Hivisrente ity Sentai touhiteseees sft
potmeninc tartans srsests

Wies ly iete

Isippestikeg ee’

Seeral

estes

ieverensteiets
pyaseesnsting

te
ees

cae en eng