(C) 2013 John T. Huber. This is an open access article distributed under the terms of the Creative Commons Attribution License 3.0 (CC-BY), which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
For reference, use of the paginated PDF or printed version of this article is recommended.
Citation: Huber JT (2013) Redescription of Mymarilla Westwood, new synonymies under Cremnomymar Ogloblin (Hymenoptera, Mymaridae) and discussion of unusual wings. ZooKeys 345: 47–72. doi: 10.3897/zookeys.345.6209
The monotypic genus Mymarilla Westwood is known only from St. Helena, a remote island in the South Atlantic Ocean. The peculiar species M. wollastoni Westwood (Mymaridae) is redescribed and illustrated from non-type material. Mymarilla is compared with Cremnomymar Ogloblinspp. from the Juan Fernández Islands in the South Pacific Ocean. Stephanodes Enock is shown to be the most likely sister genus to Mymarilla. Nesopolynema Ogloblin, syn. n., Oncomymar Ogloblin, syn. n., Scolopsopteron Ogloblin, syn. n., are placed in synonymy under Cremnomymar and their species transferred as Cremnomymar caudatum (Ogloblin 1952), comb. n., C. dipteron (Ogloblin 1957), comb. n., and C. kuscheli (Ogloblin 1952), comb. n. Wing shape and wing reductions in Mymaridae are discussed in relation to biogeography, particularly with respect island faunas and to four genera, Cremnomymar, Mymarilla, Parapolynema Fidalgo, and Richteria Girault, some or all of whose species have more or less convex fore wings.
Mymaridae, Mymarilla wollastoni, Cremnomymar, Richteria, Parapolynema, wing modifications, island faunas
The small, remote South Atlantic island of St. Helena contains a high proportion of endemic species but among the Hymenoptera only the Formicidae have been systematically studied (
Morphological terms follow
Specimens are in the following institutions.
BMNH The Natural History Museum (formerly British Museum [Natural History]), London, England (G. Broad).
CNC Canadian National Collection of Insects, Ottawa, Canada.
MRAC Musée Royal de l’Afrique Centrale, Tervuren, Belgium (E. de Coninck).
OUMNH Oxford University Museum of Natural History.
http://species-id.net/wiki/Mymarilla
Mymarilla wollastoni Westwood, by monotypy. Transferred (as genotype) to Mymarilla by
The confusion in the use of the name Mymarilla and which species should be placed in the genus continued for over 80 years.
The combination of smooth, shiny black body, extremely short mesocutum compared to much longer pronotum, and extraordinary convex and densely setose fore wing distinguish the genus and species from any other Mymaridae.
Mymarilla belongs clearly to the Polynema Halidaygroup of genera within Mymarini sensu
I propose instead that Mymarilla is derived from Stephanodes Enock, likely the most closely related genus. Four features, shared with Stephanodes, suggest this: first, the extremely smooth body without trace of microscupture on the mesosoma (Figs 1, 3, 4, 6, 9, 10); second, the slightly advanced mesothoracic spiracle about midway between the anterior apex of a notaulus and posterolateral angle of the mesoscutum (Figs 3, 4); third, the presence of a metapleural pit (Figs 9, 10); fourth, the fore wings that are held more or less horizontally. In dead specimens of Stephanodes, the fore wings are often horizontal, crossed scissor-like and covering the body, unlike other, related genera in the Polynema-group where the wings (in dead specimens) are almost always vertical, directed away from the body. The strong convexity of the fore wings of Mymarilla would appear to prevent them from being crossed scissor-like over the body. Yet they are presumably capable of enveloping the metasoma, as pointed out by Westwood who noted “... when shut [the fore wings] form a semiglobular dome over the abdomen” when the wasp is at rest. The densely hairy wing membrane with dark base around each microtrichia would allow for maximum heat absorption and retention.
Mymarilla wollastoni. 1 female dorsolateral 2 male antennae (pedicel + flagellum). Scale line = 1000 μm.
Mymarilla wollastoni. 3 mesosoma, dorsal 4 head and mesosoma, dorsolateral. Scale line = 200 μm.
Mymarilla wollastoni. 5 head and prothorax, lateral 6 head and thorax, anterolateral. Scale line = 200 μm.
http://species-id.net/wiki/Mymarilla_wollastoni
Figs 1–10Lectotype female in OUMNH (not examined), designated by
Description. Because only one species of Mymarilla is known, the generic and specific features are both described in the species description. Female. Body length 2125 (n=1). Body entirely smooth and shiny except for a few small punctures on pronotum. Colour. Head black except mandibles, mesosoma, metasoma medially, and clava; flagellum, and metasoma basally and apically dark brown; scape, pedicel, mandibles, legs, petiole, and ovipositor brownish yellow; all body setae almost white; wings brown behind venation, membrane beyond venation translucent but membrane around each microtrichia back, so wings generally appearing dark. Head. (Figs 3–6) 1.46× as wide as long, and 1.17× as wide as high (406/227/347) (n=1), in lateral view with anterior surface slightly depressed between toruli then, at level of ventral margin of eye, receding to mouth (Fig. 5). Face measured from eye to eye below toruli about as long as wide, flat above toruli but forming a shallow circular depression medially below toruli; subantennal groove and pits between toruli absent. Torulus about 1.5× its diameter from thin transverse trabecula and about mid height of eye. Preorbital groove against eye almost to ventral margin of eye (Fig. 5), then extending almost straight to anterolateral angle of mouth margin. Eye (Fig. 5) in lateral view 0.75× as long as high, not extending to back of head dorsally and separated by more than its own length from back of head ventrally. Malar sulcus absent and malar space almost 0.6× eye height. Gena narrow dorsally, very wide ventrally. Vertex in lateral view almost flat, sloping anteriorly, forming an obtuse angle to face (separated by transverse trabecula), smoothly merging posteriorly with occiput. Mid ocellus flat in shallow depression, not projecting above surface of vertex and 2× diameter of small lateral ocellus. Ocelli in high triangle, with POL (168), 1.9× LOL, and LOL (89) 3.0× OOL (30). Mandibles normal, overlapping medially when closed, with 3 teeth. Antenna. Scape about 3.9× as long as wide, with both inner and outer surfaces apparently smooth, and with radicle short but distinct; pedicel slightly shorter than fl1, about 0.41× scape length; funicle 6-segmented; clava unsegmented, slightly longer than scape. Number of mps of funicle segments and clava uncertain (not clearly visible on card mounted specimen). Measurements (n=1) length/width: scape 267/69, pedicel 109/50, fl1 129/20, fl2 267/20, fl3 198/30, fl4 149/30, fl5 149/30, fl6 149/30, clava 317/80. Mesosoma. About 2.67× as long as wide, about 1.80× as long as high, and 1.48× as wide as high (n=1). Pronotum (Fig. 3) in dorsal view clearly visible, shorter than wide (218: 267), strongly convex, with almost vertical sides flaring outward ventrally and almost horizontal at junction with propleura; pronotum in lateral view strongly triangular. Propleura (Fig. 4) in dorsolateral view tightly pressed to pronotum laterally and anteriorly, and fused to each other at neck, and, in dorsal view (Fig. 3), slightly extending lateral to pronotum. Neck long (90), clearly separating head from pronotum. Prosternum triangular, strongly appressed laterally to propleura, without median longitudinal line, anterior apex not visible but perhaps closed anteriorly. Mesonotal spiracle (Fig. 6) small, at end of short tube, midway between posterolateral angle of mesoscutum and anterior apex of notaulus. Mesoscutum (Fig. 3) smooth and shiny, in lateral view slightly convex, in dorsal view short (99), about 0.4× as long as pronotum and 0.4× as long as wide, with slightly diverging, almost straight notauli, each ending anteriorly in a distinct pit. Axillae not advanced (Figs 3, 4, 9, 10). Mesoscutellum (Fig. 3) almost as long as pronotum (267: 246), slightly overlapping metanotum, without trace of frenal line (frenum therefore not distinguishable). Prepectus (Figs 6, 9, 10) triangular, about 3× as long as dorsal width. Mesopleuron tightly appressed to prepectus, convex, not divided by suture into mesepisternum and mesipimeron. Metanotum small, triangular, separated from propodeum by wide groove. Metapleuron with a large metapleural pit at junction with mesoepisternum. Propodeum (Figs 3, 4, 7, 10) evenly convex, without carinae, with a large pit at anterior margin just anterior to spiracle, with a slightly upturned nucha covering anterior apex of petiole, and with propodeal seta almost at posterior margin. Spiracle small, round, separated by several diameters from anterior margin of propodeum. Wings. Fore wing (Figs 1, 7, 8) deeply convex, with the anterior and posterior margins strongly curving downward, the wing height in lateral view about 0.7× wing width in dorsal view. Wing behind submarginal vein very narrow, with strongly convex hind margin, abruptly widening beginning at parastigma, generally oval in dorsal view. Entire surface to wing base covered in long microtrichia, those behind venation appressed and those beyond venation semi-erect to erect. Fore wing length 1792 (n=1), width 640, length/width 2.8, venation length 287, about 0.18× forewing length. Submarginal vein black basally, brown apically, and much wider basally than apically; parastigma + stigma black, oval about 1.7× as wide as base of submarginal vein. Costal cell extremely narrow. Hind wing (Fig. 1) flat to slightly convex, with long marginal setae and erect microtrichia similar to those on forewing, and wing membrane extending almost to base of wing but very narrow behind venation. Hind wing length ca. 1535, width 77, venation black, length about 0.3× wing length. Legs. Long and slender (Figs 1, 7). Metasoma. Petiole length 180, slightly longer than metacoxa, about 6× as long as wide. Gaster (Fig. 1) smooth and shiny, narrowly oval in cross section (as seen in posterior view), wider dorsally, almost knife-like ventrally at ovipositor. Gt1 about 0.57× gaster length (870) and almost completely covering gs1 so petiole apparently attached to tergum, gt2–gt5 progressively shorter, gt6 and syntergum (gt7 + gt8) each about as long as gt2. Spiracle present on gt6, small. Ovipositor length ca. 770, slightly down turned apically, as long as gaster but not exserted beyond gaster, about 1.2× metatibia length (640).
Mymarilla wollastoni. 7 mesosoma, petiole and wings posterodorsal 8 fore and hind wing bases, dorsal. Scale line = 200 μm.
Mymarilla wollastoni. 9 mesosoma lateral 10 mesosoma, lateral, SEM. Scale line = 200 μm.
Male. Colour as in female but body dark brown (possibly due to fading), scape brownish yellow, pedicel light brown, flagellum dark brown. Body length 1843–1894 (n=2). Fore wing length 1664 (hind wing not measurable on pinned specimens), with the edges almost meeting ventrally in one specimen, giving the appearance of a hirsute cigar when seen end on in posterior view (Fig. 7). Antenna. Scape about 3.8× as long as wide, with both inner and outer surfaces apparently smooth. Measurements (n=1) length/width or length for flagellomeres: scape 218/59, pedicel 99/45, fl1 277, fl2 267, fl3 248, fl4 248, fl5 223, fl6 228, fl7 198, fl8198, fl9 178, fl10 198, fl11 198; total flagellum 2461. Fl6 length/width about 6.0, with perhaps 8 mps (not clearly visible on card mounted specimen). Metasoma. Gaster length 742–793 (n=2), in lateral view truncate apically.
SAINT HELENA. Centre. High Central Ridge, Cabbage Tree Road, 2500’, iii.1967, J. Decelle, N. & J. Leleup (2 males, MRAC); High Peak, 15°58.7'S, 5°44.0'W, ca.752m, xii.2005–1.2006, N.P. & M.J. Ashmole, H. Mendel, E.A. Thorpe, pitfall trap (1 female, BMNH).
http://species-id.net/wiki/Cremnomymar
Figs 11–14, 20–38It is improbable that the Juan Fernández Islands would have four endemic genera of Mymaridae, as treated by
I examined forty-three specimens (not the types) of Cremnomymar (including the synonyms proposed above), all from Masatierra, the largest of the Juan Fernández Islands. In some of the speciesthe fore wing is fully developed and flat (Fig. 11) but in others it is more or less reduced and convex (Figs 12–14). Depending on the extent of wing reduction the mesothorax is also reduced but the placoid sensilla on the scutellum are always widely separated. The pronotal structure varies from entire, sometimes with indication of a mediolongitutinal line, to being apparently completely divided medially by a complete longitudinal carina. The propodeum in short-winged specimens changes in ways that I consider to be at most of species-level significance, from strongly carinate and with a large, sublateral tooth bearing the propodeal seta, as in fully winged Cremnomymar (Figs 11, 12, 32) to almost smooth (Fig. 14) but still with the propodeal seta on a bump or tooth. Therefore, I propose the above generic synonymies and transfer the included species to Cremnomymar as Cremnomymar caudatum (
Cremnomymar sp., macropterous female, dorsal (fore wing flat). Scale line = 1000 μm.
Cremnomymar sp., slightly brachypterous male, dorsal (forewing slightly convex, hind wing brachypterous). Scale line = 1000 μm.
Cremnomymar sp., moderately brachypterous male, dorsal (forewing distinctly convex, hind wing micropterous). Scale line = 1000 μm.
Cremnomymar sp., strongly brachypterous male, dorsal (forewing slightly convex, hind wing absent). Scale line = 1000 μm.
St. Helena, with a land area of 122 km2, is a remote island of volcanic origin in the South Atlantic Ocean at 15°54'–16°01'S, 5°37'–5°47'W, about 1920 km W. of Africa, the nearest continent (
The Juan Fernández islands with a total land area of 124 km2, almost identical in size to St. Helena, are in the South Pacific Ocean about 600 km west of Chile. Masatierra Island (Robinson Crusoe Island), at 33°38'S, 78°51'W with a land area of about 51 km2 and a 916 m high point is 3.8–4.2 million years old. Masafuera Island (Alejandro Selkirk Island) at 33°46'S, 80°47'W with a land area of 50 km2 and a 1319 m high point is 1.0–2.4 million years old. Santa Clara, the smallest island is 1 km SW of Robinson Crusoe and 5.8 million years old (
Despite being two to three times the age of Santa Clara, the oldest of the Juan Fernández Islands, the remoteness of St. Helena from the nearest continent is likely the main reason why it has only one endemic species of Mymaridae. In contrast, the Juan Fernández Islands have several endemic species, diverse enough to have been placed in separate genera by Ogloblin, who was swayed by the most notable structural differences among species of three of the four genera, i.e., wing reduction with concomitant modifications of the mesosoma. Another reason for the difference in faunas of the two groups of islands is the harsher climate on the Juan Fernández Islands (subtropical) compared to St. Helena (tropical), causing a greater selection pressure on any terrestrial organism established there and therefore a greater likelihood of diversification into several species. Two other reasons are: 1) there is greater chance of continental species reaching the Juan Fernández Island than St. Helena, so multiple introductions could account for the diversity of Cremnomymar species on Masatierra; and 2) the fauna (and flora) of St. Helena has been much more affected by human activity, possibly resulting in the disappearance of other species of Mymaridae that may have existed there, leaving only Mymarilla wollastoni among the endemic Mymaridae.
One specimen of Parapolynema Fidalgo was seen, collected in Bolivia, La Paz, Sorata environs, 3200m, 20.iv.1997, L. Masner, sweeping open scrub (1 female, CNC).
Cremnomymar sp., macropterous female, dorsolateral (fore wing convex, hind wing brachypterous). Scale line = 1000 μm.
Richteria ara, female, lateral. Scale line = 1000 μm.
Richteria ara, male. 17 lateral, wings dorsoposterior 18 fore wing detail. Scale line = 500 μm.
The wings of Mymaridae vary considerably in shape, perhaps more than in most other families of Chalcidoidea. In a normal, fully developed (macropterous) wing the surface is flat, with the length several times greater than width and with a more or less rounded apex. Variations in shape are due partly to changes in length and width and partly to changes in outline, particularly of the posterior margin and wing apex. Depending on the genus and species a full length fore wing may become narrower so that length becomes even greater relative to width. The extremes are 2.5× as long as wide in Paranaphoidea Girault (
Wing reduction (brachyptery, microptery) or complete loss (aptery) usually occurs where there a strong selection pressure against having fully developed wings. Reduction in length may be slight shorter than normal, to extremely short, at least in one sex (usually the female) of at least one species of a genus. Species with wings reduced or absent occur on all continents, mostly in species that search for host eggs in confining places where wings would be a hindrance, e.g., in soil, leaf litter, or in tubules of bracket fungi (some Cleruchus spp.) or in windy places, e.g., small, remote oceanic islands and high elevations on mountains. All three species (100%) of Mymaridae found on small islands (Campbell I., Auckland I., South Georgia) at more than 45° south are wingless or micropterous. Each species is placed in its own genus and each genus is evidently related to a genus on the nearest mainland or is a wingless species with winged congeners in New Zealand, and South America, respectively. For example, on South Georgia (at or near sea level) Notomymar is completely wingless, whereas at 4100 m in Ecuador (
On remote oceanic islands between 30°–45° S Mymaridae are reported so far only from the Juan Fernández Islands (Cremnomymar, 7 described spp., but some probably synonyms), Norfolk Island (Cybomymar Noyes & Valentine [Lin et al. 1997], 1 sp.), and Lord Howe Island (Anagroidea sp., new record). Wingless or short-winged species represent at least 20% of the fauna, excluding genera/species likely introduced over the past few hundred years by humans. In contrast, Mymaridae from continents south of 30° S contain a low percentage of short winged or wingless species. The exception is New Zealand (a large continental island) with 17 of 42 genera that contain at least one flightless species (
Between the tropics of Cancer and Capricorn no wingless or short winged species have yet been recorded on the relatively well studied but still poorly known (for Mymaridae) Pacific oceanic islands: Micronesia (
Curvature of the fore wing so it becomes convex or dome-shaped, i.e., it is no longer flat but clearly three-dimensional, occurs in at least one species of four genera: Cremnomymar Ogloblin (Figs 12, 13), Mymarilla (Figs 1, 7), Parapolynema Fidalgo (Fig. 15), so far known from Argentina and Bolivia, and Richteria (Figs 16–18), so far known from Australia and New Zealand. On wind-swept islands convex wings may perhaps have survival value by enveloping the posterior part of the mesosoma and the metasoma, either for greater heat absorption and retention or to reduce the chance of being blown away. But species of Parapolynema and Richteria do not necessarily occur in harsh environments so there evidently are other reasons for having a more or less convex fore wing. Interestingly, the fore wings of both genera have a double dome, in contrast to the single dome of Mymarilla and some Cremnomymar.
It would be interesting to find out if Mymaridae occur on other remote Southern Hemisphere islands, particularly those south of 30°. Collecting in a diversity of microhabitats using yellow pan traps or sifting of soil and whatever vegetation is present is required because any Mymaridae present are almost certain to be wingless or short winged.
Parapolynema sp., female, dorsolateral. Scale line = 1000 μm.
Cremnomymar sp. (macropterous), micrographs. 20 head, anterior 21 head, lateral 22 head, posterior (sculpture hidden by glue except laterally and dorsally) 23 head, dorsal 24 head, ventral 25 mouthparts. Scale lines = 20 μm.
Cremnomymar sp. (macropterous), micrographs. 26 mesosoma (pronotum missing), dorsal 27 mesosoma, lateral 28 mesosoma, ventral 29 petiole, dorsal 30 petiole, lateral 31 petiole, ventral. Scale lines for 26–28= 50 μm; for 29–31 = 20 μm.
Cremnomymar sp. (macropterous), micrographs. 32 frenum + propodeum, dorsal 33 mesosoma, ventroanterior 34 metasoma, dorsal 35 metasoma, lateral 36 metasoma, ventral 37 apex of gaster, dorsal 38 apex of gaster, lateral. Scale lines for 34–36 = 100 μm; for 32, 33, 37, 38 = 20 μm.
I thank E. S. Ryder (BMNH) and E. De Coninck (MRAC) for the loan of specimens and, particularly, N. Dale-Skey Papilloud (BMNH) for sending me images of the labels on Wollaston’s Mymarilla specimens in Oxford and helping out in other ways. My technician, J. Read, is gratefully acknowledged for taking the photographs and compiling the figures and my previous technician, K. Bolte, is thanked for the scanning electron micrographs.