William Morton Wheeler.
The Identity of the Ant Genera Gesomyrmex Mayr and Dimorphomyrmex Ernest Andre.
Psyche 36:1-12, 1929.
Full text (searchable PDF, 1036K)
Durable link: http://psyche.entclub.org/36/36-001.html
The following unprocessed text is extracted from the PDF file, and is likely to be both incomplete and full of errors. Please consult the PDF file for the complete article.
PSYCHE
VOL. XXXVI MARCH, 1929 No. 1
THE IDENTITY OF THE ANT GENERA GESOMYRMEX MAYR AND DIMORPHOMYRMEX
ERNEST ANDRE.1
During the examination of a considerable number of ants collected by Dr. L. G. K. Kalshoven in the teak-forests of Java, I have come upon an undescribed species of Gesomyrmex, represented by a graded series of workers, all taken from a single colony and demonstrating that the singular Formicine genera Gesomyrmex and Dimorphomyrmex are synony- mous. I have long surmised this identity, because there are no significant differences among the worker phases, except in the proportions of the head, eyes, clypeus, frontal carinae and mandibles. Even the antennae are the same and consist of only 8 joints in the worker, instead of 12, which is the number observed in the great majority of Formicidse. But hitherto so few specimens of the two genera have been seen and the sexual forms have been so imperfectly known that no definite statements could be made in regard to their affinities.
Gesomyrmex was established sixty years ago (1868) by Mayr for a peculiar, large-eyed ant, G. hoernesi, which he found in the Baltic amber, of Lower Oligocene Age. In 1892 Ernest Andre described two extant species of ants taken by Chaper in the Kapoewas Basin, North Borneo. One of them, which he called G. chaperi, was evidently very closely related to the amber form; for the other, which was larger and had a much larger and more rectangular head and smaller eyes, he erected a new genus, Dimorphomyrmex. This form, Dimorphomyrmex janeti, he described as possessing dimorphic Wontributions from the Entomological Laboratory of the Bussey Institution, Harvard University No. 308.
================================================================================
2 Psyche [March
workers, with heads of similar structure but of different sizes. Then Emery, in 1906, discovered in the Baltic amber a form which closely resembled the Bornean janeti .and described it as D. theryi. While reviewing the ants of the Baltic amber in 1913 I was able to recognize several speci- mens of G. hoernesi and D. theryi and two intermediate forms which I described as G. annectens and D. mayri. Two years later (1916) I described a winged female from the Island of Luzon, in the Philippines as Dimorphomyrmex luzonensis, and in 1921 another species, Gesomyrmex howardi, from a couple of workers taken by Prof. C. W. Howard near Canton, China. These workers I interpreted as major and minor workers respectively and noted that the former resembled the worker of Dimorphomyrmex. The new species from Java now shows that the specimens of howardi are really media and minima workers, that those of G. hoernesi and chaperi, described by Mayr and Andre are minimae, that those of D. janeti are maximae and mediae of G. chaperi, and that the workers of D. theryi, D. mayri and G. annectens are in all prob- ability the maximse, large mediae and small mediae respec- tively of G. hoernesi. This will be clear from the following description and figures of the new species brought to light by Dr. Kalshoven.
Gesomyrmex kalshoveni sp. nov.
Worker maxima. (Fig. 1, a-d) Length 5-6.6 mm. Head subrectangular, about 1 1/4 times as long as broad, slightly narrower in front than behind, with the sides dis- tinctly concave in front of the middle; the posterior border feebly concave, the posterior corners convex and rounded; the dorsal and gular surfaces rather flat. Eyes elongate, convex, scarcely reniform, with straight or very nearly straight internal and impressed posterior orbits, one-third as long as the head and well up on its dorsal surface. Ocelli minute, unpigmented, the anterior smaller than the two posterior, which are on a level with the posterior orbits of the eyes. Mandibles stout, convex, with strongly rounded external borders, 8-toothed, the first, second, fourth, sixth a,nd eighth tooth, counting from the apex, larger than the others and rather blunt. Clypeus short and flat, its anterior
================================================================================
1 The Ant Genera Gesomyrmex and Dimorphomyrmex 3 Fig. 1. a. Gesomyrmex kalshoveni sp. nov. Maxima worker in profile; b. head of same; c. petiole from front; d. head of small maxima; e. head of large media; j. of small media; g. of large minima; 12. of small minima; i. thorax and petiole of same; k. head of male semipupa; 1. petiole of G. howardi Wheeler from front; m. same in profile.
================================================================================
4 Psyche [March
border slightly and evenly rounded, scarcely projecting in the middle; its lateral portions not reaching the anterior corners of the head, convex but not auriculate. Frontal area distinct, moderately large and triangular; only a portion of the frontal groove developed in the middle of the head just anterior to the ocelli; frontal carinse widely separated, sub- parallel, rather erect and slightly lobular anteriorly, con- tinued back to the anterior orbits of the eyes. Antennae &jointed; scapes curved basally, gradually thickened dis- tally, their tips reaching somewhat beyond the middle of the eyes; funiculi distinctly enlarged apically, first joint twice as long as broad; second and third 1 1/2 times as long as broad, remaining joints nearly as broad as long, except the terminal joint which is somewhat longer. Thorax stout anteriorly, the pro- and mesonotum convex, forming together a hemispherical mass, separated by a distinctly impressed suture, rounded above in profile, the former as broad as long with a faint median longitudinal impression behind; the latter as long as broad, subcircular with straight posterior and rather deeply impressed anterior and lateral sutures. Behind the mesonotum the thorax is divided into two regions by a transverse dorsal furrow ; the shorter ante- rior portion, representing the metamturn is depressed and bears the somewhat projecting metanotal spiracles at its sides; the posterior portion, representing the epinotum, is as broad as long, with the base somewhat convex and bear- ing the epinotal spiracles at the middle of its sides. Pos- teriorly it passes into the distinctly shorter, sloping de- clivity, with which it forms a distinct, obtuse angle. Petiole small; its node from above transversely elliptical, nearly 1 213 times as broad as long; in profile distinctly cuneate, with convex anterior and posterior surfaces, the moderately sharp superior border transverse, feebly and broadly ex- cised or sinuate in the middle. Gaster large, elongate-ellip- tical. Legs stout; all the femora somewhat thickened; spurs of middle and hind tibiae very short; tarsal claws large and stout.
Smooth and shining; mandibles coarsely punctate, with a few incomplete longitudinal rugulea; head more hely punctate; median portion of clypeus, front and cheeks finely and longitudinally striate as far back as the eyes.
================================================================================
19291 The Ant Genera Gesomyrmex and Dimorphomyrmex 5 Thorax, petiole, gaster and appendages very minutely and sparsely punctate, the gaster also very delicately and in- distinctly, transversely coriaceous.
Hairs yellowish white, absent on the body, except at the tip of the gaster and on the mandibles, clypeus and front. On these latter regions they are numerous, short, erect, stubby and obtuse. Scapes with short oblique hairs. Pubes- cence undeveloped, except on the tibiae where it is very short, dilute and appressed.
Honey yellow; gaster and legs paler and more whitish yellow, the several gastric segments very faintly infuscated posteriorly; frontal carinse and mandibles red, the latter with black teeth and dental border; impressions of the pos- terior orbits dark brown; claws and tarsal articulations reddish.
Worker media. (Fig. 1 e and f) Length 3.5-4.5 mm. Differing from the maxima mainly in the characters of the head, which is proportionately smaller and only 1 1/6 to 1 1/5 times as long as broad; its posterior portion shorter and more rounded, its anterior portion more narrowed, with proportionally larger eyes, which are more convex, and reniform, i.e. with concave internal orbits and one half as long as the sides of the head. Posterior ocelli represented by minute pits in the larger, usually absent in the smaller specimens. Mandibles 7-8 toothed and more elongate and less convex than in the maxima, with less convex external borders and only the first, fourth and fifth tooth enlarged and more acute, the third, seventh and eighth very minute. Median portion of the clypeus longer, more convex pos- teriorly, its anterior portion more advanced and lobular, depressed. All the funicular joints longer than broad, except the penultimate. Thorax somewhat more slender than in the maxima, pronoturn without longitudinal impression; meso- notum longer than broad, with more nearly parallel sides; its surface flattened and even slightly concave in profile. Petiole as in the maxima, but slightly more narrowed above. In the larger media there are distinct striae and short stubby hairs on the front, though the latter are less nu- merous; in the small media these hairs are absent and the striae are very indistinct or obsolete. Color like that of the
================================================================================
6 Psyche \ [March
maxima, except that the mandibles are yellowish, with only the teeth blackish.
Worker minima. (Fig. 1 g-i) Length 2.8-3.2 mm. Differing from the media in having a still shorter head, in the smallest minima scarcely longer than broad including the eyes, which are more convex and laterally projecting, decidedly reniform and taking up more than 3/4 of the sides of the head, so that the cheeks and postocular regions are greatly reduced, the latter without distinct corners and with slightly and evenly convex median border. Clypeus much longer and anteriorly projecting, more convex behind and somewhat more concave in front, than in the media. Ocelli represented by a pair of very minute pits or completely absent. Mandibles more elongate, flattened, with straight external borders, the teeth forming- a more even crowded series, the apical longer than the basal. Frontal carinse very short and indistinct. Antenna1 scapes extending to the pos- terior third or fourth of the eyes. Thorax more slender, pro- and mesonotum much less convex, with nearly straight out- lines in profile, the latter feebly concave as in the media. Legs slender, but with the femora distinctly enlarged basally.
Mandibles pale yellow with reddish teeth ; posterior orbits not tinged with brown. In other respects like the small media. The larger minima is intermediate between the small minima and the small media even in the structure of the mandibular teeth.
Described from 15 workers taken by Dr. L. G. K. Kal- shoven in a teak-forest at Semarang, Java. The following are the lengths of the individual (alcoholic) specimens : Maximse : 6.6 mm. ; 5 :7 mm. ; 5.3 mm. ; 5 mm. Mediag: 4.5 mm.; 4 mm.; 4 mm.; 4 mm.; 3.8 mm.; 3.8 mm. ; 3.5 mm.; 3.5 m.m.
Minimse : 3.2 mm. ; 3 mm. ; 2.8 mm.
The variations in structure in all the castes are so finely graduated that there is no hiatus between the smallest specimen measuring 2.8 mm. and the largest measuring 6.6 mm.
================================================================================
19291 The Ant Genera Gesomyrmex and Dimo~phomyrmex 7 G. kalshoveni must be very close to the form described by Andre as G. chaperi from the minima and "D. janeti" from the maxima and media worker. He gives the length of the minima as 3.5 to 4 mm., of the media as 3.5 mm. and the soldier as 6 mm. If his descriptions were drawn from cabinet specimens the fact that the largest minima was longer than the media may be explained on the supposition that the gasters of the two specimens had contracted unequally on drying. Such differences are, of course, frequent also in the gasters of living or alcoholic specimens of the same caste, owing to variable distension of the crop. The maxima of kalshoveni differs from that of chaperi in lacking the lateral auriculate expansions of the clypeus, in possessing longer funicular joints and nonpigmented ocelli, a decidedly more convex pronotum and in having the stubby hairs on the front and clypeus simple and truncate instead of bifid or multifid at their tips. Andre describes these hairs as "ap- pearing as if surmounted or even crowned with minute spines." Emery has cited G. chaperi (as D. janeti) from Sumatra, but it is not improbable that his specimens be- longed to the Javan rather than to the Bornean species. The new Gesomyrmex differs from G. howardi in the shape of the petiolar node, which in the latter (Fig. 1, 1 and m) is much thicker above, with rounded, entire border, in the epinotum which is shorter in howardi, with much less convex base and with the base and declivity subequal. The sculpture of howardi is decidedly coarser, especially on the epinotum and pleurse, so that the surface is more opaque. The color is also darker and more brownish or sordid than in kalshoveni and chaperi. There are distinct strise on the front in the media and minima of howardi. It is, of course, not im- probable that all four living species of Gesomyrmex, in- cluding G. luzonensis, which is known only from the female, may prove to be merely so many local races (sub-species or varieties), when sufficient material of these forms has found its way into our collections.
It would be interesting to revise the fossil species of Gesomyrmex in the light of the preceding discussion. Of the material which I studied in 1913 and soon afterward re- turned to the Konigsberg museums, I retained only a single block of amber containing a large and a small worker of "D.
================================================================================
8 Psyche [March
theryi" side by side and designated as K. 6397, in my paper on the "Ants of the Baltic Amber" (1914). The large worker, which measures fully 7.5 mm., is very clear, but the head of the small worker, which measures only 4 mm., is obscured by a film and some bubbles of air. I have now had the block cut down and repolished, so that the head of the small individual is more clearly visible. It proves to belong to the form which I called D. mayri, and is really a media of the same species as the large (maxima) worker. The latter has small but very distinct ocelli, but none can be detected in the companion specimen. Since G. annectens is clearly inter- mediate between D. mayri and G. hoernesi, the former is, in all probability the large minima, whereas the latter represents the small minima of Mayr's species. I am, therefore, of the opinion that all the specimens of Dimorphomyrmex and Gesomyrmex in the Baltic amber belong to a single poly- morphic species, which should be known as G. hoernesi Mayr.1 This is of considerable interest, because it shows that as far back as the Lower Oligocene one Formicine ant had not only reached a high degree of specialization in the reduction of the number of antenna1 joints of the worker from 12 to 8, but that this caste had become as polymorphic as it is in the living species of the genus now confined to Indonesia and Southern China. And conversely, the very rare and sporadic occurrence of these forms shows that they are really living fossils which have undergone no significant modification since the Early Tertiary.
The workers collected by Dr. Kalshoven were accom- panied by several milk-white larvae of different sizes and a single semipupa. The larvae resemble those of other Formi- cinae in shape, but are almost hairless. The head is very small and subglobular, the mandibles minute, with only an apical tooth, which is drawn out into a slender, acute point. The semipupa measures a little over 7 mm. and is not enclosed in a cocoon. We must infer, therefore, that the pupae of Geso- myrmex are naked as in a few other genera of Formicinse. (CEkophylla, Prenolepis, Paratrechina). The semipupa is clearly a male. It has well-developed wing-pads and rudi- ments of the genital valves, and the imaginal head (Fig. 1, k) Compare my figures 50-53 in "The Ants of the Baltic Amber."
================================================================================
19291 The Ant Genera Gesomyrmex and Dimorphomyrmex 9 is sufficiently developed to enable me to determine its main peculiarities. As will be seen from the figure, it is broader than long, with evenly rounded, convex, postocular region; the eyes are very small, for a male, not longer than the cheeks, not very convex and situated at the sides of the head; the ocelli are moderately large but flat ; the antennal insertions are very far apart and near the anterior orbits; the antennal scapes are short and abruptly bent outwards at their tips (probably a pupal character), the funiculi short, thick and distinctly 6-jointed ( !) , instead of 7-jointed as in the worker; the clypeus is rounded and slightly pro- jecting in the middle ; the mandibles small, narrow, edentate and rather blunt. The specimen is of considerable interest because no other male of the genus is known, though Mayr carefully described and figured what he took to be the male of G. hoernesi from a rather poorly preserved specimen in the Baltic amber. Although we must allow in my description of the male semipupa of G. kalshoveni for later pupal changes (especially, perhaps, in the number of funicular joints), it is clear, nevertheless, that the male of this species is deci- dedly different from Mayr's specimen. This has enormous eyes, constituting the greater portion of the head, very minute pointed mandibles, very slender 11-jointed antennae and very minute genital appendages. Although the wing- venation agrees closely with that of the female G. luzonensis, I believe that Mayr's specimen must belong to some other hitherto unidentified Formicine genus.
Two other fossil ants supposed to be allied to Gesomyr- mex may be briefly considered in this connection. The first of these is a large-eyed worker which Emery described from the Sicilian amber of Miocene age, as Gesomyrmex corniger. His description and figures, however, show that this insect cannot be included in the genus Gesomyrmex. The shape of its head, mandibles and petiole, the presence of long oblique spines on the epinotum, the very long maxillary palpi, the abundant pilosity, peculiar rugosity of the head, pro-and mesonotum, the blackish coloration and the possibility that its antennae may be really 9 or 10 instead of %jointed, led me to place it in a distinct genus, Sicelomyrmex. Apart from the peculiar recurved horns at its posterior corners, the head of this ant is more suggestive of the Neotropical
================================================================================
10 Psyche [March
Gigantiops and the Congolese Santschiella than of Geso- myrmex. The other ant is Prodimorphomyrmex primigenius which I described from the Baltic amber. It approaches Gesomyrmex much more closely, but has smaller eyes, 10- jointed antennae and a shorter thorax. Only a single imper- fect specimen was observed.
The following would now be the corrected synonymy of the known species of the genera I have been considering: Living Species
Gesomyrmex chaperi Ern. Andre. Borneo, Sumatra. Gesomyrmex chaperi Ern. Andre, Mem. Soc. Zool. France 5, 1892 p. 47, Figs. 1-3, worker minima; Emery, Gen. Insect. Formicinae 1925, p. 47, PI. 2 Fig. 3, worker media. Dimorphomy~mex janeti Ern. Andre, Mem. Soc. Zool. France 5, p. 51 Figs. 4 and 5 ; worker maxima and media. ; Emery, Ann. Soc. Ent. Belg. 43, 1894, p. 494, worker maxima.; Emery, Gen. Insect. Formicinse 1925, p. 47, worker maxima and media.
Gesomyrmex luzonensis (Wheeler). Philippines. Dimorphomyrmex luzonensis Wheeler, Proc. New England Zool. Club. 6, 1916 p. 16, Fig. 4, female; Emery, Gen. Insect. Formicinse 1925, p. 47, female.
Gesomyrmex howardi Wheeler. China.
Gesomyrmex howardi Wheeler, Psyche 28, 1921 Fig. 2, worker media and minima; Emery, Gen. Insect. Formicinse 1925, p. 47, worker media and minima.
Gesomyrmex kalshoveni Wheeler. Java.
Described above p. 2 from the maxima, media and minima worker and male semipupa.
================================================================================
19291 The Ant Genera Gesomyrmex and Dimorphomyrmex 1 1 Extinct Species.
Gesomyrmex hoernesi Mayr. Baltic Amber.
Gesomyrmex hoernesi Mayr, Beitr. Naturk. Preuss. 1, 1868 p. 52, Fig. 38 to 41, worker minima; not the male; Dalla Torre, Catalog. Hymen. 7, 1893 p. 176; Ern. Andre, Bull. Soc. 2001. France 20,1895 p. 82 ; Handlirsch, Foss. Insekt. 1908 p. 859; Wheeler, Ants, etc. 1910 p. 170, Fig. 100; Wheeler, Schrift. physikal-okonom. Gesell. Konigsberg 55, 1914, p. 108, Fig. 53, worker minima. Dimorphomyrmex theryi Emery, Bull. Soc. Ent. France 1905 p. 188 Fig. 1, worker maxima; Handlirsch Foss. Insekt. 1908 p. 868, worker maxima; Wheeler, Ants, etc. 1910 p. 173, Fig. 98, worker maxima; Wheeler, Psyche 17, 1910 p. 132 ; Wheeler, Schrift. physical-okonom. Gesell. Konigsberg 55, 1914, p. 104, Fig. 50, worker maxima. Dimorphomyrmex mayri Wheeler, ibid. p. 106, Fig. 51, worker media (large). a
Gesomyrmex annectens Wheeler, ibid. p. 107, Fig. 52, worker media (small).
Prodimorphomyrmex primigenius Wheeler. Baltic amber. Prodimorphomyrmex primigenius Wheeler, Schrif t . physikal- okonom. Gesell. Konigsberg 55, 1914, p. 112, Fig. 54, worker.
Sicelomyrmex corniger (Emery) Sicilian amber. Gesomymex corniger Emery, Mem. 1st. Bologna (5) 1, 1891, p. 581, Figs. 33-35, worker; Wheeler, Ants, etc. 1910, Fig. 101, worker.
Sicelomyrmex corniger Wheeler, Schrift. physikal-okonom. Gesell. Konigsberg 55, 1914, p. Ill, worker ; Wheeler, The Social Insects, etc., 1928, Fig. 27, worker. The affinities of Gesomyrmex and Dimorphomyrmex to other Formicine genera have been discussed by Fore1 and Emery. In his revision of the subfamily Formicinse (1925),
================================================================================
12 Psyche [Match
Emery places the two genera in his sixth tribe, or first tribe of the section Eufodcinae, which he bases on characters drawn from the gizzard (proventriculus) . Unfortunately, he named this tribe Dimorphomyrmicini (his Dimorphomyrmii of 1895). Since the genus Dimorphomyrmex now passes into the synonymy, the tribe will have to be called Gesomyrmi- cini (Ashmead, 1905). This tribe, according- to Emery, would contain besides Gesomyrmex only the genera Brachyrnyr- mex, Aphomomyrmex and Ciadomyrma. The three latter are supposed to be so closely interrelated that they form to- gether the subtribe Brachymyrmicini, leaving Gesomyrmex to represent a subtribe by itself, to which Emery gave the same name as the tribe (Dimorphomyrmicini). He evidently based his tribe on the small number of antenna1 joints (8-9 in the worker, 8-10 in the female) instead of the 12 observed in both of these castes in all Eu-formicine genera, except certain fossils (Dryomyrmex and Prodimorphomyrmex) . But the general habitus of Gesomyrmex is so different from that of Brachymyrmex, Aphomomyrmex and Cladamyrma, that its relationships would seem to be better expressed by regarding it as the representative of an independent tribe (Gesomyrmicini). This would necessitate raising- Emery's subtribe Brachymyrmicini to tribal rank. I believe that the tribe Gesomyrmicini would thus be more naturally placed as one of the series of tribes, including the Santschiellini Forel, Gigantiopini Ashmead and CEcophyllini Ashmead, which are all based on single archaic, relict genera of large-eyed ants. The fossils Prodimorphomyrmex and Dryomynnex may, perhaps, be assigned to the Brachymyrmicini, but Skelo- myrmex comiger should be regarded as the representative of an extinct tribe, the Sicelomyrmicini.
We are still in the dark in regard to the habits of Geso- myrmex. The well-developed claws in the worker and female and the not infrequent occurrence of G. hoemesi in the Baltic
Volume 36 table of contents