Research Article |
Corresponding author: Francisco Hita Garcia ( fhitagarcia@gmail.com ) Academic editor: Brian Lee Fisher
© 2014 Francisco Hita Garcia, Peter Hawkes, Gary D. Alpert.
This is an open access article distributed under the terms of the Creative Commons Attribution License (CC BY 4.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Citation:
Hita Garcia F, Hawkes P, Alpert G (2014) Taxonomy of the ant genus Proceratium Roger (Hymenoptera, Formicidae) in the Afrotropical region with a revision of the P. arnoldi clade and description of four new species. ZooKeys 447: 47-86. https://doi.org/10.3897/zookeys.447.7766
|
The taxonomy of the genus Proceratium Roger is updated for the Afrotropical region. We give an overview of the genus in the region, provide an illustrated identification key to the three clades (P. arnoldi, P. stictum and P. toschii clades) and revise the P. arnoldi clade. Four new species from the P. arnoldi clade are described as new: P. sokoke sp. n. from Kenya, P. carri sp. n. from Mozambique, and P. nilo sp. n. and P. sali sp. n. from Tanzania. In order to integrate the new species into the existing taxonomic system we present an illustrated identification key to distinguish the seven Afrotropical species of the P. arnoldi clade. In addition, we provide accounts for all members of the P. arnoldi clade including detailed descriptions, diagnoses, taxonomic discussions, distribution data and high quality montage images.
Arabuko Sokoke Forest, Gorongosa National Park, Kenya, Mozambique, Nilo Forest, Proceratium arnoldi clade, Proceratium stictum clade, Proceratium toschii clade, Sali Forest, Tanzania, taxonomy
The ant genus Proceratium Roger, 1863 contains 77 extant and 5 fossil species and is patchily distributed throughout all biogeographical regions (
The taxonomy of the genus is in a relatively good condition since
In this study we describe four new species: P. carri sp. n. from Mozambique, P. nilo sp. n. and P. sali sp. n. from Tanzania and P. sokoke sp. n. from Kenya. We place these four new species in the P. arnoldi clade sensu
The collection abbreviations mostly follow
AFRC AfriBugs, Pretoria, Gauteng, South Africa
BMNH The Natural History Museum, London, U.K.
CASC California Academy of Sciences, San Francisco, California, U.S.A.
CIRAD Centre de coopération Internationale en recherche agronomique pour le développement, Montpellier, France
MCZ Museum of Comparative Zoology, Cambridge, Massachusetts, U.S.A.
MHNG Muséum d’Histoire Naturelle de la Ville de Genève, Geneva, Switzerland
MNHN Muséum National d’Histoire Naturelle, Paris, France
SAMC Iziko Museums of South Africa, Cape Town, South Africa
The material for the new species on which this study is based was collected during several recent, still on-going ant diversity inventories in Kenya, Tanzania and Mozambique carried out independently by the three authors. The material from the new species and most of the previously known specimens can be uniquely identified with specimen-level barcodes affixed to each pin (e.g. MCZ-ENT00517081 or CASENT0235688). The series of stacked digital colour images were created either by a Canon 7D camera attached to a Leica MZ16 stereomicroscope and source images processed using Helicon Focus 5.3, or with a Leica DFC 425 camera in combination with the Leica Application Suite software. These images are all available online and can be seen on AntWeb (www.antweb.org) and AntWiki (www.antwiki.org).
The measurements were taken with a Leica MZ16 stereomicroscope equipped with an orthogonal pair of micrometers at a magnification of 100×. Measurements and indices are presented as minimum and maximum values and all measurements are expressed in mm to two decimal places. The measurements and indices used in this study are based on
EL Eye length – maximum length of eye measured in oblique lateral view
HL Head length – maximum measurable distance from the mid-point of the anterior clypeal margin to the mid-point of the posterior margin of head, measured in full-face view. Impressions on anterior clypeal margin and posterior head margin reduce head length
HLM Head length with mandibles – maximum head length in full-face view including closed mandibles
HW Head width – maximum head width directly behind the eyes, measured in full-face view
HFeL Hind femur length – maximum length of hind femur measured along its external face
HTiL Hind tibia length – maximum length of hind tibia measured on its external face
HBaL Hind basitarsus length – maximum length of hind basitarsus measured along its external face
LT3 Abdominal tergum III length – maximum length of abdominal tergum III (= length of segment III) in lateral view
LS4 Abdominal sternum IV length – maximum length of abdominal sternum IV following
LT4 Abdominal tergum IV length – maximum length of abdominal tergum IV following
PeL Petiolar length – maximum length of the petiole in dorsal view including its anterior prolongation
PeW Petiolar width – maximum width of petiole measured in dorsal view
SL Scape length – maximum length of scape shaft excluding basal condyle
TL Total body length – combined length of HLM + WL + PeL + LT3 + LT4
WL Weber’s length – diagonal length of mesosoma in lateral view from the anterior-most point of pronotal slope (excluding neck) to posteroventral margin of propodeal lamella or lobe.
CI Cephalic index – HW / HL × 100
OI Ocular index – EL / HW × 100
SI Scape index – SL / HL × 100
DPeI Dorsal petiole index – PeW / PeL × 100
ASI Abdominal segment index – LT4 /LT3 × 100
IGR Gastral reflexion index – LS4 / LT4
The morphological terminology used in this study follows
We have not been able to examine the palp formula of most species treated in this study. Due to the lack of material we did not want to risk damaging the few specimens (most of them unique types) by moving heads or dissecting mouthparts.
In addition to the Afrotropical material treated here we examined a larger number of species from other regions in order to assess general variation within Proceratium. Of special importance was the examination of the interspecific and intraspecific variation in general dimensions, the shape of the petiolar node, especially its ventral process, pilosity/pubescence and surface sculpturing. For this purpose we examined the following species, which include several undescribed species from the Malagasy region (AntWeb): P. angulinode de Andrade, 2003 [Malaysia]; P. austronesicum de Andrade, 2003 [Papua New Guinea]; P. avium Brown, 1974 [Mauritius]; P. banjaranense de Andrade, 2003 [Malaysia]; P. crassicorne Emery, 1895 [U.S.A.]; P. croceum (Roger, 1860) [U.S.A.]; P. deelemani Perrault, 1981 [Malaysia, Thailand]; P. diplopyx Brown, 1980 [Madagascar]; P. fhg-alob [Madagascar]; P. fhg-beta [Madagascar]; P. fhg-elia [Madagascar]; P. fhg-mala [Madagascar]; P. fhg-seyc [Seychelles]; P. google Fisher, 2005b [Madagascar]; P. papuanum Emery, 1897 [Papua New Guinea]; P. sulawense de Andrade, 2003 [Indonesia]; P. terraealtae de Andrade, 2003 [Malaysia]. On the basis of the examination of the above species and despite the paucity of Afrotropical material, we are very confident in using character sets that appear to be stable within other Proceratium species from other regions. Furthermore, the species P. arnoldi, P. carri and especially P. lunatum are known from at least two localities that are often several hundreds or even thousands of km apart. Despite these distances, there is little to no observable intraspecific variation, which is suggestive of a fairly high level of conservation of morphological characteristics over relatively long distances. In contrast, P. sokoke and P. nilo differ more significantly morphologically, yet their type localities are separated by only 220 km.
Nevertheless, to our surprise, one commonly used character for species level diagnostics turned out to be relatively variable. We observed a lot of intraspecific variation in the surface sculpture of several species throughout most regions. This was already pointed out by
Characters that have proven to be comparatively stable at species level are general dimensions of the head and petiole, the shape of the ventral process of the petiole, the relationship between abdominal segments III and IV, eye size and also pilosity and pubescence. Like in most ants, the shape of the head and petiole turned out to be very stable within species-specific ranges and are useful for diagnostics. The relationship between the lengths of abdominal segments III and IV proved to be valuable, too. The development of the eyes is normally not very important in Proceratium since most species have very reduced eyes consisting of one ommatidium, a tiny cluster of indistinct flat ommatidia only distinguishable at high magnifications, or no eyes at all, but a few species, such as P. burundense de Andrade, 2003, have slightly larger compound eyes. At first glance the use of pilosity/pubescence might seem challenging since most species of Proceratium are very hairy and possess different types of hairs throughout their bodies. However, despite some small variation in density, which can also be attributed to specimen processing, we could not observe any significant variability in pilosity/pubescence within species. Especially the lack or presence of abundant, longer, standing pilosity on top of the very dense mat of subdecumbent to decumbent hairs seems to be species-specific.
Proceratium arnoldi clade
Proceratium arnoldi Forel, 1913 [South Africa, Zimbabwe]
Proceratium carri Hita Garcia, Hawkes & Alpert, sp. n. [Mozambique]
Proceratium burundense de Andrade, 2003 [Burundi]
Proceratium lunatum Terron, 1981 [Cameroon, Gabon, Uganda]
Proceratium nilo Hita Garcia, Hawkes & Alpert, sp. n. [Tanzania]
Proceratium sali Hita Garcia, Hawkes & Alpert, sp. n. [Tanzania]
Proceratium sokoke Hita Garcia, Hawkes & Alpert, sp. n. [Kenya]
Proceratium stictum clade
Proceratium boltoni Leston, 1971 [Ghana]
Proceratium toschii clade
Proceratium terroni Bolton, 1995 [Cameroon]
Proceratium toschii (Consani, 1951) [Kenya]
Considering the biogeography of the genus in sub-Saharan Africa and the rarity of collections, the available data about the distribution patterns of most species is very limited. This is especially true for more than half of the species that are only known from the type locality (P. burundense, P. nilo, P. sali, P. sokoke, P. terroni and P. toschii). The data for the other four species (P. arnoldi, P. boltoni, P. carri, and P. lunatum) is a bit better, even though they are also only known from a few localities each. The widest known distribution is seen in P. lunatum, which occurs in Cameroon and Gabon, but is also found in Uganda. We expect that this species will also be found in the rainforests of the Democratic Republic of Congo, Congo, and Central African Republic with further sampling in these countries.
As mentioned above, due to the cryptobiotic lifestyle and small colony size, it is not easy to collect Proceratium, but their rarity might also be due to sampling artefacts. For example, intensive sampling in two localities in Kenya (Kakamega Forest and Mpala Research Centre, listed in
The knowledge on natural history or behaviour for the ten Afrotropical species is extremely limited.
The following key is derived from
1 | Clypeus well developed, clearly protruding anteriorly, surrounding the antennal sockets and medially impressed; vertex in full face view weakly concave; calcar of strigil with a basal spine (Fig. |
P. stictum clade |
– | Clypeus reduced, only slightly protruding anteriorly, not surrounding the antennal sockets and not medially impressed; vertex in full face view not concave, usually weakly to moderately convex; calcar of strigil without basal spine (Fig. |
2 |
2 | Frontal carinae relatively close to each other, either converging and posteriorly fused (Fig. |
P. toschii clade |
– | Frontal carinae relatively far from each other, never approaching medially and widely diverging posteriorly; lower mesopleura posteriorly inflated (Fig. |
P. arnoldi clade |
Anterior part of cephalic dorsum in full-face view showing clypeus (within white ellipse) and calcar of strigil (white arrows indicate position of calcar). A P. boltoni (CASENT0902424) (Will Ericson 2013) B P. diplopyx (Malagasy member of P. stictum clade) C P. burundense (CASENT0902427) (Will Ericson 2013) D P. sokoke (MCZ-ENT00520482).
Anterior part of cephalic dorsum in full-face view showing frontal carinae (within white ellipse). A P. terroni (CASENT0914223) (Michele Esposito 2014) B P. burundense (CASENT0902427) (Will Ericson 2013).
Diagnosis. The following diagnosis is based on
Notes.
1 | No maculae present on vertexal angles of head (Fig. |
P. carri |
– | Maculae on vertexal angles of head present and usually very conspicuous, but sometimes difficult to see (Fig. |
2 |
2 | Eyes absent (OI 0) (Fig. |
P. nilo |
– | Eyes variable in size, but always present (OI 3–8) (Fig. |
3 |
3 | Eyes larger (OI 8) (Fig. |
P. burundense |
– | Eyes always smaller (OI 3–5) (Fig. |
4 |
4 | Head, mesosoma and petiole with numerous long, fine, suberect to erect hairs on top of dense mat of much shorter decumbent to subdecumbent pubescence (Fig. |
5 |
– | Head, mesosoma and petiole with mat of short decumbent to subdecumbent pubescence only, without numerous longer, fine suberect to erect hairs (Fig. |
6 |
5 | Petiolar node in profile appearing thicker and stouter, with lower third of anterior face strongly produced anteriorly (Fig. |
P. sokoke |
– | Petiolar node higher and thinner than above, more strongly anteroposteriorly compressed, with anterior face relatively straight (Fig. |
P. sali |
6 | Head relatively longer (CI 85–87) (Fig. |
P. arnoldi |
– | Head relatively shorter (CI 92–95) (Fig. |
P. lunatum |
Vertex in posterodorsal view (black arrows indicate maculae) and abdominal segments III and IV in profile. A, D P. carri (MCZ-ENT00517081) B P. sokoke (MCZ-ENT00520482) C P. sali (CASENT0235689) (Will Ericson 2011) E P. arnoldi (CASENT0907203) (Will Ericson 2013) F P. lunatum (CASENT0005926).
Head in profile showing eye (white arrows indicate location of eye, if present). A P. nilo (CASENT0235688) (Will Ericson 2011) B P. arnoldi (CASENT0914281) (Michele Esposito 2014) C P. burundense (CASENT0902427) (Will Ericson 2013).
Head in profile showing eye (white arrows indicate location of eye) and petiole with ventral process in profile (within white ellipse). A, D P. burundense (CASENT0902427) (Will Ericson 2013) B, E P. lunatum (CASENT0005926) C, F P. arnoldi (CASENT0914281) (Michele Esposito 2014).
Head, mesosoma and petiole in profile. A P. sokoke (MCZ-ENT00520482) B P. sali (CASENT0235689) (Will Ericson 2011) C P. lunatum (CASENT0914221) (Michele Esposito 2014) D P. arnoldi (CASENT0914281).
Petiole in profile (anterior face within white ellipse). A P. sokoke (MCZ-ENT00520482) B P. sali (CASENT0235689) (Will Ericson 2011).
Head in full-face view and body in profile. A, B P. arnoldi (CASENT0914281) (Michele Esposito 2014) C, D P. lunatum (CASENT0005926).
Proceratium (Sysphincta) arnoldi Forel, 1913: 210. [Combination in Sysphincta:
Holotype, pinned worker, ZIMBABWE, Bulawayo, (MHNG: CASENT0907203).
[Note: There are two known “type” specimens of P. arnoldi, and there is some confusion about their labels and type status. De Andrade (in
SOUTH AFRICA: Natal, N of Richard’s Bay, 28.40S, 32.14E, 26.I.–11.II.1991 (A. de Kock & J.D. Majer) (BMNH: CASENT0914281); ZIMBABWE: Bulawayo, S. Rhodesia, 29.III.1913, (G. Arnold) (BMNH: CASENT0902425).
The following character combination distinguishes P. arnoldi from the other Afrotropical members of the P. arnoldi clade: eyes very small, consisting of a single dark ommatidium (OI 3–5); head clearly longer than broad (CI 85–87); maculae on vertexal angles of head well developed and conspicuous; mesopleurae weakly to moderately inflated posteriorly; petiolar node high nodiform, anteroposteriorly compressed, with anterior face relatively straight; petiole in dorsal view between 1.0 and 1.2 times wider than long (DPeI 106–114); ventral process of petiole lamelliform, subrectangular, anteroventral corner blunt and posteroventral corner conspicuously projecting posteroventrally; abdominal segment IV around 1.2 to 1.3 times longer than abdominal segment III (ASI 116–132); head, mesosoma and petiole with mat of short decumbent to subdecumbent pubescence only, without any longer, fine suberect to erect hairs.
(N=5). TL 3.27–3.56; EL 0.02–0.03; SL 0.49–0.52; HL 0.79–0.84; HLM 0.94–1.02; HW 0.69–0.71; WL 0.91–1.03; HFeL 0.59–0.67; HTiL 0.48–0.51; HBaL 0.38–0.46; PeL 0.33–0.34; PeW 0.35–0.38; DPeI 106–114; LT3 0.47–0.54; LS4 0.24–0.25; LT4 0.62–0.64; OI 3–5; CI 85–87; SI 60–63; IGR 0.38–0.41; ASI 116–132.
In full-face view head clearly longer than broad (CI 85–87), sides weakly convex, gently broadening posteriorly, vertex flat to weakly convex. Clypeus medially reduced, its anterior margin subconvex to slightly triangular, only slightly protruding anteriorly, not surrounding the antennal sockets and not medially impressed, antennal socket with broad torulus. Frontal carinae relatively very short and widely separated, not converging medially and strongly diverging posteriorly, partially covering antennal insertions; frontal carinae conspicuously raised anteriorly, much less so posteriorly. Eyes very small, consisting of one to four weak ommatidia (OI 5) and located on mid line of head. Mandibles elongate-triangular; masticatory margin of mandibles with one well developed apical tooth and a series of four denticles decreasing in size towards basal-most denticle. Mesosoma weakly to moderately convex in profile and approximately as long as the maximum head length including mandibles. Lower mesopleurae with well impressed sutures, no other sutures developed on lateral or dorsal mesosoma, mesopleurae weakly to moderately inflated posteriorly; propodeum in profile armed with very small, pointed or blunt teeth, propodeal lobes weakly to moderately developed, lamellate, subtriangular and blunt; declivitous face of propodeum between teeth and lobes concave; in posterodorsal view sides of propodeum separated from declivitous face by margin connecting propodeal lobes and propodeal teeth. Legs slender and elongate; pro- and mesotibiae with pectinate spurs; calcar of strigil without basal spine. Petiolar node in profile high, blocky nodiform, anterior face of petiole relatively straight, anterior and posterior faces approximately parallel, dorsum of node flat to very weakly convex; petiole in dorsal view between 1.0 and 1.2 times wider than long (DPeI 106–114), petiolar node in dorsal view clearly much broader than long; ventral process of petiole lamelliform, subrectangular, anteroventral corner blunt and posteroventral corner conspicuously projecting posteroventrally. In dorsal view abdominal segment III anteriorly broader than petiole; its sides diverging posteriorly; dorsum of abdominal tergum III with posteromedial, weakly developed, semitransparent, flat bulla below the integument; abdominal sternite III anteromedially with a marked subtriangular projection. Constriction between abdominal segment III and IV conspicuously impressed. Abdominal segment IV strongly recurved (IGR 0.38–0.41), conspicuously rounded on its curvature, especially posteriorly, abdominal tergum IV around 1.2 to 1.3 times longer than abdominal segment III (ASI 116–132); small, faint and semitransparent bulla situated posteromedially on abdominal tergum IV; remaining abdominal tergites and sternites relatively inconspicuous and curved ventrally. Whole body covered with dense mat of relatively short, decumbent to suberect pubescence without any abundant, much longer, suberect to erect, long, fine, standing hairs. Mandibles longitudinally rugose; most of body irregularly foveolate and/or punctate, sculpture best developed on cephalic dorsum, much weaker on remainder of body, especially weak, almost smooth on abdominal segments III and IV. Body colour uniformly yellowish to light brown.
Proceratium arnoldi worker (CASENT0914281) (Michele Esposito 2014). A Body in profile B Body in dorsal view C head in full-face view.
At present, P. arnoldi is only known from two localities in Zimbabwe and South Africa (Fig.
Proceratium arnoldi is well recognisable within the clade. Its relatively thin head in full-face view (CI 85–87) groups it close to P. carri (CI 85–86) and separates both from the other five species that have thicker heads (CI 91–95). However, P. carri is not likely to be confused with P. arnoldi. The latter possesses a mat of short decumbent to subdecumbent pubescence but without numerous much longer, fine standing hairs. These hairs are present in P. carri, which also has a much longer abdominal segment IV in relation to abdominal segment III (ASI 156–159) than P. arnoldi (ASI 116–132).
We only observed some minor, very normal variation in body size in the known material of P. arnoldi with the specimens from South Africa being somewhat larger (WL 0.98–1.03) than the ones from Zimbabwe (WL 0.91–0.92). Otherwise, there is no observable intraspecific variation.
Proceratium burundense de Andrade, in
Holotype, pinned worker, BURUNDI, Bujumbura, 4.III.77, (A. Dejean) (BMNH: CASENT0902427).
Proceratium burundense is easily distinguishable from the other Afrotropical species of the P. arnoldi clade by the following character combination: eyes larger, consisting of nine well developed ommatidia (OI 8); head slightly longer than broad (CI 91); maculae on vertexal angles of head well developed and conspicuous; mesopleurae moderately inflated posteriorly; petiolar node high nodiform, anteroposteriorly compressed, with anterior face relatively straight; petiole around 1.2 times wider than long (DPeI 121); ventral process of petiole lamelliform and subrectangular with posteroventral corner strongly pointing ventrally, almost spiniform; abdominal segment IV less than 1.1 times longer than abdominal segment III (ASI 106); head, mesosoma and petiole with mat of short decumbent to subdecumbent pubescence only, without any longer, fine suberect to erect hairs.
(N=1). TL 3.44; EL 0.06; SL 0.54; HL 0.79; HLM 0.90; HW 0.72; WL 1.02; HFeL 0.59; HTiL 0.51; HBaL 0.39; PeL 0.32; PeW 0.39; DPeI 121; LT3 0.58; LS4 0.24; LT4 0.61; OI 8; CI 91; SI 0.68; IGR 0.39; ASI 106.
[Note: the singleton holotype was examined in BMNH, but not measured. The measurements presented above are the ones given by
In full-face view head slightly longer than broad (CI 91), sides weakly convex, not broadening posteriorly, vertex flat to weakly convex. Clypeus medially reduced, its anterior margin subconvex to slightly triangular, only slightly protruding anteriorly, not surrounding the antennal sockets and not medially impressed, antennal socket with broad torulus. Frontal carinae relatively short and widely separated, not converging medially and strongly diverging posteriorly, partially covering antennal insertions; frontal carinae conspicuously raised on their anterior half, much less posteriorly. Eyes small (but larger than in remainder of group), consisting of nine well developed ommatidia (OI 8) and located on mid line of head. Mandibles elongate-triangular; masticatory margin of mandibles with four to five relatively small teeth/denticles, decreasing in size from larger apical tooth to very small basal denticle. Mesosoma clearly convex in profile and slightly longer than maximum head length including mandibles. Lower mesopleurae with well impressed sutures, no other sutures developed on lateral or dorsal mesosoma; mesopleurae moderately inflated posteriorly; propodeum in profile armed with small, pointed teeth, propodeal lobes well developed, lamellate, subtriangular and blunt; declivitous face of propodeum between teeth and lobes noticeably concave; in posterodorsal view sides of propodeum separated from declivitous face by margin connecting propodeal lobes and propodeal teeth. Legs slender and elongate; pro- and mesotibiae with pectinate spurs; calcar of strigil without basal spine. Petiolar node in profile high, blocky nodiform, anterior face of petiole relatively straight, anterior and posterior faces approximately parallel, dorsum of node flat to weakly convex; petiole in dorsal view around 1.2 times wider than long (DPeI 121), petiolar node in dorsal view clearly much broader than long; ventral process of petiole lamelliform and subrectangular with posteroventral corner strongly pointing ventrally, almost spiniform. In dorsal view abdominal segment III anteriorly broader than petiole; its sides diverging posteriorly; dorsum of abdominal tergum III with posteromedial, very conspicuous, semitransparent, flat bulla below the integument; abdominal sternite III anteromedially with a marked subtriangular projection. Constriction between abdominal segment III and IV conspicuously impressed. Abdominal segment IV strongly recurved (IGR 0.39), conspicuously rounded on its curvature, especially posteriorly, abdominal tergum IV only slightly longer than abdominal segment III (ASI 106); semitransparent bulla situated posteromedially on abdominal tergum IV; remaining abdominal tergites and sternites relatively inconspicuous and curved ventrally. Whole body covered with dense mat of relatively short, decumbent to suberect pubescence without any abundant, much longer, suberect to erect, long, fine, standing hairs. Mandibles longitudinally rugose; most of body irregularly foveolate and/or granulate, sculpture best developed on cephalic and mesosomal dorsum, less so remainder of body and especially weak on most of relatively shining abdominal tergum IV, abdominal tergum IV posteroventrally (shortly before abdominal tergum V) with irregularly rugose area; inflated, posterior part of mesopleura and declivitous face of propodeum also mostly unsculptured and relatively smooth and shining. Head, mesosoma, petiole and remaining abdominal segments brown; mandibles, antennae, and legs of lighter brown.
Proceratium burundense holotype worker (CASENT0902427) (Will Ericson 2013). A Body in profile B Body in dorsal view C head in full-face view.
The species is only known from the type locality in Burundi (Fig.
As noted above, the presence of a larger compound eye that consists of nine well developed ommatidia in the worker caste distinguishes P. burundense (OI 8) from the other six species of the clade (OI 0–5), but also from most other known Proceratium species that have either no eyes, just one ommatidium or a few very weak, almost indistinguishable ommatidia only visible under higher magnifications (
Since P. burundense is only known from the holotype, there is no information about intraspecific variation.
Holotype, pinned worker, MOZAMBIQUE, Sofala, Gorongosa National Park, 4 km NW of Chitengo, 18°57'34.1"S, 34°20'30.7"E, 41 m, sandy forest on road #2, dry soil-leaf litter, collected three bags of dry soil, misc. ants, WP113, 28.IV.2013 (G.D. Alpert) (MCZ: MCZ-ENT00517081).
MOZAMBIQUE: Tete, Moatize, Haul Road 6, 30 km, 15.97644S, 33.8557E, 336 m, closed undifferentiated woodland 13.IV.2014 (P. Hawkes & R. Mulaudzi) (AFRC: CASENT0250381), Tete, Moatize, Haul Road 6, 6 km, 15.78187 S, 33.81614 E, 303 m, closed undifferentiated woodland 14.IV.2014 (P. Hawkes & R. Mulaudzi) (AFRC: CASENT0250382).
Proceratium carri differs from the other Afrotropical members of the P. arnoldi clade by the following character combination: eyes strongly reduced, consisting of a single ommatidium (OI 5); head clearly longer than broad (CI 85–86); maculae on vertexal angles of head absent; mesopleurae weakly to moderately inflated posteriorly; petiolar node high nodiform, anteroposteriorly compressed, with anterior face relatively straight; petiole in dorsal view between 1.1 to 1.2 times wider than long DPeI 111–119; ventral process of petiole lamelliform, subrectangular, lamella weakly pointed anteriorly and strongly pointed posteriorly; abdominal segment IV around 1.6 times longer than abdominal segment III (ASI 156–159); head, mesosoma and petiole with numerous long, fine, suberect to erect hairs on top of dense mat of much shorter decumbent to subdecumbent pubescence.
(N=2). TL 2.96–3.07; EL 0.03; SL 0.48–0.51; HL 0.75–0.77; HLM 0.88–0.92; HW 0.63–0.66; WL 0.81–0.82; HFeL 0.55–0.59; HTiL 0.46–0.50; HBaL 0.34–0.35; PeL 0.26–0.29; PeW 0.31–0.32; DPeI 111–119; LT3 0.39–0.41; LS4 0.16–0.18; LT4 0.62–0.64; OI 5; CI 85–86; SI 64–66; IGR 0.25–0.29; ASI 156–159.
In full-face view head clearly longer than broad (CI 85–86), sides weakly convex, gently broadening posteriorly, vertex shallowly concave. Clypeus medially reduced, its anterior margin convex to slightly triangular, only slightly protruding anteriorly, not surrounding antennal sockets and not medially impressed, antennal socket with broad torulus. Frontal carinae relatively short and widely separated, not converging medially and strongly diverging posteriorly, partially covering antennal insertions; frontal carinae conspicuously raised on their anterior two thirds, much less posteriorly. Eyes small, consisting of a single ommatidium and located on mid line of head. Antennae 12-segmented, scapes short (SI 64–66), not reaching vertexal margin and noticeably thickening apically, first and last funicular segments longer than broad, remaining funicular segments noticeably broader than long. Mandibles elongate-triangular; masticatory margin of mandibles with five teeth/denticles, decreasing in size from larger apical tooth to basal denticles. Mesosoma weakly to moderately convex in profile and clearly shorter than maximum head length including mandibles. Lower mesopleurae with well impressed sutures, no other sutures developed on lateral or dorsal mesosoma; mesopleurae weakly to moderately inflated posteriorly; propodeum in profile armed with small, pointed teeth, propodeal lobes moderately developed, lamellate and blunt; declivitous face of propodeum between teeth and lobes noticeably concave; in posterodorsal view sides of propodeum separated from declivitous face by margin connecting propodeal lobes and propodeal teeth. Legs slender and elongate; all tibiae with pectinate spur; calcar of strigil without basal spine; pretarsal claws simple; arolia well developed. Petiolar node in profile high, blocky nodiform, anterior face of petiole relatively straight, anterior and posterior faces approximately parallel, dorsum of node flat to weakly convex; petiole in dorsal view between 1.1 to 1.2 times wider than long (DPeI 111–119), petiolar node in dorsal view clearly much broader than long and transverse; ventral process of petiole lamelliform, subrectangular, lamella weakly pointed anteriorly and strongly pointed posteriorly. In dorsal view abdominal segment III anteriorly broader than petiole; its sides diverging posteriorly; dorsum of abdominal tergum III with posteromedial, very faint, semitransparent, flat bulla below the integument; abdominal sternite III anteromedially with a marked subtriangular projection. Constriction between abdominal segment III and IV conspicuously impressed. Abdominal segment IV strongly recurved (IGR 0.25–0.29), conspicuously rounded on its curvature, especially posteriorly, abdominal tergum IV relatively long, around 1.6 times longer than abdominal segment III (ASI 156–159); moderately large, semitransparent and elongate bulla situated posteromedially on abdominal tergum IV; remaining abdominal tergites and sternites relatively inconspicuous and curved ventrally. Whole body covered with dense mat of relatively short, appressed to subdecumbent pubescence; antennal scapes and legs also with moderately abundant, much longer (several times longer than pubescence), subdecumbent to erect, long, fine, standing hairs; head, mesosoma, petiole and abdominal segments with same type of long, standing pilosity, but usually more scattered than on appendages. Mandibles longitudinally rugose; most of body irregularly foveolate and/or punctate, sculpture best developed on cephalic dorsum and abdominal tergum III, less so on sides of mesosoma and especially weak on most of relatively smooth and shining abdominal tergum IV, abdominal tergum IV posteroventrally (shortly before abdominal tergum V) with small irregularly rugose area; inflated, posterior part of mesopleura and declivitous face of propodeum also only very weakly sculptured and relatively smooth and shining. Body of uniformly yellow to light orange brown colour.
The name of the new species is a patronym dedicated to the American entrepreneur and philanthropist Gregg C. Carr. We want to honour his great accomplishments for the restoration of the Gorongosa National Park in Mozambique and his efforts in the conservation of African biodiversity.
Gorongosa National Park is geographically divided into two sections, a higher elevation section on Gorongosa Mountain (1863 m summit) with a montane rainforest and a separate lowland elevation (39 m) section within the southern end of the Great African Rift Valley of east Africa (Fig.
Proceratium carri displays a character combination that renders it easily identifiable within the P. arnoldi clade. The most conspicuous difference is the lack of maculae on the vertexal margins of the head, which are present in all examined specimens of the other six species. Nonetheless, this character can be challenging to detect sometimes, especially under low magnifications or with the wrong lighting. An additional character unique to P. carri is the relatively long abdominal segment IV, which is around 1.6 times longer than abdominal tergite III (ASI 156–159), whereas it is much shorter in the other four species, around 1.0 to 1.3 times longer (ASI 102–132). Furthermore, P. carri has a comparatively thin head in full-face view (CI 85–86), which is shared by P. arnoldi (CI 85–87), but strongly contrasts with the head shape of the other five species (CI 91–95).
We cannot observe any significant intraspecific variation between the two known workers of this species. The only difference is that the specimen from Gorongosa is of darker colour than the one from Moatize.
Proceratium lunatum Terron, 1981: 96. [see also
Holotype, pinned worker, CAMEROON, Arboretum de Mbalmayo (51 km S of Yaounde), no. 1579, 17.III.1968, (G. Terron) (CIRAD). Paratypes, one pinned paratype worker from CAMEROON, Kala (18 km W Yaounde), Ve Berlese, sp. 1, tamisage terre et terreau, 16.V.1974, (G. Terron) (CIRAD); two pinned paratype workers from CAMEROON, Mbalmayo, no. 1759, 17.III.1968 (G. Terron) (BMNH: CASENT0902426; MNHN); one pinned paratype worker from CAMEROON, UO Bikok, tamisage terre et terreau, 19.III.1974, (G. Terron) (MHNG: CASENT0914221).
CAMEROON: Sud-Ouest, Korup N. P., 6.9 km 317°NW Mundemba, 5.016 N, 8.864 E, 110 m, rainforest, 19.IV.2000 (B.L. Fisher) (CASC: CASENT0005926); GABON: Woleu-Ntem, 31.3 km 108°ESE Minvoul, 2.08N, 12.40667E, 600 m, rainforest, 11.II.1998 (B.L. Fisher) (CASC: CASENT0914280); UGANDA: Western Uganda, Kabarole, Kibale National Park, Kanyawara Biological Station, 0.56437N, 30.36059E, 1520 m, moist evergreen forest, 8.–11.VIII.2012 (B.L. Fisher et al.) (CASC: CASENT0355483).
The following character combination distinguishes P. lunatum from the other Afrotropical members of the P. arnoldi clade: eyes strongly reduced (OI 3–5), usually consisting of a single ommatidium, rarely more; head slightly longer than broad (CI 92–95); maculae on vertexal angles of head well developed and conspicuous; mesopleurae weakly to moderately inflated posteriorly; petiolar node high nodiform, anteroposteriorly compressed, with anterior face relatively straight; petiole in dorsal view between 1.2 to 1.3 times wider than long DPeI 122–129; ventral process of petiole lamelliform and approximately rectangular, lamella not significantly pointing anteriorly nor posteriorly; abdominal segment IV between 1.0 to 1.2 times longer than abdominal segment III (ASI 104–118); head, mesosoma and petiole with mat of short decumbent to subdecumbent pubescence only, without any longer, fine suberect to erect hairs.
(N=8). TL 2.81–3.43; EL 0.02–0.04; SL 0.46–0.60; HL 0.73–0.84; HLM 0.86–0.96; HW 0.68–077; WL 0.84–1.00; HFeL 0.52–0.63; HTiL 0.41–0.51; HBaL 0.31–0.42; PeL 0.27–0.30; PeW 0.35–0.38; DPeI 122–129; LT3 0.40–0.57; LS4 0.14–0.23; LT4 0.43–0.60; OI 3–5; CI 92–95; SI 63–71; IGR 0.33–0.38; ASI 104–118.
In full-face view head slightly longer than broad (CI 92–95), sides weakly convex, vertex flat to weakly convex. Clypeus medially reduced, its anterior margin subconvex to slightly triangular, only slightly protruding anteriorly, not surrounding the antennal sockets and not medially impressed, antennal socket with broad torulus. Frontal carinae relatively short and widely separated, not converging medially and strongly diverging posteriorly, partially covering antennal insertions; frontal carinae conspicuously raised on their anterior half, much less posteriorly. Eyes very small, consisting one to three or four weak ommatidia (OI 3–5) and located on mid line of head. Mandibles elongate-triangular; masticatory margin of mandibles with four relatively small teeth/denticles, decreasing in size from larger apical tooth to basal denticle. Mesosoma in profile convex and approximately as long as the maximum head length including mandibles. Lower mesopleurae with well impressed sutures, no other sutures developed on lateral or dorsal mesosoma; mesopleurae weakly to moderately inflated posteriorly; propodeum in profile armed with small, blunt teeth, propodeal lobes well developed, lamellate, rounded to subtriangular and blunt; declivitous face of propodeum between teeth and lobes noticeably concave; in posterodorsal view sides of propodeum separated from declivitous face by margin connecting propodeal lobes and propodeal teeth. Legs slender and elongate; pro- and mesotibiae with pectinate spurs; calcar of strigil without basal spine. Petiolar node in profile high, blocky nodiform, anterior face of petiole relatively straight, anterior and posterior faces approximately parallel, dorsum of node flat to weakly convex; petiole in dorsal view between 1.2 and 1.3 times wider than long (DPeI 122–129), petiolar node in dorsal view clearly much broader than long; ventral process of petiole lamelliform and approximately rectangular, lamella not significantly pointing anteriorly nor posteriorly. In dorsal view abdominal segment III anteriorly broader than petiole; its sides diverging posteriorly; dorsum of abdominal tergum III with posteromedial, very conspicuous, semitransparent, raised bulla below the integument; abdominal sternite III anteromedially with a marked subtriangular projection. Constriction between abdominal segment III and IV conspicuously impressed. Abdominal segment IV strongly recurved (IGR 0.33–0.38), conspicuously rounded on its curvature, especially posteriorly, abdominal tergum IV between 1.0 and 1.2 times longer than abdominal segment III (ASI 104–118); large, semitransparent and circular bulla situated posteromedially on abdominal tergum IV; remaining abdominal tergites and sternites relatively inconspicuous and curved ventrally. Whole body covered with dense mat of relatively short, decumbent to subdecumbent pubescence without abundant, much longer, suberect to erect, long, fine, standing hairs. Mandibles longitudinally rugose; most of body irregularly foveolate and/or granulate, sometimes more weakly developed on cephalic dorsum and anterior part of abdominal tergum IV, posteroventral part of abdominal tergum IV with conspicuous irregular rugosity; inflated, posterior part of mesopleura and declivitous face of propodeum unsculptured, smooth and shining. Head, mesosoma (excluding posteriorly inflated part of mesopleurae), postpetiole and remaining abdominal segments of light brown to brown colour, mandibles, inflated part of mesopleurae and legs always of lighter brown colour.
Proceratium lunatum worker (CASENT0914280) (Michele Esposito 2014). A Body in profile B Body in dorsal view C head in full-face view.
Proceratium lunatum is known to occur in Cameroon, Gabon and Uganda (Fig.
The recognition of P. lunatum within the P. arnoldi clade is fairly easy and straightforward. The relatively broad head in full-face view (CI 92–95) groups it together with P. burundense, P. nilo, P. sali and P. sokoke (CI 91–95) while it separates it from P. arnoldi and P. carri that have thinner heads (CI 85–87). The lack of long, standing pilosity on top of a dense mat of much shorter pubescence distinguishes P. lunatum from P. carri, P. nilo, P. sali and P. sokoke. The species closest to P. lunatum seems to be P. burundense. However, both differ in eye size, ventral process of the petiole, and propodeal armament. Proceratium lunatum has smaller eyes (OI 3–5) and shorter propodeal teeth than P. burundense (OI 8). In addition, the ventral process of the petiole has a very distinct posteroventral corner that strongly projects ventrally, whereas the process of P. lunatum is more or less rectangular without a projecting posteroventral corner.
The P. lunatum material from Cameroon and Gabon shows no observable intraspecific variation. The specimen from Uganda, however, displays some noticeable differences. It possesses longer antennal scapes (SI 71 vs. SI 63–66) and is generally larger than the western specimens (TL 3.43 vs. TL 2.81–2.94). Nevertheless, we consider these differences as intraspecific variation. The difference in body size is well within the range of other species, thus not significant, and the longer antennal scape alone is not sufficient to warrant species status. This assessment might change with further material from the eastern parts of the equatorial forest belt, such as Uganda, Rwanda and Kenya, but for the moment we keep all the material listed here as P. lunatum as one species.
Holotype, pinned worker, TANZANIA, Tanga, Korogwe, Nilo Forest Reserve, 4.91456S, 38.67712E, 1006 m, primary forest, collection code CEPF-TZ-4.1, 1.–4.IX.2005 (P. Hawkes, J. Makwati & R. Mtana) (SAMC: CASENT0235688).
Proceratium nilo can be distinguished from the other Afrotropical members of the P. arnoldi clade by the following combination of characters: eyes absent; head slightly longer than broad (CI 91); maculae on vertexal angles of head well developed and conspicuous; mesopleurae extremely inflated posteriorly; petiolar node in profile relatively low, bluntly rounded nodiform, anterior face of petiole strongly produced anteriorly on lower third and not straight; petiole in dorsal view between 1.1 and 1.2 times wider than long (DPeI 115); ventral process of petiole well developed, lamelliform and rectangular, lamella not pointed anteriorly nor posteriorly; abdominal segment IV around as long as abdominal segment III (ASI 102); head, mesosoma and petiole with numerous long, fine, suberect to erect hairs on top of dense mat of much shorter decumbent to subdecumbent pubescence.
(N=1). TL 3.31; EL n.a. (eyes absent); SL 0.56; HL 0.82; HLM 0.99; HW 0.75; WL 0.97; HFeL 0.60; HTiL 0.51; HBaL 0.40; PeL 0.34; PeW 0.39; DPeI 115; LT3 0.50; LS4 0.20; LT4 0.51; OI 0; CI 91; SI 68; IGR 0.39; ASI 102.
In full-face view head slightly longer than broad (CI 91), sides weakly convex, head not gently diverging posteriorly, vertex weakly convex. Clypeus medially reduced, its anterior margin convex to slightly triangular, only slightly protruding anteriorly, not surrounding the antennal sockets and not medially impressed, antennal socket with broad torulus. Frontal carinae relatively short and widely separated, not converging medially and strongly diverging posteriorly, partially covering antennal insertions; frontal carinae conspicuously raised on their anterior half, much less posteriorly. Eyes absent (OI 0). Mandibles elongate-triangular; masticatory margin of mandibles with four relatively small teeth/denticles, decreasing in size from larger apical tooth to basal denticle. Mesosoma weakly to moderately convex in profile and approximately as long as the maximum head length including mandibles. Lower mesopleurae with well impressed sutures, no other sutures developed on lateral or dorsal mesosoma; mesopleurae extremely inflated posteriorly; propodeum in profile armed with small, pointed teeth, propodeal lobes well developed, lamellate, rounded and blunt; declivitous face of propodeum between teeth and lobes noticeably concave; in posterodorsal view sides of propodeum separated from declivitous face by margin connecting propodeal lobes and propodeal teeth. Legs slender and elongate; pro- and mesotibiae with pectinate spurs; calcar of strigil without basal spine. Petiolar node in profile relatively low, bluntly rounded nodiform, anterior face of petiole strongly produced anteriorly on lower third and not straight, posterior face approximately straight, anterior and posterior faces not parallel, dorsum of node weakly rounded; petiole in dorsal view between 1.1 and 1.2 times wider than long (DPeI 115), petiolar node in dorsal view clearly much broader than long; ventral process of petiole well developed, lamelliform and rectangular, lamella not pointed anteriorly nor posteriorly. In dorsal view abdominal segment III anteriorly broader than petiole; its sides diverging posteriorly; dorsum of abdominal tergum III with posteromedial, very conspicuous, semitransparent, flat bulla below the integument; abdominal sternite III anteromedially with a marked subtriangular projection. Constriction between abdominal segment III and IV conspicuously impressed. Abdominal segment IV strongly recurved (IGR 0.39), conspicuously rounded on its curvature, especially posteriorly, abdominal tergum IV approximately as long as abdominal segment III (ASI 102); large, semitransparent and semicircular bulla situated posteromedially on abdominal tergum IV; remaining abdominal tergites and sternites relatively inconspicuous and curved ventrally. Whole body covered with dense mat of relatively short, decumbent to subdecumbent pubescence, and most of body with moderately abundant, much longer (several times longer than pubescence), suberect to erect, fine, standing hairs. Mandibles longitudinally rugose; most of body irregularly foveolate and/or granulate, sculpture best developed on cephalic dorsum, moderately so on mesosoma and petiole, especially weak, almost smooth, on most on anterior third of abdominal tergum IV, posterior third of abdominal tergum IV with conspicuous, longitudinal, irregular rugosity; inflated, posterior part of mesopleura and declivitous face of propodeum unsculptured, smooth and shining. Head, mesosoma (excluding posteriorly inflated part of mesopleurae), postpetiole and remaining abdominal segments of brown colour, mandibles, inflated part of mesopleurae and legs yellowish to light brown.
Proceratium nilo sp. n. holotype worker (CASENT0235688) (Will Ericson 2011). A Body in profile B Body in dorsal view C head in full-face view.
The name of the new species is derived from the type locality, the Nilo Forest Reserve in Tanzania. The species epithet is a noun in apposition and thus invariant.
Like several other species of the clade, P. nilo is only known from a singleton holotype collected in the Nilo Forest Reserve in the Tanga region of northeast Tanzania (Fig.
Proceratium nilo is a fairly conspicuous member of the clade, and possesses a unique character combination allowing an easy identification. The most noticeable difference is the total lack of eyes, which are present in all the other species of the clade. Not considering the eyes, the shape of the petiolar node groups P. nilo with P. sokoke while it separates it from P. arnoldi, P. burundense, P. carri, P. lunatum and P. sali. In the latter five the node is high nodiform, anteroposteriorly compressed and with the anterior face relatively straight, whereas the node shape of P. nilo and P. sokoke is relatively low, bluntly rounded nodiform with the anterior face strongly produced anteriorly on lower third. Despite the clear separation based on the presence/absence of the eyes, P. nilo is morphologically very close to P. sokoke. Indeed, the only significant difference is eye development, and for a short while we considered to lump them both under the same species name. However, the examination of many more species of Proceratium led us to refrain from doing so. As it seems, the presence or absence of eyes, as well as their specific development, is species-specific in the genus, which supports the separation into two species. Also, there are a few more differences. Proceratium sokoke has a longer abdominal tergum IV in relation to abdominal segment III (ASI 125) compared to P. nilo (ASI 102). In addition, the head of P. nilo does not significantly broaden posteriorly while the head of P. sokoke does so. However, based on the very limited material this could just be within a normal species-specific range. The two-species hypothesis is also supported by different habitat preferences (littoral, mixed dry forest at a very low elevation vs. submontane, primary rainforest at a medium elevation). Future sampling in East Africa might provide additional evidence for their heterospecificity or not (if eye development turns out to be variable within species), but for the moment we prefer to describe P. nilo and P. sokoke as easily identifiable species and make them both available to the taxonomic community.
Since this species is known only from the holotype there is no available information about intraspecific variation.
Holotype, pinned worker, TANZANIA, Morogoro, Ulanga, Sali Forest Reserve, 8.94497S, 36.67261E, 1150 m, primary forest, collection code CEPF-TZ-9.1, 17.–20.X.2007 (P. Hawkes, M. Bhoke & U. Richard) (SAMC: CASENT0235689).
The following character combination separates Proceratium sali from the other Afrotropical members of the P. arnoldi clade by the following combination of characters: eyes very small, consisting of three to four weak ommatidia (OI 5); CI 94; maculae on vertexal angles of head well developed and conspicuous; petiolar node high nodiform, anteroposteriorly compressed, with anterior face relatively straight; petiole in dorsal view between 1.1 and 1.2 times wider than long (DPeI 116); ventral process of petiole well developed, lamelliform and rectangular, lamella not pointed anteriorly nor posteriorly; abdominal segment IV around 1.1 times longer than abdominal segment III (ASI 108); head, mesosoma and petiole with numerous long, fine, suberect to erect hairs on top of dense mat of much shorter decumbent to subdecumbent pubescence.
(N=1). TL 3.35; EL 0.04; SL 0.55; HL 0.81; HLM 0.94; HW 0.76; WL 1.00; HFeL 0.62; HTiL 0.53; HBaL 0.41; PeL 0.31; PeW 0.36; DPeI 116; LT3 0.53; LS4 0.19; LT4 0.57; OI 5; CI 94; SI 68; IGR 0.33; ASI 108.
In full-face view head slightly longer than broad (CI 94), sides and vertex moderately convex. Clypeus medially reduced, its anterior margin convex to slightly triangular, only slightly protruding anteriorly, not surrounding the antennal sockets and not medially impressed, antennal socket with broad torulus. Frontal carinae relatively short and widely separated, not converging medially and strongly diverging posteriorly, partially covering antennal insertions; frontal carinae conspicuously raised on their anterior half, much less posteriorly. Eyes very small, consisting of three to four faint ommatidia (OI 5) and located on mid line of head. Mandibles elongate-triangular; masticatory margin of mandibles with four relatively small teeth/denticles, decreasing in size from larger apical tooth to basal denticle. Mesosoma weakly to moderately convex in profile and weakly longer than maximum head length including mandibles. Lower mesopleurae with well impressed sutures, no other sutures developed on lateral or dorsal mesosoma; mesopleurae extremely inflated posteriorly; propodeum in profile armed with very small, pointed teeth, propodeal lobes well developed, triangular and blunt; declivitous face of propodeum between teeth and lobes noticeably concave; in posterodorsal view sides of propodeum separated from declivitous face by margin connecting propodeal lobes and propodeal teeth. Legs slender and elongate; pro- and mesotibiae with pectinate spurs; calcar of strigil without basal spine. Petiolar node in profile high, blocky nodiform, anterior face of petiole relatively straight, anterior and posterior faces approximately parallel, dorsum of node flat to weakly convex; petiole in dorsal view between 1.1 and 1.2 times wider than long (DPeI 116), petiolar node in dorsal view clearly much broader than long; ventral process of petiole well developed, lamelliform and rectangular, lamella not pointed anteriorly nor posteriorly. In dorsal view abdominal segment III anteriorly broader than petiole; its sides diverging posteriorly; dorsum of abdominal tergum III with posteromedial, very conspicuous, semitransparent, flat bulla below the integument; abdominal sternite III anteromedially with a marked subtriangular projection. Constriction between abdominal segment III and IV conspicuously impressed. Abdominal segment IV strongly recurved (IGR 0.33), conspicuously rounded on its curvature, especially posteriorly, abdominal tergum IV only less than 1.2 times longer than abdominal segment III (ASI 116); large, semitransparent and semicircular bulla situated posteromedially on abdominal tergum IV; remaining abdominal tergites and sternites relatively inconspicuous and curved ventrally. Whole body covered with dense mat of relatively short, decumbent to subdecumbent pubescence; most of body with moderately abundant, much longer (several times longer than pubescence), suberect to erect, fine, standing hairs. Mandibles longitudinally rugose; most of body irregularly foveolate and/or granulate, sculpture best developed on cephalic and mesosomal dorsum, less so on mesosoma and especially weak on most of relatively smooth and shining abdominal tergum IV; inflated, posterior part of mesopleura and declivitous face of propodeum unsculptured, relatively smooth and shining. Head, mesosoma (excluding posteriorly inflated part of mesopleurae), postpetiole and remaining abdominal segments of brown colour, mandibles, inflated part of mesopleurae and legs yellowish to light brown.
Proceratium sali sp. n. holotype worker (CASENT0235689) (Will Ericson 2011). A Body in profile B Body in dorsal view C head in full-face view.
The name of the new species is derived from the type locality, the Sali Forest Reserve in Tanzania. The species epithet is a noun in apposition and thus invariant.
Proceratium sali is only known from the Sali Forest Reserve in the Morogoro region of south-central Tanzania (Fig.
Proceratium sali shares a thicker head in full-face view (CI 94) with P. burundense, P. lunatum, P. nilo and P. sokoke (CI 91–95), which contrasts with the thinner head of P. arnoldi and P. carri (85–87). In addition, P. sali (as well as P. carri, P. nilo, P. sali and P. sokoke) possesses numerous long, fine, standing hairs on top of a mat of much shorter pubescence distinguishing it from P. arnoldi, P. lunatum and P. burundense that lack this type of pilosity. The two species most similar to P. sali are P. nilo and P. sokoke, but the latter two have a lower, less compressed petiolar node with the anterior face strongly produced anteriorly on lower third. This contrasts strongly with the node of P. sali that is high nodiform and more compressed with a straight anterior face.
As for P. burundense, P. nilo and P. sokoke, P. sali is also only known from a singleton holotype, which does not allow any conclusions on intraspecific variation.
Holotype, pinned worker, KENYA, Coastal Province, Arabuko Sokoke Forest, 18°51'72"S, 39°56'26.6"E, 136 m, undisturbed and protected mixed forest, Winkler leaf litter extraction, collection code FHG00206, VI.2009 (F. Hita Garcia & G. Fischer) (MCZ: MCZ-ENT00520482).
Proceratium sokoke differs from the other Afrotropical members of the P. arnoldi clade by the following character combination: eyes strongly reduced, consisting of a single ommatidium (OI 4); CI 92; maculae on vertexal angles of head well developed and conspicuous; petiolar node in profile relatively low, bluntly rounded nodiform, anterior face of petiole strongly produced anteriorly on lower third and not straight; petiole in dorsal view between 1.1 to 1.2 times wider than long (DPeI 115); ventral process of petiole well developed, lamelliform and rectangular, lamella not pointed anteriorly nor posteriorly; abdominal segment IV around 1.25 times longer than abdominal segment III (ASI 125); head, mesosoma and petiole with numerous long, fine, suberect to erect hairs on top of dense matt of much shorter decumbent to subdecumbent pubescence.
(N=1). TL 2.47; EL 0.03; SL n.a.; HL 0.72; HLM 0.87; HW 0.66; WL 0.86; HFeL n.a.; HTiL n.a.; HBaL n.a.; PeL 0.30; PeW 0.35; DPeI 115; LT3 0.44; LS4 0.14; LT4 0.55; OI 4; CI 92; SI n.a.; IGR 0.25; ASI 125.
[Note: the holotype is partly damaged: antennae, one foreleg, one midleg and one hindleg missing, remaining hindleg broken at level of tibia].
Head longer than broad (CI 92), sides weakly convex, cephalic dorsum broader posteriorly than anteriorly; vertex in full-face view flat to weakly convex. Clypeus medially reduced, its anterior margin convex to slightly triangular, only slightly protruding anteriorly, not surrounding the antennal sockets and not medially impressed, antennal socket with broad torulus. Frontal carinae relatively short and widely separated, not converging medially and strongly diverging posteriorly, partially covering antennal insertions; frontal carinae conspicuously raised on their anterior two thirds, much less posteriorly. Eyes very small (OI 4), consisting only of one ommatidium and located on mid line of head. Mandibles elongate-triangular; masticatory margin of mandibles with four relatively small teeth/denticles, decreasing in size from larger apical tooth to basal denticle. Mesosoma weakly to moderately convex in profile and approximately as long as the maximum head length including mandibles. Lower mesopleurae with well impressed sutures, no other sutures developed on lateral or dorsal mesosoma; mesopleurae extremely inflated posteriorly; propodeum in profile armed with small, pointed teeth, propodeal lobes well developed, lamellate, rounded and blunt; declivitous face of propodeum between teeth and lobes noticeably concave; in posterodorsal view sides of propodeum separated from declivitous face by margin connecting propodeal lobes and propodeal teeth. Legs slender and elongate; pro- and mesotibiae with pectinate spurs; calcar of strigil without basal spine. Petiolar node in profile relatively low, bluntly rounded nodiform, anterior face of petiole strongly produced anteriorly on lower third and not straight, posterior face approximately straight, anterior and posterior faces not parallel, dorsum of node weakly rounded; petiole in dorsal view between 1.1 to 1.2 times wider than long (DPeI 115), petiolar node in dorsal view clearly much broader than long; ventral process of petiole well developed, lamelliform and rectangular, lamella not pointed anteriorly nor posteriorly. In dorsal view abdominal segment III anteriorly broader than petiole; its sides diverging posteriorly; dorsum of abdominal tergum III with posteromedial, very conspicuous, semitransparent, flat bulla below the integument; abdominal sternite III anteromedially with a marked subtriangular projection. Constriction between abdominal segment III and IV conspicuously impressed. Abdominal segment IV strongly recurved (IGR 0.25), conspicuously rounded on its curvature, especially posteriorly, abdominal tergum IV around 1.25 times longer than abdominal segment III (ASI 125); large, semitransparent and semicircular bulla situated posteromedially on abdominal tergum IV; remaining abdominal tergites and sternites relatively inconspicuous and curved ventrally. Whole body covered with dense matt of relatively short, decumbent to subdecumbent pubescence; most of body with moderately abundant, much longer (several times longer than pubescence), suberect to erect, long, fine, standing hairs. Mandibles longitudinally rugose; most of body irregularly foveolate and/or punctate, sculpture best developed on cephalic dorsum, less so on mesosoma and especially weak on most of relatively smooth and shining abdominal tergum IV; inflated, posterior part of mesopleura and declivitous face of propodeum also only very weakly sculptured and relatively smooth and shining. Head, mesosoma (excluding posteriorly inflated part of mesopleurae), postpetiole and remaining abdominal segments of brown colour, mandibles, inflated part of mesopleurae and legs yellowish to light brown.
The name of the new species is inspired by the type locality, the Arabuko Sokoke Forest in Coastal Kenya. The forest is the last larger remnant of the Coastal Forests of Eastern Africa in Kenya and hosts a unique fauna and flora. The species epithet is a noun in apposition and thus invariant.
Proceratium sokoke is only known from the type locality, the Arabuko Sokoke Forest in Kenya, which is a tropical dry forest adjacent to the Indian Ocean coast (Fig.
Map of sub-Saharan Africa showing the known distribution ranges of the seven Afrotropical members of the P. arnoldi clade: P. arnoldi (empty star), P. burundense (filled star), P. carri (filled circle), P. lunatum (empty circle), P. nilo (filled square), P. sali (empty square), P. sokoke (empty hexagon).
The identification of P. sokoke is straightforward within the P. arnoldi clade. Proceratium nilo and P. sokoke are the only species of the P. arnoldi clade in which the petiolar node in profile does not have a straight anterior face; instead the lower third is produced anteriorly. In the other five species the anterior face is conspicuously straight. Proceratium nilo is relatively similar to P. sokoke, but the latter has eyes that are absent in the first. Not considering presence/absence of eyes, both species could be seen as conspecific. As discussed in detail in the taxonomic notes section for P. nilo, we prefer to treat them as heterospecific in this study since eye development appears to be relatively stable within species of Proceratium. In addition to petiolar node shape, P. sokoke (CI 92) has a thicker head than P. arnoldi and P. carri (CI 85–87), and its smaller eyes (O 4) and the rectangular ventral process of the petiole distinguish it from P. burundense with its larger eyes (OI 8) and ventral process with an almost spiniform posteroventral corner. Furthermore, the presence of numerous long, fine, suberect to erect hairs on top of a dense mat of much shorter decumbent to subdecumbent pubescence is an additional character that separates P. sokoke from P. arnoldi, P. burundense and P. lunatum.
In general it is not recommendable to describe a new species based on a damaged singleton holotype. However, after detailed examination of all the material collected from Arabuko Sokoke, there was no other specimen to be found. In addition, we think that P. sokoke is a fairly distinct member of the P. arnoldi clade and even without antennae and the missing three and half legs, it can be easily separated from the remainder of the group. Consequently, we prefer to make the species available to science now than to await the discovery of additional material.
Since the species is only known from the holotype, no information about intraspecific variation exists.
The following museum curators or curatorial technicians provided important material or welcomed the authors to their collections: Mrs Suzanne Ryder and Mrs Natalie Dale-Skey Papilloud from BMNH; Dr Brian L. Fisher from CASC; Dr Henri-Pierre Aberlenc and Mrs Emma Artige from CIRAD; Mr Stefan Cover from MCZ; Dr Giulio Cuccodoro from MHNG; and Mrs Isabelle Zürcher-Pfander from NHMB. GDA is indebted to E.O. Wilson for making the fieldwork in Mozambique possible and including him in the Gorongosa Ant Project. Thanks also to Vasco Galante for logistics and transporation to remote field sites. We are thankful to Patrick McCormack (MCZ) for his support with imaging, databasing and commenting on the manuscript. GDA’s field work in Gorongosa would not have been possible without Marc Stalmans and Mateus Mutemba who organised the collection permits. In addition, GDA was financially supported by the E.O. Wilson Foundation and Carr Foundation to perform the field work in Gorongosa. FHG is thankful to Georg Fischer (Okinawa Institute of Science & Technology, Japan) for his support and collaboration in collecting and curating Kenya ants from Arabuko Sokoke and other localities. FHG also wants to thank Marcell Peters (University of Würzburg, Germany), the BIOTA East Africa project funded by the German Ministry of Education and Research (BMBF), the National Museums of Kenya and the Kenya Wildlife Service for organising the study in Arabuko Sokoke, as well as for granting collecting and export permits. FHG was granted two Ernst Mayr Travel Grants in Animal Systematics from the MCZ to visit the collections at BMNH and MCZ. The surveys during which the P. nilo and P. sali specimens were collected formed part of a project run by PGH investigating ground-living ant and beetle diversity in Tanzania, funded by the Critical Ecosystem Partnership Fund (CEPF), which is a joint initiative of l’Agence Française de Développement, Conservation International, the Global Environment Facility, the Government of Japan, the MacArthur Foundation and the World Bank. A fundamental goal of CEPF is to ensure civil society is engaged in biodiversity conservation. Funding from the CEPF is gratefully acknowledged. Thanks are also due to Joas Makwati, Renald Mtana, Upendo Richard and Munanka Bhoke, whose assistance with the fieldwork in Tanzania was invaluable, as well as Dr Bruno Nyundo of the University of Dar es Salaam for logistical assistance. Sample collection in Tanzania was carried out under COSTECH permit numbers 2005-269-NA-2005-57 and 2007-101-ER-2005-57 and the specimens exported under Tanzanian Ministry of Wildlife trophy export certificate numbers 48157 and 53211. PGH would also like to thank the staff of Baobab Resources Plc for logistical support during surveys of haul road alternatives for the Tete Pig Iron & Ferro-Vanadium Project in the Tete Region of Mozambique, and especially to Elisa Vicente for her very able assistance with export permits.