Conceived and designed the experiments: MY BLF. Performed the experiments: MY. Analyzed the data: MY BLF. Contributed reagents/materials/analysis tools: BLF MY. Wrote the paper: MY BLF.
The authors have declared that no competing interests exist.
In a male-based revision of ants of the subfamily Amblyoponinae from the Southwest Indian Ocean islands (SWIO: Comoros, Madagascar, Mauritius, Mayotte, Reunion, and Seychelles), we explore and reconsider male morphological characters that distinguish genera within the group. Our investigation redefines
Male ants are a largely untapped resource for understanding the taxonomy, phylogeny, diversity, and biology of this important insect group. Although ants are known for having distinctive sexual dimorphism, the current taxonomy of ants is for the most part based on the morphology of female workers. Males are difficult to study because they are often characterized by short emergence periods at certain times of the year, which reduces their chances of capture. Male ant morphology can be equally valuable for identifying species, and among some groups can be even more effective than female traits for distinguishing between genera or species
In the Malagasy region, we aim to complete a male-based comparative study of the major ant lineages. We have previously published male-based keys to genera and generic diagnoses for Ponerinae
A few generic keys and synopses for males of Amblyoponinae were found among previous studies. However, no existing key covers the genera in the Malagasy region, and morphological information in those studies was not sufficient to diagnose differences among the Malagasy amblyoponine genera. Emery
In recent years, molecular phylogenetic analyses have suggested new evolutionary relationships among ant genera and subfamilies, and synergies between molecular and morphological analyses promise to clarify the evolutionary development of ants as a group. Several large-scale studies aiming to clarify the trajectory of global ant evolution
1A,
In this study, we provided a key and diagnoses using reconsidered generic characters for five genera in the Malagasy region. Based on the results of a comparative study, we propose resurrecting two names,
Materials for this work were collected during arthropod surveys in Madagascar and nearby islands in the Southwest Indian Ocean conducted by B. Fisher and Malagasy ant researchers from the Madagascar Biodiversity Center in Antananarivo, Madagascar. Their work in the region includes more than 6,000 leaf litter samples, 4,000 pitfall traps, 1,000 Malaise trap collections, and 9,000 additional hand collection events throughout Madagascar from 1992 through 2009 (see Fisher
Male specimens were primarily collected with Malaise traps. Within each amblyoponine genus, specimens were sorted to morphospecies. For those species that could not be named, morphospecies codes were applied. The codes consist of a two-letter country code followed by a number, e.g.
Observations and dissections were carried out under stereoscopic microscopes (LEICA MZ12 and M125). Digital color images were created using a JVC KY-F75 digital camera. Syncroscopy Auto-Montage (v 5.0) software was used for images taken at magnifications less than 100×, and a compound microscope (Leica DM4000M) and Nikon digital camera (DXm1200) Helicon Focus version 4.10.2 software were used for images taken at magnifications greater than 100×. The images were edited in Adobe Photoshop and Illustrator. Each imaged or dissected specimen is uniquely identified with a specimen-level identifier (e.g. CASENT0003099) affixed to each pin.
The male and worker specimens listed below were examined to establish a key to genera of the subfamily. A diagnosis and analysis of morphological evolution is provided for each. Taxon names are followed by a letter code indicating the source of morphological information used to establish the key, and a CASENT specimen identifier of the dissected specimens:
[g]: male specimens that were collected from a colony and associated with workers.
[m]: male specimens that were collected alone and not associated with workers, typically in Malaise traps.
[w]: worker specimen.
[r]: information about male morphologies obtained from published studies. In these cases, the references are shown in brackets.
Only genus rank information was available [r: Emery 1911
In addition to above, several undetermined male specimens were available from Central African Republic (
Overall, our preference is for terms used generally in Hymenoptera over terms uniquely applied within Formicidae for homologous characters. Morphological terminology follows our previous work (
2A,
3A,
4,
5A,
6A,
7A,
8A,
9A,
10A,
11A,
12A,
13A,
14A–C,
In this study, tergosternal fusion is considered to occur when tergite and sternite are fused to each other in a manner different from the fusion seen in the proceeding posterior segment. The degree of fusion in males seems to be highly variable, with many intermediates. According to Bolton
We prefer the term abdominal sternum IX rather than the subgenital plate and the hypopygium, the terms used for this sclerite in previous studies, because the latter two terms are ambiguous in homology; the subgenital plate and the hypopygium are not homologous in males and females of the same taxon. Even within the same sex, the subgenital plate is not consistent between taxa. (More detailed information is provided in the terminology section and table 1 of Yoshimura & Fisher
Characters | 01 | 02 | 03 | 04 | 05 | 06 | 07 | 08 | 09 | 10 | 11 | 12 | 13 | 14 | 15 |
|
1(9) | 1(9) | 1(9) | 2(2)/3(7) | 2(1)/3(8) | 13(9) | 1(8)/0(1) | 1(5)/0(3) | 0(9) | 2(9) | 2(9) | 1(9) | 1(9) | 0(9) | 1(3)/0(5) |
|
0(12) | 1(11) | 1(11) | 4(10) | 3(9) | 13(12) | 1(8)/0(3) | 0(12) | 0(8) | 2(2)/1(9) | 2(12) | 0(12) | 0(12) | 0(8) | 0(8) |
|
0(10) | 1(9) | 0(8) | 2(7) | 2(7) | 13(9) | 0(9) | 1(8) | 0(6) | 1(8) | 1(9) | 0(9) | 0(10) | 0(6) | 0(6) |
|
1(5) | 1(5) | 1(5) | 4(4)/3(1) | 3(2)/2(3) | 13(5) | 0(5) | 0(4) | 0(5) | 2(2)/1(3) | 2(5) | 0(5) | 0(5) | 0(5) | 0(5) |
|
1(9) | 0(9) | 1(9) | 4(2)/3(7) | 3(4)/2(5) | 13(9) | 0(9) | 0(9) | 0(6) | 1(1)/0(8) | 2(9) | 0(9) | 0(9) | 0(6) | 0(6) |
Characters | 16 | 17 | 18 | 19 | 20 | 21 | 22 | 23 | 24 | 25 | 26 | 27 | 28 | 29 | 30 |
|
0(9) | 1(9) | 1(2)/0(7) | 1(9) | 1(9) | 2(7)/0(1) | 0(9) | 1(9) | 1(9) | 0(9) | 1(9) | 0(9) | 1(9) | 1(3)/0(6) | 1(6)/1(3) |
|
1(12) | 0(8) | 1(8) | 1(8) | 0(8) | 1(8) | 0(8) | 0(8) | 0(8) | 0(8) | 0(8) | 1(4)/0(4) | 0(8) | 0(8) | 0(8) |
|
0(9) | 0(6) | 1(6) | 1(3)/0(3) | 0(6) | 0(6) | 1(6) | 1(6) | 0(6) | 1(6) | 0(6) | 0(6) | 0(6) | 0(6) | 1(6) |
|
0(5) | 1(2)/0(3) | 1(4)/0(1) | 1(5) | 0(5) | 0(5) | 0(5) | 0(5) | 0(5) | 0(5) | 0(5) | 1(2)/0(3) | 0(5) | 0(5) | 1(2)/0(3) |
|
1(9) | 0(6) | 1(6) | 1(6) | 0(6) | 0(6) | 0(6) | 1(6) | 0(6) | 0(7) | 0(6) | 1(1)/0(5) | 1(6) | 1(7) | 1(7) |
For the 30 characters which seem useful for distinguishing among amblyoponine genera, character states are shown 0, 1, 2, exact number, or as 0/1 (if both states 0 and 1 were observed for each genus). The number of species in which the character states were observed is given in parentheses following the character state. Character states have been confirmed by direct observation or by dissection.
Pygostyle is used to refer to the pair of appendages on tergum X of the abdomen of male Hymenoptera. Cerci, on the other hand, should refer to appendages on abdominal tergum XI, not X (see also table 1 in Yoshimura & Fisher
We use veins, not cells, to describe wing characters and to discuss the homology of wing characters between amblyoponine genera and genera in other subfamilies. In previous studies of Amblyoponinae
In the present study, the homology of each vein across taxa was determined based on comparative studies of taxa with well-developed veins, such as species in the genus
The names of wing cells in ants have been inaccurately interpreted due to homonymy. Submarginal cells 1 and 2 in
Serna
A detailed morphological examination was carried out for 30 male characters. Some characters were chosen based on their relevance in previous studies: e.g., the presence of the pygostyles as cerci
Characters plotted on the topology of the XMAS presented in Brady
01. Frontal carinae are absent (1); present (0)
02. Dent-like projections are present on anterior margin of the clypeus (1); no such projections on the margin (0)
03. The number of mandibular teeth (
04. The number of maxillary palpal segments (
05. The number of labial palpal segments (
06. The number of antennal segments
07. Notaulus absent on the mesoscutum (1: as in
08. Epimeral lobe absent (1: as in
09. Metacoxal cavity completely fused (1); touches but is not fused (0)
10. The number of mesotibial spurs
11. The number of metatibial spurs (
12. In dorsal view, constriction between AII (petiole: the second abdominal segment) and AIII absent (1: as in
13. A transverse furrow dividing pre- and post-sclerite on AIV absent (1: as in
14. Tergosternal fusion not found on AIII (1); found on AIII (0)
15. Tergosternal fusion not found on AIV (1); found on AIV (0)
16. Pygostyles absent (1); present (0: as in
17. Distal margin of abdominal sternum IX concave (1: as in
18. Harpago separated from the basimere by a suture (1); separation between the harpago and basimere unclear (0)
19. Basal projection on the cuspis present (1: as in
20. Ventral serrate teeth on the aedeagus absent (1: as in
21. Basicoventral portion of aedeagus sharply extended ventrally (2: as in
22. On the forewing, the pterostigma is reduced in size (1: as in
23. On the forewing, radial sector is wholly or partially absent between M+Rs and 2r-rs (1: as in
24. On the forewing, radial sector is not connected with radius (1: as in
25. On the forewing, 2r-rs is connected with radial sector far distal from the pterostigma (1: as in
26. On the forewing, 2rs-m is absent (1: as in
27. On the forewing, cu-a is located at or close to a junction of media and cubitus (0: as in
28. On the hindwing, 1rs-m is absent (1: as in
29. On the hindwing, media is absent between 1rs-m and cubitus (1: as in
30. On the hindwing, media is absent apical to 1rs-m (1: as in
The electronic version of this document does not represent a published work according to the International Code of Zoological Nomenclature (ICZN), hence the nomenclatural acts contained in the electronic version are not available under that Code from the electronic edition
Results of the detailed morphological examination of males of five Malagasy genera
Males alate. Scape not reaching posterior margin of head (
Venation on forewing and hindwing varies. On forewing (
We propose that the process on the basal portion of the cuspis on the volsella (
The above character, which is difficult to confirm without dissection, is the only one unique to the subfamily Amblyoponinae. However, combinations of characters under diagnosis clearly separate the Amblyoponinae from the six other subfamilies in the Malagasy region. In the Malagasy region, Amblyoponinae differ from Cerapachyinae in lacking a bispinose abdominal sternum IX (
Amblyoponinae males have a broad, dorsal attachment of the petiole to abdominal segment III. However, this character is sometimes difficult to differentiate from the attachment in Proceratiinae, which may also appear broadly attached. The differences in the shape of abdominal segment IV mentioned above, however, will easily separate the two subfamilies.
This key may not apply outside of the Malagasy region, as variations in genus-level characters elsewhere have not been fully explored.
1. A single tibial spur present on hind leg (
-. Two tibial spurs present on hind leg (
2. Constriction between petiole and abdominal segment III indistinct in dorsal view (
-. Constriction between petiole and abdominal segment III distinct in dorsal view (
3. Pygostyles present (
-. Pygostyles absent… 4
4. Anterior margin of clypeus with dent-like projections (
-. Anterior margin of clypeus without dent-like projections (
Descriptions and diagnoses apply to species found in Madagascar. Diagnostic characters uniquely observed in each genus are given in italics. Male diagnoses for the genus
With characters of Amblyoponinae. Frontal carinae absent. Anterior margin of clypeus with dent-like projections. Antenna consisting of 13 segments. Mandible with single, blunt apical tooth (
Distal margin of abdominal sternum IX concave (
On forewing (
In full-face view, head wider than long when eyes included, ocelli well-developed, eyes well-developed and protruding laterally, situated on middle to anterior portion of lateral margin of head (as in
Mesopleural oblique furrow very weak or absent, reaching anterior margin of mesopleuron far ventrally from posteroventral corner of pronotum when the furrow is visible (
Subpetiolar process developed to various degrees, absent in several species. The distal margin of abdominal tergum VIII relatively flat, not strongly protruding on middle portion.
Body sculpture weak. Body color yellow to blackish brown.
A male synopsis of the genus
Other than characters uniquely observed in
In addition to the separable characters above, two additional characters can be useful to distinguish
New palpal formulae in the genus
With characters of Amblyoponinae. Frontal carinae present. Anterior margin of clypeus with dent-like projections. Antenna consisting of 13 segments. Mandible with single, blunt, apical tooth (
Distal margin of abdominal sternum IX convex (
On forewing (
The only unique character of males of
In practice, males of
In addition to a unique character in the aedeagus, the presence of clypeal dent-like projections, a palpal formula of 4,3, and a distinct epimeral lobe are provided as new diagnostic characters.
Several diagnostic characters provided in previous studies require updating. First, Emery described the notaulus in
With characters of Amblyoponinae. Frontal carinae present. Anterior margin of clypeus with dent-like projections. Antenna consisting of 13 segments.
Distal margin of abdominal sternum IX convex (
On forewing (
Males of
We found useful male-based generic characters of
Of these characters, the two-toothed mandible (Character 3) may not separate
Reduction in size of the basal projection on the cuspis is observed only in
16A, 16-a1-a5, Malagasy
With characters of Amblyoponinae. Frontal carinae absent. Anterior margin of clypeus with dent-like projections. Antenna consisting of 13 segments. Mandible with single, sharp, apical tooth (
Distal margin of abdominal sternum IX concave or convex (
On forewing (
We resurrect
Previous studies most often include diagnostic characters for
The following names are transferred from
The following names are transferred from
The following species, which are all restricted to Australia, New Caledonia, New Guinea, and New Zealand, remain in
With characters of Amblyoponinae. Frontal carinae absent. Anterior margin of clypeus flat,
Distal margin of abdominal sternum IX convex (
On forewing (
We resurrect
Few morphological discussions exist regarding the taxonomic status of the name
We decided that the male characters proposed above and associations between the worker lectotype and our materials are sufficient to warrant designating
A single species name,
In this study, we investigated 30 male morphological characters in five amblyoponine genera in the Malagasy region. These characters were sufficient to develop diagnoses for the genera within the Malagasy region. Ten out of 30 characters had genus-specific character states, while three characters had character states that were invariant across all Malagasy amblyoponines. Of the remaining 17 characters, character states are shared among more than one genus. It is of interest to evaluate this pattern of morphological variation in an evolutionary context using trees from recent molecular phylogenetic studies
To explore character evolution, we chose to map our data on the Brady
We focused on mapping male characters among four genera present in Madagascar:
A lack of the pygostyle (character 16) is shared by both
No morphological characters have been proposed as synapomorphies for the OCP and XMAS clades. Our morphological study provides a chance to investigate morphological characters that may support the XMAS clade. Our results suggest that the number of teeth on the mandible of the male is a unique character that defines the XMAS clade. The character state single-tooth mandible (character 03) is shared among males in
To determine whether or not this clade-defining character state holds up outside the Malagasy region, we investigated additional genera and
We also predict that males of the genus
This difference in mandibular dentition in males in the OCP and XMAS clades is also reflected in the morphology of their respective worker castes. A comparison of the mandibles of workers of Malagasy
We propose that the both SS and AD forms developed to hold larger prey, though each acquired this function in a different way as described in the mandible transformations series in
Transformation of worker mandible of Malagasy
Apical teeth curve inwards, such that they remain opposite when mandibles are open (
Mandibles become narrower, and basal denticles extend towards apex, below masticatory teeth (
Teeth are reduced in size and, at times, form bifurcating teeth when adjacent to basal denticles (
The basal-most of the basal denticles (the basal projection) becomes larger (
The basal denticles on the masticatory margin are larger than true masticatory teeth because the basal projections, and not the teeth, are used to hold prey.
Transformation of worker mandible of
Apical teeth do not curve inward and are not opposite when mandibles are open (
Mandibles become narrower, the basal denticles do not extend toward apex and remain at base of mandible (
Some teeth become larger, never fusing with basal denticles to form bifurcated teeth (
Across the Amblyoponinae, the number of teeth on the male mandible is correlated with the number of major teeth on the masticatory margin of conspecific workers. The male mandible in Amblyoponinae can be interpreted as a “reduced worker mandible” with the following features in comparison with conspecific workers: (a) smaller in length; (b) disappearance of the basal denticles; and (c) disappearance of the masticatory teeth except those which are well-developed on the worker mandible (
The point on the worker mandible used to hold prey moves distally as this transformation occurs in the AD form. The form of the anterior clypeal margin in workers of
In addition, we found a worker character uniquely observed in Malagasy
Worker mandibles similar to the SS form were also observed in workers of
The mandibular transformations in these two lineages must have occurred after the development of two characters shared between the groups. Amblyoponinae have distinctive anterior clypeal conical setae, which have transformed into a plate-like projection in
Outside the XMAS (
Advances in molecular ant systematics are creating a new opportunity and demand for morphological studies. Molecular studies will never be able to sequence all species and lineages, and therefore a need to discover morphological characters to diagnose molecular based clades remains. This study demonstrates that males have characters rich and useful for advancing our understanding of ant systematics. For example, the pattern of the number of mandibular teeth in males presents us with a possible scenario for worker mandible evolution that involves the parallel evolution of elongate worker mandibles in Amblyoponinae. We predict that
We would like to acknowledge Isabelle Zürcher-Pfander and Daniel Burckhardt (NHMB) for the loan of type materials, Tamas Szuts, Barry Bolton, Philip S. Ward, Doug Yanega, Francisco Hita Garcia, Robert Zuparko, and Steve Savage for their useful suggestions, Erin Prado and April Nobile for creating images, Michele Esposito for help with database management and for the arrangement of images, and Charles Griswold for supporting the use of the compound microscope imaging system. The fieldwork on which this study is based could not have been completed without the gracious support of the Malagasy people and the Arthropod Inventory Team (Balsama Rajemison, Jean Claude Rakotonirina, Jean-Jacques Rafanomezantsoa, Chrislain Ranaivo, Coco Randriambololona, Hanitriniana Rasoazanamavo, Nicole Rasoamanana, Clavier Randrianandrasana, Valerie Rakotomalala, and Dimby Raharinjanahary).