ENCYRTIDAE, , Trjapitzin, 1989

Noyes, John Stuart, 2023, ENCYRTIDAE OF COSTA RICA (HYMENOPTERA: CHALCIDOIDEA), 4 Subfamily Encyrtinae: tribes Arrhenophagini, Habrolepidini, Cerapterocerini, Cheiloneurini, Trechnitini, Cercobelini, Polaszekiini, Protyndarichoidini, Gahaniellini and Syrphophagini (part), mainly primary parasitoids and hyperparasitoids of Coccoidea and Psylloidea (Hemiptera), Taxonomic Monographs on Neotropical Hymenoptera (Oxford, England) 2 (11), pp. 1-921 : 17-28

CLASSIFICATION OF ENCYRTIDAE

Trjapitzin (1973a,b, 1989) has discussed the different systems of classification that have been used within the Encyrtidae . He proposed a division into two subfamilies, the Encyrtinae and Tetracneminae and this classification has been largely followed by others, e.g. Trjapitzin & Gordh (1978), Noyes & Hayat (1994), Noyes (2000, 2004) and Hayat (2004, 2006). This classification is practical, and is followed here, but recent molecular work suggests that at best the Tetracneminae is a grade-level taxon and the Encyrtinae is probably paraphyletic.

All Tetracneminae have a narrow sclerotized filament (paratergite) which more or less retains the connection of the outer plates of the ovipositor with abdominal tergite IX. The shape of the outer plates (see below) and the immovable fusion of the third valvulae to the second valvifers are diagnostic of the group. In some tetracnemines the paratergites are interrupted, but this is assumed to be secondary. A few taxa of Tetracneminae may have the third valvulae freely articulated with the second valvifers, but this is also probably secondary. In all Encyrtinae the outer plates of the ovipositor are more or less broadly triangular, separated from the ninth abdominal tergite and connected to the apical parts of the syntergum (fused abdominal tergites IX and X) only by a weak membrane. In most Encyrtinae the gonostyli (third valvulae) are articulated with the second valvifers. Where they are fused with the second valvifers this is thought to be secondarily derived, e.g. Encyrtus and some Copidosoma .

Separation of the two subfamilies by a non-expert is difficult, but the Encyrtinae can be distinguished from the Tetracneminae using the combination of characters given below.

Encyrtinae

Linea calva mostly with marginal setae differentiated with longer setae along proximal margin; filum spinosum usually present. FEMALE: outer plates of ovipositor subtriangular and apically truncate (although sometimes with an elongate tapering lateral projection, e.g. Quadrencyrtini), completely separated from syntergum; apex of hypopygium mostly not reaching apex of gaster, only in a few species extending past apex of syntergum; gonostyli, with very few exceptions, freely articulated with second valvifers.

Tetracneminae

Linea calva with marginal setae undifferentiated, more or less of equal size; filum spinosum except very rarely, not developed. FEMALE: outer plates of ovipositor relatively slender and not apically truncate, but each plate produced into a narrow filament which connects with paratergite near apex of syntergum; apex of hypopygium always reaching a little way past apex of syntergum (occasionally, if specimen has been preserved in alcohol and gaster is distorted as a result of air-drying, hypopygium may appear not to reach level with apex of syntergum); gonostyli (third valvulae), with only a few exceptions, immovably fused to second valvifers.

In the latest comprehensive treatment of Encyrtidae, Trjapitzin (1989) recognised 13 tribes within the Tetracneminae and 34 tribes within the Encyrtinae . In this classification two tribes were transferred from the Encyrtinae (Trjapitzin’s 1973b classification) to the Tetracneminae , i.e. Quadrencyrtini (including Eucoccidophagus and Quadrencyrtus ) and Oriencyrtini (including Oriencyrtus ). This was supported by the apparent presence of paratergites ( Sharkov, 1988b). This is not followed here because studies of slide-mounted material of these genera indicate that these so-called paratergites present are not homologous with those found in the Tetracneminae . There appears to be no doubt that the genera included in the Quadrencyrtini and Oriencyrtini are most closely related to Aphycoides which is retained in the Encyrtinae , tribe Microteryini by Trjapitzin (1989 b) (see discussion in Noyes & Hayat, 1994).

The genera included in the current work belong in the subfamily Encyrtinae . Including the Quadrencyrtini and Oriencyrtini, Trjapitzin (1989 b) recognised 36 tribes within the Encyrtinae . Hayat (2004, 2006) slightly modified this classification by adding a further three tribes and two new subtribes. In all these works, few of the tribes are clearly defined and therefore placement of taxa within their proposed tribal framework is difficult. It is likely that the number of tribes regarded as valid will be reduced for the currently described genera when a rigorous phylogenetic study of the subfamily as a whole can be undertaken.

Luckily, most genera dealt with in the present work can be placed within this tribal framework because of clear affinities with the type genera of the family groups recognised by these authors. The classification utilised here is modified from these classifications and can be summarised as follows:

Arrhenophagini Ashmead 1900

Tarsi 4-or 5-segmented; mandible pointed; antenna with anelliform, adpressed, indistinctly delimited funicle segments, clava relatively very large, much longer than funicle and pedicel combined; fore wing with apex of venation poorly defined; mesopleuron clearly not reaching level with posterior margin of propodeum, broadly separated from base of gaster by the propodeum and hind coxa. Primary parasitoids of armoured scale insects ( Hemiptera : Diaspididae ).

Arrhenophagus

Habrolepidini Hoffer, 1955 (= Thomsoniscini Hoffer, 1956 syn.nov.)

Mandible with a socketed peg-like structure on inner surface near ventral margin (see Figs 23, 64, 155, 187, 272, 364 – arrowed); head usually strongly triangular in profile; scutellum flat, usually with deep reticulate or striate sculpture; mesopleuron clearly not reaching level with posterior margin of propodeum, separated from base of gaster by the propodeum and hind coxa. Primary parasitoids of armoured scale insects ( Hemiptera Diaspididae ).

Habrolepidina Hoffer, 1955 Males with 2-segmented funicle and enlarged, sausage-shaped clava.

Adelencyrtus , Plagiomerus , Caenohomalopoda , Homalopoda , Pseudhomalopoda

Thomsoniscina Hoffer, 1956 (= Comperiellina Trjapitzin, 1973 syn.nov.)

Males with 6-segmented funicle, usually filiform with subequal segments that are clearly longer than broad and clothed with whorls of long setae that are mostly at least 3X as long as diameter of segments.

Coccidencyrtus , Comperiella , Neococcidencyrtus , Quetsira ,? Orixia

Cerapterocerini Hoffer, 1955

Submarginal vein with a conspicuous triangular expansion; females with a radiating pattern of infuscate and hyaline areas on fore wing; no conspicuous naked hyaline streak connecting postmarginal and stigmal veins; scape and flagellum strongly broadened and flattened; mesopleuron not reaching level with posterior margin of propodeum, broadly separated from base of gaster by propodeum and hind tibia. Males with filiform antennae with subequal funicle segments that are clearly longer than broad and clothed with whorls of long setae that are usually at least 3X as long as diameter of segments. Secondary parasitoids of soft scale insects ( Hemiptera : Coccidae ) via encyrtid primary parasitoids. Possibly should be considered as a subtribe of Cheiloneurini .

Cerapterocerus

Cheiloneurini Hoffer, 1955 (= Chalceryini Trjapitzin, 1973 syn.nov.)

Females often with fore wing infuscate and with a naked hyaline streak connecting apices of postmarginal veins and stigmal veins and which usually extends some way into disc past apex of stigmal vein; if fore wing infuscate then never with a radiating pattern of infuscate and hyaline areas; fore wing often either with parastigma conspicuously downcurved or occasionally with a triangular expansion; often with a subapical tuft of setae on the scutellum; scape and flagellum often strongly broadened and flattened; mesopleuron mostly extending to level of posterior margin of propodeum or beyond and touching, or nearly touching, base of the gaster, sometimes clearly not reaching posterior margin of propodeum and broadly separated from gaster by propodeum and hind coxa. Males dissimilar to females, almost always with hyaline fore wings and a filiform flagellum with subequal funicle segments that are clearly longer than broad and clothed with whorls of long setae that are usually at least 3X as long as diameter of segments. Mostly primary and secondary parasitoids of soft scale insects ( Hemiptera : Coccidae ), monophlebids ( Hemiptera : Monophlebidae ) and lac insects ( Hemiptera : Kerriidae ). Some species are parasitoids of Dryinidae ( Hymenoptera ).

Ammonencyrtus, Angebra , Anicetus , Anisophleps , Cheiloneurus , Diversinervus , Eusemion , Helegonatopus , Homosemion , Iceromyia , Lakshaphagus , Ludesia , Mahencyrtus , Prochiloneurus ,

Possibly also the following: Georynus , Oodia , Prionias , Stadima , Strigenia , Tico .

Trechnitini Hoffer, 1955

Body usually with a strong metallic sheen; mandible usually modified, mostly with one or two teeth and a broad truncation, or with a very small tooth and broad oblique truncation; marginal vein of fore wing mostly not longer than broad, often absent, rarely as much as 2X as long as broad. Primary parasitoids of jumping plant lice (Psylloidea).

Trechnitina Hoffer, 1955

Hypopygium reaching apex of gaster, enclosing ovipositor; marginal vein absent to punctiform, never longer than broad.

Prionomitus , Sectiliclava , Trechnites

Metaprionomitina Hoffer, 1955

Hypopygium not reaching apex of gaster; marginal vein punctiform, sometimes slightly longer than broad.

Psyllaephagus

Cercobelini Hoffer, 1953

Gaster telescopic, all sternites with relatively long anterolateral apodemes, ovipositor strongly modified, capsule-like and relatively short, outer plates attached to distinct lateral lobes near apex of syntergum; mesopleuron clearly not reaching level with posterior margin of propodeum, separated from base of gaster by the propodeum and hind coxa. Parasitoids of jumping plant lice.

Cercobelus , Arianus

Polaszekiini trib.nov.

Ovipositor structure unique, exserted part very slender and almost filamentous, outer plates slender and ribbon-like; first valvifer strongly triangular, elongate, all sides virtually straight; proximal part (semicircular sheet) of second valvifer with posterior margin almost straight, not concave or C-shaped; hypopygium elongate with posterior margin W-shaped with a strong lobe either side of a distinct median mucro; mesopleuron not quite reaching level with posterior margin of propodeum, narrowly separated from base of gaster by propodeum and hind coxa.

Polaszekius

Protyndarichoidini trib.nov.

Mandible tridentate with three equal acute teeth; fore wing with marginal vein about 3X as long as broad, stigmal vein subsessile with three placoid sensilla at apex; propodeum at least 0.2X as long as scutellum; mesopleuron not reaching level with posterior margin of propodeum, broadly to narrowly separated from base of gaster by propodeum and hind coxa; ovipositor with second valvifer curved upwards towards apex and with semicircular sheet reduced, hardly broadened; outer plate of ovipositor broadly triangular, about 1.5X as long as broad.

Protyndarichoides

Gahaniellini trib.nov.

Antenna placed high on the head with lower margins of torulus about level with lower eye margin; funicle segments subquadrate, subequal and with conspicuous longitudinal sensilla; mandible with one tooth and a broad truncation, apex of stigmal vein with only 3 placoid sensilla; mesopleuron not quite reaching level with posterior margin of propodeum, narrowly separated from base of gaster by propodeum and hind coxa; ovipositor with a short flat leaf-like extension at the apex of the second valvifer that overhangs the proximal part of the gonostylus. Probably secondary parasitoids of various scale insects ( Hemiptera : Coccoidea).

Gahaniella

Syrphophagini Trjapitzin, 1973 stat.nov.

Mandible tridentate with upper tooth transversely truncate or broadly rounded; scutellum convex, often with deeper sculpture than mesoscutum; fore wing with marginal vein at least 2.5X as long as broad (usually more than 4X as long as broad) and postmarginal vein distinct, from about half as long as stigmal vein to about as long; mesopleuron reaching about level with posterior margin of propodeum, narrowly separated from base of gaster by propodeum and hind coxa, occasionally nearly touching. Probably associated with sternorrhynchous Hemiptera as primary or secondary parasitoids or parasitoids of predators of sternorrhynchous Hemiptera . I consider the tribe to belong to a lineage of Encyrtidae that includes mainly associates of Holometabolous insects and is a possible sister group to the Copidosomatini + Bothriothoracini. It seems probable that this lineage diverged early from the main lineage of Encyrtinae that is now mostly associated with Hemiptera , especially scale insects (Coccoidea) and other sternorrhynchs.

Alkonia , Caldencyrtus , Dalek , Elaxis , Esphora , Harlarus , Hebynthus , Mendisa , Raquanus , Sosara , Syrphophagus 1

Given the above definition, the Syrphophagini may also include Diaphorencyrtus Hayat , but not Metablastothrix Sugonjaev or Aphidencyrtoides Ishii. These genera were previously included in the subtribe Syrphophagina which was included in the Microteryini by Trjapitzin (1973, 1989).

Even though the family groups included within the present revision are difficult to define or recognise, a key is presented below which should facilitate the correct placement of females within the generic framework utilised in this study.

KEY TO COSTA RICAN GENERA INCLUDED IN THIS VOLUME (Females)

1. Antennal funicle with not more than 5-segments ........................................................... 2

- Antennal funicle with 6 segments .................................................................................. 8

2 (1) Antenna ( Figs 9, 12) with funicle segments variable and closely adpressed together making them difficult to distinguish from clava; clava 1-segmented; fore wing hyaline; tarsi 4-segmented; small species about 0.5-0.6mm in length ................... ........................................................................................... Arrhenophagus (p. 28)

- Antenna with funicle segments easy to distinguish from clava; clava 3-segmented; tarsi 5-segmented; generally larger species 0.5-2.5mm in length ................................ 3

1 A revision of the Costa Rican species of Syrpophagus Ashmead will appear in the next volume of this series.

3 (2) Fore wing with stigmal vein at least slightly longer than marginal vein (e.g. Figs 1472, 1496, 1651, 1682); head not triangular in profile with frontovertex distinctly curved above top of antennal scrobes; hypopygium reaching apex of gaster; scutellum dorsally convex ....................................................................................4

- Fore wing with stigmal vein shorter than marginal vein (e.g. Figs 42, 58, 67, 75, 165); head strongly triangular in profile with frontovertex above antennal scrobes flat; hypopygium not reaching more than about two-thirds along gaster; scutellum dorsally very flat ................................................................................................... 5

4 (3) Funicle 5-segmented ........................................................................... Trechnites (p. 567) - Funicle 4-segmented ......................................................................... Cercobelus (p. 608)

5 (3) Fore wing hyaline ............................................................................. Plagiomerus (p. 41) - Fore wing with a pattern of infuscate and hyaline areas ................................................ 6

6 (5) Pronotum triangular in dorsal view, medially about half as long as mesoscutum .............

...................................................................................... Caenohomalopoda (p. 80) - Pronotum with posterior margin strongly concave, medially not more than about one-

sixth as long as mesoscutum ................................................................................ 7

7 (6) Fore wing with a distinct, single hyaline spot at apex ( Figs 141, 148, 152); flagellum distinctly flattened with all funicle segments transverse ........................................ ...................................................................................... Pseudhomalopoda (p. 74)

- Fore wing with a pair of opposite, subapical hyaline spots (e.g. Figs 74, 83, 98) or apex subhyaline (e.g. Figs 103, 111, 115); flagellum subcylindrical with at least one segment longer than broad .................................................... Homalopoda (p. 53)

8 (1) Scape not more than 2X as long as broad ....................................................................... 9 - Scape more than 2X as long as broad ........................................................................... 24

9 (8) Flagellum with segments strongly broadened and flattened (e.g. Figs 346, 372, 1240,

1252, 1261, 1262) .............................................................................................. 10 - Flagellum with segments subcylindrical ...................................................................... 15

10 (9) Apex of clava transversely truncate, the truncation formed by an enlarged sensory area that does not extend along ventral surface of clava ( Fig. 346); body dorsoventrally flattened; head distinctly triangular in profile with frontovertex forming an angle of less than 60° with face ..................................................... Comperiella (p. 142)

- Apex of clava with a strong oblique truncation formed by an enlarged sensory area that extends more than half way along ventral surface of clava (e.g. Figs 372. 1240, 1252, 1261, 1262); body not dorsoventrally flattened; head not triangular in profile ................................................................................................................. 11

11 (10) Fore wing with several sharply delimited triangular or subtriangular hyaline areas in apical half ( Figs 373, 375); some brachypterous specimens known ...................... ........................................................................................ Cerapterocerus (p. 150)

- Fore wing generally infuscate, except for hyaline apex and sometimes base, without any sharply delimited triangular hyaline areas (e.g. Figs 1245, 1253, 1272, 1280, 1303) .................................................................................................................. 12

12 (11) Infuscate area of fore wing enclosed subapically by a darker curved band (e.g. Figs 1272, 1280) .......................................................................................... Anicetus (p. 504)

- Infuscate area of fore wing not enclosed subapically by a darker curved band (e.g. Figs 1245, 1253, 1303) .............................................................................................. 13 13 (12) Hyaline apex of fore wing more or less uniformly curved (Figs 1253, 1255) ..................

.................................................................................................. Eusemion (p. 499) - Hyaline apex of fore wing subtriangular (e.g. Figs 1247, 1304, 1321) ........................ 14

14 (13) Frontofacial carina interrupted medially where frontovertex and interantennal prominence are more or less confluent (Figs 1305, 1319, 1332); antenna with F1 very broad and flat, more than 2X as wide as pedicel and at least as wide as F6 (e.g. Figs 1303, 1323, 1331) .................................................................. Ammonoencyrtus (p. 518)

- Frontofacial carina continuous across face above antennal scrobes (Fig. 1246); antenna with F1 transverse but not conspicuously broadened, only about as wide as pedicel and less than half as wide as F6 (Figs 1240, 1244) ............ Homosemion (p. 497)

15 (9) Fore wing infuscate, usually with apices of postmarginal and stigmal veins connected by a conspicuous, slender, hyaline streak which continues a little way into wing disc past apex of stigmal vein (e.g. Figs 480, 486, 497, 1029, 1034, 1112) ............. 16

- Fore wing hyaline, apices of postmarginal and stigmal veins not connected by a conspicuous, slender hyaline streak ................................................................... 19

16 (15) Either scutellum with a subapical tuft of setae (e.g. Figs 1028, 1036, 1121) or apical hyaline area of fore wing subtriangular, infuscate area not delimited by a darker curved band (e.g. Figs 1029, 1030, 1126); marginal vein always at least 1.3X as long as stigmal vein ............................................................ Cheiloneurus (p. 197)

- Scutellum without a distinct subapical tuft of setae, sometimes with a few scale-like setae ( Fig. 494); apical hyaline area of fore wing curved with infuscate area delimited by a darker band (e.g. Figs 476, 486, 497); marginal vein always at least slightly shorter than stigmal vein .....................................................................................17

17 (16) Fore wing with parastigma strongly downcurved and with a distinct triangular expansion proximally ( Figs 486, 489); sculpture of mesoscutum and scutellum similar, both with whorls of very fine, striate sculpture ( Fig. 483) ........... Anisophleps (p. 189)

- Fore wing with parastigma at most weakly downcurved, without a proximal triangular expansion ( Figs 476, 480, 497); scutellum ( Fig. 494) mostly with coarse, longitudinally elongate striate-reticulate sculpture that contrasts with the shallower imbricate-reticulate sculpture of mesoscutum .................................................. 18

18 (17) Head ( Fig. 492) at least 6X as wide as frontovertex; top of scrobes delimited by a strong transfacial carina; scutellum ( Fig. 494) with at least two, distinctly scale-like setae; fore wing ( Figs 496, 497) with differentiated scale-like setae along proximal margin of linea calva ......................................................... Lakshaphagus (p. 194)

- Head ( Fig. 471) not more than 5X as wide as frontovertex; top of scrobes not delimited by a strong transfacial carina; scutellum with scattered, uniformly bristle-like setae; fore wing ( Figs 476, 480) without differentiated scale-like setae along proximal margin of linea calva .................................................................. Angebra (p. 185)

19 (15) Fore wing ( Fig. 444) with basal cell more or less evenly setose; hypopygium reaching apex of gaster, extending very slightly past apex of syntergum; outer plates of ovipositor slender and ribbon-like ( Fig. 443) ............................. Stadima (p. 174)

- Fore wing with a large, distinct naked area in basal cell (e.g. Figs 1839, 1854, 1974); hypopygium not reaching more than 0.8X along gaster; outer plates of ovipositor narrowly triangular (e.g. Figs 1836, 1856, 1872) .............................................. 20

20 (19) Clava (Fig. 1969) with sensory part extending more than half-way along ventral margin giving it a strong, obliquely truncate appearance .......................... Elaxis (p. 710)

- Clava (e.g. Figs 1837, 1841, 1859, 1874) with apex rounded, transversely truncate or hardly obliquely truncate ................................................................................... 21

21 (20) All funicle segments longer than broad (Figs 1837, 1841) .......................................... 22 - At least some funicle segments quadrate or transverse (Figs 1859, 1866, 1874) ......... 23

22 (21) Mouth less than one-third head width; scutellum (Fig. 1846) with striate sculpture; antennal funicle (Fig. 1841) with F1 conspicuously smaller than F2; ovipositor shorter than mid tibia; hypopygium (Fig. 1840) not W-shaped, lateral arms vestigial ..................................................................................... Harlarus (p. 669)

- Mouth more than one-third head width; scutellum with cellular sculpture; antennal funicle (Fig. 1837) with F1 subequal to F2; ovipositor longer than mid tibia; hypopygium W-shaped with elongate anterior lateral arms ...... Mendisa (p. 666)

23 (21) Funicle (Figs 1859, 1860, 1866) with contrasting pale (white or pale yellow) and dark (brown) segments; mandible (Figs 1855, 1862) tridentate with two acute lower teeth and third transversely truncate, upper tooth ...................... Esphora (p. 674)

- Funicle (Fig. 1874) unicolourous pale orange; mandible (Fig. 1873) with 1 sharp tooth and a broad, upper, slightly concave truncation ............................ Sosara (p. 679)

24 (8) Brachypterous species, fore wing not reaching apex of gaster ..................................... 25 - Fully winged species, fore wing extending past apex of gaster ................................... 26

25 (24) Scutellum without a subapical tuft of setae ( Figs 456, 464) ................. Ludesia (p. 178) - Scutellum with a subapical tuft of setae ( Figs 951, 982, 985) ...... Cheiloneurus (p. 197)

26 (24) Scutellum with a subapical group of setae that are clearly more dense and longer than those elsewhere on scutellum and which form a distinct tuft (e.g. Figs 519, 653, 741, 868, 1150, 1233); ovipositor often clearly strongly exserted with exserted part at least about 2X as long as mid tibial spur ................................................ 27 [NB - The scutellar tuft may be overlooked because setae may be lost or very sparse (e.g. Figs 520, 521) - if fore wing is strongly infuscate then apices of postmarginal and stigmal veins are usually connected by a slender, naked, hyaline streak that goes someway past the apex of the stigmal vein into the wing disc (e.g. Figs 679, 694, 714.]

- Scutellum with setae evenly distributed and not forming a subapical tuft; exserted part of ovipositor rarely longer than mid tibial spur, never more than 1.5X as long as mid tibial spur ............................................................................................................30

27 (26) Mesoscutum with a distinct median tuft of setae (Fig. 1233); fore wing fully developed and with stigmal vein longer than marginal vein (Fig. 1231) ................................. ............................................................................................. Diversinervus (p. 490)

- Mesoscutum without a tuft of setae; if fore wing fully developed then stigmal vein much shorter than marginal vein (e.g. Figs 679, 694, 714, 1084) ............................... 28

28 (27) Fore wing evenly infuscate from level of marginal vein to wing apex except for a hyaline streak at apex of venation (Fig. 1238); marginal vein not more than 1.5X as long as stigmal vein (Fig. 1239) ...................................................... Iceromyia (p. 495)

- Fore wing infuscate or hyaline, if infuscate then either with conspicuous hyaline or paler areas between level of marginal vein and wing apex (e.g. Figs 679, 694, 714) or marginal vein at least 2X as long as stigmal vein .............................................. 29

29 (28) Hypopygium reaching apex of gaster; scutellum with setae forming subapical tuft arranged in distinct longitudinal rows (e.g. Figs 1160, 1173, 1214); ovipositor always conspicuously exserted with sheaths together forming a uniform, slender cylindrical tube; fore wing infuscate with distal margin of infuscation curved and subparallel to wing apex (e.g. Figs 1168, 1179, 1193, 1198) .................................. ......................................................................................... Prochiloneurus (p. 458)

- Hypopygium not reaching more than 0.8X along gaster; scutellum with setae forming subapical tuft scattered and not arranged in distinct parallel rows (e.g. Figs 884, 906, 1057); ovipositor hidden or exserted, if exserted then sheaths tapering towards apex and not forming a ovipositor slender, cylindrical tube; fore wing infuscate or hyaline, if infuscate then distal margin of infuscation rarely curved and subparallel to wing apex ............................................. Cheiloneurus (p. 197)

30 (26) Fore wing with linea calva interrupted on dorsal surface at about two-thirds across wing by about 2-4 lines of setae (e.g. Figs 174, 234, 302, 326, 1391) ....................... 31

- Fore wing with line calva not interrupted on dorsal surface, entire, although sometimes closed near posterior wing margin by one or two lines of setae ........................ 34

31 (30) Scutellum distinctly convex; antenna with funicle segments subequal in size and all with prominent linear sensilla (Figs 1390, 1396, 1399); mandible without a socketed peg near ventral margin of inner surface ............................... Gahaniella (p. 548)

- Scutellum flat; antenna with funicle segments clearly increasing in size distally, with linear sensilla not prominent and usually not present on all segments (e.g. Figs 167, 183, 246, 288, 311); mandible with a socketed peg near ventral margin of inner surface (see Figs 187, 272, arrowed) ........................................................ 32

32 (31) Fore wing infuscate, or if hyaline then mid tibia is white with a pair of dark brown rings ............................................................................ Neococcidencyrtus (p. 126)

- Fore wing hyaline; mid tibia completely white or pale yellow at most with a single dark brown band ........................................................................................................ 33

33 (32) Clava with apex rounded or pointed (e.g. Figs 224, 241, 254, 282); fore wing with postmarginal vein not longer than stigmal vein (e.g. Figs 225, 244, 255, 283) ...... ........................................................................................... Coccidencyrtus (p. 82)

- Clava with apex obliquely truncate ( Fig. 288); fore wing with postmarginal vein longer than stigmal vein ( Fig. 291) ....................................................... Quetsira (p. 123)

34 (30) Fore wing with marginal vein about as long as broad, rarely nearly 2X as long as broad (e.g. Figs 386, 404, 1415, 1519, 1694) ...............................................................35

- Fore wing with marginal vein at least 2.5X as long as broad (e.g. Figs 25, 318, 397, 515, 1821, 1889) ........................................................................................................ 40

35 (34) Fore wing with stigmal vein relatively short, not more than 2X as long as marginal vein ( Figs 386, 404); mesopleuron posteriorly reaching at least level with posterior margin of propodeum, in side view clearly separating propodeum from hind coxa .................................................................................................................... 36

- Fore wing with stigmal vein elongate, at least 2X as long as marginal vein (e.g. Figs 1415, 1519, 1694); mesopleuron clearly not reaching level with posterior margin of propodeum, falling short by at least diameter of propodeal spiracle so that in side view propodeum more or less touches hind coxa ....................................... 37

36 (35) Scutellum hardly shiny, matt; fore wing ( Fig. 406) about 2.4X as long as broad with linea calva open; stigmal vein ( Fig. 404) with 3 placoid sensilla at apex; malar suture absent ................................................................................................ Tico (p. 160)

- Scutellum strongly metallic; fore wing ( Fig. 387) about 2.8X as long as broad with linea calva closed by a single line of setae; stigmal vein ( Fig. 386) with 4 placoid sensilla at apex; malar suture present .............................. Helegonatopus (p. 152)

37 (35) Apical segment of fore tarsus very enlarged, at least as long as rest of tarsus (e.g. Figs 1419, 1426, 1432); fore wing with marginal vein absent or very short, stigmal vein usually arising from submarginal vein before it reaches anterior wing margin (e.g. Figs 1413, 1422, 1436) ......................................................... Sectiliclava (p. 555)

- Apical segment of fore tarsus not enlarged, shorter than rest of tarsus; fore wing with marginal vein always present ............................................................................. 38

38 (37) Hypopygium not reaching more than about two-thirds along gaster ................................

........................................................................................... Psyllaephagus (p. 577) - Hypopygium reaching apex of gaster ........................................................................... 39

39 (38) Mesoscutum laterally with a distinct depression anterior to tegula (Fig. 1409; see Hab. 206); scutellum with not more than about 30 setae (Fig. 1409); fore wing with postmarginal vein about half as long as stigmal vein (Fig. 1407) .......................... .............................................................................................. Prionomitus (p. 553)

- Mesoscutum laterally without a distinct depression anterior to tegula (Fig. 1696; see Hab. 252); scutellum densely setose, clothed with about 100 subequal setae (Fig. 1696); fore wing with postmarginal vein slightly longer than stigmal vein (Fig. 1694) ........................................................................................... Arianus (p. 622)

40 (34) Head triangular in profile; frontovertex above antennal scrobes almost flat; mandible ( Figs 23, 364, arrowed) with a socketed peg near ventral margin on inner surface (only visible on slide-mounted material) ........................................................... 41

- Head not triangular in profile; frontovertex above antennal scrobes distinctly convex; mandible without a socketed peg near ventral margin on inner surface ............ 43

41 (40) Sculpture on mesoscutum and scutellum of similar depth ( Figs 356, 360, 366); eye very conspicuously hairy, clothed in dark setae that are at least as long as diameter of facet ( Figs 352, 359, 364; see also Hab. 56-59); clava with apical sensory area enlarged giving apex a truncate appearance ( Figs 353, 358, 367); malar suture present and conspicuous ( Figs 359, 364) ....................................... Orixia (p. 144)

- Sculpture on scutellum distinctly deeper than that on mesoscutum ( Figs 17, 27, 37, 304, 313); eye hairy, but never very conspicuously so, setae pale and not longer than diameter of facet; clava with apex rounded or pointed (e.g. Figs 22, 34, 301, 317); malar suture absent or present and indistinct ..................................................... 42

42 (41) Mid tibia with a pair of dark brown bands; fore wing, if infuscate, with a hyaline band distad of venation (e.g. Figs 297, 326, 430); body frequently partly orange .......... .................................................................................... Neococcidencyrtus (p. 126)

- Mid tibia sometimes brown, but never with a pair of dark bands; fore wing, if infuscate, with a pair of opposite triangular areas distad of venation ( Figs 29, 38); body never partly orange ............................................................... Adelencyrtus (p. 32)

43 (40) Ovipositor hardly exserted, but with gonostylus distinctly upwardly curved; fore wing with stigmal vein subsessile without a distinct stalk, much shorter than marginal vein and with only 3 apical placoid sensilla (e.g. Figs 1348, 1357, 1382) ............. .................................................................................... Protyndarichoides (p. 535)

- Ovipositor hidden to strongly exserted, but never with gonostylus upwardly curved; fore wing with stigmal vein usually elongate with a distinct stalk, although occasionally subsessile with only 3 apical sensilla ................................................................ 44

44 (43) Apex of clava with a distinct oblique apical truncation, sensory area extending at least

one-quarter along ventral surface of clava ........................................................ 45 - Apex of clava rounded, sensory area small and at apex only ....................................... 49

45 (44) Apex of syntergum with a narrow tongue-like projection (see Hab. 69) ..........................

.................................................................................................... Prionias (p. 168) - Apex of syntergum acute or rounded, without a narrow tongue-like projection .......... 46

46 (45) Scutellum with uniform polygonally reticulate sculpture composed of fine, honeycomb-like cells that extend to sides and apex ( Fig. 420); head about 8X as wide as frontovertex .............................................................................. Georynus (p. 163)

- Scutellum with irregular polygonally reticulate to longitudinally elongate imbricate-reticulate or striate-reticulate sculpture that is often arranged in whorls, apex and sides frequently broadly completely smooth and shiny; head not more than 7X as wide as frontovertex .......................................................................................... 47

47 (46) Scutellum dorsally flat with sides and apex perpendicular and uniformly completely smooth and shiny; mandible with three acute teeth, two lower teeth relatively long and equal, upper tooth conspicuously much shorter and often vestigial (e.g. Figs 610, 620, 638, 644, 654) .................................................... Cheiloneurus (p. 197)

- Scutellum dorsally convex, sides often perpendicular but apex more gradually sloping and more broadly smooth and shiny than sides; mandible with three equal teeth, upper tooth transversely truncate or sometimes rounded (e.g. Figs 1880, 1898, 1914, 1964) ........................................................................................................ 48

48 (47) Clava as long as pedicel and funicle combined (Fig. 1968); propodeum with only 2 or 3 setae near spiracle; syntergum as long as mid tibia; hypopygium (Fig. 1967) with anterior margin more or less straight without any anterolateral or anteromedian processes ...................................................................................... Alkonia (p. 708)

- Clava shorter than pedicel and funicle combined, at most only slightly longer than funicle (e.g. Figs 1903, 1920, 1937, 1959); propodeum with at least 10 setae near spiracle; syntergum shorter than mid tibia; hypopygium with distinct anterolateral and anteromedian processes (e.g. 1890, 1910, 1935, 1956) ............ Hebynthus (p. 682)

49 (44) Mesoscutum and scutellum more or less matt with similar lineolate to striate-reticulate sculpture (e.g. Figs 507, 520, 539); posterior part of mesoscutum often with translucent or silvery setae that contrast with dark brown anterior setae; gaster (e.g. Figs 504, 522, 532, 541) with a pair of gland-like structures present on Gt1 and Gt5 of the gaster (only visible in slide-mounted material) .............................. ............................................................................................ Cheiloneurus (p. 197)

- Mesoscutum with imbricate or polygonally reticulate sculpture and shiny, with uniform setae (e.g. Figs 393, 439, 452, 1344, 1726, 1773, 1846); gaster without paired gland-like structures ........................................................................................... 50

50 (49) Sculpture of scutellum not deeper than that on mesoscutum ( Figs. 439, 1714, 1726, 1736) .................................................................................................................. 51

- Scutellum with distinctly deeper sculpture than mesoscutum (e.g. Figs 393, 452, 1344, 1758, 1796) ........................................................................................................ 52

51 (50) Head (Figs 1704, 1718, 1732) slightly less than 3X as wide as frontovertex; clypeal margin distinctly protuberant and roundly convex or even biconvex; antennal toruli situated relatively high on head, separated from mouth margin by more than their own lengths .................................................................. Caldencyrtus (p. 625)

- Head ( Fig. 436) more than 3X as wide as frontovertex; clypeal margin very slightly concave; antennal toruli situated lower on head, separated from mouth margin by less than their own lengths ........................................................ Strigenia (p. 171)

52 (50) Funicle with at least F1 and F2 quadrate or transverse ................................................. 53 - Funicle with all segments longer than broad, at most F6 quadrate ............................ 54

53 (52) Ovipositor strongly exserted, the exserted part at least 1.5X as long as mid tibial spur ...

............................................................................................... Polaszekius (p. 532) - Ovipositor not exserted ....................................................................... Raquanus (p. 671)

54 (52) Fore wing ( Fig. 394) slender, at least 3.1X as long as broad, basal cell more or less evenly setose, apical half with a median, elongate, wedge-shaped infuscate area; syntergum nearly 1.5X as long as mid tibia ........................ Mahencyrtus (p. 157)

- Fore wing (e.g. Figs 451, 1747, 1764, 1805, 1817) not more than about 2.6X as long as broad, basal cell with a distinct naked area, apical half of fore wing hyaline or weakly infuscate, but without a median, wedge-shaped mark; syntergum shorter than mid tibia ..................................................................................................... 55

55 (54) Apical half of fore wing ( Fig. 451) weakly infuscate with a poorly defined hyaline area at apex of venation and another opposite on hind margin of wing; malar suture present and conspicuous; propodeum ( Fig. 452) virtually smooth without a posterior carina above hind coxa and without a posterolateral tooth behind spiracle ........................................................................................... Oodia (p. 176)

- Fore wing (e.g. Figs 1747, 1764, 1779, 1805, 1817) hyaline, occasionally weakly suffused brown or with a small infuscate area below marginal vein, never as above; malar suture absent or, if present, very indistinct; propodeum with rough sculpture, almost always with a carina posteriorly above hind coxa that runs almost vertically finishing in a posterolateral tooth behind spiracle (see Figs 1773, 1775, 1796, 1804, 1812, 1820, 1825, 1828) .................................. Dalek (p. 635)