Bulletin of the
British Museum (Natural History)
Entomology series Vol 45 1982
British Museum (Natural History)
London 1982
Dates of publication of the parts
No 1 27 May 1982
No 2 .... 24 June 1982
No 3 . 26 August 1982
No 4 ' . ... 30 September 1982
ISSN 0524-6431
Printed in Great Britain by Henry Ling Ltd, at the Dorset Press, Dorchester, Dorset
Contents
Entomology Volume 45
Page
No 1 A catalogue and reclassification of the Ichneumonidae (Hymen-
optera) described by C. G. Thomson
M. G. Fitton . . 1
No 2 A taxonomic review of the genus Phlebotomus (Diptera: Psychodidae)
D. J. Lewis 121
No 3 Stenomine moths of the Neotropical genus Timocratica (Oecophoridae)
Vitor O. Becker 211
No 4 Afrotropical species of the myrmicine ant genera Cardiocondyla,
Leptothorax, Melissotarsus, Messor and Cataulacus (Formicidae)
Barry Bolton 307
Bulletin of the
British Museum (Natural Hisfory)
A catalogue and reclassification of the
Ichneumonidae (Hymenoptera) described by
C. G. Thomson
M. G. Fitton
Entomology series
Vo\ 45 No 1 21 May 1982
The Bulletin of the British Museum (Natural History), instituted in 1949, is issued in four
scientific series, Botany, Entomology, Geology (incorporating Mineralogy) and Zoology, and an
Historical series.
Papers in the Bulletin are primarily the results of research carried out on the unique and
ever-growing collections of the Museum, both by the scientific staff of the Museum and by
specialists from elsewhere who make use of the Museum's resources. Many of the papers are
works of reference that will remain indispensable for years to come.
Parts are published at irregular intervals as they become ready, each is complete in itself,
available separately, and individually priced. Volumes contain about 300 pages and several
volumes may appear within a calendar year. Subscriptions may be placed for one or more of the
series on either an Annual or Per Volume basis. Prices vary according to the contents of the
individual parts. Orders and enquiries should be sent to :
Publications Sales,
British Museum (Natural History),
Cromwell Road,
London SW7 5BD,
England.
World List abbreviation: Bull. Br. Mus. nat. Hist. (Ent.)
Trustees of the British Museum (Natural Histo
The Entomology series is produced under the general editorship of the
Keeper of Entomology : Laurence A. Mound
Assistant Editor : W. Gerald Tremewan
ISSN 0524-6431
British Museum (Natural History)
Cromwell Road
London SW7 5BD
Entomology series
Vol45No 1 pp 1-119
Issued 27 May 1982
/* GENERAL +
* 3 JUN nm
A catalogue and reclassification of the Ichneumonjda£ARY
(Hymenoptera) described by C. G. Thomson \^L #£
M. G.
Department of Entomology, British Museum (Natural History), Cromwell Road, London, SW7
5BD
Contents
Synopsis 1
Introduction
C.G.Thomson
Acquisition of Thomson's collections by the University of Lund . 3
Manuscript and other material associated with the collections . . 3
The collection of Ichneumonidae 3
Labelling of specimens 4
Notes on the recognition of type-material and on the selection and designation of
lectotypes g
Thomson's use of names for subgeneric categories 9
Format and arrangement of catalogue 10
Catalogue 10
Nomenclatural summary g7
Species incorrectly attributed to Thomson 100
Acknowledgements 100
References 100
Index 105
Synopsis
The 957 nominal species of Ichneumonidae (all from the western Palaearctic region) described by C. G.
Thomson are catalogued. An attempt is made to account for the type-material of all species and the generic
placements of the species to which the names apply are established after study of the types. Types of 74
species are lost and 9 names remain nomina dubia. Lectotypes are designated for 116 species and 103 new
combinations are established. One neotype is designated and one replacement name is proposed.
Introduction
The Ichneumonidae is one of the largest families of animals. More than 10000 species have been
described from the western Palaearctic region alone. Because of their parasitic habits they are of
great economic importance and biological interest. However, studies of their 'biology' depend
upon a sound and accurate knowledge of their taxonomy. It is unfortunate that the taxonomy of
the western Palaearctic fauna is currently more confused and in need of attention than that of any
other zoogeographical region. The main reason for this is that the results of the outstanding work
over the past forty years by Henry Townes and his co-workers, on the taxonomy and classifi-
cation of the family, have now been applied to all other regions and have wrought order from
chaos. There is a firm base for future systematic studies on the family in these areas. A similar
base, in the form of comprehensive modern 'catalogues', is needed for the western Palaearctic.
The word catalogue is used with some reservation because it tends to convey the wrong im-
pression of the kind and quality of studies needed to produce such works, for a group as large
and as difficult as the Ichneumonidae. This paper on the Thomson species is intended as a
contribution to a complete catalogue of the western Palaearctic Ichneumonidae.
Bull Br. Mus. not. Hist. (Ent.) 45(1): 1-119 Issued 27 May 1982
2 M. G. FITTON
C. G. Thomson is generally acknowledged to have been one of the most able hymenopterists of
his period. He had a talent for distinguishing closely related species and he described a very large
number of new species, including 957 Ichneumonidae, all from Europe and mainly from Sweden.
However, his ability is not fully demonstrated in his publications; he lacked a type-concept; and
he neglected the proper labelling of material. The existence of these deficiencies perhaps helps to
explain how he was able to be so prolific; and, together with the recent revolutionary changes in
the classification and taxonomy of the Ichneumonidae, they now limit seriously the use which can
be made of his work. This paper attempts to place all of the species of Ichneumonidae described
by Thomson in the currently-accepted generic classification of the family. This sort of work must
precede revisionary studies because, if such studies of genera or higher taxa are to have a lasting
value, one of the essential prerequisites is a knowledge of the described species which belong to
them. Because of the vast literature this problem has bedevilled taxonomic work on many groups
of European insects, but it is especially severe in the large and difficult families of parasitic
Hymenoptera such as the Ichneumonidae.
That the work of correctly placing the already-described species of western Palaearctic Ichneu-
monidae cannot be achieved successfully, as revisionary studies are undertaken, can be dem-
onstrated easily by reference to the Thomson species. For instance, in a revision of Dichrogaster,
a small distinctive genus with nine species in Europe, Horstmann (19736) included only two of the
four Thomson species which belong in it (Horstmann, 19766). Thomson originally described
three of these species in Hemiteles and one in Phygadeuon. Recognition of the genuine types of
Thomson's species has also caused problems (the reasons for which are fully explained in the
sections on labelling of specimens and recognition of types). Of about 400 specimens designated
as lectotypes or recognised as holotypes of Thomson species between 1966 and 1978 more than
twenty-five can now be shown not to have been original material of the species concerned and
therefore to be invalid. For example, Aubert (19766: 271) designated as lectotype of Mesoleius
frontatus Thomson a specimen labelled '50', the significance of which was not stated. However,
Aubert had the handwritten label upside down; it was really 'OG', an abbreviation for
Ostergotland. Since the species was described from Ystad in Skane this specimen could not be a
type. These sorts of problems can only be solved by comprehensive studies of all species described
by an author and of his methods, collections and idiosyncracies.
The generic classification of the Ichneumonidae which is the basis of the placements given in
this paper is that published by Townes (1969; 19700; 19706; 1971). This work does not cover the
subfamily Ichneumoninae, in which case Townes, Momoi & Townes (1965) and Perkins (1959;
1960) are followed. Placements of species of Anomaloninae and Ophioninae were made by I. D.
Gauld and follow his work on these groups (Gauld, 1976; 1979). The classification of parts of the
Phygadeuontinae and Tersilochinae takes into account some changes and new genera proposed
by Horstmann (19716; 19746; 1976a; 1978). Aubert (19766), Frilli (1973) and Horstmann (1979a)
have made particular studies of the Thomson species originally described in Mesoleius, Phyga-
deuon and Hemiteles respectively. In these three genera, where I have not felt the need to check,
the generic placements are credited to these authors. All species synonymies are given on the
basis of the published opinions of competent workers (to which references are given).
C. G. Thomson
The following biographical information, relating particularly to Thomson's work on the Hyme-
noptera, is taken mainly from the obituary by Bengtsson (1900).
Carl Gustav Thomson was born in the province of Skane on 13 October 1824 and died in
Lund on 20 September 1899. He succeeded Dahlbom as curator of the entomological collections
at the University of Lund. He was extremely productive: his first paper appeared in 1851 and his
total entomological publications exceeded 8800 pages. Coleoptera were his initial interest but he
soon became involved with the work on Hymenoptera which occupied him until his death. He
was a popular teacher of entomology and students were sometimes given specimens, from his
collections, of the species dealt with in his lectures.
The Proctotrupoidea was the subject of his first important work on Hymenoptera. Between
THE ICHNEUMONIDAE DESCRIBED BY C. G. THOMSON 3
1871 and 1879 he published the five volumes of Hymenoptera Scandinaviae. The Opuscula En-
tomologica, issued in 22 parts between 1869 and 1897, included all of his major work on the
groups of Hymenoptera not covered in Hymenoptera Scandinaviae. He paid for the printing of the
Opuscula Entomologica himself. Publication ceased in 1897 because problems with his eyesight
put an end to his taxonomic work. Of the Hymenoptera, only the Formicidae and Mymaridae
did not receive his attention.
Although he described over 2400 new species (including more than 2100 Hymenoptera) he
apparently regarded his work on higher classification as more important. He dealt mainly with
the Swedish fauna and collected most of the material on which he worked himself. The 'biology'
as well as the morphology of the species interested him and he spent a lot of time in the field. In
summer he went on walking tours, mainly in southern Sweden (including most parts of Skane).
He also visited Blekinge, Halland, Smaland, Oland and Gotland. He was often accompanied by
C. D. E. Roth. At the end of the 1860s he went twice to Norrland and in 1871 visited Jamtland.
He travelled abroad to Germany and in 1872 made a long trip to Germany, Austria, Switzerland
and France, during which he visited many museums and saw important collections (including
those of Fabricius and Hartig).
Thomson did not spend much time preparing specimens, which were often pinned alive in the
field. At the time of his death his collection of Hymenoptera comprised about 80 000 specimens
representing about 7000 species and was housed in 78 cabinet drawers.
Acquisition of Thomson's collections by the University of Lund
Thomson's collections were his own private property. He himself sold his collection of Coleopt-
era (but not the 'duplicate' collection) to Berlin (see Charpentier, 1972). In November 1899, after
his death, his daughter offered the remaining collections for sale to the university. She said that
Thomson had valued the collections at between 20000 and 30000 Kr. but she asked for only
8000 Kr. Together with the written offer to the university she included a synopsis of the col-
lections. The Ichneumonidae occupied 35 cabinet drawers and there were about 30 boxes of
'duplicate' material. The collections were purchased by the university on 23 January 1900.
Manuscript and other material associated with the collections
Thomson's correspondence is deposited in the main university library in Lund. He was in contact
with workers in Sweden and in Europe, including most contemporary ichneumonid specialists.
Many of the letters are accompanied by lists of species.
The library of the Zoological Institute has Thomson's personal copies of the Opuscula En-
tomologica, etc. They contain marginal notes made by Thomson. The notes are more numerous
in the earlier parts and include new synonymy and descriptions of new species. Unfortunately,
they have not proved helpful in tracing the type-material that is apparently missing from the
collection.
The Entomology Museum of the Zoological Institute has little manuscript material that is
relevant to the Ichneumonidae. It includes the letter from Thomson's daughter offering the
collections to the university (see above) and a few lists, including one of ichneumonids from
Holmgren's collection.
The collection of Ichneumonidae
Thomson's 'main' collection of Ichneumonidae is contained at present in parts of two cabinets
(numbered 395 and 396). It occupies 50 drawers (numbered 31 to 80). The arrangement of the
collection follows the Opuscula Entomologica, thus: Ichneumonidae (drawers 31-41), Cryptidae
(41-50), Pimplidae (50-56), Agriotypidae (56), Ophionidae (56-65) and Tryphonidae (66-80).
The 'duplicate' ichneumonid material is contained in various cabinet drawers (in cabinets 398,
399, 403 and 404) and in numerous separate boxes (cigar boxes etc.). The boxes are kept in
cupboard 324. Parts of the duplicate collection (boxes as well as drawers) are arranged taxo-
4 M. G. FITTON
nomically, with labels for genera and species. In some parts the arrangement is tidy and it is
possible to relate specimens to particular labels. In other parts there is a confusion of material.
Some boxes contain material from a single collector (e.g. Lethierry); others contain an assort-
ment. The contents of some boxes are partly sorted and named. The duplicate collection includes
Dahlbom, Ljungh, Holmgren, Wesmael and Zetterstedt material. There is type-material of
Thomson species and there may be type-material of other workers species (notably Holmgren
and perhaps Wesmael). Thomson apparently received a collection of Wesmael ichneumonids
(currently in two drawers in cabinet 399). It is still 'as received', each specimen bearing a number
(1-249). No key to the numbers, and thus Wesmael's identifications, has been found.
The 'main', formal collection was arranged in its present form by Simon Bengtsson. Bengtsson
was appointed curator of the entomological collections in 1900 and one of his first duties was to
take care of the then newly-acquired Thomson collections and transfer the Hymenoptera to three
new cabinets. It is known that the formal collection corresponds to Thomson's own 'main'
collection but that the arrangement may have been changed (if necessary) to correspond with the
Opuscula Entomologica. The 'duplicate' collection appears to be as Thomson left it.
The only significant curatorial work on the collection since Bengtsson's time has been the
recognition and labelling of type-material by specialists and, more recently, the addition of labels
(in the form '1978 329') to specimens sent out on loan. These labels are not removed when the
specimens are returned to the collection and they form, in conjunction with the 'loan journal', a
useful record of borrowers of material. Some years ago a few specimens were labelled 'typ' or
'typi' (e.g. Cteniscus genalis) as part of a curatorial exercise attempting to identify types/syntypes
in the museum collections (not just the Thomson collection) (H. Andersson, pers. comm.).
There are surprisingly large numbers of specimens missing from the collection (deduced from
information on localities, sexes and specimens given in the Opuscula Entomologica). There are
several possible explanations for this but none is supported by more than circumstantial evi-
dence.
There is some Anthrenus damage in the collection but very little evidence of attacks in the
present cabinets. Perhaps badly damaged specimens, including type-material, were discarded at
some time by Thomson, or by Bengtsson at the time of the transfer to the present cabinets.
The collection was undoubtedly a 'working' collection and Thomson may have redetermined
material at various stages, changing its position in the collection. Some such displaced specimens
have already been identified as types.
It is probable that Thomson exchanged material with other European workers. As far as is
known, however, he did not give any of his Swedish material to other Swedish workers (H.
Andersson, pers. comm.), with the possible exception of specimens of common species given to
students (Bengtsson, 1900). He may have returned to other Swedish workers specimens which
they sent to him. He returned to Jensen and to Drewsen material, including types now in
Copenhagen, which they collected in Jutland and Zealand.
It is difficult to assess the effects of the transfer to new cabinets and associated curatorial work
by Bengtsson. Thomson's arrangement was almost certainly not as precise and neat as the
present one. Bengtsson replaced Thomson's handwritten 'cabinet' labels by type-written ones. In
the case of generic names Thomson's labels were concealed beneath the new ones. The species
name labels were folded and transferred to the pin of the first specimen in the species series. The
type-written labels follow the Opuscula Entomologica exactly and include the typographical
errors, e.g. Microcryptus 'arrideus' instead of 'arridens' as on Thomson's own cabinet label and
Exetastes 'guttiferri instead ot'guttifer'. In some cases Thomson's cabinet label name differs from
the published one, e.g. Catoglyptus fusiventris" instead oi'fusiformis1 — presumably he changed his
mind between writing the label and writing the manuscript for publication.
Labelling of specimens
Apart from Thomson's cabinet labels (added to specimen pins by Bengtsson (see note above) and
of no significance whatsoever in the recognition of types) material in the collection is usually very
poorly labelled. The specimen labels are generally small squares of paper with an abbreviated
THE ICHNEUMONIDAE DESCRIBED BY C. G. THOMSON 5
locality name and sometimes a date. The locality labels may be handwritten or printed. Oc-
casionally the locality name is given in full. Some specimens have no locality name or abbrevi-
ation but instead have a small square of coloured paper. Unfortunately, the meaning of only one
colour is known: green indicates Ringsjon. There are two kinds of green squares, very small dark
ones and slightly larger (up to about 3x4 mm) paler, brighter ones. Interpretation of the small
dark labels as indicating Ringsjon is now generally accepted (for example, Huggert, 1973: 107;
Aubert, 1976a: 154) and I have discovered specimens bearing both a green square and a printed
label 'Ringsio'. Similar systems of coloured labels were used by other contemporary and earlier
workers, for example Zetterstedt (R. Danielsson, pers. comm.).
Specimens sometimes also have other labels, usually of one or more of the following four
kinds: a sex sign (printed); the name of a collector or collection; an additional locality label
giving a province or country; an identification.
The style of label with a locality abbreviation used by Thomson was also popular with other
contemporary collectors — most notably C. D. E. Roth who often accompanied Thomson on his
collecting trips. Thomson's and Roth's handwriting styles were similar and most labels are
Ibrekovl* B5stad« •Osb-a Karup
narcjret'el'orp
• Rossjoholm
Kungsmarken* •Reuen
-•Lund •frS3elsan3
Kallby^RSby •TornaHallestod'Ovedskl aster
•Atnarp •Dolby
yddingesj6n/^»Holmeja
rsjo* '-'•Bokeberg
Map 1 The Swedish province of Skane showing the type-localities of species of Ichneumonidae
described by C. G. Thomson.
6 M. G. FITTON
difficult to identify with certainty as the work of Thomson, although Roth's labels usually have a
date (day/month) below the locality abbreviation. The labels are poorly written and hard to
interpret until one is familiar with the forms of individual letters and the locality names from
which the abbreviations are derived.
A list of abbreviations used for Swedish localities is given below. It is not exhaustive and relates
mainly to ichneumonid type-material. The spelling used by Thomson is given first followed by
the modern equivalent where this differs. A [?] indicates that there is doubt about the form of the
abbreviation, its equivalence to the locality given or both. [Note. The Swedish letters a, a and 6
properly follow z in the Swedish alphabet but for the purposes of alphabetical order are treated as
a, a and o respectively in this list.] The localities in Skane are shown on Map 1.
Abbreviation Locality
Alnp Alnarp, Skane
Alp see 'Alnp'
Ar Arrie, Skane
Are Areskutan, Jamtland
Bast Bastad, Skane
Bgs Bogestad = Bokestad, Skane
Bkbg Bokeberg, Skane
Bl Blekinge
Boh Bohuslan
Bohl see 'Boh'
Bok see 'Bkbg'
Boks see 'Bgs'
Bor Borringe, Skane
Bs see 'Bgs'
Dby Dalby, Skane
Deg Degeberga, Skane
Dg see 'Deg'
Dgb see 'Deg'
Ekh Ekeshult, Skane
Esp Esperod = Asperod, Skane
Fg Fogelsang = Fagelsang, Skane
Fsg see 'Fg'
G Gottland = Gotland
Gbg Goteborg
Gotl see 'G'
Gott see 'G'
Hall Halland
Hbg Helsingborg, Skane
Hels see 'His'
Hg see 'Hbg'
Hhm [?] Hassleholm [?], Skane
Hkl Herrevadskloster, Skane
Him Holmia = Stockholm area
His Halsingland
Hma Holmeja, Skane
Hme see 'Hma'
Holm see 'Him'
lisp Ilstorp, Skane
Jtl Jemtland = Jamtland
Kalm Kalmar, Smaland
Kas Kaseberga, Skane
Kfge Kjeflinge = Kavlinge, Skane
Kgsm Kungsmarken, Skane
Kpe Kempinge = Kampinge, Skane
Krp Ostra Karup, 'Halland' [ = Skane]
La Lomma, Skane
Lap Lappland = Lapland [usually assumed to be Swedish Lapland]
Lapp see 'Lap'
THE ICHNEUMONIDAE DESCRIBED BY C. G. THOMSON 7
Ld Lund, Skane
Lhn Lindholmen, Skane
Loma see 'La'
Lop Loparod, Skane
Lpl see 'Lap'
Marg Margretetorp, 'Halland' [= Skane]
Mark Markaryd, Smaland
Mol Molle, Skane
Mrki Markiehage, Skane
Norl Norrland
Norr see 'Norl' [This abbreviation is easily confused with 'Norv' = Norvegica =
Norway.]
O Oland
Oel see 'O'
O.G. Ostergothland = Ostergotland
O Got see 'O.G.'
Oke Ofvedskloster = Ovedskloster, Skane
Ort Ortofta, Skane
Pal Palsio = Palsjo, Skane
Rab Raby, Skane
Raml Ramlosa, Skane
Rfn Reften, Skane
Rhm Ryssjoholm = Rossjoholm, Skane
Ron Ronnemolla, Skane
Rost Rostanga, Skane
Rota see 'Rost'
Rshm see 'Rhm'
Rsio Ringsion = Ringsjon, Skane
Rsjo see 'Rsio'
Sbg Sofdeborg = Sovdeborg, Skane
Scan Skane
Sk see 'Scan'
Skan Skanor, Skane
Skb Skabersjo, Skane
Sm Smaland
Smol see 'Sm'
Snor see 'Skan'
Ste Stehag, Skane
Steh see 'Ste'
Stkm Stockholm
Tbg Trelleborg, Skane
Tkov, Torekov, Skane
Tn [?] Torringelund [?], Skane
Tor Torringe, Skane
Trkv see Tkov'
Tve Tvedora = Torna Hallestad, Skane
V.G. Vestergothland = Vastergotland
V.W. Vestra Wram = Vastra Vram, Skane
Witt Wittsio = Vittsjo, Skane
Wml Vermland = Varmland
W.W. see 'V.W.'
Yd Yddinge, Skane
Ydd see 'Yd'
Ys Ystad, Skane
When labelling material Thomson apparently 'bracketed' together adjacent localities. Thus,
specimens of a species stated in the description to come from Palsjo may be labelled 'Hbg'
(= Helsingborg), Palsjo being a district of Helsingborg. (Other specimens are actually labelled
'Pal'.) The terms Norrland and Lappland were used rather imprecisely and sometimes inter-
changeably by Thomson. Norrland is the area of Sweden north of and including the provinces
8 M. G. FITTON
Harjedalen and Halsingland. Species stated to come from Lappland are often labelled Norrland
and vice versa. A list of the localities which it is assumed Thomson 'bracketed' is given below.
Helsingborg includes Palsjo
fKlinta
Rmgsjon includes < _ ,
(Stehag
... . fHolmeja
Yddinge includes < „
[Bokeberg
Norrland includes
Lappland
Jamtland
Halsingland
Specimens from particular collectors or collections were known by Thomson to come from a
particular locality. For instance, Ljungh specimens come mainly from Smaland and if Smaland is
given as a locality for a species the relevant specimens may actually be labelled 'Col. L-gh' but
without a locality label. Rudolphi collection specimens originate mainly from Halsingland and
Fallen specimens from Asperod (near Kivik).
Notes on the recognition of type-material and on the
selection and designation of lectotypes
Thomson did not have a type concept in any modern nomenclatural sense. He made no attempt
to preserve or label in any particular way the specimens which were the bases of his descriptions
of new species.
Usually Thomson gives no direct details of specimens with original descriptions, only the
localities or more general areas where the species had been found, such as 'Funnen vid Holmeja i
narheten af Yddingesjon' and 'Funnen vid Degeberga i Skane'. For several species much less
precise locality information is given, for example, 'Ej sallsynt i norra och medlersta Europa', or
sometimes none at all. Localities outside Skane are often only given at the level of province
('Oland' or 'Norrland', for example) or country. In the latter case Thomson may not have been
able to decipher abbreviations, read handwritten labels or ascertain the correct geographical
positions of the localities of material obtained from foreign workers. Information in addition to
the locality, when any is given, includes habitat, date, collector, an indication of abundance and
host data. Examples, 'Funnen vid Kjeflinge i barrplanteringen', 'Funnen i September vid Ortofta
nara Lund', 'Funnen talrikt vid Ilstorp i Skane af Conservator C. D. E. Roth', 'Sallsynt; funnen
pa sandmarker pa Oland' and 'I Munchen utklackt ur Thecla Betulae af D : r Kriechbaumer'.
More precise information about the specimens themselves is given only rarely, and often when
there was only one, for example, 'Exemplaret, en hona, ar funnet pa Gottland' and 'Endast ett ex.
fran sodra Frankrike (Coll. Lethierry)'.
Thus Thomson's lack of attention to original material; the inadequate published information;
the poor labelling of specimens; and changes in the arrangement of the collection subsequent to
publication all hinder recognition of type-material. Indeed, for most species it is impossible to be
absolutely certain which specimens are types. On the other hand this combination of poor
information gives wide scope in deciding which specimens are possible types. It also leaves open
the possibility that a specimen chosen as a primary type may be shown, at some later date and in
the light of further evidence, not to have been used as the basis of an original description.
However, practical considerations justify the selection of single primary type-specimens (usually
lectotypes) to serve as stable bases for the nomenclature of the species.
Of the specimens standing under the name of a Thomson species in his collection I have
recognised as syntypes all those which are in agreement with the description and with the other
information (on localities, etc.) given in the original publication. This latter qualification would
perhaps be better expressed as lack of disagreement because, for example, where Thomson cites a
THE ICHNEUMONIDAE DESCRIBED BY C. G. THOMSON 9
locality within a province and there are specimens labelled only with the name of the province
they have been accepted as syntypes. Thomson's subsequent references to his own species are
sometimes helpful in deciding on the limits of a syntype series. Syntypes of some species are in
other collections. For a few species specimens under other names have been regarded as syntypes,
usually on the basis of precise agreement with descriptions and other information, and evidence
of changes in that part of the collection subsequent to the relevant publication.
It is not always easy, using the above criteria, to decide whether or not a particular specimen
should be included in a syntype series. Problems are caused by minor disagreements with
descriptions; illegibility of labels; interpretation of locality abbreviations on labels and of col-
oured tags; and differences between localities as published and as given on labels. Agreement of
specimens with descriptions calls for a subjective judgement, sometimes open to alternative
opinions. The questions relating to labelling and localities are dealt with in the previous section
on labelling of specimens. Each case has to be judged on its individual merits, paying attention to
the utility of the particular situation.
Thomson described many new species from single specimens (holotypes). In some cases there
can be no doubt, for example, 'Ett exemplar fran Smaland'; in others it must be inferred from the
published information together with the fact that there is only one specimen in the collection.
Some workers object to recognition of specimens of the latter kind as holotypes. In cases which I
consider doubtful I have cited such specimens as single surviving syntypes.
Any member of a syntype series is eligible for designation as lectotype of the nominal species
concerned. Lectotypes have already been designated (published) for a lot of Thomson's species.
Many others have been selected (labelled) but not yet published. Lectotypes already selected but
not published are designated in the present work, but where no lectotype has been selected details
of the syntype series are given. Only a few lectotypes have been selected and designated by me,
usually in cases where previous designations are invalid for one reason or another. Lectotype
designations are best made in the context of a complete revision of the group to which the species
concerned belongs. However, for reasons of practicality it was thought desirable to publish here
the selections already made by other workers (some as long ago as 1954).
A number of the lectotype designations already published call for comment. Several are casual
in the extreme and need careful consideration of their validity in relation to the relevant provis-
ions of the International Code of Zoological Nomenclature. Others, such as Aubert's lists (1966;
1968; 1972) of indiscriminately chosen species, have been published as ends in themselves. Much
more attention must be paid to the publication of lectotype designations. There are too many
references such as 'lecto. des Townes, 1958', not meaning that there is a designation published by
Townes in 1958 but that the specimen was labelled by Townes in that year. It often happens that
such a reference to the label is the first publication of a lectotype for that species. Are such
references valid designations? For practical reasons they are best accepted as such but the
situation is far from satisfactory. Editors of journals, as well as authors, must take some of the
blame for this state of affairs.
Thomson's use of names for subgeneric categories
Thomson's use of names for 'subgenera' was inconsistent and is difficult to interpret. This incon-
sistency is not surprising in work published over a long period but, as a consequence, determining
the original subgeneric (and sometimes generic) placements of species is not always easy.
Even when Thomson used subgenera clearly he often prefixed the name of each species with
the initial letter of the subgeneric and not the generic name, for example, in Mesochorus (1886a:
327-344). Workers who have not studied Thomson's work properly have been misled by this
practice into citing incorrectly original generic placements.
Of more importance is Thomson's habit of giving genus-group names in parentheses at various
points in his keys to species. The names used in this way are mainly those originating from
Foerster's work (see Perkins, 1962: 387). In some cases the rank of such names as subgenera is
clear because Thomson gives a 'Conspectus subgenerum' at the beginning of the genus, some-
times including synonymy, for example, Megastylus (18886: 1310). In other cases Thomson does
10 M. G. FITTON
not give such an indication or the names are given so as to apply to groups of species within
subgenera, for example, the use of Myriarthrus within Megastylus subgenus Megastylus
(1888b: 1314). Thus, it is not always clear whether the names apply to formal subgenera; formal
infra-subgeneric categories ; informal groups of species ; or are included for purposes of synony-
my. They are important because they are often the first association of species with Foerster
generic names (Perkins, 1962). Subsequently the names have been cited most frequently as
subgenera. Except for those used by Thomson infrasubgenerically they are treated uniformly as
subgenera in the present work, but because of the doubt about many of them the catalogue
section is arranged in alphabetical order of binominal, and not trinominal, combinations.
Format and arrangement of catalogue
The catalogue section is arranged in alphabetical order of the original binominal combination
(that is, disregarding any subgeneric component of the name). The nomenclatural summary
which follows the catalogue and the index provide entry into the catalogue via species names,
subgeneric placements and current combinations.
For each nominal species the entry is arranged in the following sequence :
Name; date and page reference of original publication; status and sex of primary type(s); type-
locality; type-depository; lectotype designation or reference to previous valid type-restriction
(when necessary).
Details of the labels on the specimen(s).
Notes.
A statement, prefixed 'Identity', on the generic placement and synonymy of the species.
The following points should be noted with regard to these data.
The name is given as published except that the orthography is altered to comply with Articles
26, 27, 28 and 32 of the Code when necessary (in which cases the form of the name as published is
also given).
Swedish type-localities are given as cited by Thomson with the addition of the names of the
country and province (when necessary) and the modern spelling (when different).
Names of type-depositories are abbreviated as in the list below. Where types are lost this is
stated in place of a depository. The collections from which Thomson described species are given
after each depository.
CM, Norwich Castle Museum, Norwich, England (Bridgman collection)
NM, Goteborg Naturhistoriska Museet, Goteborg, Sweden (G. F. Moller collection)
NR, Stockholm Naturhistoriska Riksmuseet, Stockholm, Sweden (Holmgren collection)
UZI, Lund Universitetets Zoologiska Institutionen, Lund, Sweden (Thomson collection)
ZM, Copenhagen Zoologisk Museum, Copenhagen, Denmark (Drewsen, Jensen and Wustnei
collections)
ZSBS, Munich Zoologische Sammlung des Bayerischen Staates, Munich, West Germany
(Kriechbaumer and Foerster collections)
For specimens from Thomson's own collection details are not given of labels added since
Thomson's death (except for the cabinet labels which were transferred to the first specimen in
each series by Bengtsson). For each label an indication is given (in square brackets) of whether it
is handwritten or printed. Except for a few species with large syntype series, all primary type-
specimens have individual modern labels showing their identity and status.
Catalogue
Acanthocryptus nigriceps, 1883: 868. Syntype 1 cJ, SWEDEN: Smaland, Calmar [= Kalmar], Hossmo (UZI,
Lund).
Labels. Hossmo 4/6 70 [hand] ; Kalmar [hand] ; nigriceps [Thomson cabinet label].
Gravenhorst's specimen (1829ft: 676. Phygadeuon quadrispinus Var. 1. £) [not examined] is also a
syntype of this species. Thomson mis-spelled the name of the Gravenhorst species as quadrispinosus
instead of quadrispinus.
Identity. ? Rhembobius nigriceps (Thomson).
THE ICHNEUMONIDAE DESCRIBED BY C. G. THOMSON 1 1
Acanthocryptus nigricollis, 1883: 868. Lectotype $, SWEDEN: Skane, Bastad (UZI, Lund), by designation of
Aubert, 1966: 129.
Labels. Bast [hand] ; nigricollis [Thomson cabinet label].
Identity. ? Rhembobius nigricollis (Thomson).
Adelognathus (Adelognathus) aciculatus, 1883: 879. Type(s) 9, SWEDEN : Skane, Stehag (lost).
Identity. Adelognathus aciculatus Thomson.
Adelognathus (Adelognathus) dlmidiatus, 18886: 1276. Lectotype 9, FRANCE: Raismes (UZI, Lund), by
designation of Aubert, 1972: 147.
Labels. Raismes. [hand] ; dimidiatus [hand].
Identity. Adelognathus dimidiatus Thomson.
Adelognathus (Adelognathus) facialis, 1883: 880. Holotype?, SWEDEN : Norrland (UZI, Lund).
Labels. Norl. [printed] ; facialis [Thomson cabinet label].
Identity. Adelognathus facialis Thomson.
Adelognathus (Adelognathus) fasciatus, 1883: 878. Holotype?, SWEDEN : Skane, Sofdeborg [= Sovdeborg]
(UZI, Lund).
Label. Sbg 23/7 [hand].
Identity. Adelognathus fasciatus Thomson.
Adelognathus (Adelognathus) laevicollis, 1883: 878. Syntypes 4 ?, 1 <$, SWEDEN: Skane, Ringsion [ =
Ringsjon] (UZI, Lund).
Labels. Rsio [printed] (2$). Scan [printed] (1$). [small green square]; laevicollis [Thomson cabinet
label] (1 9)- [small green square]; <$ [printed] (1 $).
Identity. Adelognathus laevicollis Thomson.
Adelognathus (Adelognathus) limbatus, 1888ft: 1275. LECTOTYPE 9, SWEDEN: Skane, Palsjo (UZI, Lund),
here designated (selected by J. F. Aubert).
Label. Pal [hand].
Identity. Junior synonym of Adelognathus brevicornis Holmgren (Perkins, 1943ft: 95, 104).
Adelognathus (Adelognathus) nigriceps, 1888ft: 1274. Type(s)$, FRANCE (lost).
From the description it seems likely that there was only one specimen, which may have been returned
to Lethierry. There are no specimens under this name in the Thomson collection.
Identity. Adelognathus nigriceps Thomson (Perkins, 1943ft: 99).
Adelognathus (Adelognathus) nigricornis, 1888ft: 1276. Type(s) 9, FRANCE (lost).
The comments on A. nigriceps apply to this species also.
Identity. Adelognathus nigricornis Thomson (Perkins, 1943ft: 100).
Adelognathus (Cnemischus) pilosus, 1888ft: 1277. Holotype9, SWEDEN : Skane, Alnarp (UZI, Lund).
Labels. Alnarp [printed] ; pilosus [Thomson cabinet label].
Identity. Adelognathus pilosus Thomson.
Adelognathus (Adelognathus) puncticoltis, 1883: 877. Holotype9, SWEDEN : Smaland (UZI, Lund).
Labels. Smol [printed] ; puncticollis [Thomson cabinet label].
Identity. Adelognathus puncticollis Thomson.
Adelognathus (Adelognathus) punctiventrls, 1883: 877. Lectotype 9, SWEDEN: Skane, Torekov (UZI, Lund),
by designation of Jussila, 1965: 31.
Label. Tkov 23/7 [hand] [not Tkro' as stated by Jussila].
Identity. Adelognathus punctiventris Thomson.
Adelognathus (Adelognathus) punctulatus, 1883: 879. Syntype 1 9, SWEDEN: Skane, Ringsion [= Ringsjon]
(UZI, Lund).
Label. Rsio [printed].
Identity. Adelognathus punctulatus Thomson.
b,
Adelognathus (Adelognathus) scabriculus, 1883: 877. Lectotype 9, SWEDEN: Lappland (UZI, Lund), by
designation of Jussila, 1965: 30.
Label. Lpl. [printed].
Identity. Junior synonym of Adelognathus tetracinctorius (Thunberg) (Jussila, 1965 : 30).
12 M. G. FITTON
Aethecerus graniger, 1891 : 1641. Syntype 1 9, SWEDEN: Skane, Ringsjon (UZI, Lund).
Label, [small green square].
The head is missing from the syntype 9-
Identity. Aethecerus graniger Thomson.
Aethecerus palttcoxa, 1891 : 1640. Syntypes 4 9, 8 & SWEDEN: Skane (UZI, Lund).
Labels. Hbg [hand] ; pallicoxa [Thomson cabinet label] (1 9, 2 $ all on one pin). Pal [hand] (3 9, 6s.iV.fio Thalia Tr.rr» 1QH1- ^8
54 M. G. FITTON
Mesochorus (Mesochorus) picticrus, 1886a: 340. LECTOTYPE 9, SWEDEN: Skane, Palsjo (UZI, Lund), here
designated (selected by R. Hinz).
Label. Pal [hand].
Identity. Mesochorus picticrus Thomson.
Mesochorus (Astiphrommus) plagiatus, 1886a: 332. Syntypes 1 ^, SWEDEN: Skane, Helsingborg (UZI,
Lund); 1 3, ENGLAND (CM, Norwich).
Labels. Hbg. [hand] ; plagiatus [Thomson cabinet label] (Lund specimen). 546 [hand, on the specimen
mount]; G. C. Bignell April 1882 from Apanteles from Odontopera bidentata [hand, on the underside of
the specimen mount] ; plagiatus Thorn [hand] ; 3 [hand] (Norwich specimen).
The Norwich specimen is in the J. B. Bridgman collection. From Bridgman's paper (1886: 335, 353 and
354) and the label with the number (3) it is virtually certain that this specimen was sent to Thomson and is
a syntype.
Identity. Astiphromma plagiatum (Thomson).
Mesochorus (Mesochorus) punctipleuris, 1886a: 334. Lectotype 9, SWEDEN: Skane, Ringsjon (UZI, Lund),
by designation of Aubert, 1966: 131.
Label. Rsio [printed].
Identity. Mesochorus punctipleuris Thomson.
Mesochorus (Mesochorus) solids, 1886a: 338. LECTOTYPE 9, SWEDEN: Skane, Ringsjon (UZI, Lund),
here designated (selected by W. Schwenke).
Labels. Rsio [printed] ; Salicis [Thomson cabinet label].
Identity. Mesochorus salicis Thomson.
Mesochorus (Astiphrommus) simplex, 1886a: 334. LECTOTYPE 9, SWEDEN: Skane, Yddinge (UZI, Lund),
here designated (selected by W. Schwenke).
Labels. Yd [hand] ; simplex [Thomson cabinet label].
Identity. Astiphromma simplex (Thomson).
Mesochorus (Mesochorus) stigmaticus, 1886a: 341. LECTOTYPE 9, DENMARK: Maribo (UZI, Lund), here
designated (selected by W. Schwenke).
Labels. 28/7 77 Maribo ex Microgaster [hand] ; stigmaticus [Thomson cabinet label].
Identity. Junior primary homonym of Mesochorus stigmaticus Brischke, 1880. Replacement name
Mesochorus orgyiae Dalla Torre, 1901 : 56.
Mesochorus (Mesochorus) temporalis, 1886a: 336. Syntype 1 9, ENGLAND (CM, Norwich).
Labels. Bred from filipendulae 25.7.78 G. C. Bignell [hand, on the underside of the specimen mount] ; 48
[hand] ; temporalis Thn [hand].
The syntype is in the Bridgman collection. For the reasons why it is considered as such see the notes
under M . plagiatus above.
Identity. Mesochorus temporalis Thomson.
Mesochorus (Astiphrommus) tenuicornis, 1886a: 332. LECTOTYPE 9, SWEDEN: Skane, Palsjo (UZI, Lund),
here designated (selected by R. Hinz).
Labels. Pal [hand] ; tenuicornis [Thomson cabinet label].
Identity. Astiphromma tenuicornis (Thomson).
Mesochorus (Mesochorus) tenuiscapus, 1886a: 341. LECTOTYPE 9, SWEDEN: Lappland, Lund (UZI,
Lund), here designated (selected by W. Schwenke).
Labels. Lund 3 Ag. [hand] ; Lpl. [printed] ; tenuiscapus [Thomson cabinet label].
Identity. Mesochorus tenuiscapus Thomson.
Mesochorus (Mesochorus) tuberculiger, 1886a: 333. Lectotype ^, SWEDEN: Skane, Torekov (UZI, Lund), by
designation of Townes, Momoi & Townes, 1965: 345.
Labels. Trkv [hand] ; tuberculiger [Thomson cabinet label].
Identity. Mesochorus tuberculiger Thomson.
Mesocryptus nigriventris, 1896: 2384. Holotype 9, SWEDEN : 'Halland' [Skane], Margretetorp (UZI, Lund).
Labels. Hall, [printed] ; nigriventris m [Thomson cabinet label].
Margretetorp is in northern Skane, not southern Halland as stated by Thomson. It is near the bound-
ary between the two provinces, so Thomson's error is easily explained.
Identity. Oresbius nigriventris (Thomson) comb. n.
THE ICHNEUMONIDAE DESCRIBED BY C. G. THOMSON 55
Mesocryptus ochrostomus, 1896: 2384. Holotype 9, SWEDEN : Skane, Palsjo (UZI, Lund).
Labels. Pal [hand] ; ochrostomus m [Thomson cabinet label].
Identity. Aptesis ochrostomus (Thomson) comb. n.
Mesoleius (Alexeter) albilabris, 1894: 2025. Syntypes 3 9, 3
(Diptera : Psychodidae)
D. J. Lewis
c/o Department of Entomology, British Museum (Natural History), Cromwell Road, London
SW7 5BD
Contents
Synopsis 121
Introduction 122
General 122
Fossil Phlebotominae 124
Distribution 125
Biology 126
Relation to disease 126
Explanation of terms 127
Various 127
Names of collectors mentioned 128
Depositories, actual, probable or original 129
Keys, citations, distribution and notes 129
Genus Phlebotomus Rondani & Berte 129
Key to the subgenera of Phlebotomus 130
Tibia 3 in certain species 131
Subgenus Spelaeophlebotomus Theodor 131
Subgenus Idiophlebotomus Quate & Fairchild 133
Subgenus Australophlebotomus Theodor 135
Subgenus Phlebotomus Rondani & Berte 137
Subgenus Paraphlebotomus Theodor 142
Subgenus Synphlebotomus Theodor 148
Subgenus Larroussius Nitzulescu 150
Subgenus Adlerius Nitzulescu 163
Subgenus Euphlebotomus Theodor 168
Subgenus Anaphlebotomus Theodor 170
Subgenus Kasaulius subgen. n 172
Nomen nudum 172
Discussion 173
Leg ratios 173
Evolution of Phlebotominae 175
Aspects of leishmanial evolution in relation to that of Phlebotominae . . .177
Acknowledgements 191
References 191
Index 207
Synopsis
The 11 subgenera (one new), 96 species (one new) and 17 subspecies of Phlebotomus are reviewed and keys
are provided for their identification. Accounts are given of fossil sandflies and of the role of Phlebotomus in
the transmission of disease. Taxonomic citations are provided for each species and subspecies, and ah
annotated distribution list referring to a map. For some species further notes are given, including references
to transmission of disease. It is suggested that 'leg ratio' is worth recording as a measure of leg length in a
readily comparable form, and that it provides additional information about certain genera, subgenera,
species and infraspecific forms. Evolutionary hypotheses are put forward to explain features of the present
distribution of Phlebotominae and leishmaniasis.
Bull. Br. Mus. not. Hist. (Ent.) 45 (2): 121-209 Issued 24 June 1982
122 D. J. LEWIS
Introduction
General
Phlebotomus Rondani & Berte is one of the two Old World genera of Phlebotominae and
includes all the habitual mammal-biters and the vectors of human leishmaniasis in the Old
World. Disease of this group have recently increased in several countries and epidemics have
followed interruption of malaria control, so that renewed concern about the diseases and new
research programmes demand up to date information about the vectors. During the past 80 years
intensive study has yielded many widely scattered publications about Phlebotomus, particularly
from leishmaniasis areas (Anonymous, 1977), and reviews of the genus in three zoogeographical
regions have been published. Many species occur in all three of them (Lewis, 19786: 311), and a
general survey is required. The present work deals with some basic aspects of Phlebotomus
throughout the Old World.
The classification of Phlebotomus and the Phlebotominae has been discussed by Abonnenc
(1972), Fairchild (1955), Lewis et al (1977), Theodor (1948; 1958) and others. I recognize the
division of the living Phlebotominae into five genera, Phlebotomus and Sergentomyia Franga &
Parrot in the Old World and Warileya Hertig (Fairchild, 1955: 183; Lewis et al., 1977: 325),
Brumptomyia Franga & Parrot and Lutzomyia Franc, a in the New World. Ready et al. (1980)
have stressed the undoubted importance of subgenus Psychodopygus Mangabeira of Lutzomyia
and treated it as a genus. Lewis et al. (1977: 324) gave reasons against such a course which would
involve the elevation to generic rank of several, much more distinctive, Old World subgenera of
sandflies and could lead to a general multiplication of genera. Such questions are among the
'pitfalls of perfection' (Nelson, 1978) and are 'handicaps of the human need to compress into
linear form the three dimensional world of nature' (Campbell, 1974: 15). Taking a world- wide
perspective, I regard Psychodopygus as an important subgenus without changing its rank.
Publications (most with keys) dealing with Phlebotomus in various areas include the following.
The Old World: Artemiev (1979: 19, Euphlebotomus; 1980, Adlerius), Lewis (1973), Theodor
(1948).
The Palaearctic Region: Artemiev (1978, key with figures for Afghanistan), Croset (1978: 713,
key with figures for Tunisia), Lewis & Biittiker, 1981, Saudi Arabia), Nadim & Javadian (1976,
Iran), Perfil'ev (1968, key with figures for the U.S.S.R.), Theodor (1958, key and figures for the
region).
The Afrotropical Region: Abonnenc (1972, key with figures), Quate (1964, Sudan).
The Oriental Region: Lewis (19786).
The Australian Region : papers by Lewis and Dyce are being completed.
The taxonomic characters are easily seen in flies mounted in gum-chloral medium which may
be ringed with Glyceel (Kevan, 1955: 417, 418; Southey, 1970: 51, 53, 56; Tribe, 1972). Potash
was hardly ever used for maceration because it weakens intersegmental membranes and makes
specimens difficult to remount. It was occasionally used for treating the tip of the abdomen to
clarify the spermathecae although it may distort the ducts.
The characters used are described by Abonnenc (1972), Artemiev (1978: 1-8), Forattini (1973),
Lewis (1973; 19786: 219), Perfil'ev (1968), Theodor (1958), Young (1979: 5-8) and many others.
Lewis's (1973) account is being amplified to include recently introduced characters, some of
which are mentioned below.
Head length may be measured from the tip of the clypeus to the most posterior parts of the
head, and eye length to include the fore and hind facets. The inter-ocular suture is of some use,
but mainly for American species. The inter-arcal area lies between the cibarial chitinous arch and
the cibarial teeth. The labrum is measured to include the anterior sensilla. The antennal papillae
(Parrot, 1953) were discussed by Wirth & Navai (1978 : fig. 5, 47). The dental depth is the distance
from the tip of the maxilla to the most proximal tooth.
The relative lengths of various leg segments have been used for classification in several groups
of insects, including Lepidoptera (Imms, 1964: 555, 556), aphids (Eastop, 1972: 173), Culicidae
(Reid, 1953: 75), Ceratopogonidae (Wirth et al, 1977: 621), Chironomidae (Pinder, 1978: 11, 19;
Saether, 1976), Mycetophilidae (Hutson & Kidd, 1975: 29; Hutson et al, 1980: 42), Cecidomy-
iidae (Panelius, 1965: 5, 132), and Phoridae (Borgmeier, 1964; Schmitz, 1957: 431, couplet 8;
THE GENUS PHLEBOTOMUS 123
1958). For the Phlebotominae, Franga (1919: 125) pointed out that leg-segment lengths of each of
the species then known varied within narrow limits, and since then many authors have recorded
the actual lengths of several or a few segments, mainly in species of Lutzomyia. French writers
have measured the hind leg of many species. Raynal (1934: 350) indicated the value of the hind
tibia-femur ratio for separating two species of Phlebotomus, and Zariquiey (1937: 417) used the
lengths of basitarsus 1 (longer or shorter than femur 1) and of all tibiae of certain species of
Phlebotomus. Theodor (1958: 4) remarked that the legs were particularly short in Palaearctic
Sergentomyia, Artemiev (1978: 4) referred to various measurements of the hind leg, and Young
(1979: 7) mentioned tibia length in Lutzomyia. L. W. Quate often recorded leg measurements
regardless of sex, implying that the sexes are similar in this respect, and other publications
indicate that differences are usually small.
In recent years some authors have recorded lengths of leg segments but not always the same
ones, some have ceased to make such records, and others have never done so. It is now time to
appraise the value of leg characters and of the time spent in measuring them. In the present work,
therefore, the lengths of the long segments of each leg, of females when possible, are recorded in a
way to allow quick comparison of species. The legs were measured at x 60, with occasional use
of x 120 to locate extensions into preceding segments, which were included. Legs detached from
the body could usually be recognized as first, second or third because tibia 2 is nearly always
longer than 1, and 3 than 2. All lengths are expressed in units of which 100 are the length of femur
1 of a particular species, and the relative lengths of the nine long segments of one side, usually of
one fly, are followed by the actual length in mm of femur 1, and of the wing in some cases. Leg
diagrams, first drawn on the scale of one unit to one centimetre (examples in Figs 1 5-24) are
useful for comparing species and picking out features of individual species for additional
measurements.
The aedeagus comprises two side pieces fused at the base (Perfil'ev, 1968: 32, 42) and protects
the tips of the sperm tubes. According to Theodor (1958: 5) these tubes are the true aedeagus, and
the 'aedeagus' strictly speaking is the aedeagus sheath. Some authors have recorded the length of
the aedeagus but without indicating the basal point from which it was measured. The most
convenient point is usually the dorsal hind end near the bases of the coxites, and if other points
are used in certain cases they can be indicated.
The last abdominal segment or proctiger of male sandflies is the ninth (Just, 1973: 314, 315,
316, 332) and shows some specific differences. Isaev (1935: 98) noted three types, in P. papatasi, in
P. sergenti and a species of Sergentomyia, and in P. chinensis, characterized by the length of the
surstyles, the nature of their junction to the segment, and the ventral shape of the latter. Appreci-
able differences are shown by the six species illustrated in Figs 8 to 14. Surstyle is a convenient
name for the lateral lobes of the ninth tergite.
Keys to the subgenera and their species are provided and should be used in conjunction with
descriptions. Taxonomic citations serve as a guide to literature on the genus, subgenera, species
and subspecies. Distribution lists of all species show the sources of information for the maps.
References to disease transmission by known or possible vector species show many which are or
may be important, and indicate publications on biology as well as disease.
For some species full lists of taxonomic citations would be unduly complex and long, and early
references are confined to a few of historic interest.
Where the original or later depository of a holotype, syntypes or other type-material is not
shown by a describer, later author or other source, it is deduced (with a query) either from the
original paper or another publication which is indicated. Some syntypes have been located with
the aid of Abonnenc (1972) although he refers to them as holotypes. Information about the
depositories of some types from Afghanistan is given by Artemiev (1978: 23). Types of species
described by Professor O. Theodor were kept in the Hadassah Hebrew University Medical
School, Jerusalem, until the collection was purchased from the University by the British Museum
(Natural History) in 1981.
Distribution data, on which the maps are based, are of three kinds, viz. information about
types, publications which give detailed information and often earlier references, and previously
unpublished records indicated by collectors' initials or 'BMNH'.
124 D. J. LEWIS
Some Chinese records were not available when this work was being prepared, and are being
assembled for publication by Professor Leng Y.-j. They include the description of 'P. major wui\
for which a preliminary note is included below under P. major, and records of P. longiductus from
Xinjiang (Wu etal., 1979).
Fossil Phlebotominae
It is appropriate to consider the fossils of Phlebotominae and their ancestors because they help to
explain the relation of Phlebotomus to other genera and to the evolution of leishmaniasis. Leish-
mania probably arose from a monoxenic flagellate parasite of the ancestors of sandflies, so there is
likely to be a phylogenetic relationship between the leishmaniae and their vectors (Saf yanova,
19776:281).
The hopping flight of sandflies doubtless caused many to be trapped in resin, and some
excellent fossil specimens have survived in several of the sources of insects in amber (Hennig,
1973: 6). Their approximate ages in MYA (millions of years ago) quoted below were supplied by
Dr P. E. S. Whalley or taken from the British Museum (Natural History) (1972) time scale, Riek
(1970) or the work of Smith & Briden (1977) which was also consulted for continental move-
ments. Wings of the following species are illustrated (Figs 20-33) to give an impression of the
groups mentioned here and later: Permotipula patricia Tillyard, 1929: 779 (Rohdendorf, 1974: 6),
Phlebotomites brevifilis Hennig, 1972: 40, 62, Phlebotomus tipuliformis Meunier (Fig. 27 after
Hennig), Warileya nigrosacculus Fairchild & Hertig, P. (Spelaeophlebotomus) minteri, P. (Idio-
phlebotomus) frondifer, Lutzomyia paterna (Quate, 1963), Brumptomyia galindoi (Fairchild &
Hertig), P. (Paraphlebotomus) sergenti, Lu. (Dampfomyia) permira (Fairchild & Hertig), Sergen-
tomyia (Neophlebotomus) gombaki (Lewis & Wharton), S. (Sergentomyia) bedfordi (Newstead), S.
(Sergentomyia)fallax (Parrot) and S. (Parvidens) lesleyae (Lewis & Kirk).
370 MY A, Devonian
The earliest known insect, a wingless form, was living about this time (Riek, 1970: 168).
230 MYA, Upper Permian
The mecopteran Permotipula exemplifies a primitive wing to which that of Nemopalpus Mac-
quart, though unrelated (Rohdendorf, 1974: 6), is remarkably similar. Nemopalpus is probably
among the most primitive living Diptera, close to the basic stock of the Psychodidae and to the
Phlebotominae in the matter of venation (Fairchild, 1955: 182; Lewis et al., 1977: 323).
The original Diptera, present at this period, were probably biting flies feeding on insects or
vertebrates and contemporaneous with the beginning of the reptile age, when the theromorph
ancestors of mammals existed before the origin of birds (Downes, 1971 : 241, 261, 262).
220 MY A, Lower Triassic
The infraorder Dictyodipteromorpha of the dipterous suborder Archidiptera was probably in
existence; it flourished in the Upper Triassic and was apparently the ancestral group which gave
rise to two branches, the infraorder Tipulimorpha Rohdendorf and all other later Diptera (Roh-
dendorf, 1974: 27, 55, 129, 136, 289, 329).
760 MYA, Middle Jurassic
The Tipulimorpha were established (Rohdendorf, 1974: 3, 291, 292) and included the tipulid
family Tanydophryneidae Rohdendorf which appears to have been ancestral to 'superfamily'
Psychodidea [fossil Psychodidae] (Rohdendorf, 1974: 3, 53, 219, 228, 291-293). This ancient
group, distinguished from all other Tipulimorpha by primitive larval features, has retained a
complex wing venation but its members have become smaller and thus been able to colonize
microhabitats (Rohdendorf, 1974: 53, 58, 292). Before the end of the Jurassic the ancient group of
the Phlebotominae, among the smallest of Diptera, must have come into existence (Hennig, 1972:
38, 55, 58), in which the origin of R2 + 3 has been displaced towards the wing tip so that the vein
seems to come from /?4, and R2 has been reduced (Hennig, 1969: 385). R2 + 3 is usually branched
only in the most primitive Diptera (Colless & McAlpine, 1970: 664). Hennig's important 1972
paper was probably based largely on previous work unpublished owing to the second world war
(Schlee, 1978:382).
THE GENUS PHLEBOTOMUS 125
720 MY A, Lower Cretaceous
The first known two species of Phlebotominae existed in what is now the Lebanon and was south
of the Tethys Sea (Hennig, 1972: 38; Melville, 1967: 293). The small, evidently primitive Phleb-
otomites longifilis Hennig, 1972: 40, 62, and Phlebotomites brevifilis Hennig, 1972: 40, 62
(Stuckenberg, 1975: 459), had wings with a broad distal half and broadly rounded tip which may
have accounted for a displacement of the origin of R2 + 3 beyond that of/?5 (Hennig, 1972: 8, 27,
39, 43, 51). Although these species show few very striking differences from some recent forms they
were included in a new genus because close relationship to Phlebotomus was not indicated.
Hennig (1972: 21, 28) considered that they might belong to the ancestral group of the Phleb-
otominae or to his probably monophyletic 'Phlebotominae s. str.' which comprises Phlebotomus,
Sergentomyia, Brumptomyia and Lutzomyia. Stuckenberg drew attention to the short palpal
segment 5 of Phlebotomites brevifilis which is like that of some American sandflies. The two
Cretaceous species and the present-day Neotropical Warileya have a similar type of wing struc-
ture and may be the sole remnants of an early movement from Africa to South America or vice
versa across a south Atlantic connection in the Lower Cretaceous or earlier (Hennig, 1972: 38, 39,
44).
30 MY A, probably Upper Eocene
One poorly described species is known from Baltic amber (Rohdendorf, 1974: 275), P. (Phleb-
otomiella) tipuliformis Meunier, 19056 [as P. tipuliformis]; 1906: 103 [as Phlebotomiella tipuli-
formis']; 1912: 71 [as P. (Phlebotomiella) tipuliformis'] (Fairchild, 1955: 183-187; Hennig, 1972:
51-55; Stuckenberg, 1975) and may have lived in the amber forest and fed on thin-skinned
reptiles (Larsson, 1978: 92, 93). Hennig regarded it as a member of his Phlebotominae s. str. and
perhaps of genus Phlebotomus and of subgenus Euphlebotomus or Anaphlebotomus, which showed
that splitting of the ancient Phlebotominae was already far advanced. Stuckenberg referred to the
short palpal segment 5 and primitive wing of P. tipuliformis and considered it to be congeneric
with Phlebotomites and somewhat intermediate between it and 'Phlebotominae s. str.\
26 M YA, Miocene
Lu. paterna (Quate, 1963: 114) (Hennig, 1972: 56, 59, 62, fig. 41) is the first known phlebotomine
with a narrow wing and is related to living reptile-feeding species.
One M YA to the present day
Philaematus pungens Loew, 1845: 8 (Parrot, 1951 : 28; Duckhouse & Lewis, 1980: 99) from copal
of unknown origin, ' Phlebotomus pungens' Meunier, 1905a: 209 (Duckhouse & Lewis, 1980: 99)
from Zanzibar copal, and S. succini (Stuckenberg, 1975: 456) (Lewis et al., 1977: 326; Duckhouse
& Lewis, 1980: 105) from copal, possibly East African, may be less than one MYA and represent
living species of Sergentomyia. Several specimens from African copal examined proved to belong
to this genus, and one, treated with xylol and mounted in Euparal, clearly shows pharyngeal
teeth, antennal ascoids and palpal sensilla.
Distribution
Quate (1962: 169, 170) regarded the Phlebotominae as tropical with northern intrusions.
Sandflies occupy most of the Old World other than cold regions and oceanic islands, and they are
absent from the Seychelles (Scott, 1933: 369), and Phlebotomus from Madagascar (Brygoo, 1974).
Sandflies are considered to need at least 50 days a year with a temperature not less than 20°C
(Perfil'ev, 1968: 98). Map 1, showing the general distribution of the subfamily in the Old World, is
based on data cited by Lewis (1974) and Leger & Rodhain (1978). In western Europe P. per-
niciosus and P. mascittii occur about as far north as 49°N, and in Asia P. chinensis is the most
northerly species (Perfil'ev, 1968: 89), reaching about 48°N (Beklemishev & Dolmatova, 1948:
354). In Canada sandflies are known from about 50°39'N near Kamloops, from 49°39'N at
Coulee Creek in Alberta, and at 44°41'N near Ottawa. The southern boundary of sandflies in the
Old and New Worlds is about 40°S (Perfil'ev, 1968: 90). The maps illustrate a mainly northern
distribution of Phlebotomus, which is discussed later. It is exemplified by the northern distri-
bution of Larroussius, and therefore of kala-azar, in Tunisia (Croset et al, 1978: 744), and by the
126 D J. LEWIS
presence of five Phlebotomus species out of six Phlebotominae in France, and two out of 26 in
Zaire (Vattier & Bimangou, 1974: 92; Vattier & Trouillet, 1975: 2; 1978: 701). Some 35 species,
including P. orientalis, have marked eastern or western limits.
Biology
Numerous publications dealing with this extensive subject may be located by reference to Abon-
nenc (1972), Lewis (1973; 1974a; 1977; 1978a; 19786), Perfil'ev (1968) and others, and notes on
various species in the present work. The following brief note refers to a few aspects.
Sandfly larvae are difficult to find and many live in soil or burrows of animals. Development
from egg to adult takes weeks or months according to temperature, and larvae undergo diapause
in some northern and other areas. Many adults of both sexes feed on sugar and the females take
vertebrate blood. Adults are active at night and rest in various shelters by day. Movement varies
from short hops to flights of a few hundred metres and occasionally nearly 2 km, and is usually
stopped by moderate wind. Palaearctic species tend to have one or two generations a year, and
some tropical ones flourish in either the dry or the wet season.
The genus Phlebotomus includes all the habitual mammal-biters and therefore all the sandfly
vectors of human disease in the Old World.
Relation to disease
The following summary of relation to disease in the Old World is supplemented by notes on some
species. The leishmaniases are the main group of sandfly-borne vertebrate infections. It seems
probable that Leishmania Ross, 1903; Wenyon, 1926: 396, having arisen as an insect parasite,
came to infect reptiles and eventually mammals (Lewis, 19780: 94; Telford, 1979: 322; Wilson &
Southgate, 1979: 243), so that sandflies may be regarded as the primary hosts (Lainson & Shaw,
1979: 2). This phylogenetic priority is not only of historical interest for it is reflected in present-
day associations which have a practical significance. Lizard leishmanias now occurs in the Old
World and possibly in the New World (Lainson & Shaw, 1979: 34). No Leishmania is known in
birds (Adler, 1964:42).
Many forms of Leishmania are transmitted among mammals by species of Phlebotomus. Basi-
cally, each causes a zoonosis into which man may enter to a varying extent, so that human
involvement ranges from sporadic cases to a purely man-sandfly infection. Probably in Asia
leishmaniae caused enzootics in canids and rodents which led to certain anthroponotic forms
which spread to some other Palaearctic areas (Garnham, 1971 : 482, 488; 1977: 18; Hoogstraal &
Heyneman, 1969: 1185; Lysenko, 1971: 515-518).
The forms of Leishmania are now being classified by means of objective biochemical, serologi-
cal and other studies of their intrinsic characters (Chance, 1979; Chance et al, 1977; Garnham,
1976: 536; Lumsden, 1977: 47; de Raadt, 1977: 314; Taqi & Evans, 1978: 56; Williams & Coelho,
1978; Zuckerman & Lainson, 1977: 89), and many forms will probably be recognized.
For a long time the leishmaniae were grouped, according to their normal (Lainson & Shaw,
1971 : 21) effect on the (secondary) human host, into visceral leishmaniasis (VL) or kala-azar and
cutaneous or dermal leishmaniasis (CL) which causes oriental sore and other diseases. This
grouping is unsatisfactory (Chance et al., 1977: 53, 56) but, despite rapidly changing concepts, is
still of some practical value. It is used here, where the taxonomic names of the parasites are taken
mainly from Lumsden (1977a: 46, 49; 19776).
VL is caused by forms of the Le. donovani (Laveran & Mesnil, 19030; 19036: 958) complex
which occur largely in wild Canidae and are transmitted mainly by species of the subgenera
Larroussius and Adlerius. The anthroponotic leishmaniasis of eastern India is due to Le. d.
donovani which has no dog or other animal reservoir and is transmitted by a species of Euphlebo-
tomus. Le. d. infantum Nicole, 1908, probably spread from Asia via Transcaucasia into the
Mediterranean area where it attacks dogs and children rather than adults. VL probably spread
eastwards via the Gobi Desert to eastern China (Beklemishev & Dolmatova, 1948: 351). The east
African VL is transmitted by a species of Synphlebotomus, and may infect animals as secondary
hosts (Lysenko, 1971: 518).
CL is due largely to members of the Le. tropica (Wright, 1903) group. The wild hosts, if any, are
usually rodents, and most of the sandfly vectors belong to the subgenera Phlebotomus and
THE GENUS PHLEBOTOMUS 127
Paraphlebotomus. Le. t. major Yakimov: 1915: 501; Zuckerman & Lainson, 1977: 67 occurs
largely in central Asia (Lysenko, 1971: 518; Lysenko & Belaev, 1977: 250, map) where it infects
Rhombomys opimus Lichtenstein, 1823, and some other rodents, and causes 'moist sore' in man.
Le. t. tropica (= minor Yakimov) was possibly derived (Hoogstraal & Heyneman, 1969: 1184)
from Le. t. major, occurs from the Mediterranean area to India (Lysenko, 1971 : 58), is largely
urban, causes 'dry sore' in man and infects dogs. Le. aethiopica Bray, Ashford & Bray is a hyrax
parasite which causes disseminated CL in Ethiopia and is transmitted by a species of Larroussius.
VL and CL usually occur in different areas (Lysenko, 1971 : 518, 519; Lysenko & Beliaev, 1977:
250; Theodor, 1964: 487), largely owing to the distribution of their vectors.
The recognition of a vector is a complex process involving many subjects which include sandfly
taxonomy, distribution, host choice and other aspects of ecology, determination of flagellates
found in wild flies, development of leishmaniae ingested by flies in the laboratory, and experimen-
tal transmission. Proof that a species is a vector can seldom be obtained, and it applies only to a
particular place and time; de Raadt (1977: 314) pointed out that detailed study of epidemiology
only gives an instantaneous reflection of a process continuing over a long period. The significance
of a vector may alter (Lysenko & Beliaev, 1977ft: 263; Sergiev, 1977: 283). There are many
gradations from occasional to habitual minor and major vectors. It is therefore impossible to
draw up a simple list of vectors, but a list of vectors and suspected vectors is of some value,
especially if followed by a summary of the evidence related to each species. In the present work
this is confined to references to the literature.
Killick-Kendrick (1978: 299, 300) listed 52 taxa, 28 of them Old World form, of Phlebotomus,
known or suspected of being vectors of leishmaniasis. The Old World taxa, listed in relation to
types of the disease in man are : visceral (Synphlebotomus) celiae, martini, vansomerenae, (Larrouss-
ius) ariasi, kandelakii kandelakii, longicuspis, major s. 1., orientalis, perniciosus perniciosus, tobbi,
(Adlerius) chinensis chinensis, ch. halepensis, longiductus, simici, (Euphlebotomus) argentipes; cu-
taneous (with four marked '+ VL?' which may transmit kala-azar locally): P. (Phlebotomus)
bergeroti, duboscqi, papatasi (+ VL?), salehi, (Paraphlebotomus) alexandri, caucasicus (+ VL?),
chabaudi, mongolensis ( + VL?), sergenti sergenti, (Synphlebotomus) ansarii, (Larroussius) longipes,
pedifer, perfiliewi ( + VL?). P. rossi is a recent suspect.
In addition to known vectors some sandflies presumably transmit VL among animals in large
areas of Africa where the human disease occurs but is rare (Gigade, 1978: 239), and in part of the
Sudan (Hoogstraal & Heyneman, 1969: 1141) and elsewhere where the infection is present with
no apparent vector.
Many apsects of vectors have been discussed by Adler (1964: 48, 80), Bray (1974: 91), Hoog-
straal & Heyneman (1969), Killick-Kendrick (1978; 1979), Lewis (1971; 1974; 19780), Minter
(1972), Molyneux (1977: 43-53), Safyanova (1967), Sergiev (1967: 26; 1979) and Williams &
Coelho(1978).
Sandfly fever virus, transmitted by P. papatasi and possibly other species (PerfiFev, 1968: 128),
occurs mainly in the Mediterranean area.
Most vector species are difficult, and some impossible, to control. Domestic species were
largely controlled by house-spraying against malaria vectors but have increased where this has
been stopped, in India, for instance, and in Greece where VL and CL increased when mosquito-
spraying ceased (Leger et al., 1979: 12). Sandflies have shown little resistance to insecticides
(Killick-Kendrick, 1978: 304) till recent instances in India.
Explanation of terms
Various
Antenna 3 etc. Antennal segment 3 etc.
Chahar Mahal Part of Bakhtiar va Chahar Mahal Province, Iran
Chinese, romanization The Pinyin system, adopted in the 1980 edition of The Times Atlas,
of spelling is used here
CL Cutaneous leishmaniasis
Gamma The distance between the origin of wing- veins R2 + 3 and R* and the
origin of R5
128 D. J. LEWIS
Gruziya Georgian S.S.S.R. (Georgia)
ICZN International Code of Zoological Nomenclature (1964)
and Amendments (1973)
Inverted commas Places in distribution lists not located
Kosovo i Metohija Present name for Kosmet, Yugoslavia
Le. Leishmania
Leg segments 100 units = length of femur 1
Lu. Lutzomyia
Map symbol underlined Locality approximate
MYA Millions of years ago
P. Phlebotomus
Palp 3 etc. Palp segment 3 etc.
R2 etc. Radius branch 2 and other wing veins
S. Sergentomyia
Sperm pump and tubes Genital pump and filaments
Transcaucasia Historic name for U.S.S.R. area south of Caucasus (now Armenia,
Azerbaydzhan and Gruziya)
VL Visceral leishmaniasis
WL Wing length in mm
Names of collectors mentioned
A. E. E. A. E. Eaton
C. A. V. B. C. A. V. Barkhuus
D. J. L. D. J. Lewis
D. M. A. D. M. Ackland
D. M. M. D. M. Minter
E. K. S. E. K. Saliba
E. M. Unknown
G. B. W. G. B. White
G. S. G. Shidrawi
H. C. B. H. C. Barnett
H. W. L. H. W. Leathern
J. A. S. J. A. Sinton
J. O. C. J. Omer-Cooper
J. P. M. J. P. McMahon
J. P. T. B. J. P. T. Boorman
J. P. D. J.-P. Dedet
J. W. J. Waterston
J. Wn. Jane Wilson
K. B. K. Behbehani
K. K. K. Kertesz
K. Z. D. K. Zein el Dine
L. E. S. L. E. Stephen
L. Y.-j. Leng Yan-jia
M. A. M. Ashraf
M. A. R. M. A. Rifa'at
N. L. C. N. L. Corkill
P. A. B. P. A. Buxton
P. Petrie
R. A. B. R. A. Bolt
R. E. D. B. R. E. Drake Brockman
R. L. C. R. L. Coe
R. P. L. R. P. Lane
R. W. A. R. W. Ashford
S. A. S. Adler
S. A. S. S. A. Smith
S. J. R. S. J. Rahman
S. T. S. Taussig
V. D. V. Dhanda
Y. S. Y. Schlein
THE GENUS PHLEBOTOMUS
129
Depositories, actual, probable
ANIC, Canberra
BMNH
BPBM, Honolulu
CA, Los Banos
CFHS, Nanking
CIH, Sydney
CRI, Kasauli
EM, Montpellier
FM, Paris
IH, Scoplje
IP, Algiers
IP, Paris
IPH, Tehran
L, Bastia
LSHTM, London
NM, Vienna
MB, Corales
MC, Kweiyang
MH, Sinferopol
MI, Moscow
MR AC, Tervuren
NM, Nairobi
PIPD, Shantung
SAIMR, Johannesburg
TI, Dushanbe
TI, Tbilisi
TM
U, Moscow
U, Pavia
U, Vienna
US, Tashkent
ZSI, Calcutta
ZI, Leningrad
or original
Australian National Insect Collection, Commonwealth Scientific and
Industrial Research Organisation, Canberra.
British Museum (Natural History)
Bernice P. Bishop Museum, Honolulu
College of Agriculture, Los Banos, Philippines
Central Field Health Station, Nanking
Commonwealth Institute of Health, Sydney, Australia (till 1980 School of
Public Health and Tropical Medicine)
Central Research Institute, Kasauli
Laboratoire d'Ecologie, Universite de Montpellier, France
Laboratoire de Parasitologie, Faculte de Medicine, Paris
Institute of Hygiene, Skoplje, Yugoslavia
Institut Pasteur, Algiers
Institut Pasteur, Paris
School of Public Health and Institute of Public Health, Tehran
Lycee, Bastia, Corsica
London School of Hygiene and Tropical Medicine, London
Naturhistorisches Museum, Vienna
Musee Bocage, Colares, Portugal
Medical College, Kweiyang, China
Military Hospital, Sinferopol, U.S.S.R.
Institute of Tropical Medicine and Parasitology, Moscow
[Location of some holotypes mentioned by Artemiev, 1978: 23.]
Musee Royal de 1'Afrique Centrale, Tervuren, Belgium
National Museum of Kenya, Nairobi
Provincial Institute of Parasitic Diseases, Shandong
South African Institute of Medical Research, Johannesburg, South Africa
Tropical Institute of Tadzhiskaya S.S.S.R., Dushanbe
Tropical Institute, Tbilisi
T. Maa's collection
University, Moscow
University of Pavia, Italy
University, Vienna
Protozoology Division, Uzbekistan Sanitary and Biological Institute,
Tashkent
Zoological Survey of India, Calcutta
Zoological Institute, Academy of Sciences of the U.S.S.R., Leningrad
Keys, citations, distribution and notes
Genus PHLEBOTOMUS Rondani & Berte
Flebotomus Rondani & Berte in Rondani, 1840: 12. Type-species : Bibio papatasi Scopoli, by monotypy.
Phlebotomus Rondani & Berte; Loew, 1845: 9 [emendation; first use of this name and Phlebotomidae
mentioned by Lewis in Lewis et al., 1977: 321 incorrect; spelling fixed under suspension of rules by ICZN,
1954, Opinion 256: 199]; Summers, 1911;Theodor, 1948:96; 1958:316; 1965: 179; Fairchild, 1955: 188;
Quate, 1964: 237, 238; Lewis, 1967: 14; 1973: 162; 1978ft: 233; Perfil'ev, 1968: 218; Abonnenc, 1972: 75,
92; Lewis, Young, Fairchild & Minter, 1977: 321, 326; Abonnenc & Leger, 1977: 71, 76; Duckhouse &
Lewis, 1980:99.
Cibarium of female usually without a row of teeth but often having a group of spicules, pigment patch
usually absent. Antenna 3 usually long, three or more segments of male with two ascoids. Mesanepisternum
usually with a few antero-ventrad hairs (Abonnenc & Leger, 1977: 71, 72). Abdominal tergites 2-6 with
many erect hairs. Spermathecae usually segmented. Style with three to five spines, only one or two terminal.
Paramere often complex. Species often large and pale.
There are a few omissions from the keys because only the female is described for P. sejunctus,
teshi, tubifer (male found), pexopharynx, betisi and somaliensis, and only the male for P. buccina-
130 D. J. LEWIS
tor, papuensis, trifilis, katangensis,fantalensis, chadlii, langeroni, mariae, perfiliewi galilaeus, coma-
tus (female found) and caudatus, and because species A, B, C and D are not yet described, the
females of P. brevis brevis, P. chinensis halepensis and P. ch. kyreniae and the male of the latter are
not sufficiently described, and the descriptions of 'P. major wuf and P. (Eu.) autumnalis Artemiev
were not available in time. Suitable descriptions of the missing forms could lead to improved
keys.
Key to the subgenera of Phlebotomus
1 Distance between bases of R4 and Rs relatively short, not more than a quarter of width of wing.
A pair of rods present next to genital pump. Palpal sensilla not spatulate.
Antenna 3 very long and much longer than palp. Palp short, with segment 5 shorter than or
equal to 3. Style very long. Spermathecal ducts usually short and wide 2
Distance between bases of R4 and R5 relatively long, at least a third of width of wing. Genital
pump without adjacent rods. Palpal sensilla spatulate 3
2 Vein M1 + 2 forking at level of radio-median cross-vein, before base of/?4. Cibarium of female
unarmed. Antenna 3 = 2-3 to 2-5 times length of labrum. Palp segment 3 not enlarged at base,
with sensilla scattered on flat surface. Style with four spines and a long hair. Afrotropical
Region Subgenus SPELAEOPHLEBOTOMUS (p. 131)
Vein M1 + 2 forking beyond level of radio-median cross- vein, beyond base of R4. Cibarium of
female with teeth covering a large area. Antenna 3 about three or more times length of
labrum. Palp segment 3 enlarged at base, with sunken patch of sensilla. Style with three to five
thick spines and sometimes several thick hairs. Oriental, Palaearctic and Australian Regions
Subgenus IDIOPHLEBOTOMUS (p. 133)
3 Style with three spines.
Female with row of about five to ten cibarial teeth, few or no hypopharyngeal teeth, and
thin- walled spermathecae. Male with genital filaments short or very short, paramere simple
and beak-like, and coxite with simple hair pattern
Subgenus AUSTRALOPHLEBOTOMUS(p. 135)
Style with four or more spines 4
4 Coxite with hairy process near base. Genital filaments short, 1-3 to 2-3 times length of pump . 5
Coxite without such process. Genital filaments 3 to 1 1 times length of pump .... 7
5 Coxite 0-37 to 0-74 mm long, its process very small. Style long and cylindrical with three distal
spatulate spines and two other spines. Paramere with two upward processes. Surstyle with
distal spines. Pharyngeal armature of female comprising either a network of lines or scales.
Spermatheca with nearly equal segments and a refractive membrane near the distal one
Subgenus PHLEBOTOMUS (p. 137)
Coxite 0-20 to 0-33 mm long, its process usually large, and having a brush of long hairs. Style
not long, with four or five spines. Paramere simple, distal upper surface flat and elliptical with
short hairs. Surstyle without distal spines. Pharynx of female with teeth or scales. Sperma-
theca sometimes with differentiated rounded end-segment 6
6 Style with four long spines, two near the tip and two near the base. Pharynx of female with large
backwardly directed teeth Subgenus PARAPHLEBOTOMUS (p. 142)
Style with five long spines, two at the tip and three near the middle. Pharynx of female with
irregular scales or punctiform teeth . . . Subgenus SYNPHLEBOTOMUS (p. 148)
7 Style with four long spines, one distal, one subterminal, and two near middle.
Paramere with one or two extra lobes, with or without accessory spine. Aedeagus some-
times conical. Pharynx of female with a small group of teeth in middle, and behind it some
concentric lines. Spermatheca segmented, end-segment not enlarged
Subgenus ANAPHLEBOTOMUS (p. 170)
Style with five long spines 8
8 Paramere with one or two extra lobes, with or without accessory spine. Pharynx of female as in
Anaphlebotomus.
Spermatheca with differentiated end-segment . . Subgenus EUPHLEBOTOMUS (p. 168)
Paramere without extra lobes. Pharyngeal armature otherwise 9
9 Paramere truncated.
Antenna 3 and legs long, and wings narrow. Spermatheca moniliform. Haltere of male with
broad stalk. Paramere with adjacent rod .... Subgenus KASAULWS (p. 172)
- Paramere not truncated 10
THE GENUS PHLEBOTOMUS 131
10 Pharynx of female and male with punctiform teeth (large in wenyoni), except in mascittii which
has large irregular teeth. Spermatheca segmented, with long finger-like neck except in soma-
liensis, which has a rather long end-segment, and mascittii, which has a spermatheca with
transverse striations often in distal part, a small head, little or no narrowing, and a wide duct.
Genital filaments three to five times as long as pump . . Subgenus LARROUSSWS (p. 150)
- Pharynx of female with triangular or rounded group of medium-size teeth. Spermatheca incom-
pletely segmented. Genital filaments usually very long, 6-6 to 1 1-0 length of pump
Subgenus ADLERIUS(p. 163)
Tibia 3 in certain species
The following records of relative lengths of tibia 3 (femur 1 = 100 units, females unless males
indicated) are placed here for species about which no other taxonomic information is given.
P. aculeatus (Kenya) <$, 182 (2-26 mm); betisi, 199; gibiensistf, 212 (2-30 mm); guggisbergi, 188; kandelakii
kandelakii, 164; longicuspis (Algeria) .). Pakistan: Rawalpindi (1959, H. C. R; all 11 ^ Adlerius examined in
1979 were this species). Nepal: Chobhar (1976, J. Wn.).
NOTE. In Afghanistan P. hindustanicus occurs in low rocky mountains (Artemiev, 1978: 23).
Phlebotomus (Adlerius) kabulensis Artemiev
(Map 10)
Phlebotomus (Adlerius} kabulensis Artemiev, 1978: 21 [9 £]; 1980: 1188. Holotype • subova/is
H ven/furcata
•7 - Ik
12°C
24°C
Fig. 33 Geographical and ecological distribution of the albella-group of Timocratica.
Timocratica venifurcata sp. n.
(Figs 21, 33, 78, 120, 121)
c? 16-17 mm. Head white. Second segment of labial palpus white, basal two-thirds dark grey externally;
third segment white, with grey scales near apex. Legs white, distal half of fore tibia, and tarsi dark fuscous
below. Fore wing with costa gently arched, apex rounded, termen and tornus obliquely rounded; R4 and
Rs, and CuAr and CuA2, stalked; basal half of costa with fuscous and golden-yellow scales below. Hind
wing white. Abdomen golden-ochreous, first tergite, anal tuft and sternites white.
246 V. O. BECKER
GENITALIA <£ (Figs 120, 121). Uncus wide, lateral margins nearly parallel, apex bifurcate. Apex of gnathos
narrow, pointed. Digitate processes of juxta flat, triangular, distal half with long sparse setae. Valva with
basal half wide, narrowing progressively towards apex. Aedeagus bent ventrad, vesica with strong, short
cornutus and many smaller spines opposite.
REMARKS. T. venifurcata is the only white species which has an ochreous abdomen and #4 and
R5 stalked on the fore wings (Fig. 21). It can be easily distinguished from all others with an
ochreous abdomen by the lack of ochreous coloration on its palpi and legs.
DISTRIBUTION (Fig. 33). Brazil (Central Plateau), in Tropical Premontane Moist Forest.
MATERIAL EXAMINED
2 <$ (1 <$ genitalia preparation).
Holotype & Brazil: Distrito Federal, Planaltina, 1000 m, 9.xi.l977 (Becker) (MN).
Paratype. Brazil: 1
SW7 5BD
Contents
Synopsis 307
Introduction 307
Measurements and indices 308
Abbreviations of museums . 309
Cardiocondyla Emery ........ 309
Synonymic list of Afrotropical Cardiocondyla species 311
Key to species (workers) .312
Leptothorax Mayr ......... 319
Synonymic list of Afrotropical Leptothorax species 323
Key to species (workers) 323
Melissotarsus Emery 333
Synonymic list of Afrotropical Melissotarsus species 335
Key to species (workers) 335
Messor Forel ........... . 338
Synonymic list of Afrotropical Messor species 342
Key to species (medium to large workers) 342
Cataulacus F. Smith 354
Key to species (workers) 354
Appendix 354
Acknowledgements 365
References 365
Index ... 369
Synopsis
The Afrotropical species of the myrmicine ant genera Cardiocondyla Emery, Leptothorax Mayr,
Melissotarsus Emery and Messor Forel are revised and keyed, and a revised key to Cataulacus F. Smith is
presented. At genus-level Loncyda Santschi, Dyclona Santschi and Prosopidris Wheeler are newly
synonymized with Cardiocondyla; Nesomyrmex Wheeler and Tetramyrma Forel with Leptothorax; and
Veromessor with Messor. The current synonymy of Aphaenogaster Mayr, a genus very close to Messor, is
listed with the inclusion of Brunella Forel as a new synonym. At species-level nine Cardiocondyla (four new),
1 1 Leptothorax (one new), three Melissotarsus and 12 Messor (one new) are recognised in the regional fauna.
New species-level synonymy includes 10 names in Cardiocondyla, three in Leptothorax, four in Melissotarsus
and 14 in Messor, most of the last being of former infraspecific names. Five former infraspecific names in
Messor are given new status here as valid species. In Cataulacus six new species are described and four
previously synonymized names are reinstated as valid species.
Introduction
This paper is presented as a further contribution towards a revision of the subfamily Myrmicinae
in the Afrotropical region which, for the purpose of this study, excludes the fauna of the Malagasy
region. Previously issued parts of this series include studies of the genera Epitritus Emery (Bolton,
1972), Cataulacus F. Smith (Bolton, 1974), Decamorium Forel, Rhoptromyrmex Mayr and
Triglyphothrix Forel (Bolton, 1976), Tetramorium Mayr (Bolton, 1980), Meranoplus F. Smith,
Bull. Br. Mus. nat. Hist. (Ent.) 45 (4): 307-370
Issued 30 September 1982
308 B. BOLTON
Dicroaspis Emery and Calyptomyrmex Emery (Bolton, 198 la), Ankylomyrma Bolton,
Atopomyrmex Andre, Baracidris Bolton, Cyphoidris Weber, Ocymyrmex Emery, Pristomyrmex
Mayr and Terataner Emery (Bolton, 19816).
With the inclusion of the four genera treated in this paper a total of 20 of the region's 43
presently recognized myrmicine genera have been revised in the present series. The Afrotropical
fauna of some myrmicine genera has been studied by Brown who, beside revising Rhoptromyrmex
(Brown, 1964), has also analysed the genera of the myrmicine tribe Dacetini and revised its main
genera on a world-wide basis. In the case of sub-Saharan Africa this included the genera
Serrastruma Brown (Brown, 1952), Smithistruma Brown (Brown, 1953) and Strwnigenys F. Smith
(Brown, 1954).
Prior to these studies very little synthesising work had been carried out on the Afrotropical
myrmicines, the only notable contributions being the series of papers produced by Arnold
between 1916 and 1926 on the fauna of South Africa, and a catalogue of species by Wheeler
(1922) who also included a key to world genera. This key is now very much out of date, is difficult
to use and cannot be trusted. Similarly Arnold's (1916) key to the South African myrmicine
genera has, through subsequent synonymies and descriptions of new genera, become unusable.
More recently Bolton (1973) presented a subfamilial and generic key for the Afrotropical region
but again detailed investigation of the individual genera mentioned above has already rendered
this partially obsolete. A key to the 19 myrmicine genera in which the antennal club is restricted
to two segments has been constructed by Bolton (19816) and a key to the remaining genera is
presently being built up.
The four genera newly revised in this paper, which are discussed in more detail under their
individual sections, constitute a relatively minor proportion of the regional fauna in terms of
number of species. Melissotarsus and Cardiocondyla are arbitrarily regarded as small genera, with
three and nine species respectively in the region, whilst Leptothorax with 1 1 and Messor with 12
species are of moderate size. Apart from Melissotarsus, which is restricted to sub-Saharan Africa
and Madagascar, most species of the other three genera are primarily distributed elsewhere, the
Afrotropical fauna merely representing the few species which have successfully invaded the region
from the north.
Measurements and indices
Total Length (TL). The total outstretched length of the individual, from mandibular apex to
gastral apex.
Head Length (HL). The length of the head proper, excluding the mandibles, measured in a
straight line from the anteriormost point of the median clypeal margin to the mid-point of the
occipital margin, in full-face view. (In species with strongly concave occipital margin the head
length is measured to the mid-point of a line connecting the posterolateral corners.)
Head Width (HW). The maximum width of the head in full-face view, measured behind the eyes.
HW x 100
Cephalic Index (CI). —
HL
Eye Length (EL). In Cataulacus; the maximum length of the eye in full-face view.
EL x 100
Ocular Index (OI). In Cataulacus;
HW
Scape Length (SL). The maximum straight-line length of the antennal scape excluding the basal
constriction or neck. (In Cataulacus the SL usually measured in profile view with the scape in its
scrobe, as it is usually in this position in mounted specimens.)
c _ SL x 100
Scape Index (SI).
Pronotal Width (PW). The maximum width of the pronotum in dorsal view.
AFROTROPICAL MYRMICINE ANT GENERA 309
Alitrunk Length (AL). The diagonal length of the alitrunk in profile from the point at which the
pronotum meets the cervical shield to the posterior base of the metapleural lobes or teeth. (In
Melissotarsus measured to posteroventral corner of alitrunk as metapleural lobes absent.)
Abbreviations of museums
AMNH, New York American Museum of Natural History, New York, U.S.A.
BMNH British Museum (Natural History), London, U.K.
IE, Bologna Istituto di Entomologia del'Universita, Bologna, Italy.
MCSN, Genoa Museo Civico di Storia Naturale 'Giacomo Doria', Genoa, Italy.
MCZ, Cambridge Museum of Comparative Zoology, Cambridge, Massachusetts, U.S.A.
MHN, Geneva Museum d'Histoire Naturelle, Geneva, Switzerland.
MNHN, Paris Museum National d'Histoire Naturelle, Paris, France.
MNHU, Berlin Museum fur Naturkunde der Humboldt-Universitat, Berlin, Germany (D.D.R.).
MR AC, Tervuren Musee Royal de 1'Afrique Centrale, Tervuren, Belgium.
NM, Basle Naturhistorisches Museum, Basle, Switzerland.
NM, Bulawayo National Museum, Bulawayo, Zimbabwe. (Hymenoptera from this museum are
now deposited in SAM, Cape Town.)
NM, Vienna Naturhistorisches Museum, Vienna, Austria.
SAM, Cape Town South African Museum, Cape Town, South Africa.
USNM, Washington United States National Museum, Washington, D.C., U.S.A.
ZM, Kiev Zoological Museum, Institute of Zoology, Academy of Sciences of Ukrainian
S.S.R., Kiev, U.S.S.R.
CARDIOCONDYLA Emery
(Figs 1-7)
Cardiocondyla Emery, 1869: 20. Type-species: Cardiocondyla elegans Emery, 1869: 21, by monotypy.
Emeryia Forel, 1890: ex. Type-species: Emeryia wroughtonii Forel, 1890: cxi, by monotypy. [Synonymy by
Forel, 1892: 313.]
Xenometra Emery, 1917:96. Type-species: Xenometra monilicornis Emery, 1917:96. ( = Cardiocondyla
emeryi Forel), by monotypy. [Synonymy by Urbani, 1973: 199.]
Loncyda Santschi, 1930: 70 [as subgenus of Cardiocondyla]. Type-species: Cardiocondyla (Loncyda)
monardi Santschi, 1930: 70, by monotypy. Syn. n.
Dyclona Santschi, 1930: 70 [as subgenus of Cardiocondyla]. Type-species: Monomorium cristatum Santschi,
1912: 163, by original designation. Syn. n.
Prosopidris Wheeler, 1935: 40 [as subgenus of Cardiocondyla]. Type-species: Cardiocondyla (Prosopidris)
sima Wheeler, 1935: 41, by original designation. Syn. n.
Prosopidris Wheeler; Reiskind, 1965: 80. [Raised to genus.]
DIAGNOSIS OF WORKER. Small to minute monomorphic myrmicine ants. Mandibles with 5 teeth which
decrease in size from apical to basal. Palp formula 5, 3 (16 species examined). Clypeus with flattened and
prominent projecting lateral portions which are fused to the raised projecting median portion to form a shelf
which projects forward over the mandibles (Fig. 2). Sometimes the lateral portions of the clypeus extend
further forward than the median so that the anterior margin of the projecting shelf is concave medially.
Median portion of clypeus posteriorly broadly inserted between small narrow frontal lobes. Frontal carinae
and antennal scrobes absent. Eyes present, generally large and conspicuous, situated in front of the
midlength of the sides. Antennae with 1 1-12 segments, usually with a distinct 3-segmented club but the first
club segment may be relatively small. Promesonotal dorsum flattened to evenly convex in profile, the dorsal
alitrunk without sutures but the metanotal groove commonly (but by no means universally) impressed.
Pronotal corners in dorsal view broadly rounded to bluntly angular and projecting. Propodeal spiracle
small, situated approximately at the midlength, often low down on the side but not shifted back towards the
margin of the declivity. Propodeum unarmed to strongly bispinose. Metapleural lobes low and rounded.
Petiole nodiform with a moderate to long, usually slender, anterior peduncle. Postpetiole dorsoventrally
flattened in profile, in dorsal view very broad, much broader than the petiole node. Sting large and strongly
developed, knife blade-like and broad in profile, without lamelliform appendages. Dorsal surfaces of body
usually hairless.
310
B. BOLTON
The genus Cardiocondyla contains about 40 species, mostly distributed in the Old World.
Discounting tramp species only two have been described from the New World (ectopia Snelling
and venmtula Wheeler) but it is quite possible that both represent introductions, although to the
present no conspecific forms have been found among Old World material of the genus.
Cardiocondyla contains several very successful tramp species which are easily and apparently
frequently spread by human commerce. Such tramps include the cosmopolitan emeryi,
tropicopolitan wroughtonii and the Pacific island-hopping nuda (Mayr), which sometimes reaches
Figs 1-7 Cardiocondyla workers. 1, profile of shuckardi. 2, head of shuckardi. 3-7, alitrunk and pedicel
segments of (3) monardi, (4) wroughtonii, (5) emeryi, (6) weserka, (7) neferka.
AFROTROPICAL MYRMICINE ANT GENERA 311
the Pacific coast of North America. The fauna of the Afrotropical region includes 9
Cardiocondyla species. Of these six are found only in this region, two are the common tramps
emeryi and wroughtonii, and one also occurs on Madagascar (shuckardi). One of the six endemic
species, zoserka, described from a series of females, is suspected of being the first inquiline to be
found in this genus.
The majority of species of the world are known only from workers; a few queens are known
and these are quite normal apart from having the wing venation much reduced. The peculiarity of
Cardiocondyla lies in the males, which are known to be dimorphic in several species. Ordinary
alate males are known for a fair number of species but in some (emeryi, wroughtonii, elegans,
batesii Forel) dealate, highly ergatoid males are also produced; such peculiar males were
responsible for two of the generic names in the synonymy above, Emeryia and Xenometra. In a
further species, papuana (Reiskind), the only known male is an ergatoid. The problem is that the
extent of ergatoid male production among the species, and the reasons for the production of such
males, is unknown. It may well be that all species of Cardiocondyla are capable of developing
both normal and ergatoid males, given the right conditions, but it may be that some species only
have normal alate males, some only have ergatoid males, and some have both. It is certainly an
intriguing problem and deserves further investigation.
Recent studies of Cardiocondyla include the works of Wilson & Taylor (1967) on the Pacific
species, and of Bernard (1956) on the Palaearctic fauna; the species of sub-Saharan Africa have
not been dealt with previously.
To the present Cardiocondyla has occupied its own tribe, the Cardiocondylini, characterized
primarily by its prominent clypeus and broad postpetiole in the worker, and the reduced
venation in the female. Other features noted by Emery (1922a) and Wheeler (1922) have been
eroded away by subsequent discoveries of species not then known. Nevertheless, the tribal status
has remained as such since 1922 although Urbani (1977) has recently pointed out the similarity
between C. monardi and Leptothorax. He interpreted this as convergence but I consider that a
real relationship exists between Leptothorax and Cardiocondyla and that the latter belongs in
tribe Leptothoracini. Comparing the two genera there is broad agreement in head shape,
dentition, high palp formula, position of eyes, antennal segmentation, size and shape of frontal
lobes, broad insertion of the posterior clypeus between the frontal lobes, lack of scrobes and
frontal carinae, size and position of propodeal spiracle, and form of the metapleural lobes. The
presence of all these characters together in both genera argues strongly that they are genuinely
closely related and I propose the dissolution of Cardiocondylini and the incorporation of its sole
genus in the Leptothoracini. Within the tribe Cardiocondyla is still separated from Leptothorax
and its close relatives (as discussed under that genus) by the characters devised by Emery and
Wheeler, namely the specialized form of the anterior clypeus (although this is hinted at in some
Leptothorax), the characteristic form of the postpetiole and the reduced wing venation of the
females. A further character distinguishing the two is the specialized blade-like sting of
Cardiocondyla, not seen in Leptothorax.
Synonymic list of Afrotropical Cardiocondyla species
emeryi Forel
emeryi var. rasalamae Forel syn. n.
emeryi subsp. mahdii Karavaiev syn. n.
monilicornis Emery
nuda subsp. nereis Wheeler
mauritia Donisthorpe syn. n.
monardi Santschi
neferka sp. n.
nilotica Weber
sekhemka sp. n.
shuckardi Forel
globinodis Stitz syn. n.
badonei Arnold syn. n.
wassmanni [sic] Santschi syn. n.
312 B. BOLTON
wasmanni var. sculptior Santschi syn. n.
brevispinosa Weber syn. n.
fusca Weber syn. n.
weserka sp. n.
wroughtonii (Forel)
wroughtonii var. hawaiensis Forel
emeryi subsp. chlorotica Menozzi syn. n.
zoserka sp. n.
Key to species (workers)
Note. C. zoserka, described from suspected inquiline females, is omitted from the key.
1 With alitrunk in profile the dorsum without trace of a metanotal groove or impression (Fig. 3).
Propodeum unarmed. Postpetiole in dorsal view distinctly longer than broad. (Angola)
mortar di (p. 314)
With alitrunk in profile the dorsum with a distinct metanotal groove or impression (Figs 1, 4-7).
Propodeum sharply angulate to bispinose. Postpetiole in dorsal view as broad as to markedly
broader than long 2
2 With the head in full-face view the scapes, when laid back, distinctly exceeding the occipital
corners. (Sudan) nilotica (p. 315)
With the head in full-face view the scapes when laid back, either failing to reach or just reaching
the occipital corners, never exceeding them 3
3 Dorsal surfaces of head and alitrunk smooth and glossy, unsculptured everywhere except for
widely separated minute punctulae on the head. Head relatively broad and scapes short, CI 86,
SI 74. (Ghana) . sekhemka (p. 315)
Dorsal surfaces of head, alitrunk or both finely and densely sculptured, the sculpture usually
conspicuous. Scapes longer, SI > 80. Head with CI usually < 80, rarely otherwise ... 4
4 Propodeum in absolute profile bluntly angulate to bidenticulate (Fig. 1), never with a pair of
strong teeth or spines which are longer than their basal width in profile and which are as long
as half the distance separating their bases in dorsal view. Scapes relatively long, Si in range
93- 100. (Widespread in sub-Saharan Africa; Madagascar) .... shuckardt (p. 316)
Propodeum in absolute profile strongly bidentate to bispinose (Figs 4-7), the teeth or spines
longer than their basal width in profile and at least as long as half the distance separating their
bases in dorsal view. Scapes relatively short, SI in range 8 1-94 5
5 With alitrunk in profile the propodeal dorsum approximately flat behind the metanotal groove
and more or less level with the promesonotal dorsum, the propodeal dorsum not showing a
long gradual slope down to the spines (Fig. 6). (Cameroun) weserka (p. 317)
With alitrunk in profile the propodeal dorsum convex behind the metanotal groove and then
showing a long gradual slope down to the spines (Figs 4, 5, 7) 6
6 With alitrunk in profile the mesonotal dorsum abruptly changing slope posteriorly and
descending steeply to the metanotal groove (Fig. 4). Petiole node in dorsal view subglobular,
usually slightly broader than long. Head relatively broad, CI in range 79-86. (Pantropical
tramp species) wroughtonii (p. 317)
With alitrunk in profile the mesonotal dorsum curving evenly into the meianotal groove, without
an abrupt change of slope posteriorly (Figs 5, 7). Petiole node in dorsal view not subglobular,
usually quite distinctly longer than broad. CI in range 72-79 7
7 Pronotal corners bluntly but conspicuously angular in dorsal view. Propodeal spines relatively
long and slender (Fig. 7). (Ghana, Cameroun) neferka (p. 314)
Pronotal corners rounded in dorsal view. Propodeal spines relatively short and stout (Fig. 5).
(Cosmopolitan tramp species, very common) emeryi (p. 312)
Cardiocondyla emeryi Forel
(Fig. 5)
Cardiocondyla emeryi Forel, 1881: 5. Syntype workers, VIRGIN Is.: St Thomas I.. 1878 (MHN, Geneva)
[examined].
Cardiacondyla emeryi var. rasalamae Forel, 1891: 161. Syntype workers, MADAGASCAR: Imerina (P.
Camboue) (MHN, Geneva) [examined]. Syn. n.
AFROTROPICAL MYRMICINE ANT GENERA 313
Cardiocondyla emeryi subsp. mahdii Karavaiev, 1911:8. Syntype workers, SUDAN: Khartoum, Sirdargarten,
no. 1900 ( V. Karavaiev) (ZM, Kiev) [examined]. Syn. n.
Xenometra monilicornis Emery, 1917: 96. Holotype ergatoid male [not female], VIRGIN Is.: St Thomas I.
(MCSN, Genoa). [Synonymy by Urbani, 1973: 200.]
Cardiocondyla nuda subsp. nereis Wheeler, 1927: 140. Syntype workers, females, NORFOLK I.: 1915 (A. M.
Lea) (MCZ, Cambridge). [Synonymy by Wilson & Taylor, 1967: 53.]
Cardiocondyla mauritia Donisthorpe, 1946: 776. Holotype and paratype workers, MAURITIUS: 1941-45, no.
102 (R. Mamet) (BMNH) [examined]. Syn. n.
WORKER. TL 1.7-2.1, HL 0.45-0.52, HW 0.34-0.38, CI 72-78, SL 0.30-0.36, SI 86-94, PW 0.22-0.28, AL
0.48-0.58 (40 measured).
Antennal scapes of moderate length (SI, above), when laid back on the head usually failing to reach the
occipital corners but in a few samples just reaching them; never distinctly exceeding the occipital corners.
Maximum diameter of eye 0.10-0.12, about 0.28-0.32 x HW and with 8-10 ommatidia in the longest row.
Head always conspicuously longer than broad in full-face view, CI > 80 in all samples examined. With the
alitrunk in dorsal view the pronotal corners narrowly but evenly rounded, not produced into angular
shoulders. In profile the alitrunk with the promesonotal dorsum forming an even shallow convexity from
front to back, the slope of the dorsum not changing radically just in front of the metanotal groove.
Metanotal groove sharply and conspicuously impressed, the propodeal dorsum convex behind the groove,
then entering a long slope down to the propodeal spines. In profile the propodeal spines short and stoutly
constructed but longer than their basal width. In dorsal view each spine longer than half the distance
separating their bases. Petiole and postpetiole shaped as in Fig. 5, the petiole node showing some variation
in shape but in dorsal view always at least as long as broad and usually distinctly longer than broad.
Peduncle of petiole moderately long, the sternite of the postpetiole showing a blunt anteroventral
prominence or bulge. Postpetiole in dorsal view much broader than long, with a shallowly concave anterior
margin and evenly convex sides. Dorsal surfaces of head and alitrunk usually with scattered fine punctures,
the surface between them finely and densely shagreened or granular. In some specimens the punctures are
very small or widely scattered and inconspicuous, in which case the entire surface appears shagreened to
granular. Occasionally the granular ground-sculpture is reduced leaving the fine punctures on a more or less
smooth surface. Sculpture on the dorsal head is frequently stronger and better defined than on the dorsal
alitrunk. Hairs absent except on mouthparts and around gastral apex but a fine appressed pubescence is
present all over the body, being more conspicuous on the darkly coloured gaster than elsewhere. Head and
alitrunk yellow to light brown, sometimes orange-brown; gaster much darker, blackish brown to black and
contrasting strongly with the head and alitrunk.
A well known highly successful tramp-species, emeryi has been spread widely over the earth's
surface, mainly by human commerce. In the tropics and subtropics it survives outside, but in the
temperate zones it is more or less restricted to constantly heated buildings and greenhouses. The
presence of two very closely related species in West Africa, neferka and weserka, implies that the
Afrotropical region is most probably the place of origin of emeryi.
Like a few other species emeryi is known to have dimorphic males (see discussion, p. 311). The
species usually produces normal winged males but sometimes also develops highly ergatoid
males which may be found wandering alone, far from any nest.
MATERIAL EXAMINED
Afrotropical region. Ghana: Polcoase (W. Bellfield); Kibi (D. Leston). Nigeria: Gambari (B. Bolton); Bussa
(J. T. Medler). Cameroun: Nkoemvon (D. Jackson). Angola: Luanda (G. R. Gradwell & D. Snow). Sudan:
Khartoum (V. Karavaiev). Uganda: Ruwenzori, Semliki Forest (D. S. Fletcher). Kenya: Embu, Ishiara (V.
Mahnert & J.-L. Ferret). Tanzania: Lindi (D. V. Fitzgerald); Manyara Nat. Park (M. E. Irwin & E. S. Ross);
Zanzibar (L. F. Brown). Zimbabwe: Bembesi (G. Arnold). Botswana: Shorobe (A. Russell-Smith). South
Africa: Durban (C. B. Cooper); Nelspruit (M. Samways).
Other regions. Madagascar: Joffreville (J. M. Betsch); Imerina (P. Camboue). Seychelles: Little Sister I. (U.
M'iiller). Aldabra: South I. (B. Cogan & A. M. Hutson). Chagos Archipelago: Diego Garcia (A. M. Hutson).
Ascension I. (E. A. G. Duffey). Egypt: Gizeh (F. Morey); Siwa (J. Omer-Cooper); Zegawa (J. Omer-Cooper).
Madeira: Funchal (N. L. H. Krauss). Cape Verde Is.: Fogo (Lindberg); Fogo (Groh); S. Vincente (Lindberg);
S. Tiago (Lindberg); Nicolau (Lindberg); St Helena (Wollaston). Virgin Is.: St Vincent I. (H. H. Smith); St
Thomas I. West Indies: Anguilla (A. G. Parker). Puerto Rico: Mayaquez (M. R. Smith). Norfolk I. (A. M.
Lea). Mauritius (R. Mamet).
For Pacific distribution see Wilson & Taylor (1967); for Neotropical distribution see Kempf (1972).
314 B. BOLTON
Cardiocondyla monardi Santschi
(Fig. 3)
Cardiocondyla (Loncyda) monardi Santschi, 1930: 70, fig. 5. Syntype workers, ANGOLA: Rio Mbale,
ix.1928-i.1929 (A. Monard) (NM, Basle) [examined].
WORKER. TL 2.7, HL 0.58, HW 0.46, CI 79, SL 0.49, SI 107, PW 0.33, AL 0.68.
Antennal scapes relatively long, SI > 100; when laid back on the head exceeding the occipital corners.
Maximum diameter of eye 0.14, about 0.30 x HW and with approximately 14 ommatidia in the longest row.
Pronotal corners in dorsal view broadly and evenly rounded. Alitrunk in profile with the dorsum forming a
single uninterrupted surface, without trace of a metanotal groove or impression. Propodeum unarmed, the
dorsum rounding broadly, smoothly and evenly into the declivity. Petiole in profile with a very long anterior
peduncle and a long low feebly convex node. Petiole node in dorsal view subglobular, only very slightly
longer than broad. Postpetiole in dorsal view somewhat longer than broad, narrow (c. 0.13) at its junction
with the petiole, then rapidly broadening posteriorly to a maximum width of c. 0.26 at about its midlength,
and behind this narrowing again to a posteriormost width of c. 0.20. Dorsal length of postpetiole about 0.30,
of petiole peduncle plus node about 0.40. All dorsal surfaces of head, alitrunk, petiole, postpetiole and first
gastral tergite reticulate-punctate. Whole of body dorsally with glinting silvery pubescence which is mostly
set within the punctures. Colour yellow with glinting silvery highlights due to the pubescence.
This very distinctive species should not be confused with any other African form. It is quickly
separated from all its congeners in the Afrotropical region by its long scapes, lack of a metanotal
groove or impression, absolutely unarmed propodeum, elongate pedicel segments and glinting
silvery pubescence on a yellow background.
MATERIAL EXAMINED
Angola: Rio Mbale (A. Monard).
Cardiocondyla neferka sp. n.
(Fig. 7)
HOLOTYPE WORKER. TL 1.8, HL 0.48, HW 0.36, CI 75, SL 0.32, SI 89, PW 0.26, AL 0.49.
Antennal scapes of moderate length (SI 87-91 in type-series), when laid back on the head not reaching the
occipital corners in full-face view. Maximum diameter of eye 0.11, about 0.31 x HW and with 9-10
ommatidia in the longest row. Head conspicuously longer than broad, CI < 80. Pronotum in dorsal view
with narrowly rounded, somewhat prominent corners, giving the ant a conspicuously square-shouldered
appearance. With the alitrunk in profile the promesonotum forming an even shallow convexity from front to
back which grades into the metanotal groove without passing through an abrupt change of slope.
Metanotal groove shallowly impressed, the propodeal dorsum shallowly convex behind the groove, then
sloping downwards posteriorly towards the spines. Propodeal spines elongate and narrow, in profile much
longer than their basal width; in dorsal view the spines slightly incurved and each as long as the distance
separating their bases. Shape of pedicel segments as in Fig. 7. In dorsal view the petiole node longer than
broad, the postpetiole distinctly broader than long and broadest at its midlength. Dorsum of head
shagreened-granular, the sculpture very fine and dense, blanketing the surface. Promesonotal dorsum very
finely and densely superficially shagreened and mat, but the propodeal dorsum with only vestigial sculpture
and glossy, much less densely sculptured than the promesonotum. Dorsal surfaces of petiole, postpetiole and
gaster unsculptured except for a faint and patchy superficial patterning. Hairs absent except on mouthparts
but a fine appressed pubescence is present which is most apparent on the gaster. Colour uniform light
brownish yellow, the dorsum of the head slightly darker than the sides; sides of the first gastral tergite a rich
darker brown.
PARATYPE WORKERS. TL 1.80-1.81, HL 0.46-0.48, HW 0.35-0.37, CI 76-79, SL 0.32-0.33, SI 87-91, PW
0.24-0.27, AL 0.48-0.51 (3 measured).
Maximum diameter of eye 0.10-0.11, about 0.27-0.31 x HW and with 9-10 ommatidia in the longest
row. As holotype but in a couple the darker colour of the sides of the first gastral tergite extends onto the
dorsum.
Holotype worker, Ghana: Mampong, 10.ii.1970 (P. Room) (BMNH).
Paratypes. 3 workers with same data as holotype (BMNH; NM, Basle; MCZ, Cambridge).
Non-paratypic material examined. Cameroun: Nkoemvon (D. Jackson).
AFROTROPICAL MYRMICINE ANT GENERA 315
The Cameroun material differs from the type-series only in colour as here the dorsum of the head
is conspicuously much darker than the sides and the gaster is uniformly dark brown. This is
merely an intensification of the condition seen in the type-series and has no significance at
species-level.
C. neferka is closest related to emeryi but is quickly separable by its elongate narrow propodeal
spines and conspicuously square-shouldered appearance when the pronotum is seen in dorsal
view.
Cardiocondyla nilotica Weber
Cardiocondyla nilotica Weber, 1952: 8, fig. 13. Holotype worker, SUDAN: White Nile R., Ed Dueim, lat. 14U
00' N., 2.vii.l939, no. 1234 (N. A. Weber) (not in AMNH, New York ; presumed lost).
The only known representative of this species cannot be found in AMNH, New York and must
be presumed lost. However, Weber's original description contains enough information to give a
reasonable picture of this species, and it appears distinct from all other species of the Afrotropical
region. The following diagnostic characters are taken from Weber's description.
WORKER. TL 2.5. Antennal scapes when laid back distinctly exceeding the occipital corners. Metanotal
groove broad and rounded-concave. Propodeum armed with a pair of short triangular tubercles. Peduncle
of petiole slender. Petiole node in dorsal view broader than long, the postpetiole slightly broader than long
(taken from Weber's fig. 13, where the postpetiole appears subglobular in dorsal view). Densely and finely
punctate on head and alitrunk, gaster smooth and shining. Colour bright ferruginous, the head with a dark
area dorsally; appendages pale and gaster dark brown.
The overall picture which emerges is of a relatively large species closely related to shuckardi but
with decidedly longer scapes, narrower postpetiole and lighter colour, although a few pale
coloured individuals of shuckardi are known.
Cardiocondyla sekhemka sp. n.
HOLOTYPE WORKER. TL 1.8, HL 0.44, HW 0.38, CI 86, SL 0.28, SI 74, PW 0.12, AL 0.32.
Head relatively short and broad, scapes relatively short (CI and SI, above). When laid back on the head
the scapes failing to reach the occipital corners in full-face view. Projecting median portion of clypeus and
flattened prominent lateral parts of clypeus closely fused and forming a more or less evenly semicircular
projecting lobe which hides most of the mandibles in full-face view (only the two apicalmost teeth of the
right mandible can be seen in the holotype). Eyes relatively large, maximum diameter 0.12, about
0.32 x HW and with 10-11 ommatidia in the longest row. Shape of eye irregular in profile, narrowed and
drawn out anteroventrally, rounding the lower curve of the sides and onto the margins of the ventral surface
of the head. Pronotal corners rounded in dorsal view. With alitrunk in profile the promesonotum evenly
convex from front to back, sloping posteriorly to the feebly impressed metanotal groove. Propodeal dorsum
more shallowly convex than promesonotum and on a much lower level so that there is a distinct step-down
from the promesonotum to the propodeum. Posteriorly the propodeal dorsum sloping down to a pair of
broad blunt and very low tubercles which are much shorter than the metapleural lobes and which are
shorter than their basal widths. In dorsal view the tubercles distinctly shorter than half the distance
separating their bases. Petiole in profile with a short peduncle and rounded node. In dorsal view the petiole
node subglobular, slightly broader than long. Postpetiole in dorsal view much broader than long, with a
shallowly concave anterior margin and evenly convex sides. Dorsum of head sculptured with widely
scattered superficial minute punctulae, the surface between the punctulae smooth and shining. Remainder of
body unsculptured, smooth and shining. Hairs absent except on mouthparts and gastral apex. Colour
uniform glossy blackish brown, the legs and antennae lighter.
Holotype worker, Ghana: Tumu, 24.xii. 1969 (P. Room) (BMNH).
This small, virtually unsculptured darkly coloured species is easily recognised by its relatively
short scapes, broad head, characteristically shaped eyes, lack of developed propodeal spines and
feebly impressed metanotal groove followed by a depressed propodeum. In the Afrotropical
region only wroughtonii approaches the CI value of sekhemka, but in that species the propodeal
spines are long and strongly developed. Only shuckardi has the propodeal armament as feebly
developed as in sekhemka but here the head and body are usually strongly sculptured, the eye is
not drawn out anteroventrally, and the dimensions are very different.
316 B. BOLTON
Cardiocondyla shuckardi Forel
(Figs 1,2)
Cardiocondyla shuckardi Forel, 1891: 161. Syntype workers, MADAGASCAR: Imerina, Antananarivo
(Cambou'e) (MHN, Geneva) [examined].
Cardiocondyla globinodis Stitz, 1923: 154. Syntype workers, SOUTH WEST AFRICA: Omaruru, 22.vi.191 1 (W.
Michaelsen) (MNHU, Berlin) [examined]. Syn. n.
Cardiocondyla badonei Arnold, 1926: 225, fig. 64. Syntype workers, MOZAMBIQUE: Amatongas Forest,
ii.1917 (G. Arnold) (BMNH; MCZ, Cambridge) [examined]. Syn. n.
Cardiocondyla wassmanni [sic] Santschi, 1926: 241. Holotype worker, CAMEROUN: Gr. Batanga (R. P. E.
Wasmann) (NM, Basle) [examined]. Syn. n.
Cardiocondyla wasmanni var. sculptior Santschi, 1926: 241. Holotype worker, GABON: Samkita (F. Faure)
(NM, Basle) [missing from mount]. Syn. n.
Cardiocondyla brevispinosa Weber, 1952: 6. Holotype worker, ZAIRE: Beni, lat. 0° 24' N., long. 29° 24' E.,
24.ii.1948, no. 2116 (N. A. Weber) (not in AMNH, New York; presumed lost). [Junior secondary homo-
nym of Pheidole brevispinosa Donisthorpe 1947: 593 (= Cardiocondyla paradoxa Emery); synonymy by
M. R. Smith, 1955: 305.] Syn. n.
Cardiocondyla fusca Weber, 1952: 7. Holotype worker, UGANDA: Jinja, 15.viii.1939, no. 1495 (N. A. Weber)
(not in AMNH, New York ; presumed lost). Syn. n.
WORKER. TL 2.0-2.6, HL 0.50-0.60, HW 0.38-0.46, CI 75-79, SL 0.36-0.45, SI 93-100, PW 0.27-0.35, AL
0.54-0.69 (35 measured).
Antennal scapes when laid back on the head in full-face view either just failing to reach or just reaching
the occipital corners, never distinctly surpassing them; the scapes moderately long, SI > 90. Maximum
diameter of eye 0.1 1-0.14, about 0.26-0.30 x HW and with 9-12 ommatidia in the longest row. Head always
obviously longer than broad, CI < 80 in material examined. Pronotal corners in dorsal view broadly and
evenly rounded. With the alitrunk in profile the promesonotal dorsum forming an even shallow convexity
from front to back, sloping evenly into the metanotal groove. Metanotal groove impressed but the depth of
the impression varying between samples. To some extent the apparent variation in depth is caused by the
convexity of the propodeum behind the groove as in some cases it rises more steeply and is more convex
than in others. Propodeal dorsum behind the convex portion sloping downwards posteriorly to the junction
with the declivity. Propodeal armament very reduced, at best represented only by a pair of minute triangular
denticles which may be acute or blunted, or by a pair of tubercles, or merely bluntly angular; never with
developed teeth or spines (Fig. 1). In dorsal view the propodeal armament scarcely visible, the length of each
component constituting only a fraction of the distance separating their bases. Petiole node in dorsal view
subglobular, usually broader than long but in some only about as broad as long. Postpetiole distinctly
broader than long. In profile the petiole and postpetiole as in Fig. 1, the petiolar dorsum convex and
somewhat variable in length. Sculpture of dorsal head and alitrunk usually of fine, very dense blanketing
shagreening or granulation, but this may be reduced on the alitrunk or even on the head, though less
frequently on the latter than on the former. In extreme cases the dorsal alitrunk may be almost smooth.
Hairs absent except on mouthparts and gastral apex. Colour varying from medium brown to blackish
brown, sometimes black.
The commonest and most widespread endemic species in the Afrotropical region, shuckardi is
recognised by its dimensions and extremely reduced propodeal armament. Other species in the
region with reduced propodeal armament include monardi, sekhemka and nilotica. In the first of
these the metanotal groove is absent and the pedicel segments are very elongate (Figs 1, 3). C.
sekhemka is a much smaller species with shorter scapes and a broader head, and nilotica has
longer scapes than shuckardi and a narrower postpetiole.
MATERIAL EXAMINED
Ghana: Kibi (D. Lesion); Mampong (P. Room); Mole G. R. (J. C. Greig). Nigeria: Ibadan (K. Whitney};
Ibadan (B. Critchley). Cameroun: Nkoemvon (D. Jackson); Batanga (Wasmann). Zimbabwe: Umtali (G.
Arnold). Botswana: Shorobe (A. Russell-Smith). South West Africa: Okahanja (P. Hammond); Omaruru (W.
Michaelsen). South Africa: Transvaal, Plaston (M. Samways); Nelspruit (M. Samways); Natal, Ubombo (W.
L. & D. E. Brown); Illovo (P. Atkinson). Mozambique: Amatongas Forest (G. Arnold). Madagascar: Mont
d'Ambre (J. M. Betsch); Antananarivo (Camboue).
AFROTROPICAL MYRMICINE ANT GENERA 317
Cardiocondyla weserka sp. n.
(Fig. 6)
HOLOTYPE WORKER. TL 1.9, HL 0.46, HW 0.35, CI 76, SL 0.32, SI 91, PW 0.25, AL 0.48.
Antennal scapes moderately long but when laid back on the head failing to reach the occipital corners in
full-face view. Maximum diameter of eye 0.12, about 0.34 x HW and with 9-10 ommatidia in the longest
row. Pronotum in dorsal view with the corners narrowly rounded but not prominent. With the alitrunk in
profile the promesonotum with its dorsum almost flat, rounding broadly into its anterior declivity but
running into the metanotal groove almost in a straight line, with only the feeblest of curves. Metanotal
groove narrowly but quite distinctly impressed. Behind the metanotal groove the propodeal dorsum more
or less flat and on a slightly higher level than the posterior part of the promesonotum; the propodeal
convexity behind the metanotal groove followed by a long slope down to the spines, which is characteristic
of most species of the region, is absent here. Propodeal spines elongate and narrow, much longer than their
basal width in profile; in dorsal view the spines somewhat incurved, each spine easily as long as the distance
separating their bases. Shape of pedicel segments in profile as in Fig. 6. In dorsal view the petiole node
conspicuously longer than broad, its dorsal surface narrow. Postpetiole much broader than long, its anterior
face slightly concave, its sides convex. Dorsum of head blanketed by a fine dense granular sculpture or
shagreening. Dorsal promesonotum more lightly shagreened than head, the sculpture here being extremely
fine and very dense indeed. Propodeal dorsum with same sculpture as promesonotum but somewhat weaker
and appearing shiny in places. Petiole and postpetiole very finely and superficially shagreened. Hairs absent
except on mouthparts but a fine appressed pubescence is present, most easily visible on the first gastral
tergite. Alitrunk medium brown, the appendages slightly lighter. Head dorsally and gaster blackish brown
to black. Pedicel segments intermediate in shade between alitrunk and gaster.
Holotype worker, Cameroun: Nkoemvon, 1980, no. M35 (D. Jackson) (BMNH).
Among the species of the region in which the metanotal groove is impressed, weserka is
immediately distinguished by the shape of the propodeal dorsum. In general the propodeal
dorsum is convex behind the groove and then enters a long slope down to the tubercles, spines or
teeth (Figs 1, 4, 5, 7), but in weserka the dorsum is almost flat and does not conform to this usual
shape (Fig. 6).
Cardiocondyla wroughtonii (Forel)
(Fig. 4)
Emeryia wroughtonii Forel, 1890: cxi. Holotype male [ergatoid, not worker], INDIA: Poona (Wroughtori)
(MHN, Geneva) [examined].
Cardiocondyla wroughtonii (Forel) Forel, 1892: 313.
Cardiocondyla wroughtonii var. hawaiensis Forel, 1899: 119. Syntype workers, HAWAII: Molokai (MHN,
Geneva). [Synonymy by Wilson & Taylor, 1967: 56.]
Cardiocondyla emeryi subsp. chlorotica Menozzi, 1930: 84. Syntype workers, female, SOMALI REPUBLIC:
Duca Abruzzi, x.1926 (G. Paoli & A. Chiaromonte) (IE, Bologna) [examined]. Syn. n.
WORKER. TL 1.6-1.9, HL 0.42-0.50, HW 0.34-0.40, CI 79-86, SL 0.30-0.36, SI 81-89, PW 0.24-O.28, AL
0.46-0.55 (25 measured).
Small species with relatively broad head and short scapes, CI and SI above. When laid back on the head
the scapes failing to reach the occipital corners in full-face view. Maximum diameter of eye 0.09-0.11, about
0.26-0.30 x HW and with 9-11 ommatidia in the longest row. Pronotal corners rounded in dorsal view.
With the alitrunk in profile the promesonotum forming a shallow convexity from front to back but the slope
changing sharply posteriorly and becoming quite steep where it slopes down to the strongly impressed
metanotal groove; this change in slope very conspicuous in absolute profile. Propodeal dorsum behind the
metanotal groove convex in profile, then entering a long downward slope to the propodeal spines.
Propodeal spines enlongate and narrow in profile, longer than their basal width; in dorsal view each spine as
long as the distance separating their bases. Petiole node in dorsal view subglobular, as broad as or slightly
broader than long. Postpetiole distinctly broader than long. Dorsal surfaces of head and alitrunk blanketed
by fine shagreening or punctulate shagreening. Petiole and postpetiole finely superficially shagreened. Hairs
absent except on mouthparts and gastral apex but a sparse appressed pubescence is present, easiest seen on
the first gastral tergite. Head, alitrunk and appendages yellow to yellowish brown, colour of gaster variable.
Frequently the gaster is the same colour as the head and alitrunk but in some the sides of the tergite are
318 B. BOLTON
darker than the dorsum. In others the darker colour has also extended across the posterior portion of the
first tergite and in some the gaster is uniformly dark.
A tramp species probably originating in South East Asia, wroughtonii is now widespread in the
tropics and subtropics. Amongst the Afrotropical region species wroughtonii is recognizable by its
small size, relatively short scapes and broad head, subglobular petiole node in dorsal view, and
the characteristic shape of the promesonotum in profile. In terms of CI it is approached only by
sekhemka, but this species is uniformly dark in colour, has much shorter scapes (SI 74), and has a
differently shaped alitrunk.
MATERIAL EXAMINED
Afrotropical Region. Somali Republic: Duca Abruzzi (Paoli & Chiaromonte). Tanzania: Dar es Salaam (A.
J. Halstead); Zanzibar (M. J. Way).
Other regions. West Malaysia: Alor Star (G. H. Lowe); Gombak (B. Bolton). Australia: Qld, Mackay
(R. E. Turner). Japan: Chichi-jima, Ogasahara (M. Tanaka). Hawaii: Molokai (R. C. L. Perkins). Sri Lanka:
Peradeniya (A. Rutherford); Nawalapitiya. India: Poona (Wrought on); Pusa (S. D. Agarwala). Thailand.
U.S.A.: Fla, Dade Co., Tamiami Trail (W. F. Bur en).
Cardiocondyla zoserka sp. n.
HOLOTYPE FEMALE. TL 3.3, HL 0.68, HW 0.55, CI 81, SL 0.46, SI 84, PW 0.47, AL 1.04.
With the head in full-face view the outer margins of the mandibles conspicuously sinuate, passing through
a right-angle apically and forming a flat transverse anterior margin along to the apical tooth. Masticatory
margin of mandible with the usual five teeth but the apical tooth considerably enlarged, the three basalmost
teeth very small. Form of clypeus more Leptothorax-like than is usual in the genus, with a broadly and
evenly convex anterior lobe which projects over the base of the mandibles and with an impressed area
between the frontal lobes behind the posterior margin of the clypeus. Funicular segments of antennae with
bizarre modification and highly characteristic. In dorsal view funicular segment 1 slightly longer than broad,
2 slightly broader than long, but thereafter segments 3-10 short and very broad, becoming even broader
apically and with segments 8-10 extremely broad. The apical funicular segment swollen-conical in dorsal
view. In ventral view the funiculus even more bizarre. Segments 1-5 appearing the same as in dorsal view,
segments 6-7 flattened dorsoventrally, segment 8 slightly transversely concave, the very broad segment 9
strongly transversely concave and segment 10 so concave that the strongly arched ventral surface appears
almost to touch the dorsal at the point of maximum concavity. Apical segment invaginated and forming a
cup-shaped hollow which extends deep into the segment. Ocelli distinct. Maximum diameter of eye 0.24,
about 0.44 x HW. With alitrunk in dorsal view the mesoscutum slightly broader than long, the rounded
pronotal corners visible anteriorly. In profile the propodeal dorsum sloping down posteriorly to a pair of
small acute denticles. Petiole and postpetiole nodes both distinctly broader than long in dorsal view. Dorsal
surfaces of head, mesoscutum and scutellum granular to shagreened, with scattered punctures, the
mesoscutum also with very faint striate vestiges longitudinally. Dorsal propodeum with ground-sculpture
vestigial to absent, with a few feeble transverse rugulae. Petiole, postpetiole and gaster with scattered minute
punctulae dorsally. Hairs absent except on mouthparts but the body with a fairly dense and quite
conspicuous appressed pubescence which is most easily visible on the first gastral tergite. Colour dark
brown to blackish brown, the appendages lighter.
PARATYPE FEMALES. TL 2.9-3.3, HL 0.62-0.67, HW 0.51-0.55, CI 82-84, SL 0.42-0.46, SI 82-85, PW
0.42-0.46, AL 0.90-1.00 (4 measured).
As holotype but may be slightly lighter in colour. Sculpture reduced in some, the propodeal dorsum
almost smooth and the dorsal alitrunk less intensely sculptured. Maximum diameter of eye 0.21-0.24, about
0.41-0.44 x HW.
Holotype female, Nigeria: nr Abuja, Gurara Falls, 20.iii.1972 (E. Classey) (BMNH).
Paratypes. 4 females with same data as holotype (BMNH; NM, Basle; MCZ, Cambridge).
Although it is not usual practice to describe ant species from isolated females I make an exception
in this case for two reasons. Firstly, the modification of the mandibles, clypeal structure and
antennal funiculi lead me to suspect that this species is an inquiline. Secondly, the bizarre
modification of the funiculi renders the species immediately recognizable. To the best of my
knowledge no other ant has funiculi even remotely resembling this one, and certainly they cannot
be confused with any other member of Cardiocondyla. Assuming that I am correct in my
AFROTROPICAL MYRMICINE ANT GENERA 319
supposition that zoserka is an inquiline species (which makes it the first one known in the genus),
it is interesting to speculate what its host might be. Apart from the modifications of the head and
its appendages the overall appearance of zoserka is very like that of shuckardi females. The two
are definitely closely related and it may be that shuckardi represents the host of zoserka.
LEPTOTHORAX Mayr
(Figs 8-22)
Leptothorax Mayr, 1855: 431. Type-species: Formica acervorum F., 1793: 358, by subsequent designation of
Bingham, 1903:214.
Temnothorax Mayr, 1861: 68. Type-species: Myrmica (Leptothorax) recedens Nylander, 1856: 94, by
monotypy. [Synonymy by Forel, 1890 7 by increase of the denticle series.]
Propodeal spiracle shifted back and down, set
behind the midlength; the spiracle usually in the
posteroventral quadrant of the side of the
propodeum.
Apart from the few African species revised below the taxonomy of most of the Old World
fauna of Leptothorax is in a poor condition. Only the faunas of North America (Creighton, 1950;
Brown, 1955) and of the Neotropical region (Kempf, 1959; Urbani, 1978) have been studied in
any detail. The west European fauna is mostly covered by Bernard (1968), Collingwood (1978;
1979) and Kutter (1977) but the remainder of the Old World remains unstudied by modern
methods.
Most of the generic synonymy noted above is straightforward and needs no further comment
here; a few, however, require further explanatory notes, as follows.
Temnothorax, synonymized long ago by Forel (1890a) on the grounds that it graded into
Leptothorax, has frequently been resurrected by European authors and treated either as a
subgenus of Leptothorax or even as a separate genus (most recently by Bernard, 1968). The
reason for this is not hard to find for among the west European species recedens, the type-species
of Temnothorax, stands out as an oddity as it does not belong to any of the usual west European
species-groups. However, when the extensive North African fauna is considered recedens is seen
as a fairly unexceptional Leptothorax species, and when the world fauna is taken into
consideration it seems decidedly mundane. The truth of the matter appears to be that recedens,
along with a few other species, really belongs to the North African fauna but has managed to
establish itself north of the Mediterranean. Urbani (1971) has discussed the validity of
Temnothorax and concluded that Forel's approach was the only logical one. I agree completely
and thus the original synonymy of Forel stands.
Tetramyrma, originally described as a subgenus of Dilobocondyla and later transferred into the
Tetramoriini, was recognized by Bolton (1976) to be only dubiously separable from Leptothorax.
On closer study it has not proved possible to find any genus-level characters to keep the name
separate. The type-species of Tetramyrma, braunsi, seems odd at first sight because of its domed
petiole and rounded, unarmed propodeum, but these developments are foreshadowed in maximus
Santschi and its allies. L. simoni, the only other species ever placed in Tetramyrma, provides a
good link back into the main mass of Leptothorax species, showing as it does a pair of propodeal
teeth whilst otherwise resembling braunsi very closely.
Nesomyrmex, with its own set of earlier synonyms (Caulomyrma, Goniothorax, Limnomyrmex),
is here formally synonymized with Leptothorax for the first time. Brown (1973) placed it as a
possible synonym in his world list of genera. Some members of this predominantly tropical group
appear very odd as a number of them have the petiole node denticulate, others have dentate
pronotal corners and many have very prominent clypeal lobes. However, there do not appear to
be any characters, either alone or in combination, which can serve to keep the former
Nesomyrmex species separate from the mass of Leptothorax. The largest representation of this
group occurs in South America and has been revised by Kempf (1959). His definition does not
separate Nesomyrmex from Leptothorax and one of his stated characters, the 5,3 palp formula,
seems universal in the genus. Species formerly placed in Nesomyrmex show considerable
variation in form and grade into more ordinary Leptothorax in all their specialized characters.
322 B. BOLTON
In my opinion all the earlier synonymy quoted above is valid and none of the included names
is deserving of further recognition as none of the characters invoked to separate them is
consistent or particularly functional. In fact, the similarities so enormously outweigh the
supposed differences, and the assumed diagnostic characters are so variable both within and
between the supposed subgenera, that the subgeneric system used in Leptothorax was at best
artificial, at worst misleading.
The only remaining subgeneric name in Leptothorax is Macromischa Roger ( = Antillaemyrmex
Mann, = Croesomyrmex Mann). Until recently this was treated as a good genus but Urbani
(1978), in his revision of the group, showed that the more exotic species (formerly in
Macromischa) graded into the more ordinary Leptothorax groups without it being possible to
draw any meaningful dividing line. However, instead of sinking Macromischa he chose to treat it
as a subgenus, though with considerable apprehension as some of the characters used are also
demonstrable, as Urbani says, elsewhere in Leptothorax, whilst others are not consistent through
Macromischa itself. The implication is that Macromischa is best regarded as a synonym of
Leptothorax.
The closest relatives of Leptothorax include many small inquiline or dulotic genera, all of
which are derived directly from Leptothorax. These genera are Chalepoxenus Menozzi,
Harpagoxenus Forel, Epimyrma Emery, Leonomyrma Arnoldi, Myrmoxenus Ruzsky,
Doronomyrmex Kutter, Formicoxenus Mayr, Myrmetaerus Soudek, and Symmyrmica Wheeler. Of
these Epimyrma is characterized by a reduced palp formula of 4,2 or 3,2 and usually a reduced
dentition; the genus may be valid. Harpagoxenus and Chalepoxenus both have strong frontal
carinae and short scrobes. The two are basically very similar and retain the standard
leptothoracine palp formula count of 5,3. The difference of antennae 11 -segmented versus
12-segmented which is used to separate them is not convincing as both antennomere counts
occur in Leptothorax (and several other myrmicine genera). The relationship of these two needs
further study for, although Chalepoxenus was revised quite recently (Kutter, 1973) its standing
with relation to Harpagoxenus was not discussed. The older separation based on mandibular
dentition, with Chalepoxenus having dentate and Harpagoxenus edentate mandibles works for
Europe, but the North American Harpagoxenus species have teeth.
Doronomyrmex, with its two parasitic species pads Kutter and pocahontas Buschinger, seems
indefensible as a genus. Its specialized features all result from inquiline syndrome characters
common to numerous parasitic but otherwise unrelated ants. The same appears to be true of
Myrmetaerus and Myrmoxenus, although further study of all these is needed. More information is
also required of Leonomyrma and Symmyrmica as both genera have short but fairly prominent
frontal carinae. The former also has the eyes shifted back on the head and the latter has 6-dentate
mandibles although this is not unknown in Leptothorax.
Finally Formicoxenus. Because of their very specialized inquiline lifeways in the nests of much
larger formicine ants Formicoxenus species have always presented a problem. Until recently the
genus only contained the two Palaearctic species nitidulus (Nylander) and orientalis Dlussky, and
was separated from Leptothorax by its possession of a strongly dentate subpostpetiolar process.
This postpetiolar development is a common feature in many unrelated inquilines from all parts of
the Myrmicinae and is a recognized character of the inquiline syndrome. It should not, by itself,
be regarded as being of generic significance. Dissection of nitidulus has, however, shown that the
mandibles are apparently consistently 6-dentate and the palp formula is reduced to 4,3. These
characters, coupled with the 11 -segmented antennae (again not a strong character when taken
alone) combine to form a reasonable case for maintaining Formicoxenus as a genus. An
observation in support of this comes from the decision of Buschinger (1979) to transfer the
American species hirticornis Emery and diversipilosus M. R. Smith from Leptothorax to
Formicoxenus on the grounds that their social organization is the same as in the European
nitidulus, and despite the fact that they lack a strong subpostpetiolar process. Dissection of
hirticornis has shown a 4,3 palp formula and 6-dentate mandibles as in nitidulus. I have not been
able to dissect any diversipilosus but a similar dentition and palp formula there would reinforce
the case for maintaining Formicoxenus as a genus separate from Leptothorax.
AFROTROPICAL MYRMICINE ANT GENERA 323
Synonymic list of Afrotropical Leptothorax species
angular us Mayr
angulatus st. ilgii Forel syn. n.
latinodis Mayr syn. n. (provisional)
angulatus var. concolor Santschi syn. n.
braunsi (Forel) comb. n.
cenatus sp. n.
denticulatus Mayr
evelynae Forel
grisoni Forel
humerosus Emery
innocens (Forel)
megalops Hamann & Klemm
simoni (Emery) comb. n.
stramineus (Arnold)
Key to species (workers)
1 With the alitrunk in absolute profile the dorsum forming a single uninterrupted surface which is
evenly flat or slightly convex, without trace of a metanotal impression and not having the
propodeum depressed (Fig. 8) 2
With the alitrunk in absolute profile the dorsum with the metanotal groove impressed even if
only feebly so, or the propodeum depressed below the level of the promesonotum, or both
(Figs 14-22) ... 3
2 Head and body uniform blackish brown to black. (Ghana, Zaire) .... grisoni (p. 329)
Head and body uniform yellow. (Extremely widespread) angulatus (p. 324)
3 First gastral tergite everywhere with blunt standing hairs 4
First gastral tergite either without standing hairs at all or at most with a single transverse row at
theapexofthesclerite 9
4 Petiole node narrow in profile, not denticulate (Figs 14-16). Antennal scapes longer, SI 85-1 10.
Eyes larger, maximum diameter 0.30-0.38 x HW 5
Petiole node broad in profile, denticulate (Figs 18-20). Antennal scapes shorter, SI 68-74. Eyes
smaller, maximum diameter 0.24-0.29 x HW 7
5 Anterior pronotal angles projecting as a pair of acute teeth in dorsal view; sides of pronotum
sharply marginate. Petiole node sharply triangular in profile (Fig. 16). Scapes relatively shorter
and head broader (Fig. 12), SI 85, CI 83. (' East Africa ') humerosus (p. 329)
Anterior pronotal angles evenly bluntly rounded in dorsal view; sides of pronotum not
marginate. Petiole node not sharply triangular in profile (Figs 14, 15). Scapes relatively longer
and head narrower (Figs 1 1, 13), SI 107-1 10, CI 70-78 .... 6
6 Eyes larger, maximum diameter 0.38 x HW. Petiole node in profile without a strongly
differentiated posterodorsal angle (Fig. 14). Mandibles almost smooth, with only vestiges of
sculpture. (Sudan) megalops (p. 331)
Eyes smaller, maximum diameter 0.30-0.31 x HW. Petiole node in profile with a strongly
differentiated posterodorsal angle (Fig. 15). Mandibles with strong but fine longitudinal
rugular sculpture. (Kenya) cenatus (p. 327)
7 Subpetiolar process a tooth anteriorly followed by a long cuticular flange which runs back to
the postpetiolar junction (Fig. 19). Eyes with 10-11 ommatidia in the longest row. Larger
species, HW 0.62-0.68, PW 0.46-0.52. (South Africa) . . . denticulatus (p. 328)
Subpetiolar process an anteriorly situated simple tooth or denticle (Figs 18, 20). Eyes with 7-8
ommatidia in the longest row. Smaller species, HW 0.49-0.53, PW 0.35-0.38 . 8
8 Propodeal spines short and broad, in profile about as long as their basal width, the declivity
between the spines and the metapleural lobes concave (Fig. 20). Dorsum of head densely and
sharply reticulate-punctate, with traces of fine rugulae. (Zaire) . . . innocens (p. 330)
Propodeal spines long and narrow, in profile distinctly longer than their basal width and
slightly downcurved, the declivity between the spines and the metapleural lobes straight
(Fig. 18). Dorsum of head weakly superficially reticulate-punctate, without trace of rugulae.
(South Africa) stramineus (p. 332)
9 Propodeum unarmed (Fig. 22). (South Africa) . . braunsi (p. 325)
- Propodeum armed with a pair of spines or teeth (Figs 17, 21) 10
324 B. BOLTON
10 Eye with only 7-8 ommatidia in the longest row. Alitrunk shaped as in Fig. 17. Small yellow
species with longer scapes, HW < 0.60, SI > 90. (Ghana, Zaire) .... evelynae (p. 328)
Eye with 15-16 ommatidia in the longest row. Alitrunk shaped as in Fig. 21. Large reddish
species with darker gaster and shorter scapes, HW > 0.85, SI < 85. (South Africa) . simoni (p. 331)
The few species constituting the Afrotropical fauna of Leptothorax apparently represent outliers
derived from a number of different species-groups of extralimital origin, one or two species from
each of which have managed to enter the region and to survive there. Because of the
unsatisfactory state of the taxonomy of Leptothorax the species-group limits have not been
worked out, but the 1 1 species occurring in sub-Saharan Africa aggregate as follows.
L. angulatus and grisoni. Metanotal groove absent. SI > 85. Eyes large, with 15 or more ommatidia in the
longest row. Pronotal corners acute. Petiole node large, with a short anterior peduncle; the node sculptured
but not denticulate. Frontal carinae absent. Median clypeal lobe more or less evenly convex.
L. denticulatus, innocens and stramineus. Metanotal groove present. SI < 75. Eyes relatively small, with
7-10 ommatidia in the longest row. Pronotal corners blunt. Petiole node large and denticulate, with a
moderately long anterior peduncle. Frontal carinae absent and the median clypeal lobe more or less evenly
convex.
L. braunsi and simoni. Metanotal groove present and the propodeum somewhat depressed below the level
of the promesonotum. SI in intermediate range, 78-83. Eyes large, with 16-18 ommatidia in the longest row.
Pronotal corners rounded. Petiole node massive and domed, not denticulate and with a moderately long
narrow peduncle. Clypeal lobe conspicuously produced; frontal carinae absent.
L. evelynae, cenatus and megalops. Metanotal groove present but shallow, sometimes very shallow.
SI > 90. Eyes relatively small to moderate, with 7-12 ommatidia in the longest row. Pronotal corners
bluntly angular to evenly rounded. Petiole node small, without denticles and with a moderately long
peduncle. Frontal carinae very feeble to absent and the median clypeal lobe more or less evenly rounded.
L. humerosus. Metanotal groove present. SI 85. Eyes large, with 14-15 ommatidia in the longest row.
Pronotal corners sharply dentate, the sides of the pronotum sharply marginate. Petiole node acutely
triangular, not denticulate, with a short peduncle. Feeble frontal carinae present and the median clypeal lobe
conspicuously produced.
Leptothorax angulatus Mayr
(Figs 8, 9)
Leptothorax angulatus Mayr, 1862: 739. LECTOTYPE worker, EGYPT: 'auf der sinaitischen Halbinsel' (R.
v. Frauenfeld) (NM, Vienna), here designated [examined].
Leptothorax angulatus st. ilgii Forel, 1894: 82. Holotype worker, ETHIOPIA: ' Sudabessinien ' (A. Ilg) (MHN,
Geneva) [examined]. Syn. n.
Leptothorax latinodis Mayr, 1895: 130. Holotype worker, MOZAMBIQUE: Delagoa Bay (H. Brauns) (not
found, presumed lost). Syn. n. (provisional).
Leptothorax angulatus var. concolor Santschi, 1914a: 107, fig. 15. Syntype workers, KENYA: M6mbasa, st.
no. 3, x.1911 (Alluaud & Jeannel) (NM, Basle) [examined]. Syn. n. [Data labels on syntypes read L.
(Goniothorax) angulatus var. concolor. ,]
WORKER. TL 3.1-3.8, HL 0.70-0.90, HW 0.56-0.74, CI 75-85, SL 0.50-0.66, SI 88-97, PW 0.40-0.56, AL
0.82- 1.08 (65 measured).
Mandibles delicately but densely longitudinally striate, the striation usually distinct but sometimes
superficial. Median clypeal lobe extensive, broad, covering the bases of the mandibles and having its anterior
margin conspicuously arched-convex. Median clypeal carina fine, not strongly developed but usually
discernible, only rarely the carina partially or wholly effaced. Antennal scrobes absent. Frontal carinae
absent but in some the frontal lobe followed on one or both sides by a weak rugular line which runs back on
the head. Maximum diameter of eyes 0.17-0.22, about 0.27-0.33 x HW and with 13 or more ommatidia in
the longest row. With the head in full-face view the sides narrower in front of the eyes than behind, slightly
convergent anteriorly. Sides of head behind eyes shallowly convex, slightly convergent posteriorly and
meeting the occipital margin in a blunted angle. Occipital margin transverse to very shallowly concave, with
a slightly projecting rim above the occipital foramen which is visible in full-face view. With the alitrunk in
profile the dorsum forming a single shallowly convex to almost flat surface, without trace of a metanotal
impression. Propodeum armed with a pair of triangular teeth or short broad spines of variable size. In
general the teeth are about as long as their basal width and slightly upcurved, but individuals with spines
AFROTROPICAL MYRMICINE ANT GENERA 325
longer than their basal width are fairly common. Specimens with the propodeal armament reduced to short
broad teeth, where they are shorter than the basal width, are less common. Metapleural lobes low and
rounded. In dorsal view the alitrunk with angulate to weakly dentate pronotal corners. Mesonotum
narrower than pronotum and the sides of the propodeum diverging to the level of the spiracle and then
converging to the bases of the propodeal teeth. Petiole in profile shaped as in Fig. 8, with a short anterior
peduncle which has a triangular dentiform anteroventral process. Dorsal surface of peduncle with a
denticulate process in front of the level of the spiracle on each side. Anterodorsal angle of node quite sharply
defined, the posterodorsal angle much broader and bluntly rounded. Petiole node in dorsal view variable in
shape and size. Usually the node about as broad as long, rarely slightly longer than broad but quite
commonly obviously broader than long, in some cases approaching the postpetiole in width. Dorsum of
head covered with a fine dense reticulate-punctulate ground-sculpture which in some samples is superficial
and granular in appearance. Superimposed on this are very fine irregular rugulae which frequently form a
narrow reticulum occipitally and sometimes also on the sides of the head. Dorsal surfaces of alitrunk, petiole
and postpetiole with fine granular or punctulate ground-sculpture and with disorganized fine rugulae. The
rugular sculpture is usually distinctive but in some individuals may be partially effaced. Base of first gastral
tergite generally with a superficial reticular pattern but sometimes almost completely smooth. All dorsal
surfaces of head and body with numerous short stout blunt hairs ; such hairs absent from the appendages.
Colour yellow, frequently with the antennal club darker.
L. angulatus is the most widely distributed and commonest species of this genus in sub-Saharan
Africa. It is easily identified by its yellow colour and lack of any trace of a metanotal groove or
impression. Only one other species in the region lacks a metanotal groove, grisoni, but in this
species the full adult colour is uniform blackish brown or black.
Arnold (1916: 259) noted that he only found angulatus on the trunks of trees but personal
observation has shown that it also occurs in leaf litter samples and log mould. However, the
species does seem to prefer to nest clear of the ground when possible, as colonies are often found
in West Africa in cocoa pods which are still attached to the tree, and the sample from Malawi
noted below was collected in Swartzia pods.
MATERIAL EXAMINED
Egypt: Sinai (Frauenfeld). Ghana: Legon (D. Lesion); Tafo (B. Bolton); Tafo (C. A. Collingwood); Adeiso
(P. Room); Adeiso (D. Lest on). Nigeria: Gambari (B. Taylor). Ethiopia: ' Sudabessinien ' (A. Ilg). Sudan:
Equatoria (N. A. Weber); Port Sudan (N. A. Weber); Nile above Khartoum (N. A. Weber). Kenya: Nairobe
(Patrizi); Mombasa (Alluaud & Jeannel); Tana Riv., Wema (V. Mahnert & J.-L. Ferret). Tanzania: Dar es
Salaam (N. L. H. Krauss). Malawi: nr Salima (B. J. S.). Zimbabwe: Victoria Falls (G. Arnold); Melsetter (G.
Arnold); Khami Riv. (G. Arnold). Botswana: Maxwee (A. Russell-Smith). South Africa: Natal, St Lucia (J. C.
Faure).
Leptothorax braunsi (Forel) comb. n.
(Figs 10, 22)
Dilobocondyla (Tetramyrma) braunsi Forel, 1912: 767. Holotype worker, SOUTH AFRICA: Cape Colony,
Willowmore (H. Brauns) (BMNH) [examined].
Tetramyrma braunsi (Forel) Forel, 19136: 122. [See also Bolton, 1976: 291.]
WORKER. TL 5.2-5.9, HL 1.20-1.36, HW 1.00-1.16, CI 83-86, SL 0.82-0.94, SI 79-83, PW 0.78-0.96, AL
1.44- 1.62 (9 measured).
Mandibles finely longitudinally striate, the spaces between striae finely punctulate or shagreened; the
striate sculpture sometimes inconspicuous. Median lobe of clypeus prominent (Fig. 10), its anterior margin
shallowly and evenly convex. Frontal carinae and antennal scrobes absent, the scapes of moderate length (SI
above). Maximum diameter of eye 0.28-0.31, about 0.26-0.29 x HW and with 16-18 ommatidia in the
longest row. In full-face view the head shaped as in Fig. 10. Alitrunk and pedicel segments in profile as in
Fig. 22, the promesonotum evenly convex, the metanotal groove not or only slightly impressed but the
propodeal dorsum distinctly depressed below the level of the promesonotum. Propodeum absolutely
unarmed, the dorsum rounding evenly into the declivity. In dorsal view the pronotal corners rounded, the
promesonotum narrowing posteriorly. Metapleural lobes rounded. Node of petiole in profile massive, with a
relatively narrow anterior peduncle which has a dentiform anteroventral process. In dorsal view the petiole
node subglobular, slightly broader than long; postpetiole broader than long and broader than the petiole.
Dorsum of head longitudinally rugulose with a few cross-meshes, occipitally a weak reticulum may be
326
B. BOLTON
17
19
22
Figs 17-22 Leptothorax workers. Alitrunk and pedicel segments of (17) evelynae, (18) stramineus, (19)
denticulatus, (20) innocens, (21) simoni, (22) braunsi.
AFROTROPICAL MYRMICINE ANT GENERA 327
formed. Sides of head above and behind eyes generally more obviously reticulate than the dorsum. Dorsal
alitrunk irregularly rugose, the sculpture quite strong, usually forming a reticulum on the propodeum and
anterior pronotum. Petiole and postpetiole irregularly reticulate-rugose. First gastral tergite densely
punctulate or shagreened, the sculpture generally strongest basally and usually traces of very fine
longitudinal costulae may be seen. A few short inconspicuous erect hairs present on dorsum of head but the
dorsal alitrunk, petiole and postpetiole hairless. First gastral tergite without standing hairs but with a short
fine appressed sparse pubescence. Appendages without standing hairs. Head and gaster dark brown tinged
with red to reddish black; alitrunk and pedicel segments red, the two colours strongly contrasting in fresh
specimens.
This large and conspicuous South African species is easily recognized by its large size, unarmed
propodeum, lack of hairs on alitrunk and first gastral tergite and depressed propodeal dorsum.
The closest related species in sub-Saharan Africa is simoni, but here the propodeum is distinctly
bidentate.
MATERIAL EXAMINED
South Africa: Cape Prov., Willowmore (//. Brauns).
Leptothorax cenatus sp. n.
(Figs 11, 15)
HOLOTYPE WORKER. TL 3.6, HL 0.78, HW 0.60, CI 77, SL 0.64, SI 107, PW 0.47, AL 0.98.
Mandibles finely but strongly longitudinally rugulose. Anterior clypeal margin convex and concealing the
basal tooth of the mandibles. Median clypeal carina feebly developed, weaker than the more laterally
situated clypeal carinae, which converge anteriorly. The anteriormost clypeal carina runs across the clypeus
in an unbroken transverse arc just behind the anterior margin, terminating at the antennal fossa on each
side. Frontal carinae represented by a pair of feeble meandering rugula-like ridges which run back from the
narrow frontal lobes to a point behind the level of the posterior margins of the eyes; these carinae scarcely
stronger than the rugular sculpture of the head and merging with that sculpture posteriorly. Antennal
scrobes absent, the scapes relatively long, SI > 100. Eyes quite large, maximum diameter 0.18, about
0.30 x HW and with 1 1 ommatidia in the longest row. With the head in full-face view the occipital margin
shallowly transversely convex, the occipital corners rounded. With alitrunk in dorsal view the pronotal
corners rounded. With alitrunk in profile the promesonotum shallowly evenly convex, the metanotal area
broadly but shallowly impressed and the propodeum armed with a pair of acute narrow spines. Metapleural
lobes very low, rounded. Petiole in profile with a moderately long anterior peduncle, the dorsal surface of
which is confluent with the anterior face of the node, the two not separated by an angle. Node with well
developed antero- and posterodorsal angles, the dorsum between them more or less flat. In dorsal view the
dorsum of the petiole node broader than long, and the petiole narrower than the postpetiole. Dorsum of
head with fine, widely spaced, irregular rugulae which are predominantly longitudinal. Occipitally the
rugulae are more sharply defined and have a few cross-meshes, although no reticulation is developed. Spaces
between rugulae smooth or at most with only vestiges of ground-sculpture. Sides of head above eyes
sculptured as dorsum but both in front of and behind the eyes the rugulae are more crowded and tend to
form a loose reticulum. Promesonotal dorsum finely and predominantly longitudinally rugulose, with very
sparse cross-meshes. The rugulae widely spaced and with a ground-sculpture of extremely fine superficial
punctulae between them, which in places is almost effaced. Rugulae present on propodeal dorsum but
weaker than on promesonotum. Petiole and postpetiole with fine superficial shagreening and a few weak
inconscpicuous rugulae. First gastral tergite unsculptured except for hair-pits. All dorsal surfaces of head
and body with numerous stout blunt mainly straight hairs which are shorter and more erect on the head
than on the first gastral tergite. Legs and scapes without such hairs. Dorsum of head brown, remainder of
body dull yellow with a brown tint, especially on the petiole and postpetiole which are somewhat darker
than the alitrunk and gaster but not as dark as the head.
PARATYPE WORKER. TL 3.2, HL 0.74, HW 0.58, CI 78, SL 0.62, SI 107, PW 0.42, AL 0.90.
As holotype but slightly 'smaller, maximum diameter of eye 0.18, about 0.31 x HW and with 10
ommatidia in the longest row.
Holotype worker, Kenya: Lake Nakuru, Nat. Park, 6.xi.l974, leaf litter (V. Mahnert) (MHN, Geneva).
Paratype. 1 worker, Kenya: Nakuru, Lake Elmenteita, 7.xi.l977, 1800 m (V. Mahnert & J.-L. Ferret)
(BMNH).
328 B. BOLTON
L. cenatus is closest related to megalops but differs in having smaller eyes and a differently shaped
petiole node (compare Figs 11,13 and 14, 15).
Leptothorax denticulatus Mayr
(Fig. 19)
Leptothorax denticulatus Mayr, 1901: 5. Syntype workers, female, SOUTH AFRICA: Cape Prov., Port
Elizabeth (H. Brauns) (NM, Vienna) [examined].
WORKER. TL 3.1-3.5, HL 0.74-0.84, HW 0.62-0.68, CI 81-85, SL 0.46-0.48, SI 71-74, PW 0.46-0.52, AL
0.82-0.94 (8 measured).
Mandibles finely shagreened to virtually smooth. Anterior margin of median lobe of clypeus evenly
arched-convex; median clypeal carina present and usually quite distinct. Frontal carinae absent; antennal
scrobes absent. Maximum diameter of eye 0.16-0.19, about 0.26-0.29 x HW and with 10-11 ommatidia in
the longest row; the eye in profile only very slightly longer than high. With the head in full-face view the
occipital margin straight to feebly convex, rounding evenly into the sides; the latter slightly narrower in
front of the eyes than behind and feebly convergent anteriorly. With the alitrunk in profile the
promesonotum shallowly convex dorsally, the metanotal groove impressed and the propodeal dorsum
convex. Propodeum armed with a pair of strong spines which are longer than their basal width. Metapleural
lobes rounded. In dorsal view the alitrunk with the pronotal corners bluntly angular to narrowly rounded.
Petiole node in profile large and blocky (Fig. 19), the upper sides and dorsum with numerous peaks or
denticles from which hairs arise. Peduncle of petiole short and broad, subtended by an extensive ventral
process which takes the form of a triangular denticle or tooth anteriorly, followed by a long cuticular ridge
which runs back to the junction with the postpetiole. In ventral view the subpetiolar ridge is seen to fork at
about its midlength, forming an inverted Y-shape. With the pedicel segments in dorsal view the denticles
conspicuous on the sides of both the petiole and postpetiole; both segments broader than long, the latter
somewhat broader than the former. Dorsum of head covered with a blanket of fine dense punctulate
ground-sculpture which is overlaid everywhere by dense and very fine rugular sculpture. On the dorsum the
rugulae are close and longitudinal but on the sides, above the eyes and occipitally there is a tendency for a
narrow reticulum to be formed. Dorsal alitrunk reticulate-punctate and with fine rugulae which are
predominantly longitudinal; on the promesonotum a reticulum may be formed anteriorly and in some the
rugulae are quite strongly developed. Sculpture of petiole and postpetiole dorsally predominantly
reticulate-punctate but a few fine rugulae may be present. Base of first gastral tergite superficially reticulate
to almost smooth. All dorsal surfaces of head and body densely and evenly clothed with short blunt hairs;
the appendages without such hairs. Colour uniform yellow, sometimes the posterior half of the gaster darker
than the anterior half.
Among the species in which the metanotal groove is impressed three, denticulatus, innocens, and
stramineus, have the petiole node bearing denticles from which hairs arise. Of the three
denticulatus is recognized by its strongly developed subpetiolar process, dense pilosity, larger eyes
and larger size.
MATERIAL EXAMINED
South Africa: Cape Prov., Barrydale (H. V. Daly); Port Elizabeth (H. Brauns).
Leptothorax evelynae Forel
(Fig. 17)
Leptothorax (Goniothorax) evelynae Forel, 1916: 423. Syntype workers, female, ZAIRE: St Gabriel (Kohl)
(MHN, Geneva) [examined].
WORKER. TL 2.5-2.9, HL 0.58-0.70, HW 0.47-0.54, CI 77-81, SL 0.44-0.52, SI 92-98, PW 0.35-0.44, AL
0.65-0.82 (7 measured).
Mandibles finely shagreened. Median clypeal lobe evenly arched-convex. Median clypeal canna present
but fine, incomplete in a few specimens. Frontal carinae and antennal scrobes absent. Antennal scapes
relatively long, SI > 90. Maximum diameter of eye 0.12-0.16, about 0.26-0.29 x HW and with 7-8
ommatidia in the longest row. With the alitrunk in profile the metanotal groove shallowly but
conspicuously impressed, the promesonotum evenly shallowly convex and the propodeal dorsum almost flat
to shallowly convex. Propodeal spines straight, distinctly longer than their basal width. Metapleural lobes
AFROTROPICAL MYRMICINE ANT GENERA 329
low and rounded. In dorsal view the alitrunk with the pronotal corners angulate and the sides ot the
promesonotum bluntly marginate. The dorsal surface gradually narrows from front to back but the sides of
the mesonotum are slightly convex and the sides of the propodeum diverge from the metanotal groove to
the level of the spiracle and then converge to the bases of the spines. Petiole in profile with a high narrow
node (Fig. 17) which is not equipped with denticles. Ventral process of peduncle a simple small tooth,
anteriorly situated. In dorsal view the petiole node broader than long. Dorsum of head with fine superficial
reticulate-punctate ground-sculpture which is overlaid by a very fine narrow reticulate-rugulose net
everywhere except in the area immediately behind the frontal lobes. Dorsal alitrunk with superficial
punctulate ground-sculpture overlaid by fine rugulae. In specimens from Zaire this rugular sculpture is faint
and weakly developed, forming a reticulum only on the anterior pronotum, but in material from Ghana the
rugulae are more strongly developed everywhere and reticular meshes are frequent. Petiole and postpetiole
finely punctulate or granular dorsally, sometimes with one or two fine rugulae. Base of first gastral tergite
very lightly shagreened or with a superficial reticular pattern. Dorsum of head with scattered short stout
blunt hairs. Similar hairs are present on the pronotum (several pairs), mesonotum (1-3 pairs), petiole and
postpetiole (1-2 pairs each) but are absent from the propodeum and absent from the first gastral tergite
except for a transverse row at the extreme apex. Gastral segments behind the first also with a transverse
apical row each. Colour yellow.
The characteristic distribution of the body hairs renders evelynae quickly recognizable among the
African Leptothorax species. The lack of hairs on the propodeum and first gastral tergite is
paralleled only in simoni and braunsi, but these are both much larger (HW > 0.85), darker
coloured species with much more massively developed petiole nodes and shorter antennal scapes
(SI < 85). The closest related species appears to be megalops, but here the eyes are larger and the
propodeum and first gastral tergite both have hairs present.
MATERIAL EXAMINED
Ghana: Tafo (B. Bolton). Zaire: St Gabriel (Kohl).
Leptothorax grisoni Forel
Leptothorax (Goniothorax) grisoni Forel, 1916: 425. Syntype workers, male, ZAIRE: St Gabriel (Kohl) (MHN,
Geneva; MCZ, Cambridge) [examined].
WORKER. TL 2.9-3.4, HL 0.76-0.86, HW 0.60-0.66, CI 77-81, SL 0.56-0.63, SI 94-98, PW 0.46-0.53, AL
0.85-1. 02 (9 measured).
Answering to the description of angulatus, differing only in colour and intensity of sculpture. In grisoni the
full adult colour is uniform blackish brown to black, as opposed to the uniform yellow found in angulatus.
The dorsum of the head and alitrunk in grisoni is blanketed by a dense reticulate-punctate ground-sculpture
which is overlaid by conspicuous fine rugulae which form a distinct reticulum on the alitrunk and on much
of the head. Basically this sculpture is the same as that seen in angulatus, but here it is more intensely and
sharply developed.
Although these seem relatively minor differences I am prepared to accept them as valid for the
present. The reason for this is that the two colour forms have ranges which only partially overlap.
The yellow angulatus is known from most of the continent from Ethiopia and Sudan to South
Africa and Botswana, but the dark grisoni has only been found in Ghana and Zaire. L. angulatus
also occurs in Ghana in the same areas where grisoni has been discovered, but even here the two
maintain their distinctive colours, no intermediates being known.
MATERIAL EXAMINED
Ghana: Tafo (B. Bolton); Kade (J. Majer); Asamankese (P. Room). Zaire: St Gabriel (Kohl).
Leptothorax humerosus Emery
(Figs 12, 16)
Leptothorax humerosus Emery, 1896: 62. Holotype worker, 'AFRICA ORIENTALS': no loc. (Staudinger &
Bang-Haas) (MCSN, Genoa) [examined].
WORKER. TL 3.7, HL 0.90, HW 0.75, CI 83, SL 0.64, SI 85, PW 0.52, AL 0.98.
330 B. BOLTON
Mandibles almost smooth, with only vestigial traces of superficial sculpture. Median lobe of clypeus
strongly produced, roughly rectangular, its anterior margin transverse and flat; the anterior margin meeting
the sides of the lobe in a distinct angle. Median clypeal carina absent. Narrow weakly differentiated frontal
carinae present which are scarcely stronger than the remaining cephalic sculpture but which reach back well
beyond the level of the posterior margins of the eyes. Antennal scrobes absent. Antennal scapes moderately
long, SI 85. Maximum diameter of eye 0.24, about 0.32 x HW and with 14-15 ommatidia in the longest row.
With the head in full-face view the sides narrower in front of the eyes than behind and somewhat convergent
anteriorly. Sides behind eyes rounding evenly into the occipital margin, the latter shallowly and evenly
transversely convex. Pronotum sharply marginate laterally, the anterior pronotal corners dentate. With the
alitrunk in profile the promesonotum convex, the metanotal groove deeply impressed. Propodeum broadly
and evenly convex in profile, sloping down posteriorly to the long spines; the latter blunt apically and with
their dorsal margins angled (Fig. 16). Metapleural lobes low and rounded. Petiole in profile with the node
triangular, rising to an acute peak above; anterior peduncle of petiole short. In dorsal view the sides of the
petiole roughly parallel. Postpetiole much broader than petiole. Basal face of first gastral tergite transverse
except for a median concavity where it articulates with the postpetiole. Dorsum of head everywhere
sculptured with fine longitudinal rugulae and with a fine granular to punctulate superficial
ground-sculpture. Dorsal alitrunk with ground-sculpture similar to head. Pronotum also with 7-8 broad,
coarse longitudinal rugae which are almost sulcate in appearance and are most strongly developed
anteriorly. These longitudinal rugae are continuous over the length of the promesonotum and also traverse
the base of the metanotal groove, but they either fade out or become very weakly defined on the propodeum
where a punctulate ground-sculpture predominates. Petiole and postpetiole finely and densely
reticulate-punctulate, the first gastral tergite very densely finely shagreened and opaque. All dorsal surfaces
of head and body with distinctive short stout blunt hairs. Body colour more or less uniform medium brown
but the mandibles, clypeal lobe and antennae yellow. Propodeal spines yellowish, lighter in colour than the
propodeum itself.
This very distinctive species is easily separable from all other known African forms by its
flat-margined clypeal lobe, deep metanotal groove, large eyes, sharply marginate pronotum with
dentate corners and sharply triangular petiole node. It is not obviously related to any other
species of Leptothorax in the region and, as far as is known, is only represented in collections by
the holotype.
MATERIAL EXAMINED
' East Africa ' : no loc. (Staudinger & Bang-Haas).
Leptothorax innocens (Forel)
(Fig. 20)
Tetramorium (Leptothoraxl) innocens Forel, 191 3a: 317. Holotype worker, ZAIRE: Katanga, Elizabeth ville
( = Lubumbashi) (Bequaert) (MHN, Geneva) [examined].
Leptothorax innocens (Forel) Forel, 1916: 425.
WORKER. TL 2.3-2.5, HL 0.60-0.62, HW 0.49-0.50, CI 81-82, SL 0.34-0.35, SI 69-70, PW 0.35-0.36, AL
0.66-0.68 (2 measured).
Mandibles extremely finely and delicately superficially shagreened, almost smooth. Median portion of
clypeus with anterior margin evenly arcuate-convex, with a narrow cuticular apron. Median clypeal carina
vestigial to absent. Frontal carinae and antennal scrobes absent, the scapes short. Maximum diameter of eye
0.12-0.13, about 0.24-0.26 x HW and with 7-8 ommatidia in the longest row. With the head in full-face
view the sides narrower in front of the eyes than behind and somewhat convergent anteriorly. Behind the
eyes the sides very slightly convex and rounding into the occipital corners, the occipital margin itself very
feebly impressed medially to almost straight. With the alitrunk in profile the promesonotum shallowly
convex, the metanotal groove impressed and the propodeum convex, more strongly so than the
promesonotum. Propodeat spines broadly triangular and stout, about as long as their basal width.
Propodeal declivity concave between the spines and the rounded low metapleural lobes. In dorsal view the
alitrunk with the pronotal shoulders obtusely and bluntly rounded, the promesonotum narrowing to the
metanotal groove. Behind this the sides of the propodeum diverge to about the level of the spiracle, pass
through an obtuse angle and then converge again to the bases of the spines. Petiole in dorsal view with a
tubercle at either side of the node and the posterior margin of the node with a much smaller and
inconspicuous pair of tubercles. Petiole in profile with a short anterior peduncle which has a small
AFROTROPICAL MYRMICINE ANT GENERA 331
triangular process ventrally. Dorsal surfaces of head, alitrunk, petiole and postpetiole finely densely sharply
reticulate-punctate, the promesonotum and head also with traces of fine rugular sculpture. First gastral
tergite basally with very delicate superficial shagreening. All dorsal surfaces of head and body with
numerous very short blunt hairs; legs and scapes without standing hairs. Colour uniform yellow.
This small species is closest related to stramineus but the two are separated by the characters
given in the key and noted under the discussion of stramineus.
MATERIAL EXAMINED
Zaire: Katanga, Elizabeth ville (Bequaert).
Leptothorax megalops Hamann & Klemm
(Figs 13, 14)
Leptothorax (Icothorax) megalops Hamann & Klemm, 1967: 417, fig. 1. Holotype worker, and paratype
female, SUDAN : Wadi Haifa, 2811962 (H. Hamann & W. Klemm) (NM, Vienna) [examined].
WORKER. TL 2.5, HL 0.67, HW 0.47, CI 70, SL 0.52, SI 1 10, PW 0.35, AL 0.72.
Mandibles almost smooth, with faint vestiges of superficial sculpture. Median lobe of clypeus prominent,
its anterior margin evenly but shallowly convex. Median clypeal carina present, fine but distinct. Frontal
carinae and antennal scrobes absent. Head relatively longer and narrower and scapes relatively longer than
any other known species of the region (CI and SI above). Eyes relatively large, maximum diameter 0.18,
about 0.38 x HW and with 1 1-12 ommatidia in the longest row. With the head in full-face view the occipital
margin feebly indented medially, rounding broadly and evenly into the sides; the latter approximately
parallel but converging slightly anteriorly. Alitrunk in profile with the metanotal groove shallowly
impressed, the propodeal spines short, stout and straight. Metapleural lobes low and rounded, the declivity
between the spines and the metapleural lobes more or less straight. In dorsal view the pronotal angles
broadly and evenly rounded. Petiole in profile with the node bluntly triangular (Fig. 14), the anteroventral
process small and triangular. Dorsum of head with fine longitudinal rugulae the spaces between which are
smooth or at most only faintly superficially punctulate. Dorsal alitrunk with more conspicuous punctulate
ground-sculpture which is overlaid by fine, predominantly longitudinal rugulae, although these are irregular
in places. Petiole and postpetiole finely punctulate with traces of fine rugulae. First gastral tergite
unsculptured. All dorsal surfaces of head and body with quite long blunt hairs which are, however, shorter
on the head than on the alitrunk; the appendages without such hairs. Colour uniform yellow.
This very conspicuous species is easily recognized by its long narrow head, long scapes, large
eyes, rounded pronotal corners and lack of denticles on the petiole node, coupled with its
impressed metanotal groove and possession of hairs on the first gastral tergite. The closest
relatives of megalops in the Afrotropical region are evelynae and cenatus. Both are distinguished
from megalops quite easily as the former lacks pilosity on the first gastral tergite except for a
sparse apical row, and the latter has strongly sculptured mandibles, smaller eyes and a broader
head, and has the petiole node differently shaped (Figs 14, 15).
MATERIAL EXAMINED
Sudan: Wadi Haifa (Hamann & Klemm).
Leptothorax simoni (Emery) comb. n.
(Fig. 21)
Tetramorium simoni Emery, 1895ft: 35, pi. 2, fig. 22. Lectotype worker, SOUTH AFRICA: Transvaal, Makapan
(E. Simon) (MCSN, Genoa), designated by Bolton, 1976: 292 [examined].
Tetramyrma simoni (Emery) Emery, 1922a: 291. [See also Bolton, 1976: 291.]
WORKER. TL 4.7, HL 1.10, HW 0.92, CI 84, SL 0.72, SI 78, PW 0.74, AL 1.32.
Mandibles with faint longitudinal sculpture and scattered pits. Median lobe of clypeus prominent, its
anterior margin shallowly convex. Median clypeal carina distinct. Frontal carinae and antennal scrobes
absent, the scapes of moderate length. Maximum diameter of eye 0.29, about 0.32 x HW and with 17-18
ommatidia in the longest row. With the head in full-face view the occipital margin shallowly convex
centrally, more steeply convex laterally where it rounds into the sides. With the alitrunk in profile the
promesonotum evenly convex, sloping down posteriorly to the shallow metanotal groove. Propodeal
332 B. BOLTON
dorsum more shallowly convex than the promesonotum and strongly depressed below the level of the
promesonotum (Fig. 21). Propodeum armed with a pair of teeth which are slightly shorter than the rounded
metapleural lobes. Petiole node massive, domed in profile, the anterior peduncle short and narrow, equipped
with a dentiform anteroventral process. In dorsal view both petiole and postpetiole inflated, broader than
long. Dorsum of head finely and predominantly longitudinally rugulose, with scattered cross-meshes. On
the sides of the head and occipitally a loose reticulum is present. Ground-sculpture between the rugulae a
fine dense punctulation, superficial in places. Dorsal surfaces of alitrunk, petiole and postpetiole
reticulate-rugose, the meshes narrower and usually more sharply defined on the propodeum and pedicel
segments than on the promesonotum. Fine punctulate ground-sculpture present everywhere but stronger on
the pedicel segments than on the promesonotum. Base of first gastral tergite finely and very densely
reticulate-punctulate, the sculpture weakening posteriorly on the sclerite. Extreme base of first tergite, just
behind the postpetiolar articulation, with very short but strongly developed costulae. Short erect hairs very
sparse on dorsum of head, absent from all other surfaces except the petiole where a single pair is present.
Hairs absent from appendages. Sparse short decumbent to appressed pubescence present on alitrunk and
first gastral tergite; hairs present on tergites of gaster behind the first. Head and gaster dark brown with a
dull reddish tinge, alitrunk and pedicel segments dull red. Mandibles yellow.
A distinctive species characterized by its large size, depressed propodeum, lack of standing hairs
on the first gastral tergite, large eyes and short propodeal teeth. The closest relative of simoni is
braunsi, but in the latter the propodeum is unarmed (Figs 21, 22).
MATERIAL EXAMINED
South Africa : Transvaal, Makapan (E. Simon).
Leptothorax stramineus (Arnold)
(Fig. 18)
Limnomyrmex stramineus Arnold, 1948: 223, figs 10, lOa, lOb. Holotype worker, SOUTH AFRICA: Natal,
Zululand, St Lucia Lake (J. C. Faure) (NM, Bulawayo) [examined].
Leptothorax (Nesomyrmex) stramineus (Arnold) Brown, 1971 : 4.
WORKER. TL 2.8, HL 0.63, HW 0.53, CI 84, SL 0.36, SI 68, PW 0.38, AL 0.68.
Mandibles very delicately shagreened, almost smooth. Median portion of clypeus with anterior margin
evenly arcuate-convex, with a very narrow cuticular apron. Median clypeal carina vestigial. Frontal carinae
and antennal scrobes absent, the antennal scapes short. Maximum diameter of eye 0.13, about 0.25 x HW
and with 7-8 ommatidia in the longest row. Sides of head behind eyes very feebly convex, the sides
narrowing in front of the eyes so that the width immediately in front of the eyes is 0.47 and across the
clypeus at its widest is 0.38 (compare with HW 0.53). Occipital margin approximately straight and the
occipital corners evenly rounded. Alitrunk in profile with promesonotum evenly and shallowly convex, the
propodeal dorsum on the same level but shorter and more strongly convex, the two convexities separated by
the conspicuously impressed metanotal groove. Propodeal spines narrow and somewhat downcurved along
their length, longer than their basal widths. Slope of the declivity between the spines and the rounded
metapleural lobes straight. In dorsal view the alitrunk with the pronotal shoulders obtusely and bluntly
angled, the promesonotum narrowing posteriorly to the metanotal groove. Behind this the propodeum
broadening to the level of the spiracle then narrowing again to the bases of the spines; the latter divergent
and in the holotype with the right spine slightly longer than the left. Petiole node in dorsal view with a
strong lateral tubercle on each side, a pair of smaller tubercles on the posterior margin and a very feeble pair
anterodorsally which are almost effaced. Postpetiole with a low but broad lateral tubercle on each side. In
profile the petiole peduncle short, with a small triangular anteroventral process and with a very small
sub-denticulate process dorsally. The node itself higher than long, the lateral and posterior tubercles
distinct. Sides of first gastral tergite in dorsal view curving evenly away from the articulation with the
postpetiole, without a sharp, flattened appearance. Head, alitrunk, petiole and pcstpetiole very finely and
very densely superficially reticulate-punctulate, without rugulose sculpture. First gastral tergite with only
the faintest vestiges of superficial sculpture basally. Dorsal surfaces of head and body with scattered very
short blunt stout hairs; such hairs absent from appendages and sides of head but one or two may project
from the curved part of the occipital corner on each side. Colour uniform pale yellow.
Among the species which have the metanotal groove impressed stramineus is most closely related
to innocens, the two species sharing the characters of relatively small eyes and having short hairs
on the first gastral tergite (as opposed to the first gastral tergite being hairless). In fact, stramineus
AFROTROPICAL MYRMICINE ANT GENERA 333
and innocens form an extremely close species-pair and may eventually prove to be expressions of
a single species. For the present the two may be separated as the sculpture in innocens, although
punctulate as in stramineus, is much more strongly developed and sharply defined, with traces of
rugular sculpture also present at least on the head. Besides this the petiolar tubercles are not as
strongly developed in innocens as they are in stramineus, and the propodeal spines are shorter and
broader (Figs 18,20).
MATERIAL EXAMINED
South Africa: Natal, St Lucia Lake (J. C. Faure).
MEL1SSOTARSUS Emery
(Figs 23, 24)
Melissotarsus Emery, 1877: 378. Type-species: Melissotarsus beccarii Emery, 1877: 379, fig., by monotypy.
DIAGNOSIS OF WORKER. Myrmicine ants with moderate to conspicuous size variation in most nest samples,
living under bark and in wood of live trees; general appearance as in Figs 23, 24. Mandibles short, when
unworn armed with a long finger-like apical tooth followed by two much smaller teeth and sometimes also
by a minute basal denticle. With wear these gradually become an undifterentiated blunt margin. Palp
formula 0,1 (weissi). Median portion of clypeus bluntly triangular in shape and somewhat raised, not
projecting back between the frontal lobes. Lateral portions of clypeus simple and unmodified. Frontal lobes
narrow, confluent centrally and separated only by a narrow impressed line; the anteriormost parts of the
frontal lobes abut the posterior clypeal margin. Antennal scrobes absent. Frontal carinae absent. Antennae
with six segments, the scapes very short (SI 39-47), the two apical segments forming a strong club. Eyes
present, distinctly longer than broad and set in front of the midlength of the sides. Alitrunk short, fusiform
and box-like, without dorsal sutures or impressions except in the very largest individuals where rarely a
metanotal impression is shallowly present. Propodeum unarmed and rounded. Metapleural lobes absent.
Lateral portions of pronotum reduced to a narrow V-shaped wedge below the level of the conspicuous
mesothoracic spiracle. Anterior coxae small, much smaller than the massively developed middle and hind
coxae. Propodeal spiracle round, situated low on the side of the propodeum and just behind its midlength.
Metapleural gland system easily visible through the cuticle. Basitarsal segment of each leg greatly swollen,
as wide as the preceding tibia, terminating apically in a circlet of small teeth on the anterior (leading) edge on
the middle and hind basitarsi. Petiole with an anterior peduncle and a small low posteriorly situated node
which is broadly attached to the postpetiole; the latter broadly attached to the gaster. Dorsal alitrunk finely
longitudinally costulate throughout. Elongate fine hairs present dorsally on head and body, and also present
on the upper surfaces of the scapes and outer surfaces of the tibiae.
This small genus, of which only four uncommon species are presently recognized, is restricted to
the Malagasy region (1 species) where it is rare, and the Afrotropical region (3 species) where it is,
however, very widespread. The species nest in the healthy wood of living trees, apparently
tunnelling their own galleries below the surface. For this reason most collections of Melissotarsus
are made more by luck than by intent as their presence in the wood is usually not detectable on
the surface. Delage-Darchen (1972) has shown that the method of walking in these ants is very
strange; they progress on their front and hind legs with the middle pair projecting upwards, and
presumably in contact with the gallery roof. She also noted the presence of coccids inside the
galleries, also discussed by Ben-Dov (1978). It seems probable that coccid secretions form a
major, if not the main, item in the diet of Melissotarsus species.
The genus most closely related to Melissotarsus is Rhopalomastix Forel, represented by three
or four poorly defined species distributed throughout the Oriental and Indo- Australian
zoogeographical regions and utilizing the same lifeway as Melissotarsus. Since Emery (1922a) and
Wheeler (1922) produced their classifications these two small genera have always been placed
together in a tribe of their own (Melissotarsini) and it is fairly certain that they represent two
stages on a single adaptive line. Rhopalomastix is the more generalized of the two, Melissotarsus
decidedly the more specialized, but the modifications seen in the latter are foreshadowed in the
former genus. It is the accentuation of these adaptive specializations which separates the genera,
as follows.
334
B. BOLTON
23
29
Figs 23-29 23, 24, profile and head of Melissotarsus weissi. 25-29, Messor workers. 25, profile of angularis.
26-29, heads of (26) angularis, (27) striatifrons, (28) decipiens, (29) denticornis. Pilosity omitted from 24,
26-29.
AFROTROPICAL MYRMICINE ANT GENERA
335
Rhopalomastix
Antennae 10-segmented.
Lateral portion of pronotum extensive, distinctly
larger than the mesopleuron.
First coxa as large as or larger than the second and
third coxae.
Petiole sub-sessile, with a strong ventral process.
Free posterior face of petiole node long, its
articulation with the postpetiole narrow.
Basitarsal segment of each leg not swollen, without
apical circlets of teeth.
Sting long and strong.
Melissotarsus
Antennae 6-segmented.
Lateral portion of pronotum very reduced,
forming a V-shaped narrow wedge which is
smaller than the mesopleuron.
First coxa much smaller than the swollen second
and third coxae.
Petiole short-pedunculate, with feeble or no
ventral process.
Free posterior face of petiole node very short, its
articulation with the postpetiole very broad.
Basitarsal segment of each leg strongly swollen,
with apical circlets of teeth.
Sting very reduced and probably non-functional.
So little material of Melissotarsus is available at present that this survey must be regarded as
strictly preliminary. Three species are now recognized in the Afrotropical region but it is possible
that each may be compounded of more than one different sibling-species. Conversely it is by no
means impossible that further collections will bridge what appear here as species tor the
differences between them, though consistent in the few samples to hand, are relatively minor and
may well be anulled by further collecting.
For the present I define weissi as having a dark brown to black strongly sclerotized male, and a
similarly coloured female in which the postpetiole in dorsal view is quite narrow
(1.20-1.40 x broader than long) and has a rounded or even hemispherical anterior margin. The
worker of weissi has the alitrunk medium to dark reddish brown, the anterior margin of the
pronotum in dorsal view sharply defined and angular where it meets the anterior declivity, and
the sides of the alitrunk meeting the dorsum in a fairly well-defined angle.
M. emeryi and beccarii, on the other hand, have pale yellow feebly sclerotized males, and have
females in which the postpetiole in dorsal view is quite broad (1.90-2.20 x broader than long)
and lacking a rounded anterior margin, the margin instead being more or less straight or even
slightly concave. The workers are yellow to light yellowish brown and have the sides of the
alitrunk rounding bluntly into the dorsum when seen in dorsal view. Females of emeryi differ
from those of beccarii as in the former the mesoscutum is broader than long in dorsal view; it is
longer than broad in the latter. Workers of emeryi have the anterior margin of the pronotum
sharply defined and angular where it meets the anterior declivity, whereas in beccarii there is no
such sharp differentiation between dorsum and anterior declivity, instead the one surface rounds
bluntly into the other.
The shape of the alitrunk in dorsal view shows subtle but perhaps significant differences
between separate series of workers presently grouped as single species, but discovering whether
these differences are meaningful, or even consistent, will have to await the amassing of
considerably more samples than are presently available.
Synonymic list of Afrotropical Melissotarsus species
beccarii Emery
titubans Delage-Darchen syn. n.
emeryi Forel
emeryi var. pilipes Santschi syn. n.
compressus Weber syn. n.
weissi Santschi
major Santschi syn. n.
Key to species (workers)
1 With the alitrunk in dorsal view the anterior margin of the pronotum rounding evenly into the
anterior declivity, the two not meeting in a sharp angle or edge. (Ethiopia, Tanzania, South
Africa, Ivory Coast) beccarii(p. 336)
336 B. BOLTON
With the alitrunk in dorsal view the anterior margin of the pronotum separated from the anterior
declivity by a sharp angle or edge 2
2 Sides of alitrunk meeting dorsum in a fairly well-defined angle. Alitrunk colour medium to dark
reddish brown. (Ghana, Congo, Zaire) weissi(p. 337)
Sides of alitrunk rounding bluntly into the dorsum. Alitrunk colour yellow to light yellowish
brown. (Ethiopia, Sudan, Kenya, Tanzania, Zaire, Central African Republic, South Africa,
Ivory Coast, Ghana) emeryi (p. 337)
The three presently recognised species are basically so similar that to present a full description for
each would be redundant so, for the purposes of identification, a description of the type-species
beccarii is given and the other two are compared to it.
Metissotarsus beccarii Emery
Melissotarsus beccarii Emery, 1877: 379, fig. Syntype workers, ETHIOPIA: Keren (Beccari) (MCSN, Genoa;
MHN, Geneva) [examined].
Melissotarsus titubans Delage-Darchen, 1972: 216, figs 1-10. Syntype workers, females, males, IVORY COAST:
Lamto (Delage-Darchen) (probably in collection of Delage-Darchen). Syn. n.
WORKER. TL 2.3-3.3, HL 0.56-0.82, HW 0.56-0.80, CI 97-105, SL 0.24-0.34, SI 39^17, PW 0.34-0.55, AL
0.58-0.80(15 measured).
With the head in full-face view the occipital margin concave, sometimes deeply so medially, and with the
sides convex and weakly to distinctly convergent in front of the eyes. Mandibles with a long finger-like
apical tooth, worn down to nothing in some specimens; the mandibles unsculptured. Eyes much longer than
broad, strip-like in many, the maximum diameter 0.12-0.16, about 0.18-0.22 x HW. Median portion of
clypeus raised above the level of the lateral portions, not extending back between the frontal lobes; the latter
contiguous and separated only by an impressed line. Scapes very short, SI < 50. Alitrunk in dorsal view with
anterior pronotal margin rounding into the declivity, the two surfaces not separated by a sharp edge or
angle. Dorsum of alitrunk roughly rectangular longitudinally, somewhat narrower behind than in front but
not strongly so, and with the dorsum rounding into the sides. In profile the promesonotal dorsum and
anterior propodeum are more or less flat but the posterior part of the propodeum rounds very broadly and
evenly into the declivity, without trace of armament. Fore coxae small, about half the size of the strongly
swollen middle and hind coxae. Peduncle of petiole short and grading into the relatively high narrow node,
the node with a short posterior free face, broadly attached to the postpetiole. In dorsal view the petiole node
much broader than long. Postpetiole in dorsal view much broader than long, slightly broader than the
petiole and very broadly attached to the first gastral tergite without a posterior constriction. Gaster only
feebly sclerotized, crumpled in most mounted specimens. Dorsum of head with a silky superficial
ground-sculpture upon which scattered small pits are usually superimposed. The ground-sculpture may
cover the whole head but frequently it fades out occipitally. Median portion of clypeus more densely and
strongly sculptured than dorsum of head capsule. Dorsal alitrunk finely longitudinally costulate
throughout, the costulae fading out where the propodeal dorsum rounds into the declivity. Dorsal surfaces
of head, scapes, pronotum, mesonotum, pedicel segments and gastral tergites with scattered sparse long fine
hairs. Propodeal dorsum usually with one or two shorter hairs but these are frequently missing. Dorsal
(outer) surfaces of tibiae with sparse long hairs similar to those on alitrunk. Head and alitrunk dull yellowish
brown to dark yellow, the gaster lighter, usually pale dull yellow.
The key character given to separate beccarii and emeryi workers is quite weak. In most
individuals there is a reasonable visible difference between the two, with the anterior pronotal
margin rounding bluntly into the declivity in beccarii, and with the anterior pronotal margin
separated from the declivity by an angle or edge in emeryi. Having said that, however, it should
be pointed out that the difference is not so well marked in some individuals, which in
consequence are difficult to place. Both species have a pale yellow feebly sclerotized male.
Females of both species have the postpetiole in dorsal view conspicuously broader than long but
it seems that two species are present as in some the mesoscutum is longer than broad (beccarii)
but in others broader than long (emeryi). It should be admitted that very few worker-associated
females are known and further collections may annul this apparent difference. To sum up, for the
present I recognise these two as separate on the strength of the differently shaped mesoscutum in
females and the form of the anterior pronotal margin in workers, but harbour a suspicion that
only a single real species may in fact be represented here.
AFROTROPICAL MYRMICINE ANT GENERA 337
M. beccarii differs from weissi fairly consistently in all castes. The workers of weissi are darker
in colour than those of beccarii and have both the anterior pronotal margin and the sides of the
alitrunk relatively strongly marginate. The male is dark brown to black and strongly sclerotized,
and in the female the postpetiole is relatively narrow in dorsal view with an arched-convex
anterior margin which is quite different in shape from the strongly transverse form seen in
beccarii. Measurements of the postpetiolar widths of the various forms are given under the
discussion of the genus.
MATERIAL EXAMINED
Ivory Coast: Lamto Field Station (W. L. Brown); nr Abidjan (W. L. Brown). Ethiopia: Keren (Beccari).
Tanzania: Lulanguru (G. D. H. Carpenter). South Africa: Natal, Durban (C. B. Cooper); Durban (H. B.
Mar ley).
Melissotarsus emeryi Forel
Melissotarsus emeryi Forel, 1907: 133. Syntype workers, ETHIOPIA: Colba, 1905 (M. de Rothschild) (MHN,
Geneva) [examined].
Melissotarsus emeryi var. pilipes Santschi, 1914a: 71. Syntype workers, KENYA: Taveta, 750 m, st. no. 65,
iii. 1912; and TANZANIA: Kilimanjaro, Bismarckhiigel, 2740 m, st. no. 70, iii.!912(C. Alluaud& R.Jeannel)
(NM, Basle) [examined]. Syn. n.
Melissotarsus compressus Weber, 1952: 1, figs 28, 29. Holotype female, CENTRAL AFRICAN REPUBLIC:
Ubangi-Shari, Haut Mbomu, lat. 5° 30' N., long. 25° 15' E., iii. 1948, no. 2184 (N. A. Weber) (AMNH,
New York) [examined]. Syn. n.
WORKER. TL 2.5-3.4, HL 0.66-0.88, HW 0.70-0.90, CI 100-105, SL 0.30-0.38, SI 39^3, PW 0.37-0.57, AL
0.62-0.88 (13 measured).
Answering to the description of beccarii but with the anterior pronotal margin in dorsal view separated
from the anterior declivity by a well defined angle or edge.
As pointed out under beccarii a few individuals seem intermediate between those of emeryi and
those of beccarii, and in consequence are difficult to place. My suspicions are that these two
names may represent a single species but I feel unsure enough to avoid synonymizing them whilst
the apparent difference between the females remains unresolved. So, until the taxonomic value of
the shape of the mesoscutum in females can be assessed, the two must remain as separate species.
MATERIAL EXAMINED
Ethiopia: Colba (Rothschild). Sudan: Darfur, Jebel Murra (M. Steele). Kenya: Muguga (K. Njukiine);
Taveta (Alluaud & Jeannel). Zaire: Popokabaka (E. S. Ross & R. E. Leech). Ghana: Tafo (C. A.
Collingwood). South Africa: Cape Prov., Clanwilliam (Y. Ben-Dov). Central African Republic: Haut Mbomu
(N. A. Weber).
Melissotarsus weissi Santschi
(Figs 23, 24)
Melissotarsus weissi Santschi, 1910: 356, fig. 3. Holotype female, CONGO: Brazzaville (A. Weiss) (NM, Basle).
[Only gaster and one forewing remaining on mount.]
Melissotarsus major Santschi, 1919: 85. Syntype workers, ZAIRE: Penghe, 13.ii., no. 125 (Bequaert) (NM,
Basle; MR AC, Tervuren) [examined]. Syn. n.
WORKER. TL 2.3-3.0, HL 0.58-0.74, HW 0.60-0.78, CI 98-104, SL 0.27-0.34, SI 41-47, PW 0.36-0.50, AL
0.56-0.84 (14 measured).
Answering to the description of beccarii but darker in colour, the alitrunk medium to dark reddish brown;
with the anterior pronotal margin meeting the anterior declivity in a well-defined angle or edge, and with the
sides of the alitrunk meeting the dorsum in a fairly well-marked angle.
Lighter coloured workers may sometimes be difficult to separate from emeryi, but in general the
sharper marginations of the sides of the alitrunk in weissi are fairly distinct. The sexual forms of
weissi are both easily separated from those of emeryi as the male of the former is dark brown to
black (pale yellow and feebly sclerotized in the latter), and the female of weissi has the postpetiole
relatively narrow in dorsal view with an arched-convex anterior margin, as opposed to a very
broad and distinctly transverse postpetiole in emeryi.
338 B. BOLTON
MATERIAL EXAMINED
Ghana: Tafo (B. Bolton). Zaire: Kamaiembi (H. Schouteden); Penghe (Bequaeri).
MESSOR Forel
(Figs 25-32, 35-43)
Messor Forel, 1890a: Ixviii [as subgenus of Aphaenogaster Mayr]. Type-species: Formica barbara L., 1767:
962, by subsequent designation of Bingham, 1903 : 277.
Messor Forel; Bingham, 1903: 277. [Raised to genus.]
Cratomyrmex Emery, 1891: 572. Type-species: Cratomyrmex regalis Emery, 1891: 572, by monotypy.
[Synonymy by Emery, 1922a: 357.]
Veromessor Forel, 1917: 235 [as subgenus of Novomessor Emery]. Type-species: Aphaenogaster andrei
Mayr, 1886: 448, by subsequent designation of Emery, 1921 : 67. Syn. n.
Veromessor Forel; Wheeler, 1922: 680. [Raised to genus.]
Lobognathus Enzmann, 1947: 152 [as subgenus of Veromessor]. [Erroneous entry for Veromessor
lobognathus (Andrews); see Brown, 1949: 49.]
DIAGNOSIS OF WORKER. Granivorous myrmicine ants, mostly strongly polymorphic but a few monomorphic
or only weakly polymorphic. Head massively constructed in larger workers. Mandibles large and powerful,
multidentate in smaller workers (up to 15 teeth) but this number usually decreasing with increased body size
until in largest workers only a few massive teeth or an edentate crushing edge remains. Sometimes also in
small workers the teeth are worn down to an edentate margin. Palp formula predominantly 4,3 but in
largest workers usually 5,3 (30 species dissected). Median portion of clypeus broad and shield-like, broadly
inserted between the widely separated frontal lobes; both median and lateral portions of clypeus unmodified
except for a central impression of the anterior margin in some species. Frontal lobes short but conspicuous,
at least partially concealing the antennal insertions. Frontal carinae absent. Antennal scrobes absent.
Antennae 12-segmented, either filiform and without an apical club (in which case the flagellar segments
gradually increase in size apically), or with a feebly defined incipient club where the apical 3-4 segments are
slightly enlarged. Eyes present, moderate to large in size, situated at or just behind the midlength of the sides
in full-face view. Ventral surface of head with elongate ammochaete hairs which usually form a
psammophore. This may be reduced and non-functional in some species but the hairs are still conspicuous
and generally longer than those found elsewhere on the body; in a few species the psammophore is better
developed in smaller than in larger workers. With the alitrunk in profile the promesonotum swollen and
convex, frequently dome-like and sloping down steeply behind to the metanotal groove which is weakly to
distinctly impressed. Propodeum rounded to strongly bispinose posteriorly and on a much lower level than
the convex promesonotum. Promesonotal suture fused and inflexible but its track represented by a distinct
arched impression across the dorsum. Mesonotum bounded by impressions on all sides, its boundary easily
discernible except in the smallest workers of a few species. Metapleural lobes absent or at most represented
by a pair of low broadly rounded ridges. Propodeal spiracle large and conspicuous, circular to subcircular
and situated approximately at the midlength of the propodeum or sometimes slightly behind the midlength,
but never shifted conspicuously back towards the declivity. Basal posterior portion of mesopleuron just
above the middle coxa with a few hairs projecting downwards and backwards. (Whether these are
guard-hairs indicating the exit site of a gland is not known, but the hairs remain even in species where other
body pilosity is very reduced or absent.) Spurs on posterior tibiae varying from very feebly pectinate through
partially barbate and minutely barbulate to simple. Alitrunk ventrally with a strong metasternal process
which is usually large to very large (reduced but still conspicuous only in rufotestaceus (Foerster) and
vaucheri Emery out of 45 species dissected). Petiole with a long anterior peduncle, the spiracle situated at
about the midlength of the peduncle, well in front of the node. Petiole node in profile narrow and often
bluntly triangular to conical in shape, but frequently a sloping differentiated dorsal surface is present where
the anterodorsal angle is generally the highest point.
Messor is a moderately sized genus of granivorous ants occurring in grassland and savannah, and
in arid to desert situations. The main base of the genus is in the Palaearctic region where about
70-80 species occupy a broad strip of territory reaching across the whole width of North Africa
and the southern European countries, across the Near and Middle East and thence eastwards
through the U.S.S.R. to China and Japan. Compared to this the faunas of other zoogeographical
regions are relatively minor. The Afrotropical region has 12 species and Madagascar has 1; the
Oriental region has 3-4 species and the Nearctic has 8, all distributed on the western side of the
continent and formerly occupying a genus of their own, Veromessor, now synonymized. Species
AFROTROPICAL MYRMICINE ANT GENERA 339
of Messor are absent from the Neotropical region, the Indo-Australian region and Australasia,
nor do they occur on any of the Pacific island systems.
Recent studies of Messor include those of Arnoldi (1977) on the fauna of the U.S.S.R., and
Collingwood (1978) on the species of the Iberian Peninsula. The only previous synthesis of
sub-Saharan African species is that of Arnold (1920), for the then-recognized South African
forms, but no key was given in that revue. Creighton (1950) has keyed the North American
species formerly in Veromessor. Knowledge of the detailed biology of the species is sparse, but
good basic work has been done on some African species by Levieux & Diomande (1978), and
Levieux (1979).
The closest relatives of Messor are the genera Aphaenogaster and Pheidole Westwood.
Members of the latter genus are easily separated from Messor as the palp formula is reduced to
2,2, its species are dimorphic, and the antennal funiculus ends in a strongly defined 3-segmented
club. Aphaenogaster, which is absent from sub-Saharan Africa, is more difficult to differentiate as
its species, apart from being uniformly monomorphic, are very close to Messor and share most of
its diagnostic characters, including the filiform to feebly clavate funiculi and high palp formula
(PF) count. Of 55 species of Aphaenogaster dissected 31 had PF 5,3, and 24 had PF 4,3. For some
reason, although species with the higher PF apparently outnumber those with the lower count,
the zoogeographical distribution of the latter is much wider than that of the former.
Aphaenogaster species with PF 5,3 are found in the Nearctic, Palaearctic and Oriental regions;
species with PF 4,3 are also found in these three regions and in the Neotropical, Malagasy,
Indo-Australian and Australasian regions as well.
After a study of Aphaenogaster for genus-level characters, primarily a search for strong
characters to separate it from Messor, it became apparent that Brown (1973) was correct in
relegating the former subgenera of Aphaenogaster to the synonymy. These former subgenera
(Attomyrma Emery, Deromyrma Forel, N ystalomyrma Wheeler and Planimyrma Viehmeyer) have
no significance as they are founded upon minor, inconsistent or gradient character-states.
Further, it is now clear that Brown (1974) was also correct in assigning Novomessor to the
synonymy of Aphaenogaster. The only real character separating the two was the fore-wing
venation, there being one closed cubital cell in the former and supposedly two in the latter. The
same character was invoked to separate Veromessor from Messor, again the former having one,
the latter two closed cubital cells. A survey of the venation of Aphaenogaster and Messor shows
that in both genera the same finely graded series of changes in wing venation occurs (Figs 35-43),
which obviates these supposed differences in number of closed cubital cells; it is instructive to
consider both genera together.
The most complete, and therefore most primitive, venation pattern (Fig. 35, M. galld) shows
two closed cubital cells and has Rs + M dividing well in front of the level of cross- vein m — cu, so
that m — cu arises from M itself and there is a short free section of M between the point of
division of Rs + M into its constituent parts and the point where m — cu meets M.
The free section of M then contracts (Fig. 36; M. tropicorum, angularis, nigriceps Santschi; A.
geei Wheeler, schurri (Forel)) as the fusion of Rs + M lengthens outwards along the wing until the
condition shown in Fig. 37 is seen (M. rugosus (Andre); A. schurri) where there is no free portion
of M between Rs + M and the point of origin of m — cu, the veins Rs, M and m — cu all
appearing to arise from a point at the apex of Rs + M.
Next, the fusion of Rs + M advances further out along the wing so that Rs and M now
separate a short distance beyond the point of origin of m — cu, which now arises direct from
Rs + M (Fig. 38; M. intermedius Forel, angularis, himalayanus Forel, aciculatus (Smith), structor
(Latreille), regalis; A. rudis Emery, treatae Forel). Following this the fusion of Rs + M advances
further out along the wing, drawing closer to cross-veins 2r and r — m, as shown in Figs 39, 40,
this stage constituting what may be considered as the normal pheidoline venation (M. barbarus
(L.), capitatus (Latreille), structor, galla, denticornis, capensis, leubberti, muticus (Nylander),
aegyptiacus (Emery), nigriceps Santschi, semirufus (Andre), instabilis (Smith), meridionalis (Andre);
A. geei, rudis, lamellidens Mayr,famelica (Smith), fulva Roger, japonica Forel, pallida (Nylander),
huachucana Creighton, splendida (Roger), megommatus Smith, subterranea (Latreille), occidentalis
Emery, crocea Andre, gemella (Roger), senilis Mayr).
340 B. BOLTON
As the fusion of Rs + M progresses still further along the wing a critical point is reached at
which cross-vein r — m vanishes. This occurs whilst the advancing fusion is still some little
distance away from 2r. A male of A. spinosa Emery in BMNH shows the critical point as the
specimen has r — m present on the left wing but it has vanished from the right. The dissapearance
of r — m leaves the venation as in Fig. 41, which is present in a wide range of species (M.
pergandei (Mayr), lobognathus, formerly of Veromessor; A. albisetosus Mayr, formerly of
Novomessor; A. dromedarius (Emery), longiceps (Smith), pythia Forel, phalangium Emery, beccarii
Emery, araneoides Emery, sagei Forel).
Eventually the stage seen in Fig. 42 is reached where Rs and M are fused to the point of
intersection of 2r (M. andrei (Mayr); A. cockerelli Andre), and finally in A. ensifera Forel the
fusion of Rs + M has extended beyond the level of 2r so that this cross-vein now arises from
Rs + M (Fig. 43).
It should be pointed out that there is considerable variation present along this sequence within
single species and that it is by no means rare to find specimens with different venation patterns on
the left and right forewings, representing different stages in the sequence, and thus showing it to
be a dynamic rather than a static system. Also, adventitious vein-stubs frequently arise at random
from all the main veins, and from the cross-veins too on occasion.
Thus the loss of r — m cross- vein, reducing the two cubital cells to only one, rather than being
the concise taxonomic character it was thought to be in the past, can now be seen as just one step
in a long gradual sequence of venation development in both Aphaenogaster and Messor, and of
no significance in genus-level discrimination among these ants. To draw a line at any point in the
sequence and claim that it is more significant than a line drawn at any other point is thus purely
arbitrary, and as a direct consequence of the establishment of this sequence the synonymy of
Novomessor with Aphaenogaster is confirmed and the name Veromessor falls into the synonymy
of Messor, there being no other consistent character to separate them.
A side development in the history of Novomessor, following Brown's (1974) synonymy, was the
suggestion of Holldobler, Stanton & Engel (1976) that the name might be resurrected for two of
its former members (albisetosus and cockerelli) because of the presence of an exocrine gastral
glandular system which was absent from other Aphaenogaster species examined, and incidentally
absent also from the third former Novomessor species (ensifera), which was to be retained in
Aphaenogaster despite the fact that it is otherwise very close to the first two. The obvious
inference was that the presence of such a gland system merited genus-level consideration. This is
reasonable logic as far as it goes, though many would argue (myself included) that basing genera
on such features is grossly over-weighting a relatively weak single character. The discussion
would probably have rested there but Kugler (1978) published a paper indicating that such
glands occur widely in the Myrmicinae in a range of genera, including a member of
Aphaenogaster (phalangium) whose placement in that genus has never been doubted, but which is
not closely related to any of the three mentioned above. This gives rise to three possibilities.
Firstly, that the presence of such glands is highly significant and that, following the model of
Aphaenogaster-Novomessor, every species showing such structures must be assigned to a genus
separate from the parent genus, irrespective of any basic similarities they may otherwise show.
The idea is ludicrous of course, and obviously not at all what Holldobler et al intended; the
plethora of pointless generic names thus produced would be incredible and no more sensible than
selecting genera from groups of closely related species on grounds of, say, presence or absence of
hairs on the first gastral tergite. Holldobler et al. in their study found a gastral exocrine system in
one species of Ocymyrmex Emery but not in two others; they did not suggest the creation of a
separate genus here.
Secondly, we can reassign such forms with gastral exocrine glands (or indeed any other
individual specialization) when it suits us to do so, and ignore it otherwise. Thus we can utilize
such a character to prop up an otherwise poorly defined or undefinable genus which looks like
falling irrevocably into the synonymy. This idea does not hold much merit as it again leads
unerringly to the creation of swarms of peripheral genera, each with only one or two species,
which cannot be adequately separated from their closest relatives remaining embedded in the
central mass of species.
AFROTROPICAL MYRMICINE ANT GENERA 341
Finally, we can consider that the development of such gland systems in some species of a genus
but not in others, whilst uniformly stable genus-level characters span the entire range of species,
reflects a specialization in the lifeway of the ants involved and is significant at species or
species-group level but not beyond that, providing always that other genus-level characters
remain uniform throughout. This is decidedly the alternative which I favour as in the long run it
will produce strong, well-defined genera, and realistic species-groups within those genera.
To conclude the observations on the genus-level synonymy of Aphaenogaster, it is now
apparent that the monotypic genus Brunella Forel sinks as a synonym. This Malagasy species has
had a chequered career since its original description as Aphoenogaster [sic] belli Forel, 1895; 248.
(Syntype workers, MADGASCAR: Imerina, Moramanga (M. Sikora) (MHN, Geneva) [examined].)
It was later shifted by Forel (1917) out of Aphaenogaster to form the type-species of his genus
Brunella. Emery (1922a: 242) disagreed with this and synonymized Brunella under Atopula
Emery, which for him was a catch-all genus to which a number of obscure species were relegated.
During my study of the tetramoriine genera (Bolton, 1976) it transpired that the type-species of
Atopula was in fact a Tetramorium, so that the name Atopula fell into synonymy and the
remaining former occupants of Atopula were transferred to other genera. At that time I had not
examined the type-series of belli and so referred the species back to Forel's temporarily
resurrected Brunella. Now, having at last examined the types of belli, it turns out to be a fairly
unexceptional Aphaenogaster which seems to belong to the Oriental sagei-group as it has a broad
occipital margin, relatively short antennal scapes, a moderately well-developed antennal club and
distinct propodeal spines.
A summary of the current genus-level synonymy of Aphaenogaster is given in the appendix,
p. 364.
Of the two names synonymized with Messor above, Cratomyrmex was recognized as a
synonym by Emery as long ago as 1922. The separation of the two was based on the presence of
pectinate hind tibial spurs in the latter and their supposed absence in the former. This was
quickly spotted as a feeble and variable character and the status of the genus challenged
(Santschi, 1920). The form of the hind tibial spurs is in fact very variable in Messor, showing all
stages from feebly pectinate, through barbate and minutely barbulate to simple. Even in the same
series there is sometimes variation in spur form between different-sized workers.
Veromessor, which began its existence as a subgenus of Novomessor, was given generic status
by Wheeler (1922) who separated it from Messor on the venation character discussed above and
now known to be spurious. The discussion in Wheeler & Creighton (1934: 356-360) indicated
that Messor and Veromessor were extremely closely related, but no means of separating them was
given. Presumably only the venation character invoked previously by Wheeler could be found.
The present investigation has shown the two to be synonymous for, leaving aside the venation, all
characters of Messor are duplicated in Veromessor, except for the species relictus Wheeler &
Mann. This last was originally described as a member of Aphaenogaster but was transferred to
Veromessor by Wheeler & Creighton (1934), for no apparent reason. In my opinion it is an
ordinary member of Aphaenogaster, fitting the diagnostic characters of that genus and having all
the criteria required to separate it from Messor which are tabulated below; it is herewith returned
to Aphaenogaster. Finally, the fossil species sculpturatus Carpenter, originally described in
Messor (where it is a junior secondary homonym of sculpturatus Stitz), later included in
Veromessor but suggested as a possible Pogonomyrmex species by Wheeler & Creighton, is
impossible to place at present and requires further study. The living North American species now
included in Messor are andrei (Mayr) comb, n., chamberlini Wheeler, julianus (Pergande) comb, n.,
lariversi (Smith) comb, n., lobognathus Andrews, pergandei (Mayr) comb, n., smithi (Cole)
comb, n., stoddardi (Emery) comb. n.
Aphaenogaster and Messor are very closely related and certainly derive from a single parent
stock. The characters tabulated below will separate them even though a few species show
exceptions to one or another of the characters.
342
B. BOLTON
Aphaenogaster
Entirely monomorphic.
Mostly without ammochaete hairs (present in a
very few species).
Head usually slender, CI 90 at maximum,
generally much less (range 49-90 in 75 species
measured).
Metasternal process small to absent, approaching
size seen in Messor only in A. subterranea (55
species dissected).
Outer margins of mandibles not strongly curved
towards midline, the mandibles triangular in
shape and not massive.
Messor
Mostly polymorphic species (a very few feebly
polymorphic and monomorphic species known).
Mostly with ammochaete hairs present (reduced in
a few species).
Head massive and broad, in medium to large
workers CI > 90 (range 95-125 in 64 species
measured).
Metasternal process large to very large, always
very conspicuous (45 species dissected).
Outer margins of mandibles strongly curved
towards midline, the mandibles massive and
heavy.
Synonymic list of Afrotropical Messor species
angularis Santschi stat. n.
capensis (Mayr)
pseudoaegyptiaca Emery syn. n.
barbarus subsp. capensis var. schencki Forel (unavailable)
braunsi Forel syn. n.
donisthorpei Santschi syn. n.
cephalotes (Emery)
plinii Santschi syn. n.
collingwoodi sp. n.
decipiens Santschi stat. n.
barbarum r. capense var. decipiens Forel (unavailable)
barbarus subsp. capensis var. proba Forel (unavailable)
arcistriatus Santschi syn. n.
denticornis Forel
denticornis var. parvidens Forel syn. n.
denticornis var. brunni Forel syn. n.
galla (Mayr)
barbarum subsp. caduca var. galla Emery (unavailable)
barbarus subsp. semirufus var. rufa Forel (unavailable)
barbarus st. galla var. triempressa Santschi (unavailable)
barbarus st. latinodis Santschi syn. n.
barbarus r. semirufus var. rufula Forel (unavailable)
barbarus subsp. galla var. armata Emery (unavailable)
cjalla st. nobilis Santschi syn. n.
galla var. airensis Bernard syn. n.
incisus Stitz (nomen dubium)
luebherti Forel stat. n.
piceus Stitz
regalis (Emery)
regalis var. rubea Santschi syn. n.
sculpturatus Stitz syn. n.
ruginodis Stitz stat. n. (nomen dubium)
striatifrons Stitz stat. n.
tropicorum Wheeler stat. n.
denticornis var. laevifrons Stitz syn. n.
braunsi var. nigriventris Stitz syn. n.
Key to species (medium to large workers)
Note. The nomina dubia incisus Stitz and ruginodis Stitz are omitted from the key.
1 Hairs absent from first gastral tergite or at most with a single sparse transverse row at the
extreme apex of the sclerite
Hairs present on first gastral tergite, more or less evenly distributed over the whole surface of
the sclerite . ...
AFROTROPICAL MYRMICINE ANT GENERA 343
2 Dorsumofpropodeum with one or more pairs ol standing hairs 3
Dorsum of propodeum without standing hairs 4
3 Dorsum of head coarsely and densely reticulate-punctate everywhere, the mid-dorsal strip also
rugulose. (Niger, Mali) collingwoodi (p. 346)
Dorsum of head smooth everywhere except for the rugulose mid-dorsal strip; without coarse
dense reticulate-punctate sculpture. (Throughout Sahelian zone and northern East Africa,
also occurring coastally in West Africa) galla (p. 349)
4 Head sculptured everywhere with close-packed longitudinal rugulae between which is
reticulate-punctate ground-sculpture. Eyes slightly smaller, 0.15-0.18 x HW in HW range of
2.00-3.12. (Tanzania, Zimbabwe, Angola, Botswana, South West Africa, South Africa)
leubberti(p. 351)
Head smooth except usually for a short central rugular area behind the frontal lobes. Eyes
slightly larger, 0.18-0.21 x HW, in HW range of 2.00-2.76. (Kenya) . . angularis (p. 344)
5 Basal third or more of first gastral tergite strongly and conspicuously sculptured with rugulae,
costulae, coarse reticulate-puncturation, or a combination of these 6
Basal third of first gastral tergite unsculptured except for hair pits and very faint superficial
patterning. In some very large workers a few short basigastral costulae may develop but these
are restricted to the area immediately behind the postpetiole 7
6 With the head in full-face view the sides with projecting hairs. Petiole and postpetiole coarsely
closely and deeply rugose. (Nigeria, Benin Republic, Congo) regalis (p. 352)
With the head in full-face view the sides without projecting hairs. Petiole and postpetiole finely
sculptured with feeble rugulae, dense puncturation or a combination of both. (Ethiopia,
Kenya, Tanzania) cephalotes (p. 346)
7 Posterior half of clypeus between frontal lobes with a distinct, strongly raised central step or
welt. (Angola, South West Africa) tropicorum (p. 354)
Posterior half of clypeus between frontal lobes without a raised central step or welt, usually
more or less flat or even slightly concave .......... 8
8 Eyes relatively large, the maximum eye diameter 0.21-0.25 x HW, in HW range of
2.50- > 4.00. (Botswana, South West Africa, South Africa) .... denticornis (p. 349)
Eyes smaller, the maximum eye diameter 0.14-0.19 x HW, in HW range of 2.50- > 4.00 . . 9
9 In HW range 2.80- > 4.00 the sides of the head conspicuously evenly convex in full-face view
(Fig. 27). Propodeum in profile relatively long and low (Fig. 32). (South West Africa, South
Africa) stria tifrons (p. 353)
In HW range 2.80- > 4.00 the sides of the head approximately straight in full-face view, the
sides parallel or divergent anteriorly (Fig. 28). Propodeum in profile relatively short and high
(Fig. 31) '10
10 Body pilosity very dark in colour, deep red-brown to blackish. (Botswana, South Africa)
piceus (p. 352)
Body pilosity pale, white or silvery to yellowish 11
1 1 Head red in majo'r workers, contrasting in colour with the much darker alitrunk and gaster.
(Zimbabwe, Botswana, Lesotho, South Africa) decipiens(p. 348)
Head brown to black in major worker, about the same colour as the alitrunk and gaster. (South
West Africa, Botswana, South Africa) capensis (p. 345)
Among strongly polymorphic species such as these, where there is an enormous worker
size-range, the standard measurements which I have otherwise used consistently for the
Myrmicinae become meaningless and cannot be utilized. A few standard ratios have, however,
proved to be of value in some cases and these are included in the relevant descriptions. The keys
and descriptions are based on medium to large workers as these show the best discriminating
characters, the minor workers of closely related species being sometimes indistinguishable. Size
ranges covered by the descriptions are given for each species in terms of H W range.
The presence or absence of propodeal teeth or spines, which appears to be a functional
diagnostic character in other parts of the range of Messor, is not of much use in the Ethiopian
region species for, although some always have the propodeum armed (regularis, collingwoodi) and
some always have it unarmed and rounded (angularis, luebberti), the rest show a disconcerting
variability in this character, sometimes differing even in individuals from the same nest sample.
The 12 recognizable species are distributed roughly as follows in the Afrotropical region.
344
B. BOLTON
Northern (Sahelian) species: collingwoodi, galla, cephalotes (in extreme east).
Western species : regalis, galla (coastally).
Eastern species : cephalotes, angularis, leubberti (in south), galla (in north).
Southern species : denticornis, luebberti, striatifrons, tropicorum, piceus, decipiens, capensis.
The species fall into two groups in terms of pilosity. The first group, characterized generally by
reduced pilosity and virtual absence of hairs on the first gastral tergite, contains the species
angularis, collingwoodi, galla, and luebberti. In the second group pilosity is generally dense and is
evenly distributed over the first gastral tergite. Included here are the remaining eight species
noted above. Of them regalis is very conspicuous and not obviously close to any of the others. Of
the remainder the southern complex of piceus, decipiens and capensis may represent a single
species, and denticornis, striatifrons and tropicorum are closely related.
Messor angularis Santschi stat. n.
(Figs 25, 26)
Messor barbarus st. semirufus var. angularis Santschi, 1914a: 75 [unavailable name]; Santschi, 1928: 202
[galla var. angularis, first available use of name]. Syntype workers, KENYA: Naivasha, 1900 m, st. no. 14,
xii.191 1 (C. Alluaud & R. Jeannel) (NM, Basle) [examined].
MEDIUM TO LARGE WORKER. HW 2.00- > 2.75.
Anterior clypeal margin flattened to weakly and quite broadly indented medially. With the head in
full-face view the occipital margin indented medially, the indentation becoming more distinct with increased
size. In HW range 2.00-2.80 the maximum diameter of the eye is 0.42-0.52, about 0.18-0.21 x HW, and the
CI range is 104-113. Propodeum unarmed, rounded to right-angled where dorsum meets declivity and
sometimes with a reinforcing ridge or flange following the curve, especially in largest workers. Dorsum of
head with sculpture very reduced, sometimes without sculpture. Usually with a few very feeble low
longitudinal rugulae between the frontal lobes which may extend for a short distance behind them. On each
side of this median area, moving outwards towards the eyes and occipital corners, the head is unsculptured
except for a very feeble superficial reticular pattern and a few scattered faint punctulae. Pronotum and
mesonotum dorsally unsculptured to feebly densely punctulate, generally with some weak transverse
rugulae immediately behind the cervical shield. Rarely these are absent but in some they extend further back
on the pronotum than is usual. Propodeal dorsum transversely rugose, conspicuously more strongly
sculptured than the pronotum or mesonotum. First gastral tergite unsculptured and smooth, usually with a
faint superficial reticular pattern visible. Head dorsally with very reduced pilosity; apart from the strong
mouthpart hairs and those around the frontal lobes the dorsum with only 2-3 pairs, spanning the midline of
the head. With the head in full-face view the sides both in front of and behind the eyes, the occipital corners
and the occipital margin without projecting hairs except mid-occipitally where the posteriormost dorsal pair
may project on each side of the occipital impression. Psammophore strong, the J-shaped hairs very long and
conspicuous. Pronotum dorsally with 0-3 pairs of hairs, when present situated posteriorly, close to the
promesonotal junction. Mesonotum with 0-5 pairs of hairs. Some of this variation may be due to abrasion,
the mesonotal hairs in particular seem easily lost. Propodeum always hairless dorsally. Petiole with 0-1,
postpetiole with 0-2 pairs of hairs respectively. First gastral tergite hairless or at most with a sparse
transverse row at the extreme apex of the sclerite. Colour variable, usually with reddish head and alitrunk
and black gaster, but the alitrunk often with some black, the amount of which varies from sample to sample.
In extreme cases the entire body black but even here the head with a reddish tint showing through.
One of the four sub-Saharan African species which lacks hairs on most or all of the first gastral
tergite, angularis is at present known only from Kenya. The most closely related species is the
extremely widespread galla which also occurs in Kenya. The two are separated as follows.
angularis galla
Propodeal dorsum without hairs. Propodeal dorsum with one or more pairs of hairs.
Occipital margin on each side of the median Occipital margin on each side of the median
impression without projecting hairs impression with one or more pairs of projecting
hairs.
Ventral surface of hind femora without freely Ventral surface of hind femora with numerous
projecting hairs or at most with 1-2 close to the freely projecting hairs which usually occur over
trochanter. the length of the shaft but which are often
densest proximally.
AFROTROPICAL MYRMICINE ANT GENERA 345
Ventral surface of postpetiole in profile without an Ventral surface of postpetiole in profile with a
anterior prominence or at most with a feeble sharp dentiform or angular prominence
angle, the surface immediately behind this anteriorly, the surface immediately behind this
smoothly concave. irregular, not smoothly concave.
Median strip of head dorsally unsculptured or at Median strip of head dorsally usually
most with very feeble rugulae anteriorly. conspicuously rugulose, only very rarely
reduced.
MATERIAL EXAMINED
Kenya: Tana Riv. (J. L. Clark); Olikoriti (M. G. Lepage); Kajiado (J. Darlington); Bissel (J. Darlington);
Kajiado (G. Nyamasyo); Kajiado (W. Sands); Isiolo (E. S. Ross & R. E. Leech); Naivasha (Alluaud &
Messor capensis (Mayr)
(Fig. 31)
Atta capensis Mayr, 1862: 743. LECTOTYPE worker, SOUTH AFRICA: Cape of Good Hope, Novara Expd.
" D ". (NM, Vienna), here designated [examined].
Aphaenogaster pseudoaegyptiaca Emery, 1884: 384. Syntype workers, SOUTH AFRICA: Cape of Good Hope
(MCSN, Genoa) [examined]. Syn. n.
Messor barbarus subsp. capensis var. schencki Forel, 1910a: 15. Holotype worker, SOUTH WEST AFRICA:
Bethanien (Schenck) (not found in MHN, Geneva, presumed lost). [Unavailable name.]
Messor braunsi Forel, 19136: 138. Syntype workers, SOUTH AFRICA: Cape Prov., Willowmore (H. Brauns)
(MHN, Geneva) [examined]. Syn. n.
Messor donisthorpei Santschi, 1937: 51. Syntype workers, females, SOUTH WEST AFRICA: West of Maltahohe,
1500 m, 12.xii.1934 (K. Jordan) (BMNH; MCZ, Cambridge; USNM, Washington) [examined]. Syn. n.
MEDIUM TO LARGE WORKER, HW 2.35- > 3.40.
Anterior clypeal margin varying from shallowly convex to transverse, only very rarely with the faintest
vestige of a median indentation. With the head in fullface view the sides more or less straight and
approximately parallel, never evenly convex nor obviously diverging anteriorly. Occipital margin broadly
and shallowly concave to indented medially. In HW range 2.35-3.44 the maximum diameter of the eye is
0.40-0.58, about 0.15-0.19 x HW, and the CI range is 103-119. Propodeum in profile with the dorsum
rounding narrowly into the declivity in most cases; in some more broadly rounded and in a few right-angled,
but only rarely with dentiform prominences and here usually only in the largest workers. Usual sculpture of
entire dorsum of head of fine, densely packed parallel longitudinal rugulae, most commonly with fine
punctulation between them. Variation in the sculpture consists of a reduction, in density or intensity, or one
or both of these components. Sometimes the rugulae are more widely spaced and fainter than is usual, in
which case the punctulate ground-sculpture is much more obvious and may appear as the dominant
component in places. On the other hand the punctulate sculpture may fade out, leaving the rugulae sharply
defined; the rugulae may then also become less intense and leave the head only feebly sculptured. Dorsal
alitrunk usually rugose or rugulose everywhere but, as on the head, this sculpture may be reduced until it is
very faint or even absent. When distinctly present the direction of sculpture on the pronotum shows
variation. Commonly it is longitudinal but forms with the sculpture diagonal, transverse, irregular or
varying on different parts of the surface are fairly frequent. First gastral tergite unsculptured or at most with
a very faint superficial patterning. All dorsal surfaces of head and body with numerous conspicuous standing
hairs. Colour black to dark reddish brown, the head and alitrunk always the same colour, the gaster
sometimes darker.
The taxa capensis, decipiens and piceus, treated here as separate species, may in fact represent
only a single variable species. The differences invoked to distinguish the three are minor (see key)
and may eventually prove to be gradient.
Among the species in which the first gastral tergite is uniformly hairy the three taxa mentioned
above are characterized together by their relatively small eyes, lack of strong gastral sculpture,
relatively straight-sided head and short propodeum, and lack of a median prominence on the
posterior half of the clypeus.
346 B. BOLTON
MATERIAL EXAMINED
Botswana: Kuke Pan (Vernay-Lang); Gomodimo Pan (Vernay-Lang); Gomodimo (G. U. Son). South
Africa: Cape Prov., Willowmore (G. Arnold); Willowmore (H. Brauns); Grahamstown (W. L. Brown);
Grahamstown (F. Jacot-Guillarmod); Cape Town (E. Simon); Cape Town (J. C. Bridwell); Cape Town (R. E.
Turner); Addo (M. Samways); Balfour (E. S. Ross & R. E. Leech); Fish River Valley (G. Arnold); Fort
Beaufort (J. W. G.); Oudtshorn (B. Brunhuber); Port Elizabeth (B. Brunhuber). South West Africa: W. of
Malahohe (K. Jordan).
Messor cephalotes (Emery)
Stenamma (Messor) barbarum subsp. cephalotes Emery, 1895a: 179. Syntype workers, ETHIOPIA: Arussi
Galla, Ganale Gudda, 3.V.1893 (V. Bottego) (MCSN, Genoa; MHN, Geneva) [examined].
Messor cephalotes (Emery) Emery, 1908 : 443. [Raised to species.]
Messor plinii Santschi, 1912: 165. Syntype workers, KENYA: Nakuru, 1904 (C. Alluaud) (NM, Basle)
[examined]. Syn. n.
MEDIUM TO LARGE WORKER, HW 3.20- > 5.00.
Median portion of clypeus with anterior margin broadly but shallowly indented-concave. With the head
in full-face view the occipital margin more or less transverse, very shallowly impressed medially to virtually
straight, only very rarely evenly shallowly convex. Head broad and massive, very strongly transversely
convex between the eyes, CI 109-123 in HW range 3.28-5.52. Eyes fairly small, their maximum diameter
0.54-0.72, about 0.13^0.17 x HW within the above-stated HW range; the relatively smaller eyes occurring
in larger individuals. Psammophore generally more strongly developed in smaller than in larger workers,
the characteristic hooked or J-shaped hairs sparse or absent in very large workers. Propodeal dorsum
varying from rounding bluntly and evenly into the declivity to meeting the declivity in a sharp right-angle.
In either case a low reinforcing lip or flange may be present which follows the curve, but prominent blunt
teeth or lamellae are only very rarely known to develop. Dorsum of head blanketed everywhere with
extremely fine, very densely and tightly packed, parallel longitudinal costulae; the head with a silky
appearance under low magnification. The direction of the costulae is variable but usually they run straight
back from clypeus to occiput centrally on the dorsum, and tend to diverge towards the occipital corners
away from this central strip. In very large workers there is a tendency for the direction of the sculpture to be
less regular, and even loops or whorls may occur. Dorsal alitrunk densely rugulose everywhere, the
sculpture usually transverse but sometimes irregular on the propodeum. Sculpture on propodeal dorsum
generally coarser and more widely spaced than on pronotum, and always coarser on pronotum than on
dorsum of head. Petiole and postpetiole finely and densely sculptured with feeble rugulae, dense
puncturation, or a combination of both. Base of first gastral tergite extensively sculptured with exceedingly
fine close-packed scratch-like costulae, or sometimes with dense granular puncturation, or with a
combination of both. The extent of this sculpture is variable but always at least the basal third of the first
tergite is covered. Pilosity quite dense, all dorsal surfaces of head and body with standing hairs. With the
head in full-face view the sides in front of and behind the eyes, and the curved side portions of the occipital
corners, without projecting hairs; the occipital margin itself usually with conspicuous projecting hairs.
Colour red to reddish dark brown, often with the gaster somewhat darker than the head and alitrunk.
A very distinctive East African species, cephalotes is one of the only two known African forms in
which the gaster is strongly sculptured. The other, regalis, has much coarser sculpture, as noted in
the key, and also differs by having the propodeum always bidentate or bispinose, a feature only
very rarely developed in cephalotes. Beside this the anterior clypeal margin, always concave in
cephalotes, is shallowly convex and irregular in regalis, and the sides of the head, hairless in
cephalotes, have distinct standing hairs in regalis, at least behind the eyes.
MATERIAL EXAMINED
Ethiopia: Ganale Gudda (V. Bottego). Kenya: Nakuru, (E. Pinhey); Nakuru (T. J. Anderson); Nakuru (C.
Alluaud); Lake Ngunga (Allen & Brooks); Kericho (F. W. Dry); Athi Riv. (C. S. Betton); Olikoriti (M. G.
Lepage); Kajiado (J. Darlington); Kajiado (G. Nyamasyo). Tanzania: Dodoma (W. M. Mann); Umbulu
(W. M. Mann); Arusha (C. F. D.).
Messor coltingwoodi sp. n.
(Fig. 30)
HOLOTYPE WORKER, HW 2.56.
Anterior clypeal margin broadly but shallowly indented medially. With the head in full-face view the sides
more or less straight, slightly convergent anteriorly and rounding broadly and evenly into the occipital
AFROTROPICAL MYRMICINE ANT GENERA
347
margin behind. Occipital margin sharply indented medially. Maximum diameter of eye 0.52, about
0.20 x HW, and the CI 107. Propodeum armed with a pair of short but well-developed triangular spines
which are somewhat downcurved along their length. Dorsum of head sculptured everywhere. Mid-dorsal
strip of head longitudinally rugulose to level of posterior margins of eyes; behind this the rugulae rapidly
weakening. Everywhere dorsum of head finely and very densely reticulate-punctate, with superimposed very
feeble rugulae away from the more strongly sculptured median strip. Pronotum and mesonotum dorsally
31
32
Figs 30-34 30-32, Messor workers. Alitrunk of (30) collingwoodi, (31) capensis, (32) striatifrons. 33, 34,
Cataulacus workers. Profile of (33) centrums, (34) moloch.
348 B. BOLTON
transversely rugulose, the propodeum more strongly transversely rugose. Sides of pronotum less strongly
rugulose than the pleurae. First gastral tergite unsculptured except for the usual fine superficial reticular
patterning. Dorsum of head sparsely hairy. Discounting the strong pilosity on the mouthparts and around
the frontal lobes the dorsum with only a few pairs of hairs spanning the mid-dorsal strip. With the head in
full-face view the sides both in front of and behind the eyes lacking projecting hairs. Projecting hairs also
absent from occipital corners but a single hair projecting from the occipital margin on each side of the
median indentation. Psammophore strongly developed, the J-shaped hairs conspicuous. Dorsal alitrunk
without hairs on pronotum, with 4 pairs on mesonotum and one pair on the propodeum. Petiole with one
pair, postpetiole and first gastral tergite hairless. Colour uniform very dark blackish brown.
MEDIUM TO LARGE PARATYPE WORKERS, HW 2.16-2.72. As holotype but in some individuals the mid-dorsal
rugulae of the head more sharply defined. Variation in pilosity throughout the type-series shows the dorsal
head with 2-5 pairs, pronotum with 0-1 pair, occipital margin with 0-2 pairs, mesonotum with 4-6 pairs,
propodeum with 1-3 pairs, petiole with 0-3 pairs, postpetiole with 0-3 pairs of hairs. First gastral tergite
consistently hairless. Eyes fairly large, within the HW range given above the maximum eye diameter is
0.46-0.58, about 0.20-0.22 x HW. CI range is 103-1 10.
Holotype worker, Niger: Azanyares, iii.1979 (J. Newby) (BMNH).
Paratypes. 12 workers with same data as holotype (BMNH; MCZ, Cambridge; NM, Basle; MHN,
Geneva).
Non-paratypic material examined. Mali: Tessalit (P. Room).
Among the species with hairless or near hairless first gastral tergite collingwoodi is distinguished
by having propodeal hairs present, having an extensively sculptured head, and having persistent
propodeal spines. M. luebberti, which also has the head sculptured everywhere, is reddish in
colour and lacks propodeal hairs and spines. Also, the rugular cephalic sculpture is more
extensively developed than in collingwoodi. M. angularis also lacks propodeal hairs and spines
and has the head weakly or not sculptured. M. galla, which frequently develops propodeal lobes
or teeth and which also has propodeal hairs present, lacks the characteristic cephalic sculpture of
collingwoodi.
The closest relatives of collingwoodi are, however, not to be found among the other
sub-Saharan African species but among the members of the aegyptiacus-group, of which
collingwoodi seems to be the only Afrotropical species.
Messor decipiens Santschi stat. n.
(Fig. 28)
Stenamma (Messor) barbarum r. capense var. decipiens Forel, 1905: 177 [unavailable name]; Santschi, 1917:
94 [Messor capense st. decipiens, first available use of name]. Syntype workers, females, SOUTH AFRICA:
Natal (Wroughton) (MHN, Geneva) [examined].
Messor barbarus subsp. capensis var. proba Forel, 191 la: 266. Holotype worker, SOUTH AFRICA: Orange
Free State, Bothaville (H. Brauns) (MHN, Geneva) [examined]. [Unavailable name.]
Messor arcistriatus Santschi, 1928: 202. Holotype worker, SOUTH AFRICA: Natal (Wroughton) (NM, Basle)
[examined]. Syn. n.
MEDIUM TO LARGE WORKER, HW 2.64- > 4.20.
Answering to the description of capensis in most particulars. In the HW range quoted above the
maximum diameter of the eye is 0.44-0.66, about 0.14-0.18 x HW, and the CI range is 107-121, the largest
workers known for decipiens thus being somewhat larger and broader headed than those known for
capensis. Propodeum in profile with the dorsum usually meeting the declivity in a right-angle, which may
project into a broad but quite short lobe or tooth of variable shape and size; rarely the propodeum merely
narrowly rounded. In contrast the propodeum of capensis is generally rounded, only seldom with dentiform
prominences. Sculpture of head basically the same as in capensis but here the rugae tending to be more
sharply developed and more widely separated, although there is some variation. Spaces between the rugae
usually smooth, frequently glossy, much less commonly with traces of punctulate ground-sculpture. Head
usually obviously red, contrasting in colour with the alitrunk and gaster which are darker. In smaller
workers this distinction in colour is not nearly so obvious and at the lower limit of the size range considered
here (and smaller) the ant may be unicoloured.
AFROTROPICAL MYRMICINE ANT GENERA 349
Very closely related to capensis and piceus, decipiens is separated from the former only on the
weak characters mentioned above. It is even closer to the latter, being distinguished only by the
colour of the hairs as noted in the key, and the fact that piceus does not have the head distinctly
different in colour from the alitrunk in large workers. It seems very probable that more extensive
collecting of this complex will reveal that these forms represent but a single species.
MATERIAL EXAMINED
Zimbabwe: Bulawayo (G. Arnold). Botswana: Ghazi (J. Maurice). Lesotho: Mafeteng (R. Crawshay). South
Africa: Natal, Weenen (G. Arnold); Natal (Wroughton); Durban (G. Arnold); Drakensberg, Van Reenen
(R. E. Turner); Mkuzi Reserve (C. P. Peelers); no loc. (ex coll. F. Smith); Transvaal, Brakfontein (Lingnau);
Vryburg (G. Arnold); Shiluvane (Junod); Orange Free State, Bothaville (H. Brauns).
Messor denticornis Forel
(Fig. 29)
Messor denticornis Forel, 1910a: 14. Syntype workers, female, male, SOUTH WEST AFRICA: Liideritzbucht,
1903 (L. Schultze) MHN, Geneva; BMNH) [examined].
Messor denticornis var. parvidens Forel, 19100: 15. Syntype workers, SOUTH WEST AFRICA: Kubub (L.
Schultze) (MHN, Geneva) [examined]. Syn. n.
Messor denticornis var. brunni Forel, 19106: 444. Syntype workers, SOUTH WEST AFRICA: no loc. (Brunn);
and SOUTH AFRICA: Cape Prov., Steckstown (Wartmann) (MHN, Geneva) [examined]. Syn. n.
MEDIUM TO LARGE WORKER, HW 2.48- > 3.10.
Anterior clypeal margin usually evenly convex medially, only rarely with the faintest trace of a central
indentation. With the head in full-face view the sides more or less straight and diverging anteriorly, but
sometimes the sides more nearly parallel. Occipital margin broadly but shallowly concave, this feature
fading out in smaller workers where the margin is approximately transverse. In HW range 2.48-3.16 the
maximum diameter of the eye is 0.56-0.70, about 0.21-0.25 x HW, and the CI range is 100-106. Propodeum
in profile relatively long and low, resembling that of striatifrons (Fig. 32). Propodeal armament very
variable, the junction of dorsum and declivity being rounded, acutely angled or distinctly bidentate. These
variants are commonly seen in the same series and are in fact shown by the type-series of denticornis itself.
Basic sculpture of the head finely densely packed parallel longitudinal rugulae with punctulate
ground-sculpture between them. Frequently the sculpture much reduced, either by suppression of the
ground-sculpture so that the rugulae stand out from a smooth surface or by reduction of the rugulae in
number and intensity so that the head is mostly or wholly punctulate. In smaller individuals the surface may
be almost smooth. Dorsal alitrunk rugulose to rugose, the sculpture sometimes partially or totally effaced
from the pronotum. First gastral tergite smooth and shining or at most with faint superficial pattering. All
dorsal surfaces of head and body with numerous standing hairs. Colour mid-brown to black, sometimes
with the gaster darker than the head and alitrunk.
A distinctive species amongst those with uniformly distributed pilosity on the first gastral tergite,
denticornis is immediately isolated by its relatively large eyes. Only a few workers of tropicorum
approach even the lower end of its eye size range but in the latter species the clypeus has a
conspicuous posteromedian tumulus or welt and the propodeum is shorter and higher in profile
than is the case in denticornis.
MATERIAL EXAMINED
Botswana: Ghanzi (E. S. Ross & A. R. Stephen). South West Africa: Okaukuejo (E. S. Ross & R. E. Leech);
Spitzekopfe (E. S. Ross & K. Lorenzen); Ababis (R. W. Tucker); Berseba (L. O. Sordahl); Liideritzbucht (L.
Schultze); Kubub (L. Schultze). South Africa: Cape Prov., Oudtshorn (B. Brunhuber); Strydenburg (M.
Patterson); Steckstown (Wartmann).
Messor galla (Mayr)
Stenamma (Messor) barbarum subsp. caduca var. galla Emery, 1895a: 179 [unavailable name]; Mayr, 1904:
5 [Stenamma (Messor) barbarum var. galla, first available use of name]; Santschi, 1928: 201 [galla raised
to species]. Holotype worker, ETHIOPIA: Alto Duau, Boran Galla, v. 1893 (V. Bottego) (MCSN, Genoa)
[examined].
Messor barbarus subsp. semirufus var. rufa Forel, 1910c: 250. Syntype workers, ETHIOPIA: Nefassit (K.
Escherich) (MHN, Geneva) [examined]. [Unavailable name.]
350 B. BOLTON
Messor barbarus st. galla var. triempressa Santschi, 1917: 92. Syntype workers, CHAD: Baguirmi, Techeckna;
ETHIOPIA: no loc.; SENEGAL : Casamance (Clavaux). (NM, Basle) [examined]. [Unavailable name.]
Messor barbarus st. latinoda Santschi, 1917: 93, fig. 2. Syntype workers, 'EAST AFRICA': no loc.
(Reichensperger) (NM, Basle) [examined]. Syn. n.
Messor barbarus r. semirufus var. rufula Forel, 1918: 156. [Unnecessary replacement name for rufa Forel,
1910c: 250, above.] [Unavailable name.]
Messor barbarus subsp. galla var. armata Emery, 19226: 98. Syntype workers, GHANA: no loc. (MCSN,
Genoa) [examined]. [Unavailable name.]
Messor galla st. nobilis Santschi, 1928: 201. Syntype workers, female, ETHIOPIA: Bisa Tint, 1200m
(Reichensperger) (NM, Basle) [examined]. Syn. n.
Messor galla var. airensis Bernard, 1950: 286, Syntype workers, NIGER: Air Dist., Dabaga, 600 m; Mt
Baguezans, 1500 m; Agadez, 525 m (Chopard & Villiers). [Not found in MNHN, Paris, presumed lost.]
Syn. n.
MEDIUM TO LARGE WORKER, HW 2.40- > 3.70.
Median portion of clypeus with anterior margin broadly but shallowly concave to more or less entire.
With the head in full-face view the sides very shallowly convex to roughly straight, usually slightly
convergent in front of the eyes. Occipital margin broadly indented medially. In HW range 2.40-3.76 the
maximum diameter of the eye 0.44-0.68, about 0.17-0.20 x HW, and the CI range 102-114. Propodeum
showing great variation; frequently with the dorsum rounding into the declivity but sometimes with a pair
of broad teeth or lamellae. Between these two extremes is a range of intermediates including forms with a
narrow to broad rim or flange following the curve of the surface, forms with a small to large salient angle
and forms with the angle or flange projecting to various degrees. Dorsum of head smooth and shining, away
from the median strip sculptured only with very widely scattered small pits or a faint superficial patterning.
Median strip of head behind clypeus with longitudinal rugular sculpture which usually extends back at least
as far as the level of the posterior margins of the eyes, and often distinctly further back than this; only very
rarely is the rugular strip shorter. Intensity of rugulae on the median strip very variable and the width of the
strip not usually exceeding the width across the frontal lobes and often narrower, only rarely slightly wider.
Pronotum dorsally with weak transverse rugulae which may sometimes be very feeble or even partially
effaced. Mesonotum varying from almost smooth to faintly rugulose. Propodeal dorsum generally sharply
transversely rugose but in some samples the rugae diagonal, irregular or interrupted. First gastral tergite
unsculptured but often showing a faint superficial patterning. With the head in full-face view the sides
without projecting hairs, the occipital margin with 0-4 hairs on each side of the median impression.
Generally hairs are present occipitally, specimens with zero count are very few and may be the result of
abrasion. Dorsum of head sparsely hairy, the psammophore conspicuous ventrally. Parts of dorsal alitrunk
with pilosity as follows; pronotum with 0-4 pairs, mesonotum with 4-10 pairs, metanotal groove with 1-2
pairs at least in large workers, propodeum with 1-5 pairs. Petiole with 1-3, postpetiole with 3-6 pairs of
hairs. First gastral tergite without hairs or with a sparse transverse row at the extreme apex of the sclerite.
Ventral surfaces of hind femora usually with hairs all along the shaft but in some they are denser proximally
than distally. Colour reddish brown to blackish brown, usually with the gaster darker than the head and
alitrunk. In some samples the head slightly more reddish than the alitrunk.
Without doubt the commonest, most successful and most widely distributed Messor species in
the northern half of sub-Saharan Africa, galla ranges throughout the Sahelian zone across the
entire width of the continent. On the eastern side it is found as far south as Kenya, and in the west
it occurs coastally as well as in the drier northern parts of the West African states. Of the four
species in the region which lack dense gastral pilosity galla is separated from collingwoodi and
luebberti by the extensive cephalic sculpturing of the last two. The separation of galla from its
closest African relative, angularis of Kenya, is tabulated under that name. Some aspects of the
biology of galla have been investigated by Levieux & Diomande (1978) and Levieux (1979).
MATERIAL EXAMINED
Ethiopia: Addis Ababa, Entoto Hills (K. Guichard); Addis Ababa (V. O. De Mass/); Boran Galla (V.
Bottego); Lake Zwai, Sucsuci (J. 0. Cooper); Wachacha Ravine (H. Scott); Bisa Tint (Reichensperger); Mt
Monagasha (Cloudsley-Thompson); Gondar (Cloudsley-Thompson); Tisisat Falls (Cloudsley-Thompson);
Holetta; Dessie (E. S. Ross); Nefassit (K. Escherich); Barentu (Muller); Tessenei (Muller); Amba Derho
(Muller); Om Agar (Muller); Ghinda (K. Escherich); no loc. (G. McCreagh). Somali Republic: Alabla Balleh
(P. E. Glover). Kenya: Nakuru (N. A. Weber); Marsubit (Rift Valley Expd.); Tsavo East (J. Darlington);
Maralal (M. E. Irwin & E. S. Ross). Sudan: Kadugli (C. Sweeny); Khartoum (N. A. Weber); Khartoum (R.
Cottom); Khartoum (H. H. King); Kulme (H. Lynes); Lake Kellek (C. Sweeny); Dilling-El Obeid Rd. (C.
AFROTROPICAL MYRMICINE ANT GENERA 351
Sweeny); Sennar (B. Hocking); Imatong Mts (N. A. Weber); Equatoria (N. A. Weber). Niger: Niamey (P.
Room); Niamey (J. Levieux); Ayorou (P. Room); Assode (J. Levieux). Mali: Gao (B. Malkin); Anefis (P.
Room). Upper Volta : Ougadougou (P. Room); Banfora (Betbeder). Senegal: Dakar (W. L. Brown); Dakar (N.
L. H. Krauss); Casamance (Clavaux). Ivory Coast: Korhogo (R. Lucius); Ferkessedougou (J. Levieux).
Ghana: Lawra (W. Cook); Bolgatanga (P. Room); Tamale (Anipare); Tumu (P. Room); Navrongo (C. A.
Collingwood); Dawhwenya (D. Lesion); Dawhwenya (C. A. Collingwood); Nyankpala; Prampram (W.
Belfield); Achimota (W. Belfield); Nungua (W. Belfield); Accra (C. A. Buckman). Nigeria: Kalkala (F. D.
Golding); Illela (Lelean); Katsina (J. T. Medler); Zaria (A. S. Ahman); Maiduguri (E. R. Ross & K.
Lorenzen).
Messor incisus Stitz nomen dubium
Messor incisus Stitz, 1923: 149. Holotype female, SOUTH WEST AFRICA: Okosongomingo Farm, vii-viii.1912
(H. Thomsen) [not found in MNHU, Berlin, presumed lost].
Described from a single female which has since been lost, the identity of incisus cannot be
ascertained accurately at present. In his original description of incisus Stitz compares it to the
female of denticornis. So few females of denticornis are known that it is possible for incisus to fall
within the range of variation of that species. On the other hand incisus may be the female of
striatifrons or indeed be a separate species. Considerably more samples of Messor females will be
necessary before any attempt at placing incisus can be made.
Messor luebberti Forel stat. n.
Messor barbarus subsp. lubberti Forel, 1910a: 13. Syntype workers, SOUTH WEST AFRICA: Okahandja
(Peters), and no loc. (Lubbert) (MHN, Geneva) [examined].
MEDIUM TO LARGE WORKER, HW 2.00- > 3.00.
Anterior clypeal margin flattened to slightly indented medially. With the head in full-face view the sides
more or less straight, roughly parallel or weakly convergent anteriorly. Occipital margin distinctly indented
medially in large workers but the indentation becoming obliterated with reduced size. In HW range
2.00-3.12 the maximum diameter of the eyes 0.38-0.50, about 0.15-0.18 x HW, and the CI is 100-112. With
the propodeum in profile the dorsum rounding narrowly into the declivity to meeting the declivity in a
right-angle; propodeal armament never developed. Dorsum of head everywhere finely and densely
longitudinally rugulose, the rugulae approximately parallel and becoming finer away from the mid-dorsal
strip. Ground-sculpture of minute punctulation is present between the rugulae but this is less conspicuous in
some samples than in others. Pronotal dorsum weakly and faintly to quite strongly transversely rugulose,
but always with a fairly distinct punctulate component between the rugulae. Mesonotum smooth with only
vestigial traces of sculpture to irregularly granular, only rarely with a rugular component. Propodeal
dorsum transversely rugulose to rugose, with punctures between the rugulae. First gastral tergite
unsculptured except for the fine superficial reticular patterning which is usual in the genus. With the head in
full-face view the sides and occipital margin lacking projecting hairs. Projecting hairs very sparse to absent
on dorsum of head but present on mouthparts and between frontal lobes. Psammophore strongly
developed. On dorsal alitrunk the pronotum with 0-4 pairs of hairs, the mesonotum with 2-6 pairs; the
propodeum, petiole and postpetiole lacking hairs. First gastral tergite without hairs or at most with 2-3 at
the extreme apical margin of the sclerite. Colour usually red with a blackish gaster but in some the gaster the
same shade of red as the head and alitrunk. Shade of red of head and alitrunk varying from bright, almost
orange, to very dull.
This very distinctive species is extremely widespread in the southern half of the African continent.
It is immediately recognizable by its strongly sculptured head and very reduced pilosity. Of the
sparsely hairy species of Africa only collingwoodi from Mali and Niger has the head anywhere
near as strongly sculptured as luebberti, but in that species the propodeum has hairs and the
junction of propodeal dorsum and declivity is armed with a pair of short spines.
MATERIAL EXAMINED
Tanzania: Dodoma (A. Loveridge). Zimbabwe: Bulawayo (G. Arnold); Springvale (G. Arnold). Botswana:
Damara Pan (G. U. Son); Kuke Pan (G. U. Son); Gomodimo (G. U. Son); Xani Pan (A. Russell-Smith).
Angola: Cahama (E. S. Ross & R. E. Leech). South West Africa: Gemsbok Pan (G. U. Son); Okahandja
(Peters); no loc. (Lubbert); Windhoek (Ross & Stephen). South Africa: Transvaal, Shiluvane (Junod);
Malagieskraal (Lingnau); Pretoria; Pietersburg (E. S. Ross & R. E. Leech).
352 B. BOLTON
Messorpiceus Stitz
Messor piceus Stitz, 1923: 150. Syntype workers, female, SOUTH AFRICA: Transvaal (Ulrich) [not found in
MNHU, Berlin, presumed lost].
MEDIUM TO LARGE WORKER, HW 3.28- > 4.20.
Answering to the description of capensis, but differing mainly in the colour of the body pilosity which is
white to yellowish in capensis but very deep red-brown to blackish in piceus. Apart from this the anterior
clypeal margin is indented medially in piceus; the propodeum varies from narrowly rounded through
right-angled to broadly and bluntly dentate, and the largest known workers are larger than those of
capensis. In the HW range 3.28-4.20 the maximum diameter of the eye is 0.54-0.64, about 0.15-0.17 x HW,
and the CI range is 106-1 19. The maximum known for capensis is HW 3.44 but this may not be the largest
worker of the species, merely the largest available for study at present. Relative size of eye and CI fall within
the range of capensis.
Unfortunately the type-series of piceus appears to be lost, but three short series from Transvaal
match the original description tolerably well and show the dark pilosity noted by Stitz. I am
therefore applying the name piceus to these specimens and to two other short series, from Natal
and Botswana, noted under material examined.
M. piceus is a very closely related to capensis and decipiens; these three names may ultimately
prove to represent only a single species.
MATERIAL EXAMINED
Botswana: Gomodimo (Vernay-Lang). South Africa: Natal, Pietermaritzburg (Akerman); Transvaal,
Sabie; Kimberley (G. Arnold); Oliphants River, Grootdraai (H. Lang).
Messor regalis (Emery)
Cratomyrmex regalis Emery, 1891: 572, pi. 15, fig. 16. LECTOTYPE female, NIGERIA: Benue (Staudinger)
(MCSN, Genoa), here designated [examined].
Cratomyrmex regalis var. rubea Santschi, 1913: 308. Holotype worker, BENIN REPUBLIC: no loc. (Le Moult)
(NM, Basle) [examined]. Syn. n.
Cratomyrmex sculpturatus Stitz, 1916: 377, fig. 2. Syntype workers, CONGO: Fort Possel-Fort Crampel,
xi.1910 (Schubotz); and Chutes de la Nana, 'bei Fort Crampel', 7.xi.l910 (Haberer) (MNHU, Berlin)
[examined]. Syn. n.
Messor regalis (Emery) Emery, 1922a: 357.
MEDIUM TO LARGE WORKER, HW 3.00- > 4.50.
Median portion of clypeus with anterior margin shallowly convex to somewhat flattened, irregular
because of strong sculpture but not strongly impressed-concave. In HW range 3.00-4.40 the maximum
diameter of the eye is 0.48-0.70, about 0.16-0.17 x HW, and the CI range is 109-115. With the head in
full-face view the sides in front of the eyes more or less straight, roughly parallel or slightly convergent
anteriorly. Behind the eyes the sides rounding very broadly and evenly into the occipital margin ; the latter
usually shallowly indented medially. Propodeum armed with a pair of short triangular spines. Dorsum of
head densely sculptured everywhere with coarse parallel longitudinal rugulae. On the median strip behind
the frontal lobes the rugulae tend to run straight back on the head; on each side of this strip they diverge
towards the occipital corners. Pronotal dorsum coarsely sharply and irregularly rugose, frequently
reticulate-rugose in places and generally with a strip of strong transverse rugae immediately behind the
cervical shield. Remainder of dorsum and also sides of alitrunk strongly and generally sharply rugose
everywhere, the sculpture stronger than on the dorsum of the head. Tergal portions of petiole and
postpetiole very closely and coarsely irregularly rugose, the surfaces with a crumpled and wrinkled
appearance. First gastral tergite rugulose to sharply costulate basally, the sculpture extending at least over
the basal third of the sclerite and becoming finer posteriorly. All dorsal surfaces of head and body with
numerous standing hairs, pilosity also dense on legs. With the head in full-face view projecting hairs are
present on the sides behind the eyes, on the broad curve of the occipital corners and on the occipital margin
itself. One or two hairs usually also project from the sides in front of the eyes. Psammophore conspicuously
developed. Colour dull red to reddish brown, the gaster sometimes with an orange tint.
A species of West and Central Africa regalis is easily characterized by its blanketing coarse
rugose sculpture. No other species in the region has sculpture approaching that found in regalis.
This feature coupled with the dense pilosity and persistent propodeal spines renders the species
AFROTROPICAL MYRMICINE ANT GENERA 353
quickly recognizable. Only cephalotes and regalis have extensive sculpture on the first gastral
tergite; characters separating the two are given under cephalotes.
Some aspects of the biology of regalis have recently been investigated by Levieux & Diomande
( 1 978) and Levieux( 1979).
MATERIAL EXAMINED
Nigeria: K. State, N. Bussa (J. T. Medler); Mokwa (C. Longhurst); Olokemeji (Bridwell); Benue
(Staudinger). Benin Republic: no loc. (Le Moult). Congo: Fort Crampel (Schubotz).
Messor ruginodis Stitz stat. n., nomen dubium
Messor barbarus st. ruginodis Stitz, 1916: 374, fig. 1. Syntype workers, CONGO: Fort Crampel,
xi.l910-6.i.!91 1 (Schubotz) [not found in MNHU, Berlin, presumed lost].
Apart from the very distinctive regalis this is the only other species of Messor recorded from the
Congo. It is possible to decide from Stitz's description that ruginodis is related to capensis and its
allies, but further placement cannot be attempted without the types as the description alone is not
good enough. It must suffice for the present to state that, apart from regalis, no Messor species is
known to extend its range into the Congo, so ruginodis remains an enigma.
As the species, whatever it really is, is definitely not closely related to barbarus, I have raised it
to species-level here.
Messor striatifrons Stitz stat. n.
(Figs 27, 32)
Messor denticornis var. striatifrons Stitz, 1923: 149. Syntype workers, SOUTH WEST AFRICA: no loc. (Scheben)
(MNHU, Berlin) [examined].
MEDIUM TO LARGE WORKER, HW 2.84- > 3.75.
Anterior clypeal margin usually shallowly convex medially but sometimes a weak central indentation of
the margin is present. With the head in full-face view the sides convex. Generally the convexity is distinct
(Fig. 27) in larger workers but tends to be less marked in smaller individuals; infrequently the reverse is true
and medium sized workers show the convexity more strongly than larger specimens. Occipital margin
shallowly indented medially, the indentation best developed in large workers and slowly disappearing with
decrease in size. Within the HW range 2.84-3.76 the maximum diameter of the eye is 0.52-0.68, about
0.16-0.18 x HW, and the CI range is 104-114. Propodeum in profile relatively long and low (Fig. 32),
usually rounded at the junction of dorsum and declivity but quite frequently right-angled or projecting into
a broad short tooth which is really no more than a projection of the right-angle. Dorsum of head sculptured
with extremely fine dense longitudinal rugulae which in the strongest sculptured individuals are very close
packed. Spaces between the rugulae with fairly conspicuous ground-sculpture of fine punctures. In medium
sized workers, and quite frequently in maximum sized workers also, the sculpture on the dorsal head is
modified by a weakening of the rugular component and an intensification of the punctures, so that in some
the rugular component is supressed and the head appears reticulate-punctate everywhere or almost
everywhere. Dorsal alitrunk rugulose, the direction of the sculpture variable but usually stronger on the
propodeum than elsewhere. First gastral tergite unsculptured or at most with the faint superficial patterning
so commonly seen in this genus. All dorsal surfaces of head and body with numerous standing hairs; evenly
distributed hairs conspicuous on first gastral tergite. Colour medium to dark brown, commonly uniform but
often with the gaster darker, blackish brown.
A fairly distinctive member of the group of species centring on capensis, striatifrons is
characterized by its relatively long low propodeum and convex head sides. The shape of the head
is not duplicated in other African species but denticornis has a similarly proportioned
propodeum. However, in this last-named species the eyes are larger, with a range of
0.21-02.5 x HW.
MATERIAL EXAMINED
South Africa: Cape Prov., Victoria West (G. Arnold); Steinkop (G. Arnold); Springbok (E. S. Ross & R. E.
Leech); Picketberg (E. S. Ross & R. E. Leech); Citrusdal (E. S. Ross & R. E. Leech); Papendrop (E. S. Ross &
K. Lorenzen); Clanwilliam (£. 5. Ross & R. E. Leech). South West Africa: no loc. (Scheben).
354 B. BOLTON
Messor tropicorum Wheeler stat. n.
Messor barbarus subsp. capensis var. tropicorum Forel, 1910fo: 444 [unavailable name]; Wheeler, 1922: 805
{capensis var. tropicorum, first available use of name]. Syntype workers, ANGOLA: Mossamedes (Baum &
Van der Kellen) (MHN, Geneva) [examined].
Messor denticornis var. laevifrons Stitz, 1923: 148. Syntype workers, SOUTH WEST AFRICA: Usakos,
iv.-vi.1911; and Grootfontein, 7-ll.vi.1911 (W. Michaelsen} |(MNHU, Berlin) [examined]. Syn. n.
Messor braunsi var. nigriventris Stitz, 1923: 150. Syntype workers, SOUTH WEST AFRICA: Grootfontein,
7-1 l.vi.191 1 (W. Michaelsen) (MNHU, Berlin) [examined]. Syn. n.
MEDIUM TO LARGE WORKER, HW 3.00- > 3.80.
Anterior clypeal margin entire or at most with a feeble median indentation. With the head in full-face view
the sides approximately straight, more or less parallel or feebly diverging anteriorly. Occipital margin
usually broadly and shallowly concave but this becomes less apparent with decreased size. Centre of
posterior half of clypeus, between the frontal lobes, with a conspicuously raised tumulus or welt in large
workers, this feature decreasing in intensity with reduced size and not present in smaller workers. In the HW
range 3.00-3.84 the maximum diameter of the eye is 0.64-0.72, about 0.19-0.21 x HW, and the CI range is
102-1 1 1. Propodeum in profile relatively short and high, like that of capensis (Fig. 31). Propodeal dorsum
either rounding into declivity, or meeting it in a right-angle, or armed with a pair of short triangular teeth ;
variation occurs within series. Dorsum of head sculptured with narrow fine longitudinal rugulae. In
strongest sculptured individuals the rugulae are dense and conspicuous, but often they are much reduced or
partially to entirely effaced away from the central strip. Between the rugulae the ground-sculpture is of a fine
superficial punctulation, often completely effaced. Dorsal alitrunk rugulose to rugose, the sculpture
frequently weak on the pronotum or even absent in places. First gastral tergite unsculptured or at most with
a faint superficial reticular pattering. All dorsal surfaces of head and body with numerous conspicuous
standing hairs. Head and alitrunk reddish brown, the gaster darker.
Larger workers of tropicorum are quickly isolated by their possession of a strong prominent welt
or tumulus posteromedially on the clypeus, but this character fades with reduced worker size.
The eyes are quite large, approaching the lower limit of the range seen in denticornis, but in the
latter the propodeum is longer and lower in profile.
MATERIAL EXAMINED
South West Africa: Kabiras (R. W. E. Tucker); Usakos (W. Michaelsen), Grootfontein (W. Michaelsen).
Angola: Mossamedes (Baum & Van der Kellen).
CATAULACUS F. Smith
Cataulacus F. Smith, 1853: 225. Type-species: Cataulacus taprobanae F. Smith, 1853: 225, by subsequent
designation of Bingham, 1903: 120.
For diagnosis of genus, current synonymy and generic revision see Bolton (1974). For some time I
have been unhappy about the treatment which I gave to some species in the C. tenuis-group of
Africa (Bolton, 1974). It has become apparent, with the acquisition of more material and with
further experience of the group, that I was wrong to synonymize some of the names. The
opportunity to rectify these mistakes now presents itself and the changes from the previous
system are summarized below. Following this six new species are described and a revised key to
the Afrotropical species is provided which reflects these additions and changes, and which
includes also the two African species recently described by Snelling (1979). The new key only
deals with the Afrotropical fauna; it excludes the Malagasy species which were incorporated in
the former (1974) key. For identification of such species the reader is referred back to the earlier
study.
Key to species (workers)
1 Dorsal alitrunk without standing hairs of any description or at most with only 1-2 very short
hairs at the highest point of the pronotum. Generally hairs absent from alitrunk but rarely
sparse strongly appressed hairs may be present
Dorsal alitrunk with numerous standing hairs which are usually conspicuous. If the standing
hairs are very short they are more or less evenly distributed over the dorsum and are not
restricted to the highest point of the pronotum 10
AFROTROPICAL MYRMICINE ANT GENERA 355
2 Propodeum completely unarmed, without trace of spines or teeth. (Zaire) . . . inermis Santschi
Propodeum armed with a pair of spines or teeth .3
3 Dorsal alitrunk strongly sulcate throughout. Appressed hairs present on the dorsal alitrunk.
(Ghana) . adpressus Bolton
Dorsal alitrunk reticulate-punctate to reticulate, usually also with fine rugulae or a
rugoreticulum present ; never sulcate. Appressed hairs absent from dorsal alitrunk . . 4
4 Petiole dorsally strongly transversely rugose or sulcate everywhere 5
Petiole dorsally variously sculptured but never transversely strongly rugose or sulcate . . 7
5 First gastral sternite laterobasally with a longitudinal margination or carina which parallels the
laterobasal margination of the first tergite. Femora of hind legs not excessively
anteroposteriorly compressed ........... . 6
First gastral sternite laterobasally without a longitudinal margination or carina which parallels
the laterobasal margination of the first tergite. Femora of hind legs strikingly
anteroposteriorly compressed, narrow and very deep. (Sierra Leone, Cameroun, Equatorial
Guina, Congo, Zaire, Uganda) kohli Mayr
6 Sides of head behind eyes irregular, either denticulate, crenulate or otherwise jagged. Relatively
broader-headed species, CI > 125, the head strongly broadened behind the eyes. Laterally
projecting hairs on sides of head behind eyes long and conspicuous. (Sierra Leone, Ghana,
Nigeria, Cameroun, Uganda, Zaire, Zambia) huberi Andre
Sides of head behind eyes regular, smooth, neither denticulate nor crenulate. Relatively
narrower-headed species, CI 120 or less, the head not strongly broadened behind the eyes.
Laterally projecting hairs on sides of head behind eyes minute and inconspicuous or absent.
(Ghana, Nigeria, Cameroun, Uganda, Congo, Zaire) egenus Santschi
7 Petiole and postpetiole in dorsal view strongly longitudinally sulcate. Postpetiole dorsally
divided into two projecting lobes by a deep median longitudinal cleft. (Cameroun, Zaire)
lobatus Mayr
Petiole and postpetiole in dorsal view not strongly longitudinally sulcate. Postpetiole dorsally
not divided into two projecting lobes by a deep median longitudinal cleft .... 8
8 Lateral pronotal margination with 2 teeth. Dorsal and lateral surfaces of petiole and postpetiole
with numerous tubercles and small angular prominences, presenting a multi-peaked and
irregular surface. (Cameroun, Congo, Zaire, Kenya) pullus Santschi
Lateral pronotal margination with 0-1 teeth. Dorsal and lateral surfaces of petiole and
postpetiole not equipped with tubercles and small angular prominences .... 9
9 With the head in full-face view the lateral margins behind the eyes without a row of short
projecting hairs. Lateral pronotal margination without teeth. (Ghana, Cameroun, Guinea,
Zaire) tar dux Santschi
With the head in full-face view the lateral margins behind the eyes with a row of short projecting
hairs. Lateral pronotal margination with a single tooth on each side, close to the anterior
pronotal corner. (Zaire) theobromicolus Santschi
10 Petiole and postpetiole strongly transverse, much flattened dorsoventrally and without nodes,
both very broadly thickly V-shaped in dorsal view. Propodeum armed only with a pair of
small teeth or tubercles which are inconspicuous. (Sierra Leone, Liberia, Ghana, Nigeria,
Cameroun, Zaire) . • mocquerysl Andre
Petiole and postpetiole nodiform, not strongly transverse nor flattened, not broadly V-shaped
in dorsal view. Propodeal spines well developed and conspicuous .... . 11
1 1 Hairs on clypeus and usually also on remainder of cephalic dorsum bizarre, strongly clavate or
stalked-suborbicular. In most the apex of each hair is very strongly swollen whilst the stem is
narrow; sometimes the stem may be short or very short . . . . . . . 12
Hairs on clypeus and remainder of cephalic dorsum simple, usually stout cylindrical and blunt
but sometimes very short and stubble-like, sometimes elongate and fine and occasionally
gradually increased in thickness from base to apex, but not strongly clavate or
stalked-suborbicular ............ 20
1 2 With the alitrunk in dorsal view the pronotal margin on each side without an unbroken series of
denticles which project laterally between the pronotal corner and the site of the promesonotal
junction . 13
With the alitrunk in dorsal view the pronotal margin on each side with an unbroken series of
denticles which project laterally between the pronotal corner and the site of the promesortotal
junction ................ 15
13 First gastral tergite regularly longitudinally sulcate throughout. (Cameroun) . . jacksoni (p. 360)
356 B. BOLTON
First gastral tergite reticulate-punctate or with tine rugulae overlying reticulate-punctate
ground-sculpture, never longitudinally sulcate 14
14 Propodeal dorsum longitudinally rugulose. (Nigeria, Cameroun) . . . . vorticus Bolton
Propodeal dorsum transversely rugose. (Nigeria) ....... boltoni Snelling
15 Bizarre hairs on dorsum of head behind clypeus with a very short basal stem, appearing
stud-like, the swollen apices set very close to the cephalic surface 16
Bizarre hairs on dorsum of head behind clypeus with an elongate basal stem, never short and
stud-like, the swollen apices conspicuously raised well clear of the cephalic surface . . . 17
16 Larger species, HW 0.80 or more. (Tanzania, Zimbabwe, Angola, South Africa)
brevisetosus Forel
Smaller species, HW < 0.80. (Ivory Coast, Ghana, Cameroun, Uganda, Kenya, Tanzania,
Angola) jeannett(p. 358)
17 Dorsal alitrunk with weak rugulose sculpture and a blanketing dense reticulate-punctate
ground-sculpture which is very conspicuous between the rugulae, the surface matt and dull . 1 8
Dorsal alitrunk with strong dense rugose sculpture the spaces between which are unsculptured
or at most contain some feeble superficial ground-sculpture, the surface glossy . . . 19
18 Denticles on lateral pronotal margins minute and inconspicuous in dorsal view, much smaller
than the tooth at the pronotal corner. (Cameroun) satrap(p. 363)
Denticles on lateral pronotal margins large and conspicuous, at least as large as the tooth at the
pronotal corner, sometimes larger. (Ghana, Nigeria, Cameroun, Zaire) . . . lujae(p. 358)
19 Smaller species, HW 0.80 or less. Body hairs relatively short (Fig. 34). Basal quarter of first
gastral tergite without strong rugulae, either punctate or with feeble rugulae caused by
alignment of punctures. Propodeal spines in profile evenly feebly curved. (Ghana, Nigeria)
moloch(p. 361)
Larger species, HW > 0.90. Body hairs relatively long (Fig. 33). Basal quarter of first gastral
tergite with strong longitudinal rugulae which are independent of the underlying
puncturation. Propodeal spines in profile with basal third elevated and apical two-thirds
recurved. (Cameroun) centrurus (p. 359)
20 Erect hairs on dorsal surfaces of head, alitrunk and gaster abundant, dense, very long narrow
and fine, curved or even sinuate, the entire ant with a softly pilose appearance rather than the
bristly appearance usually associated with this genus 21
Erect hairs on dorsal surfaces of head, alitrunk and gaster relatively sparse, short broad and
blunt, coarse and usually straight, the entire ant with a bristly or stubbly appearance . . 22
21 Propodeal dorsum longitudinally rugulose or rugose. Larger species, HL > 0.90, HW > 0.85.
(Ghana, Cameroun, Angola) elongatus Santschi
Propodeal dorsum transversely rugulose. Smaller species, HL < 0.90, HW < 0.85. (Zaire)
pilosus Santschi
22 Head relatively broad or very broad, the eyes small, CI > 112, OI < 30. In dorsal view the
posterolateral portion of the pronotal margin produced into a large spine or triangular
prominence. Propodeal spines long and very strong, not dorsoventrally flattened ... 23
Head relatively narrow and eyes larger, CI 1 10 or less, OI > 32. In dorsal view the
posterolateral portion of the pronotal margin usually armed with a short tooth or a denticle.
When a short tooth is present in this position it is usually comparable in size with others on
the pronotal margin. Propodeal spines usually dorsoventrally flattened, only rarely otherwise
. . . . 24
23 Sculpture of dorsal alitrunk a very distinct rugoreticulum with strongly reticulate-punctate
interspaces. Lateral margins of mesonotum usually with one or more denticles. (Liberia,
Ghana, Cameroun, Equatorial Guina, Gabon, Congo, Zaire). . . . erinaceus Stitz
Sculpture of dorsal alitrunk variable in intensity but consisting essentially of a longitudinal
rugation or sulcation which may be irregular or sinuate. Lateral margins of mesonotum
usually without denticles. (Ivory Coast, Liberia, Ghana, Nigeria, Cameroun, Equatorial
Guinae, Zaire, Uganda) guineensis F. Smith
24 Posterior one-quarter of first gastral tergite coarsely longitudinally sulcate, rugose or striate,
this sculpture always very distinct and usually extending to the apex of the tergite ... 25
Posterior one-quarter of first gastral tergite reticulate-punctate or finely superficially sculptured
and shining; a few fine scattered longitudinal rugulae formed by the fusion of the margins of
aligned punctures may sometimes be present 27
25 Smaller species, HW < 0.90, with relatively large eyes, OI 50 or more. (Zaire, Kenya)
striativentris Santschi
AFROTROPICAL MYRMICINE ANT GENERA 357
Larger species, HW > 0.95, with relatively smaller eyes, OI in range 34-48 .... 26
26 Dorsal surfaces of head and alitrunk with numerous conspicuous relatively long stout hairs.
Eyes slightly larger, OI range 43-48. (Kenya, Mozambique, South Africa) . . wissmanni Forel
Dorsal surfaces of head and alitrunk with relatively few inconspicuous very short stubbly hairs.
Eyes slightly smaller, OI range 34—40. (Ethiopia, Somali Republic, Kenya, Tanzania, Zambia,
Malawi, Zimbabwe, Mozambique, Angola, South West Africa, South Africa)
intrudens (F. Smith) (part)
27 Occiput with a distinct deeply incised transverse groove above the foramen. Below this the
remaining strip of the occiput juts out as a shield over the dorsal rim of the foramen itself.
(Uganda) impressus Bolton
Occiput without a deeply incised transverse groove above the foramen 28
28 Subpetiolar process complex, anteroventrally with a prominent broadly rounded angle and
posteroventrally with an extended heel or spur; the surface between these two usually
concave. Postpetiole with a strongly developed simple long digitiform ventral process . . 29
Either the subpetiolar process simple, a rectangular or subrectangular lobe without the above
configuration or with a feebly prominent acute angle or small tooth posteroventrally; if the
latter then the postpetiole with a short blunt or short tooth-like ventral process ... 35
29 Eyes relatively small, OI < 50 30
Eyes relatively large, OI 50 or more 32
30 Propodeal spines long, 0.40 or more in profile (in HW range 1.10-1.26), strongly divergent and
markedly elevated; in profile the spines distinctly longer than the maximum length of the
petiole. (Cameroun, Zaire) greggi Bolton
Propodeal spines short, < 0.25 in profile (in HW range 0.90-1.04), not strongly divergent nor
markedly elevated; in profile the spines distinctly shorter than the maximum length of the
petiole 31
31 Stout hairs on cephalic dorsum extremely dense, appearing as a bristly pelt in profile. A line
across the dorsum at the midlength of the eyes with many more than 10 hairs. Hairs on
dorsum of head more or less cylindrical, not spatulate; the hairs truncated apically, their sides
more or less straight and parallel. (Zaire) cestus (p. 360)
Stout hairs on cephalic dorsum sparse, not giving the appearance of a bristly pelt in profile. A
line across the dorsum at the midlength of the eyes with at most 10 hairs. Hairs on dorsum of
head conspicuously spatulate, broadly convex apically, their sides shallowly convex and
convergent basally. (Kenya) kenyensis(p. 358)
32 Most or all of stout hairs' on clypeus and dorsum of head increasing markedly in thickness from
base to apex, frequently 2-3 times broader at apex than at base. (Sierra Leone, Ghana,
Cameroun, Chad, Zaire) pygmaeus Andre
Most or all of stout hairs on clypeus and dorsum of head cylindrical or nearly so, not increasing
markedly in thickness from base to apex ; in some cases the hairs may broaden approximately
to their midlength and then continue at that width to their apices 33
33 Mesonotal and propodeal dorsa with very fine superficial low irregular weak wandering
rugulae, feeble or faded out in places but never evenly spaced nor regularly longitudinal.
Spaces between these fine rugulae densely strongly reticulate-puntate and dull. (Ghana,
Congo, Zaire) weissi (p. 358)
Mesonotal and propodeal dorsa with conspicuous strong broad longitudinal rugae which may
be parallel but which are never faded out in places. Spaces between the rugae weakly
superficially sculptured or unsculptured, the surfaces shining 34
34 Entire body exceptionally highly polished and very shiny. Longitudinal rugae on posterior half
of mesonotum and on propodeum very broad, subsulcate and parallel, without anastomoses
on the prodpodeum. (Nigeria) taylori(p. 364)
Dully shining, not obviously highly polished. Longitudinal rugae on posterior half of
mesonotum and on propodeum not subsulcate, not parallel, tending instead to diverge and
converge slightly along their lengths or to be weakly wavy; on the propodeum with
anastomoses. (Benin Republic) diffidlis Santschi
35 Hairs on dorsum of head exceptionally short, forming only a minute stubble on the surface.
Dorsum of head usually meeting occipital surface in a marked angle or edge, the one not
rounding evenly into the other 36
Hairs on dorsum of head conspicuous and quite dense, not represented only by a minute
stubble on the surface. Dorsum of head rounding into occipital surface 37
358 B. BOLTON
36 Tooth on mesokatepisternum large, long and acute, projecting anteriorly and usually clearly
visible in dorsal view, projecting beyond the margins of the mesonotum. (South Africa)
mi cans Mayr
Tooth on mesokatepisternum small and short, usually a mere denticle or acute angle,
sometimes not even as strong as this ; not visible in dorsal view. (Ethiopia, Somali Republic,
Kenya, Tanzania, Malawi, Zimbabwe, Mozambique, Angola, South West Africa, South
Africa) intrudens (F. Smith)
37 Larger species, HW > 1.10, PW > 0.90. (Zaire) bequaerti Forel
Smaller species, HW< 1.10, PW< 0.90 rife* . . 38
38 Pronotum laterally with a number of irregular rounded tuberculiform projections, without a
regular series of denticles although some of the projections appear to consist of 2 or more
denticles fused together. (South Africa) fricatidorsus Santschi
Pronotum laterally with a more or less regularly spaced series of denticles • 39
39 Dorsal surfaces of mesonotum and propodeum extremely finely and very densely more or less
evenly longitudinally rugulose, the rugulae so close together that the spaces between them are
wide enough for only 1-2 rows of punctures. (Cameroun) mckeyi Snelling
Dorsal surfaces of mesonotum and propodeum coarsely rugose, the rugae predominantly
longitudinal but with some strong cross-meshes, breaks or irregularities. The rugae widely
spaced so that the spaces between most of them accomodate many more than 2 rows of
punctures. (Ghana, Nigeria, Cameroun, Zaire, Sudan, Uganda, Tanzania, South Africa)
traegaordhi(p. 358)
Cataulacus lujae Forel sp. rev.
Cataulacus lujae Forel, 191 Ib: 311. Syntype workers, ZAIRE: Kasai, Kondue (Luja) (MHN, Geneva)
[examined]. [Wrongly synonymized with brevisetosus Forel by Bolton, 1974: 31.]
C. lujae var. gilviventris Forel should be included as a synonym oflujae, not of brevisetosus.
Cataulacus jeanneli Santschi sp. rev.
Cataulacus jeanneli Santschi, 1914a: 108, fig. 16. Holotype worker, KENYA: Gazi, 20 km S. of Mombasa, st.
no. 6, xi. 1911 (C. Alluaud & R. Jeannel) (NM, Basle) [examined]. [Wrongly synonymized with
brevisetosus by Bolton, 1974: 31.]
The names pygmaeus st. degener Santschi and janneli [sic] var. loveridgei Santschi should be
included in the synonymy of jeanneli, not of brevisetosus. The types of loveridgei still have not
been found; the holotype of brevisetosus has now been located in MHN, Geneva.
Cataulacus weissi Santschi
Cataulacus weissi Santschi, 1913: 310. Holotype worker, CONGO: Brazzaville, 1907 (A. Weiss) (NM, Basle)
[examined].
Cataulacus jeanneli var. aethiops Santschi, 1924: 220. Syntype workers, ZAIRE: Kidada-Kitobola,
14-25.ii.1922 (H. Schouteden) (MRAC, Tervuren) [examined]. Syn. n.
Cataulacus kenyensis Santschi stat. n.
Cataulacus jeanneli st. kenyensis Santschi, 1935: 272, figs 6a-c. Syntype workers, KENYA: Nairobi, st. 2,
1660 m, 1932-33 (C. Arambourg, P. Chappuis & R. Jeannel) (NM, Basle) [examined]. [Wrongly
synonymized with weissi by Bolton, 1974: 39.]
Cataulacus traegaordhi Santschi sp. rev.
Cataulacus traegaordhi Santschi, 19146: 24, fig. 3. Syntype workers, female, male, SOUTH AFRICA: Natal,
Zululand, Dukudu, 27.vii.1905 (/. Tragardh) (NM, Basle) [examined]. [Wrongly synonymized with
pygmaeus Andre by Bolton, 1974: 48.]
AFROTROPICAL MYRMICINE ANT GENERA 359
Of those names formerly included as synonyms under pygmaeus, the forms C. tragardhi [sic] var.
ugandensis Santschi, C. marleyi Forel (types in MHN, Geneva, not previously seen), and C.
pygmaeus subsp. suddensis Weber should now be included in the synonymy of traegaordhi, not of
pygmaeus.
Cataulacus centrums sp. n.
(Fig. 33)
HOLOTYPE WORKER. TL 3.9, HL 1.00, HW 0.92, CI 92, EL 0.47, OI 51, SL 0.48, SI 52, PW 0.70, AL 1.10.
With the head in full-face view the lateral margins of the head behind the eyes denticulate, terminating
posteriorly in a short tooth at the occipital corner. Occipital crest absent, the dorsum of the head rounding
evenly but narrowly into the occipital surface; the occipital margin itself unarmed except for a small tooth
situated close to the tooth at the corner. Eyes relatively large, OI > 50. Alitrunk with promesonotum both
longitudinally and transversely convex. In profile the highest point at about the midlength of the pronotum,
the remainder sloping evenly downwards posteriorly to the base of the propodeal spines. Anterior strongly
curved portion of pronotal dorsum with a number of minute peaks or tubercles from which hairs arise; such
peaks absent elsewhere on alitrunk. Tooth at base of mesokatepisternum developed. Propodeal spines in
profile with the basal third elevated at an angle of about 45°, the apical two-thirds back-curved. Metapleural
lobes low and rounded. With the alitrunk in dorsal view the pronotal corners denticulate and the lateral
margins of the pronotum armed with a series of 6-7 regularly spaced triangular denticles. Lateral margins of
mesonotum with a pair of small denticles whose bases are fused, situated at approximately the midlength.
Following the metanotal identation of the margin the sides of the propodeum are equipped with 2-3 small
tubercles. Propodeal spines in dorsal view broad and evenly divergent. Petiole in profile rising to a sharp
peak dorsally, behind which the surface slopes evenly downwards to the postpetiolar junction. Subpetiolar
process with a bluntly rounded anterior lobe and a weakly developed posteroventral tooth. Postpetiole in
profile with its dorsal and posterior surfaces tuberculate and its ventral process simple, short digitiform.
First gastral tergite not marginate laterally, conspicuously longer than broad. Dorsum of head finely and
evenly reticulate-rugulose, the recticular meshes of irregular size and the rugulae low and rounded.
Ground-sculpture in the meshes reduced to an inconspicuous vestigial superficial shagreening, without
punctulae. Pronotal dorsum similarly but somewhat more strongly sculptured, the reticulum breaking down
on the mesonotum so that the longitudinal component predominates and the cross-meshes are reduced or
incomplete. Propodeal dorsum more strongly and predominantly longitudinally rugose, irregular centrally.
Transverse rugae are present between the bases of the propodeal spines. Ground-sculpture of alitrunk
mostly as head but the mesonotum with some minute and virtually effaced punctulae. Petiole in dorsal view
longitudinally rugose, the sculpture converging posteriorly. Postpetiole dorsum irregularly rugulose. First
gastral tergite blanketed by fine dense reticulate-punctate sculpture, the basal quarter also with widely
spaced fine longitudinal costulae. Behind this level the tergite with scattered short longitudinal rugulae
which are very fine and irregular and formed by the alignment of the margins of adjacent punctures. First
gastral sternite reticulate-punctate. Sides of pronotum obliquely sulcate, the mesopleuron transversely
sulcate and the sides of the propodeum more or less vertically so behind the level of the spiracle. Sides of
petiole and postpetiole longitudinally sulcate-rugose. Discounting the long simple hairs which arise round
the eyes the entire dorsum of the head thickly clothed with stalked-suborbicular hairs, the stems of the hairs
long and fine and holding the suborbicular distal portions well clear of the surface of the head. Occipital
surface with a number of elongate narrowly clavate hairs. All remaining dorsal surfaces of body densely
clothed with moderately long stout cylindrical simple hairs which are truncated apically; those on the
alitrunk and petiole straight, those on the postpetiole and first gastral tergite weakly back-curved. Colour
uniform black, glossy ; the scapes, tibiae and tarsal segments dull yellow.
Holotype worker, Cameroun: Nkoemvon, 1979 (D. Jackson) (BMNH).
As indicated by the stalked-suborbicular cephalic hairs centrums belongs to the complex of
species centring on brevisetosus, and is most closely related to the smaller moloch. In the latter
species the simple pilosity of the alitrunk and gaster is very short and stubble-like, whereas in
centrums it is long and conspicuous (Figs 33, 34). The specialized cephalic hairs of moloch are
sparser than in centrums, have the basal stems of the hairs shorter and the apices less strongly
expanded. With the head in profile the specialized hairs immediately in front of the eye have the
basal stem longer than the swollen apex in centrums, shorter than the swollen apex in moloch. In
profile the propodeal spines of centrums have the basal third elevated and the apical two-thirds
360 B. BOLTON
recurved, a feature not seen in moloch where the spines are exceedingly feebly but evenly curved
along their length. Finally, the shape of the subpetiolar process differs in the two species, that of
moloch having the posteroventral angle more salient and the ventral surface more concave than
in centrums.
Cataulacus cestus sp. n.
HOLOTYPE WORKER. TL 4.0, HL 1.00, HW 0.99, CI 99, EL, 0.45, OI 45, SL 0.48, SI 48, PW 0.76, AL 1.10.
Sides of head behind eyes denticulate, terminating in a larger denticle at the occipital corner. Occipital
crest absent but the occipital surface shallowly concave above the foramen and meeting the dorsum in an
angle, the two surfaces not evenly rounded together. Occipital margin unarmed except for a denticle or short
tooth close to the one at the corner. Eyes relatively small, OI < 50. With the alitrunk in profile the dorsum
evenly shallowly convex between the more steeply sloped anterior portion of the pronotum and the base of
the propodeal spines. Pronotal and propodeal surfaces beset with small peaks or tubercles in profile, the
mesonotal dorsum also having such peaks but they are here more scattered and much lower, having the
appearance of minute irregularities in the outline. Mesokatepisternal tooth small. Metapleural lobes
rounded. Propodeal spines in profile short, more or less straight, only very slightly elevated. Alitrunk in
dorsal view with the pronotal corners denticulate, the lateral marginations of the pronotum behind the
corners with 6-7 sharp triangular denticles projecting laterally. Sides of mesonotum with 1-2 small denticles
and sides of propodeum also with 1-2, occurring on the convexity over the spiracle. Propodeal spines short
and broad, widely divergent. Petiole node in profile rising to an acute peak dorsally. The subpetiolar process
with a rounded and slightly prominent anteroventral lobe and a triangular projecting posteroventral tooth
or heel; the ventral surface between the two angles feebly concave. Postpetiole in profile high, its dorsal
surface with a number of conspicuous peaks or tubercles and its ventral process short-digitiform. Dorsum of
head irregularly reticulate-rugose, the meshes of varying size and the rugae low and rounded. Many of the
reticular meshes incomplete or with their walls broken. Ground-sculpture within the meshes a very fine
superficial shagreening or granular roughening of the surface, not reticulate-punctate. Dorsal alitrunk
irregularly reticulate-rugose everywhere, many of the rugular meshes incomplete or broken and very
irregular in shape. Ground-sculpture finely reticulate-punctate to densely shagreened. Petiole node in dorsal
view strongly longitudinally rugose, the rugae converging posteriorly. Postpetiole irregularly rugulose and
finely densely punctulate. First gastral tergite coarsely and densely reticulate-punctate everywhere, the
whole surface also loosely covered with anastomosing fine irregular superficial rugulae which are strongest
basally and fade out apically on the sclerite. First gastral sternite similarly sculptured. Entire dorsum of head
covered with a dense pelt of short straight erect bristly blunt hairs which are cylindrical to subcylindrical in
shape. All remaining dorsal surfaces of body with similar dense bristly pilosity. Colour uniform black; the
scapes, tibiae and tarsi dull yellow.
PARATYPE WORKERS. TL 4.0-^.1, HL 0.98-1.02, HW 0.98-1.02, CI 98-100, EL 0.45-0.48, OI 46-^7, SL
0.48-0.50, SI 49-51, PW 0.76-0.86, AL 1.08-1.16 (4 measured).
As holotype but in some the gastral rugulae are less strongly developed and in one the gastral rugulae are
effaced. The ventral surface of the subpetiolar process may be more strongly concave than is the case with
the holotype.
Holotype worker, Zaire (B. Congo on data label): Ituri For., Beni-Irumu, ii.1948, no. 2122 (N. A. Weber)
(MCZ, Cambridge).
Paratypes. 1 worker with same data as holotype; 1 worker with same data as holotype but no. 2120; 2
workers with same data as holotype but no. 21 19. (MCZ, Cambridge; BMNH).
Cataulacus jacksoni sp. n.
HOLOTYPE WORKER. TL 3.5, HL 0.98, HW 0.94, CI 96, EL 0.46, OI 50, SL 0.49, SI 52, PW 0.68, AL 0.98
(cephalic measurements approximate as head crushed).
With the head in full-face view the sides behind the eyes minutely denticulate. Occipital crest absent, the
dorsum rounding into the occipital margin. Head of holotype crushed behind level of eyes and the surface
fractured; the fracture also running forward on the head along the inner margin of the right eye to the
clypeus. With the alitrunk in profile the dorsal outline rising steeply to about the midlength of the
pronotum. Behind this the remainder of the dorsum evenly shallowly convex to the bases of the propodeal
spines, the outline not interrupted by superficial peaks or tubercles. Mesokatepisternal tooth small and
broadly triangular. Propodeal spines in profile strongly downcurved along their length. Metapleural lobes
AFROTROPICAL MYRMICINE ANT GENERA 361
very small. With the alitrunk in dorsal view the pronotal corners angular, the angle slightly projecting. Sides
of pronotum behind this not marginate, without a regular series of laterally projecting denticles. Instead the
sides with only a blunt tubercle at the point of junction of the pronotum and mesonotum and with one or
two minute irregularities, too low, small and blunt to be called tubercles or denticles, situated behind the
corner. Sides of mesonotum and propodeum unarmed and immarginate, the latter with a low salient welt at
the site of the spiracle. Propodeal spines in dorsal view curved, bowed outwards along their length. Petiole
in profile blunt above, not rising to a sharp peak. Subpetiolar process with the anteroventral angle rounded,
the posteroventral angle acute and slightly projecting. Postpetiole in profile very high and narrow, with a
flat anterior face and a long simple ventral process. In dorsal view the postpetiole with the sides converging
dorsally so that the node narrows from base to apex. Dorsum of head to level of posterior margins of eyes
finely longitudinally rugose, behind this level the head with very heavy broad strong sulci. Ventral surface of
head longitudinally sulcate. Dorsal alitrunk regularly strongly longitudinally sulcate except for the area
between the bases of the propodeal spines where the sulci are arched-transverse. Propodeal declivity
transversely sulcate. Coxae, femora and tibiae of legs all longitudinally sulcate. Anterior face of petiole node
transversely sulcate, the dorsum with U-shaped sulci. Upper half of anterior face of postpetiole vertically
sulcate. Sides of alitrunk diagonally sulcate from anteroventral to posterodorsal on each sclerite except on
the mesokatepisternum where they run from posteroventral to anterodorsal. First gastral tergite and first
sternite covered with strong parallel longitudinal sulci throughout. Dorsum of head with abundant
stalked-suborbicular hairs which have slender basal stems. Remainder of dorsal surfaces of body with sparse
fine curved hairs which are very feebly clavate apically. Colour uniform black but scapes, anterior tibiae and
tarsi, and tarsi of middle and hind legs dull yellow.
Holotype worker, Cameroun: Nkoemvon, 1980 (D. Jackson) (BMNH).
The characteristic strong sulcate sculpture of this species, coupled with its possession of
immarginate and unarmed lateral pronotal margins, stalked-suborbicular cephalic hairs, and
propodeal spines which are bowed outwards in dorsal view and downcurved in profile, make
jacksoni very easily recognisable.
Cataulacus moloch sp. n.
(Fig. 34)
HOLOTYPE WORKER. TL 3.4, HL 0.90, HW 0.80, CI 89, EL 0.43, OI 54, SL 0.42, SI 53, PW 0.60, AL 0.94.
With the head in full-face view the sides behind the eyes denticulate, ending in a low triangular tooth at
the occipital corner. Occipital margin with a small tooth close to the one at the corner but otherwise
unarmed; the occipital crest absent, the dorsum rounding evenly into the occipital surface. Eyes relatively
large, OI > 50. Alitrunk with promesonotum both longitudinally and transversely convex. In profile the
alitrunk with its highest point at about the midlength of the pronotum, behind which the dorsum is evenly
shallowly convex to the base of the propodeal spines. The steeply sloping anterior portion of the pronotal
dorsum with a number of minute peaks or tubercles from which hairs arise, such peaks absent elsewhere on
the dorsum. Tooth on mesokatepisternum moderately developed, distinct. Propodeal spines in profile
scarcely elevated and almost straight, only very feeble downcurved along their length; not having the basal
portions elevated and the distal portions recurved. With the alitrunk in dorsal view the pronotal corners
with prominent acute dentiform angles. Pronotal margin behind the corners with 5-6 triangular, laterally
projecting denticles which are quite evenly spaced. Sides of mesonotum with two small denticles, the sides of
the propodeum convex over the site of the spiracles, with one or two minute tubercles. Propodeal spines
broad in dorsal view and evenly divergent. Petiole in profile rising to an acute peak dorsally. Subpetiolar
process complex, with a blunt and strongly prominent anteroventral angle and a tooth-like projecting
posteroventral angle, the two separated by a conspicuously concave ventral margin. Postpetiole in profile
with the dorsal and posterior surfaces distinctly denticulate, the subpostpetiolar process elongate-digitiform.
Dorsum of head irregularly reticulate-rugulose, the reticular meshes of varying size and the rugulae
themselves low and rounded. Ground-sculpture within the meshes reduced to an inconspicuous vetigial
shagreening, without punctures. Pronotal dorsum similarly but more strongly sculptured, with a few low
broad transverse rugae anteriorly but with the longitudinal component predominating behind this. On the
mesonotum and propodeum the longitudinal component of the sculpture predominates, the rugae being
broader and more strongly developed ; many of the cross-meshes are feeble or incomplete. Rugae between
bases of propodeal spines transverse. Petiole in dorsal view regularly longitudinally rugose, the rugae
converging posteriorly; the postpetiole irregularly rugose. Ground-sculpture of alitrunk as on head. First
gastral tergite blanketed by dense reticulate-punctation, without strong basigastral rugulae but here and
362
B. BOLTON
r- m
-m
r-m
r-m
r-m
2r
Rs*m J Rs
J Rs
VM
Figs 35-43 Semi-diagrammatic representation of principal venation development on forewing of Messor
and Aphaenogaster. For explanation see text, pp. 339-340.
AFROTROPICAL MYRMICINE ANT GENERA 363
there with feeble short rugulae formed by the alignment of margins of adjacent punctures. First gastral
sternite reticulate-punctate. Sides of pronotum transversely sulcate. Discounting the long hairs which arise
around the eyes the dorsum of the head with numerous stalked-suborbicular hairs, the basal stems of which
are quite short. Occipital surface with longer hairs which increase in thickness from base to apex. All
remaining dorsal surfaces of body with numerous short stout blunt hairs. On the alitrunk some of these
hairs are slightly thicker apically than basally, these hairs straight everywhere except on the base of the first
gastral tergite where they are slighly back-curved. Colour uniform black; the scapes, tibiae and tarsi dull
yellow.
PARATYPE WORKERS. TL 2.8-3.2, HL 0.74-0.86, HW 0.68-0.72, CI 86-92, EL 0.39-0.42, OI 54-57, SL
0.38-0.40, SI 53-56, PW 0.50-0.60, AL 0.78-0.92 (3 measured).
As holotype but averaging slightly smaller.
Holotype worker, Ghana: Pankese, 24.ix.1968 (C. A. Collingwood) (BMNH).
Paratype. Ghana: 1 worker with same data as holotype. Nigeria: 2 workers, Onipe, CRIN, 1 l.vi.1975, tree
47-16 (A63.1), black pod project (B. Taylor] (BMNH).
C. moloch is closest related to centrums, the differences between them are noted under the latter
name.
Cataulacus satrap sp. n.
HOLOTYPE WORKER. TL 3.5, HL 0.87, HW 0.82, CI 94, EL 0.44, OI 53, SL 0.40, SI 49, PW 0.56, AL 0.96.
With the head in full-face view the sides behind the eyes minutely denticulate, the denticles partially
concealed by the thickened short hairs which project above them; the row of denticles ends in a small tooth
at the occipital corner. Occipital crest absent, the dorsum of the head rounding into the occipital surface.
Occipital margin unarmed except for a small tooth close to the one at the corner. Eyes relatively large,
OI > 50. In profile the anterior outline of the pronotal dorsum sloping steeply, the surface equipped with a
number of low peaks or tubercles. Behind this the remainder of the alitrunk shallowly but evenly convex,
sloping down posteriorly to the base of the propodeal spines. Mesokatepisternal tooth prominent,
moderately well developed. Metapleural lobes low and rounded. Propodeal spines in profile straight, only
slightly elevated. With the alitrunk in dorsal view the pronotal corners angular and projecting. Sides of
pronotum behind the corners only weakly marginate and with a series of 4-5 projecting denticles, all of
which are small and widely spaced. In the holotype the right pronotal margin with 5, the left with 4 denticles.
On both sides the posteriormost denticle the largest, the anteriormost distinctly smaller; the 2-3 between
them minute and inconspicuous. Sides of mesonotum and propodeum without differentiated denticles.
Propodeal spines in dorsal view broad and feebly divergent. Petiole in profile rising to an acute peak above.
Subpetiolar process simple, with a bluntly rounded anteroventral angle and an acute, weakly projecting
posteroventral angle, the two separated by a flat ventral surface. Postpetiole dome-like and high in profile,
with two feebly developed peaks dorsally; the subpostpetiolar process short-digitiform and blunt. Dorsum
of head irregularly reticulate-rugulose, the reticular meshes of uneven size and irregular shape, the rugulae
low and rounded. Ground-sculpture of the rugular meshes a fine dense reticulate-puncturation. Dorsal
alitrunk densely covered in fine rugulae which are low and rounded, reticulate in places but predominantly
longitudinal behind the pronotum. Entire dorsum of alitrunk also blanketed by a fine dense and very
conspicuous reticulate-punctate ground-sculpture. Petiole and postpetiole with dense reticulate-punctate
sculpture, the former also with longitudinal rugae in dorsal view, the latter only with a few vestigial irregular
rugulae. First gastral tergite strongly and densely reticulate-punctate everywhere. Dorsum of head with
numerous distinctive stalked-suborbicular hairs, those situated anteriorly on the dorsum more strongly
expanded apically than those situated behind the level of the eyes. All remaining dorsal surfaces of body with
many very short thick blunt hairs. Colour uniform black, dull; the scapes, tibiae and tarsi dull yellowish
brown.
PARATYPE WORKER. TL 3.4, HL 0.88, HW 0.80, CI 91, EL 0.43, OI 54, SL 0.40, SI 50, PW 0.57, AL 0.96.
As holotype but propodeal spines slightly less divergent and the subpetiolar process with the
anteroventral and posteroventral angles separated by a feebly concave ventral surface. On the pronotal
margins the anteriormost denticle behind the corner is no larger than those following it (except for the last in
the row, which is the largest); and the left side of the pronotum with 5 denticles, the right side with 4.
Holotype worker, Cameroun: Nkoemvon, 1970, M12 (D. Jackson) (BMNH).
Paratype. 1 worker with same data as holotype (BMNH).
Related to vorticus which it resembles closely, satrap is immediately separated by its possession of
denticles on the lateral pronotal margins.
364 B. BOLTON
Cataulacus taylori sp. n.
HOLOTYPE WORKER. TL 3.2, HL 0.82, HW 0.76, CI 93, EL 0.42, OI 55, SL 0.44, SI 58, PW 0.60, AL 0.90.
With the head in full-face view the sides behind the eyes denticulate, ending in a tooth at the occipital
corner. Occipital crest absent, the dorsum rounding evenly into the occiput; the occipital margin unarmed
except for a smaller tooth close to the one at the corner. Eyes relatively large, OI > 50. With the alitrunk in
profile the highest point of the dorsum at about the midlength of the pronotum. In front of this the dorsum
slopes down to the cervical shield and a few scattered minute peaks occur on the outline. Behind the highest
point the dorsum is shallowly convex and slopes evenly downwards towards the bases of the propodeal
spines. Mesokatepisternal tooth developed. Metapleural lobes low and rounded. Propodeal spines in profile
narrow, slightly downcurved along their length. Alitrunk in dorsal view with the pronotal corners
denticulate, the lateral margins of the pronotum with 6-7 projecting triangular denticles. Sides of
mesonotum and propodeum each with a single projecting denticle, the latter also with the sides convex at
the site of the spiracle. Propodeal spines narrow and evenly divergent in dorsal view. Petiole in profile rising
to a sharp peak above. Subpetiolar process complex, with a narrow rounded projecting blunt anteroventral
angle and a spur-like posteroventral angle, the ventral surface between the two angles strongly concave.
Postpetiole node with dorsal surface denticulate, the ventral process narrow and digitiform. Dorsum of head
feebly reticulate-rugulose, the rugulae very weak, fine, low and rounded, the reticular meshes mostly
incomplete and irregular in shape and size. Ground-sculpture in the meshes almost completely effaced, the
surface glossy. Dorsal alitrunk predominantly longitudinally rugose, with some anastomoses on the
pronotum but behind this the rugae straight and parallel, quite broad and without cross-meshes. Spaces
between the rugae glossy and almost smooth, with only the faintest vestiges of ground-sculpture. Rugae on
declivity between bases of spines transverse. Petiole and postpetiole longitudinally rugose, the rugae
converging posteriorly. First gastral tergite shiny, with superficial fine reticulate-puntulate sculpture
everywhere and with a weak pattern of very fine longitudinal irregular rugulae. Stronger longitudinal
rugulae present on the basal one-fifth of the tergite. First gastral sternite similarly but even more delicately
sculptured. Dorsum of head with numerous short stout straight cylindrical hairs which are blunt apically.
All remaining dorsal surfaces of body with similar pilosity, the longest hairs occurring on the base of the first
gastral tergite where they are slightly recurved. Colour uniform glossy jet black; the scapes, tibiae and tarsi
dull yellow.
PARATYPE WORKER. TL 3.5, HL 0.88, HW 0.81, CI 92, EL 0.45, OI 56, SL 0.46, SI 57, PW 0.67, AL 0.96.
As holotype but slightly larger, its subpostpetiolar process shorter and broader than in the holotype. The
rugae on the dorsal alitrunk not running straight back as in the holotype but slightly skewed to the left
posteriorly.
Holotype worker, Nigeria: Gambari, CRIN, 24. v. 1976, black pod project (B. Taylor) (BMNH).
Paratype. Nigeria: 1 worker, Onipe, CRIN, 25.vii.1975, black pod project (B. Taylor) (BMNH).
Appendix
The current genus-level synonymy of Aphaenogaster is as follows.
APHAENOGASTER Mayr
Aphaenogaster Mayr, 1853: 107. Type-species: Aphaenogaster sardoa Mayr, 1853: 107, by subsequent
designation of Bingham, 1903: 270.
Deromyrma Forel, 1913c: 49 [as subgenus of Ischnomyrmex Mayr]. Type-species: Aphaenogaster
(Ischnomyrmex) swammerdami Forel, 18866: cvi, by monotypy. [Synonymy by Brown, 1973: 180.]
Planimyrma Viehmeyer, 1914: 604 [as subgenus of Aphaenogaster}. Type-species : Stenamma (Ischnomyrmex)
loriai Emery, 1897: 563, by original designation. [Synonymy by Brown, 1973: 184.]
Attomyrma Emery, 1915: 70 [as subgenus of Aphaenogaster}. Type-species: Formica subterranea Latreille,
1 798 : 49, by original designation. [Synonymy by Brown, 1 973 : 1 78.]
Novomessor Emery, 1915: 73. Type-species: Aphaenogaster (Ischnomyrmex) cockerelli Andre, 1893: 150, by
original designation. [Synonymy by Brown, 1974: 47.]
Nystalomyrma Wheeler, 1916: 215 [as subgenus of Aphaenogaster}. Type-species: Myrmica longiceps F.
Smith, 1858: 128, by original designation. [Synonymy by Brown, 1973: 183.]
Brunella Forel, 1917: 234. Type-species: Aphoenogaster [sic] belli Forel, 1895: 248, by original designation.
Syn. n.
AFROTROPICAL MYRMICINE ANT GENERA 365
Acknowledgements
My sincere thanks go to the following for the loan of types and other material during the course
of this study.
Dr Claude Besuchet (MHN, Geneva); Mrs Cathy A. Car (NM, Bulawayo); Dr Jean Decelle
(MRAC, Tervuren); Mrs Marjorie Favreau (AMNH, New York); Dr Max Fischer (NM,
Vienna); Dr F. Koch (MNHU, Berlin); Prof Egidio Mellini (IE, Bologna); Mr Alfred Newton Jr
(MCZ, Cambridge); Dr Roberto Poggi (MCSN, Genoa); Dr A. G. Radchenko (ZM, Kiev); Miss
Helen Rae and Dr A. J. Prins (SAM, Cape Town); Dr David R. Smith (USNM, Washington); Dr
Cesare Baroni Urbani (NM, Basle); Mme Janine C. Weulersse (MNHN, Paris).
Finally my thanks to Ms Dorothy Jackson of Oxford University for collecting Cataulacus
material for me at my request, and to Mr David Morgan for lettering Figs 35-43.
References
Andre, E. 1893. Description de quatre especes nouvelles de fourmis d'Amerique. Revue Ent. 12: 148-152.
Arnold, G. 1916. A monograph of the Formicidae of South Africa, part 2. Ann. S. Afr. Mus. 14: 159-270, figs.
— 1920. A monograph of the Formicidae of South Africa, part 4. Ann. S. Afr. Mus. 14: 403-578.
— 1926. A monograph of the Formicidae of South Africa, appendix. Ann. S. Afr. Mus. 23: 191-295, figs.
— 1948. New species of African Hymenoptera, no. 8. Occ. Pap. natn. Mus. Sth. Rhod. 14: 213-250, 23 figs.
Arnoldi, K. V. 1977. Revision of the harvester ants of the genus Messor in the fauna of the U.S.S.R. [In
Russian.] Zoo/. Zh. 56: 1637-1648. 3 figs.
Ben-Dov, Y. 1978. Andaspis formicarum n. sp. associated with a species of Melissotarsus in South Africa.
Insectes soc. 25: 315-321, 1 fig.
Bernard, F. 1950. Contribution a 1'etude de 1'Air (mission L. Chopard et A. Villiers). Mem. Inst.fr. Afr. noire
10:284-294.
— 1956. Revision des fourmis palearctiques du genre Cardiocondyla Emery. Bull. Soc. Hist. nat. Afr. N. 47:
299-306, 6 figs.
1968. Faune de I' Europe et du Bassin Mediterraneen 3. Les fourmis d'Europe occidentale et
septentrionale. 411 pp. Paris.
Bingham, C. T. 1903. Fauna of British India, including Ceylon and Burma. Hymenoptera 2, Ants and Cuckoo
Wasps. 506 pp., 161 figs, 1 pi. London.
Bolton, B. 1972. Two new species of the ant genus Epitritus from Ghana, with a key to the world species.
Entomologist's mon. Mag. 107: 205-208, 4 figs.
— 1973. The ant genera of West Africa: a synonymic synopsis with keys. Bull. Br. Mus. nat. Hist. (Ent.) 27:
317-368, 1 fig.
— 1974. A revision of the palaeotropical arboreal ant genus Cataulacus F. Smith. Bull. Br. Mus. nat. Hist.
(Ent.) 30: 1-105, 41 figs.
- 1976. The ant tribe Tetramoriini. Constituent genera, review of smaller genera and revision of
Triglyphothrix Forel. Bull. Br. Mus. nat. Hist. (Ent.) 34: 281-379, 73 figs.
- 1980. The ant tribe Tetramoriini. The genus Tetramorium Mayr in the Ethiopian zoogeographical
region. Bull. Br. Mus. nat. Hist. (Ent.) 40: 193-384, 145 figs.
— 198 la. A revision of the ant genera Meranoplus F. Smith, Dicroaspis Emery and Calyptomyrmex Emery
in the Ethiopian zoogeographical region. Bull. Br. Mus. nat. Hist. (Ent.) 42: 43-81, 44 figs.
1981ft. A revision of six minor genera of Myrmicinae in the Ethiopian zoogeographical region. Bull. Br.
Mus. nat. Hist. (Ent.) 43: 245-307, 43 fig.
Brown, W. L. 1949. Synonymic and other notes on Formicidae. Psyche, Camb. 56:
- 1952. Revision of the ant genus Serrastruma. Bull. Mus. comp. Zool. Harv. 107: 67-86.
- 1953. Revisionary studies in the ant tribe Dacetini. Am. Midi. Nat. 50: 1-137, 10 figs, 3 pis.
- 1954. The ant genus Strumigenys F. Smith in the Ethiopian and Malagasy regions. Bull. Mus. comp.
Zool. 112:1-34, 1 fig.
- 1955. The ant Leptothorax muscorum (Nylander) in North America. Ent. News 66: 43-50.
- 1964. Rhoptromyrmex, revision of and key to species. Pilot Reg. Zool. cards 11-19, figs.
- 1971. Characters and synonymies among the genera of ants, part 4. Some genera of subfamily
Myrmicinae. Brevioa no. 365: 1-5.
- 1973. A comparison of the Hylean and Congo-West African rain forest ant faunas, pp. 161-185. In
Meggers, B. J., Ayensu, E. S., Duckworth, W. D., Tropical forest ecosystems in Africa and South America, a
review. Washington, D.C.
366 B. BOLTON
— 1974. Novomessor manni a synonym of Aphaenogaster ensifera. Ent. News 85: 45-53.
Buschinger, A. 1979. Functional monogyny in the American guest ant Formicoxenus hirticornis (Emery)
( = Leptothorax hirticornis). Insectes soc. 26: 61-68.
Collingwood, C. A. 1978. A provisional list of Iberian Formicidae with a key to the worker caste. Eos, Madr.
52:65-95.
— 1979. The Formicidae of Fennoscandia and Denmark. Fauna entomologica scand. 8: 1-174, 268 figs.
Creighton, W. S. 1950. The Ants of North America. Bull. Mus. comp. Zool. 104: 1-585, 57 pis.
Delage-Darchen, B. 1972. Une fourmi de Cote d'lvoire: Melissotarsus titubans Del., n. sp. Insectes soc. 19:
213-226, 10 figs.
Donisthorpe, H. St J. K. 1946. New species of ants from the island of Mauritius. Ann. Mag. nat. Hist. (1 1) 12:
776-782.
— 1947. A third instalment of the Ross collection of ants from New Guinea. Ann. Mag. nat. Hist. (1 1) 14:
589-604, 1 fig.
Emery, C. 1869. Enumerazione dei formicidi che rinvengonsi nei contorni di Napoli. Annali Accad. Aspir.
Nat. Napoli (2) 2:1-26,1 pi.
- 1877. Catalogo delle formiche esistenti nelle collezioni del Museo Civico di Geneva. Parte prima.
Formiche provenienti dal viaggio dei signori Antinori, Beccari e Issel nel Mar Rosso e nel paese dei
Bogos. Annali Mus. civ. Stor. nat. Genova 9: 363-381, figs.
- 1884. Materiali per lo studio della fauna Tunisia, raccolti da G. e L. Doria. 3. Rassegna delle formiche
della Tunisia. Annali Mus. civ. Stor. nat. Genova 21 : 373-386.
— 1891. Voyage de M Ch. Alluaud dans le territoire d'Assinie (Afrique occidentale) en juillet et aout 1886.
Annls Soc. ent. Fr. 60: 553-574, 1 pi.
- 1895a. Esplorazione del Giuba e dei suoi affluenti compiuta dal Cap. V. Bottego durante gli anni
1892-93 sotto gli auspicii della Societa Geografica Italiana. Risultati zoologici. Annali Mus. civ. Stor. nat.
Genot?fl(2)15[35]:177-184.
- 1895fe. Beitrage zur Kenntniss der nordamerikanischen Ameisenfauna. Zool. Jb. (Syst.) 8: 257-360,
Ipl.
— 1896. Studi sulle formiche della fauna neotropica. Boll. Soc. ent. ital. 28: 33-107, pi. 1.
- 1897. Viaggio di Lamberto Loria nella Papuasia orientale, 8. Formiche raccolte nella Nuova Guinea
dal Dott Lamberto Loria. Annali Mus. civ. Stor. nat. Genova 38: 546-593, 1 pi.
— 1908. Beitrage zur Monographic der Formiciden des palaarktischen Faunengebietes. Dt. ent. Z. 1908:
437-465, 13 figs.
- 1915. Definizione de genere Aphaenogaster e partizione di esso in sottogeneri. Parapheidole e
Novomessor nn. gg. Re. Sess. Accad. Sci. 1st. Bologna 19: 67-75.
- 1917. Questions de nomenclature et synonymies relatives a quelques genres et especes de formicides.
Bull. Soc. ent. Fr. 1917: 94-97.
- 1921. In Wytsman, P. A. G. Genera Insect. Hym. fam. Formicidae subfam. Myrmicinae. fasc. 174a:
1-94.
- 1922a. In Wytsman, P. A. G. Genera Insect. Hym. fam. Formicidae subfam. Myrmicinae. fasc.
174b-174c: 95-397.
1922/>. Messor barbarus (L.). Appunti di sinonimia e di sistematica. Boll. Soc. ent. ital. 54: 92-99.
Enzmann, J. 1947. New forms of Aphaenogaster and Novomessor. Jl N.Y. ent. Soc. 55: 147-152, pi. 8.
Fabricius, J. C. 1793. Entomologia systematica emendata et aucta 2:519 pp. Hafniae.
Forel, A. 1881. Die Ameisen der Antille St Thomas. Mitt, munch, ent. Ver. 5: 1-16.
— 1886a. Especes nouvelles de fourmis Americaines. Cr. Soc. ent. Belg. 30: xxxviii-xlix.
- 1886ft. Diagnoses provisoires de quelques especes nouvelles de fourmis de Madagascar, recoltees par
M Grandidier. Cr. Soc. ent. Belg. 30: ci-cvii.
— 1890a. Fourmis de Tunisie et de 1'Algerie orientale. Cr. Soc. ent. Belg. 34: Ixi-lxxvi.
- 1890ft. Aenictus-Typhlatta decouverte de M Wroughton. Nouveaux genres de formicides. Cr. Soc. ent.
Belg. 34: cii-cxiii.
- 1891. In Grandidier, A., Histoire physique, naturelle, et politique de Madagascar 20. Histoire naturelle
des Hymenopteres, part 2 Les Formicides : 237 pp. 7 pis. Paris.
- 1892. Die Ameisenfauna Bulgariens (Nebst biologischen Beobachtungen). Verh. zool. hot. Ges. Wien
42: 305-3 18, pi. 5.
- 1894. Abessinische und andere afrikanische Ameisen, gesammelt von Herrn Ingenieur Alfred Ilg, von
Herrn Dr Liengme, von Herrn Pfarrer Missionar P. Berthoud, Herrn Dr Arth. Miiller, etc. Mitt, schweiz.
ent. Ges. 9: 64-100.
— 1895. Nouvelles fourmis de 1'Imerina oriental (Moramanga etc.). Annls Soc. ent. Belg. 39: 243-251.
1899. Fauna Hawaii. 1. Part 1, Hymenoptera Aculeata, Heterogyna (Formicidae): 1 16-122. Honolulu.
AFROTROPICAL MYRMICINE ANT GENERA 367
— 1905. Miscellanea myrmecologiques 2. Annls Soc. ent. Fr. 49: 155-185.
- 1907. Fourmis d'Ethiopie. Revue Ent. 26: 129-144.
- 1910a. In Schultze, L. S. Zoologische und anthropologische Ergebnisse einer Forschungsreise im
westlichen und zentralen Sudafrika. Formicidae: 1-30.
- 19106. Note sur quelques fourmis d'Afrique. Annls Soc. ent. Belg. 54: 421-458.
- 1910c. Ameisen aus der Kolonie Erythraa. Gesammelt von Prof Dr K. Escherich (nebst einigen in
West-Abessinien von Herrn A. Ilg gesammelten Ameisen). Zoo/. Jb. (Syst.) 29: 243-274.
- 191 la. Die Ameisen des K. Zoologischen Museums in Miinchen. Sber. buyer. Akad. Wiss. 1911:
249-303.
- 19116. Ameisen des Herrn Prof v. Ihering aus Brasilien (Sao Paulo u. s. w.) nebst einigen anderen aus
Sudamerika und Afrika. Dt. ent. Z. 1911 : 285-312.
- 1912. Descriptions provisoires de genres, sous-genres et especes de formicides des Indes orientales.
Revue suisse Zoo/. 20: 761-774.
- 1913a. Formicides du Congo Beige recoltes par MM Bequaert, Luja etc. Revue Zoo/. Bot. Afr. 2:
306-351.
- 19136. Fourmis de Rhodesia etc., recoltees par M G. Arnold, le Dr H. Brauns et K. Fikendey. Annls
Soc. ent. Belg. 57: 108-147.
— 191 3c. Wissenschaftliche Ergebnisse einer Forschungsreise nach Ostindien, ausgefiihrt im Auftrage der
Kgl. Preuss. Akademie der Wissenschaften zu Berlin von H v. Buttel-Reepen. 2. Ameisen aus Sumatra,
Java, Malacca und Ceylon. Zoo/. Jb. (Syst.) 36: 1-148, figs.
- 1914. Formicides d'Afrique et d'Amerique nouveaux ou peu connus. Bull. Soc. vaud. Sci. nat. 50:
211-288.
- 1915. Formicides d'Afrique et d'Amerique nouveaux ou peu connus 2. Bull. Soc. vaud. Sci. nat. 50:
335-364.
— 1916. Fourmis du Congo et d'autres provenances recoltees par MM Hermann Kohl, Luja, Mayne etc.
Revue suisse Zoo/. 24: 397-460.
- 1917. Cadre synoptique actuel de la faune universelle des fourmis. Bull. Soc. vaud. Sci. nat. 51 : 229-253.
1918. Quelques fourmis de Madagascar recoltees par le Dr Friederichs et quelques remarques sur
d'autres fourmis. Bull. Soc. vaud. Sci. nat. 52: 151-156.
Hamann, H. H. F. & Klemm, W. 1967. Ergebnisse der zoologischen Nubien-Expedition 1962. Annln naturh.
Mus. Wien 70: 41 1^21, 1 fig.
Holldobler, B., Stanton, R. & Engel, H. 1976. A new exocrine gland in Novomessor and its possible
significance as a taxonomic character. Psyche, Camb. 83: 32^41, 5 figs.
Karavaiev, V. 1909. Nachtrag zu meinen "Ameisen aus Transcaspien und Turkestan." Ent. Obozr. 9:
268-272, 3 figs.
— 191 1. Ameisen aus Aegypten und dem Sudan. Ent. Obozr. 11 : 1-12, 3 figs.
Kempf, W. W. 1959. A synopsis of the New World species belonging to the Nesomyrmex-group of the ant
genus Leptothorax Mayr. Studia ent. 2: 391-432, 31 figs.
- 1972. Catalogo abreviado das formigas da regiao Neotropica. Studia ent. 15: 3-344.
Kugler, C. 1978. Pygidial glands in the myrmicine ants. Insectes soc. 25: 267-274.
Kutter, H. 1973. Zur Taxonomie der Gattung Chalepoxenus. Mitt, schweiz. ent. Ges. 46: 269-280, 8 figs.
- 1977. Insecta Helvetica Fauna 6 Hymenoptera Formicidae: 298 pp., 627 figs. Zurich.
Latreille, P. A. 1798. Essai sur I'histoire des fourmis de la France. 50 pp. Brive.
Levieux, J. 1979. La nutrition des fourmis granivores. Cycle d'activite et regime alimentaire de Messor galla
et de Messor ( = Cratomyrmex) regalis en saison des pluies fluctuations annuelles. Insectes soc. 26:
279-294.
Levieux, J. & Diomande, T. 1978. La nutrition des fourmis granivores. Cycle d'activite et regime alimentaire
de Messor galla et de Messor ( = Cratomyrmex) regalis. Insectes soc. 25: 127-139.
Linnaeus, C. 1767. Systema Naturae ed. 13. Tom. 1 pars 2: 533-1327. Vindobonae.
Mayr, G. 1853. Beitrage zur Kenntniss der Ameisen. Verh. zool.-bot. Ver. Wien 3: 101-1 14.
- 1855. Formicina austriaca. Beschreibung der bisher im osterreichischen Kaiserstaate aufgefunden
Ameisen nebst Hinzufiigung jener in Deutschland, in der Schweiz und in Italien vorkommenden Arten.
Verb, zool.-bot. Ver. Wien 5: 273-478, 1 pi.
— 1861. Die europaischen Formiciden. 80 pp., 1 pi. Wien.
- 1862. Myrmecologische Studien. Verh. zoo/, hot. Ges. Wien 12: 649-776, pi. 19.
- 1866. Myrmecologische Beitrage. Sber. Akad. Wiss. Wien 53: 484-517, 1 pi.
1886. Die Formiciden der Vereinigten Staaten von Nord-amerika. Verh. zoo/. 6o/. Ges. Wien 36:
419^64.
— 1887. Sudamerikanische Formiciden. Verh. zoo/. 6or. Ges. Wien 37: 510-632.
368 B. BOLTON
— 1895. Afrikanische Formiciden. Annln nat. Mus. Wien 10: 124-154, 3 figs.
- 1901. Siidafrikanische Formiciden gesammelt von Dr Hans Brauns. Annln nat. Mus. Wien 16: 1-30, 4
figs.
1904. In Jagerskiold, A. L. K. E., Results of the Swedish Zoological Expedition to Egypt and the White
Nile (Dec. 1900-July 1901). Part 1, section 6 Formiciden: 1 1 pp. Uppsala.
Menozzi, C. 1924. Res mutinensis. Formicidae. Atti Soc. Nat. Mat. (6) 8: 22-47, 3 figs.
— 1930. Formiche della Somalia Italiana meridionale. Mem. Soc. ent. ital. 9: 76-130, 4 figs, 3 pis.
Milne-Edwards, A. 1879. Description de quelques Crustaces nouveaux. Bull. Soc. philomath. Paris (7) 3:
103-110.
Nylander, W. 1856. Synopsis des formicides de France et d'Algerie. Annls Sci. nat. (Zool.) (4) 5: 51-109.
Reiskind, J. 1965. A revision of the ant tribe Cardiocondylini 1. The genus Prosopidris Wheeler. Psyche,
Camb. 72: 79-86, 8 figs.
Roger, J. 1863. Verzeichniss der Formiciden-Gattungen und Arten. Berl. ent. Z. 7 Supplement: 1-65.
Ruzsky, M. D. 1904. The ants of the Archangel District. Zap. imp. russk. geogr. Obshch. obshch. Geogr. 41:
287-294.
Santschi, F. 1910. Formicides nouveaux ou peu connus du Congo fransais. Annls Soc. ent. Fr. 78 (1909):
349^00, 20 figs.
— 1912. Fourmis d'Afrique et de Madagascar. Annls Soc. ent. Belg. 56: 150-167, figs.
— 1913. Glanures de fourmis africaines. Annls Soc. ent. Belg. 57: 302-314, 5 figs.
— 1914a. In Voyage de Ch. Alluaud et R. Jeannel en Afrique orientate 191 1-1912. Insectes Hymenopteres
2 Formicidae: 43-148, 2 pis, 30 figs. Paris.
- 1914b. Meddelanden fran Goteborgs Musei Zoologiska Afdeling no. 3. Fourmis du Natal et du
Zululand recoltees par le Dr I. Tragardh; avec un appendice. Notes biologiques par Ivar Tragardh.
Goteborgs K. Vetensk.-o. vitterh Samh. Handl. 15: 1-47, 10 figs.
— 1917. Races et varietes nouvelles du Messor barbarus L. Bull. Soc. Hist. nat. Afr. JV. 8: 89-94, 2 figs.
— 1919. Nouvelles fourmis du Congo Beige du Musee du Congo Beige, a Tervuren. Revue Zool. Bot. Afr.
7:79-91.
— 1920. Formicides africains et americains nouveaux. Annls Soc. ent. Fr. 88 (1919): 361-390, 16 figs.
- 1924. Descriptions de nouveaux Formicides africains et notes di verses 2. Revue Zool. Bot. Afr. 12:
195-224, 10 figs.
— 1926. Description de nouveaux formicides Ethiopiens. Revue Zool. Bot. Afr. 13: 207-267, 6 figs.
— 1928. Descriptions de nouvelles fourmis ethiopiennes (suite). Revue Zool. Bot. Afr. 16: 191-213, figs.
— 1930. Formicides de 1'Angola. Revue suisse Zool. 37: 53-82, 10 figs.
- 1935. In Jeannel, R., Mission scientifique de I'Omo 2, Zoologie, fasc. 15, Hymenoptera 1, Formicidae:
255-277, 7 figs. Paris.
— 1937. Glanure de fourmis ethiopiennes. Bull. Annls Soc. ent. Belg. 77: 47-66, 15 figs.
1939. Fourmis de Rhodesia et du Congo. Bull. Annls Soc. ent. Belg. 79: 237-246, 8 figs.
Smith, F. 1853. Monograph of the genus Cryptocerus, belonging to the group Cryptoceridae- Family
Myrmicidae-Division Hymenoptera Heterogyna. Trans, ent. Soc. Lond. 2: 213-228, pis 19-21.
- 1858. Catalogue of hymenopterous insects in the collection of the British Museum, part 6, Formicidae:
216 pp., 14 pis. London.
Smith, M. R. 1950. On the status of Leptothorax Mayr and some of its subgenera. Psyche, Camb. 57: 29-30.
- 1955. The correct taxonomic status of Pheidole (Pheidolacanthinus) brevispinosa Donisthorpe. Proc.
ent. Soc. Wash. 57: 305.
Snelling, R. R. 1979. Three new species of the palaeotropical arboreal ant genus Cataulacus. Contr. Sci. 315:
1-8, 26 figs.
Stitz, H. 1916. Ergebnisse der Zweiten Deutschen Zentral Afrika Expedition, 1910-1911. 1. Zoologie:
369-405, 13 figs. Leipzig.
- 1923. Beitrage zur Kenntnis der Land- und Susswasserfauna Deutsch-Sudwestafrikas. Ergebnisse der
Hamburger deutsch-siidwestafrikanischen Studienreise 191 1. Hymenoptera 7 formicidae: 143-167, 2 figs.
Hamburg.
Urbani, C. B. 1971. Catalogo delle specie di Formicidae d'ltalia. Memorie Soc. ent. ital. 50: 5-287.
- 1973. Die Gattung Xenometra, ein objektives Synonym. Mitt, schweiz. ent. Ges. 46: 199-201.
- 1978. Materiali per una revisione dei Leptothorax neotropicali appartenenti al sottogenere
Macromischa Roger. Entomologica basil. 3: 395-618, 223 figs.
Viehmeyer, H. 1914. Mayr's Gattung Ischnomyrmex, nebst Beschreibung einiger neuer Arten aus anderen
Gattungen. Zool. Jb. (Syst.) 37: 601-61 1, figs.
Weber, N. A. 1952. Studies on African Myrmicinae, 1. Am. Mus. Novit. no. 1548: 1-32, 36 figs.
AFROTROPICAL MYRMICINE ANT GENERA
369
Wheeler, W. M. 1910. Three new genera of myrmicine ants from tropical America. Bull. Am. Mus. nat. Hist.
28: 259-265, 3 figs.
- 191 1. A list of the type-species of the genera and subgenera of Formicidae. Ann. N.Y. Acad. Sci. 21 :
157-175.
— 1916. The Australian ants of the genus Aphaenogaster Mayr. Trans. R. Soc. S. Aust. 40: 213-223, 2 pis.
- 1922. Ants of the Belgian Congo, parts 1-8. Bull. Am. Mus. nat. Hist. 45: 1-1139, 23 pis, 76 figs.
- 1927. The ants of Lord Howe Island and Norfolk Island. Proc. Am. Acad. Arts Sci. 62: 121-153, 12 figs.
1935. New ants from the Philippines. Psyche, Camb. 47: 38-52, 3 figs.
Wheeler, W. M. & Creighton, W. S. 1934. A study of the ant genera Novomessor and Veromessor. Proc. Am.
Acad. Arts. Sci. 69: 341-387, 2 pis.
Wilson, E. O. & Taylor, R. W. 1967. The ants of Polynesia. Pacif. Insects Monogr. 14: 1-109, 84 figs.
aethiops 358
airensis 350
angularis 344
angulatus 324
Aphaenogaster 364
arcistriatus 348
armata 350
Attomyrma 364
badonei 316
beccarii 336
braunsi (Leptothorax) 325
braunsi (Messor) 345
brevispinosa 316
Brunella 364
brunni 349
capensis 345
Cardiocondyla 309
Cataulacus 354
Caulomyrma 319
cenatus 327
centrums 359
cephalotes 346
cestus 360
chlorotica 317
collingwoodi 346
compressus 337
concolor 324
Cratomyrmex 338
decipiens 348
denticornis 349
denticulatus 328
Deromyrma 364
Dichothorax 319
donisthorpei 345
Dyclona 309
emeryi (Cardiocondyla) 312
emeryi (Melissotarsus) 337
Emery ia 309
evelynae 328
Index
Synonyms are in italics,
fusca 316
galla 349
globinodis 316
Goniothorax 319
grisoni 329
hawaiensis 317
humerosus 329
Icothorax 319
ilgii 324
incisus 351
innocens 330
jacksoni 360
jeanneli 358
kenyensis 358
laevifrons 354
latinoda 350
latinodis 324
Leptothorax 319
Limnomyrmex 319
Lobognathus 338
Loncyda 309
luebberti 351
lujae 358
mahdii 313
major 337
mauritia 313
megalops 331
Melissotarsus 333
Messor 338
moloch 361
monardi 314
monilicornis 313
Mychothorax 319
Myrafant 319
Myrmammophilus 319
370
B. BOLTON
neferka 314
nereis 313
Nesomyrmex 319
nigriventris 354
nilotica 315
nobilis 350
Novomessor 364
Nystalomyrma 364
parvidens 349
piceus 352
pilipes 337
Planimyrma 364
p/iw'i 346
profta 348
Prosopidris 309
pseudoaegyptiaca 345
rasalamae 312
regalis 352
rwftea 352
ru/oi 349
ru/w/a 350
ruginodis 353
satrap 363
schencki 345
sculptior 316
sculpturatus 352
sekhemka 315
shuckardi 316
simoni 331
stramineus 332
striatifrons 353
taylori 364
Temnothorax 319
Tetramyrma 319
titubans 336
traegaordhi 358
triempressa 350
tropicorum 354
Veromessor 338
316
weissi (Cataulacus) 358
weissi (Melissotarsus) 337
weserka 317
wroughtonii 317
Xenometra 309
zoserka 318
British Museum (Natural History)
Chance, change & challenge
Two multi-author volumes from one of the foremost scientific institutions in the world.
General Editor : P. H. Greenwood
The Evolving Earth
Editor:L.R.M. Cocks
The Evolving Biosphere
Editor : P. L. Forey
In the first volume, The Evolving Earth, twenty scientists have been asked to review the present
state of knowledge in their particular field, ranging from the origin of the Earth, through ocean
sediments and soils to continental drift and palaeogeography.
In the companion volume, The Evolving Biosphere, museum scientists have chosen an evolution-
ary concept — speciation, coevolution, biogeography etc. and related this to the group of animals
or plants in which they are specialising. Thus beetles and birds exemplify sympatric and
allopatric speciation, butterflies mimicry and certain fishes explosive evolution.
In both volumes the text is supplemented by over one hundred specially commissioned pieces of
two-colour artwork.
These two books will be invaluable to all sixth-form and undergraduate biology and geology
students.
The Evolving Earth: 276 x 219 mm, 280pp, 138 line illustrations, 42 halftones
The Evolving Biosphere: 216 x 219 mm, approx. 320pp, 133 line illustrations
Published: May 1981
Co-published by the British Museum (Natural History), London and Cambridge University
Press, Cambridge.
Titles to be published in Volume 45
A catalogue and reclassification of the Ichneumonidae (Hymenoptera) described by C. G.
Thomson.
By M. G. Fitton
A taxonomic review of the genus Phlebotomus (Diptera : Psychodidae).
By D. J. Lewis
Stenomine moths of the Neotropical genus Timocratica (Oecophoridae).
By V. O. Becker
Afrotropical species of the myrmicine ant genera Cardiocondyla, Leptothorax,
Melissotarsus, Messor and Cataulacus (Formicidae).
By Barry Bolton
Typeset by Santype International Ltd., Salisbury and Printed by Henry Ling Ltd., Dorchester.
BOUND
15 DEC 1987
J