Descriptive notes on a new and other little known species of Anagrus Haliday, 1833 (Hymenoptera: Mymaridae) from the New World tropics and subtropics

Recibido: 11-ix-2001
Aceptado: 11-iii-2002
Correcciones devueltas por el autor: 10-vi-2002

Resumen

Introduction

This paper presents new information mainly on the little known Neotropical species of the fairyfly genus Anagrus Haliday, 1833, which has gained some additions since the relatively recent review by Triapitsyn (1997) of the taxa occurring in America south of the United States. Three new, very distinctive, Neotropical species were described after that publication, two from Argentina and Brazil (Triapitsyn 1999) and one from Guatemala and Mexico (Triapitsyn 2000). It has been found that the ranges of some of those species extend into the subtropics of the Nearctic, where the fauna of Anagrus was revised by Chiappini et al. (1996). The available keys do not include some of the previously described species because of lack, or poor quality, of the type material. Two such species, A. armatus (Ashmead, 1887) and A. urichi Pickles, 1932, are redescribed and diagnosed below based on recently discovered specimens. Many new distribution and host records are added to the knowledge about other species included in this review and an updated key to the described Neotropical species of Anagrus is provided.

Several species of Anagrus are important egg parasitoids of various pests belonging to Auchenorrhyncha (Homoptera) as well as of a few mirid and tingid bugs (Heteroptera: Miridae and Tingidae, respectively). They were successfully used in a number of biological control programs (Triapitsyn and Beardsley 2000). Identification of Anagrus species is difficult because of their minute size and also because of generally poor preservation techniques of the voucher specimens. Thus, most of the earlier identifications and especially catalog records of Anagrus species in the New World need confirmation.

Many new Neotropical taxa in this genus await description, that must be preceded by careful preparation and examination of several thousand specimens accumulated in CNCI, UCRC, and other major collections of Mymaridae. Unfortunately, most of these were collected by sweeping or various trapping methods and thus lack host information. Because of the enormity of the task, only one such species is described here as new, and only because it represents a very distinctive and previously unknown form of the “stethynioides” species group of Anagrus, which is defined herein.

Terminology for morphological features used in the key and descriptions is that of Gibson (1997). Measurements are given in microns (µm) as length or, if applicable, as length/width; when more than two specimens were measured, the average is followed by the range in parentheses. Abbreviations for depositories of specimens are as follows: CNCI, Canadian National Collection of Insects, Ottawa, Ontario, Canada; CSCA, California State Collection of Arthropods, Sacramento, California, USA; FSCA, Florida State Collection of Arthropods, Gainesville, Florida, USA; IMLA, Fundación e Instituto Miguel Lillo, San Miguel de Tucumán, Tucumán, Argentina; UCRC, Entomology Research Museum, University of California, Riverside, California, USA; USNM, National Museum of Natural History, Washington, D.C., USA. An abbreviation used in the text is: F = funicular (flagellar in males) segment.

Anagrus Haliday, 1833

Synonyms: Pteratomus Packard, 1864, Packardiella Ashmead, 1904, Paranagrus Perkins, 1905, Anagrella Bakkendorf, 1962. For the latest diagnoses of the genus, see Chiappini and Lin (1998) and Triapitsyn and Beardsley (2000). The complete list of references on Anagrus and the diagnoses of its three subgenera, Anagrus Haliday, 1833 s. str., Paranagrus Perkins, 1905 and Anagrella Bakkendorf, 1962, as well as of the species groups of Anagrus s. str., were given by Chiappini et al. (1996).

The “stethynioides” species group of the subgenus Anagrus s. str. is defined here based on A. stethynioides sp. n. described below. This species group is characterized by the clava with six sensory ridges (Figure 1) and by the relatively wide (4.6-4.9 x as long as wide) forewing (Figures 2, 3) which superficially resembles in shape the forewing of Stethynium Enock, 1909, a genus closely related to Anagrus. The longest marginal cilia on the forewing of A. stethynioides sp. n. are subequal to or slightly longer than the maximum forewing width. In all other described species of Anagrus, the forewing is more than 5.5 x as long as wide and the longest marginal cilia are much longer than the maximum forewing width. Besides A. stethynioides sp. n., this group also includes one undescribed species, known to me from a single female specimen from Ecuador.

Members of Anagrella have not yet been formally recorded from the Neotropical region, but I have seen several undescribed species of this subgenus from specimens collected in Colombia, Costa Rica, Ecuador, Mexico, and Panama. The key below deals primarily with the already described Neotropical species of Anagrus, thus excluding six diagnosed and keyed, but unnamed species from Argentina (Triapitsyn 1999). Also excluded from the key are A. takeyanus Gordh, 1977, mistakenly recorded from Mexico (Triapitsyn 1997) from a specimen which was described later as a new species, A. raygilli (Triapitsyn 2000), as well as A. incarnatus Haliday, 1833 and A. optabilis (Perkins, 1905) because their respective records from Juan Fernández Islands (De Santis 1979) and Ecuador (De Santis and Fidalgo 1994) could not be confirmed (no voucher specimens are available). Annotations to the key include, besides of the described and redescribed species, those taxa for which new information (e.g., new distribution and host records) is presented.

Key to the Neotropical species of Anagrus, females

1	Ocelli on a well-defined triangular stemmaticum............................2

1'	Ocelli not on a stemmaticum, which is absent (subgenus *Anagrella* Bakkendorf, 1962) ......................A. (Anagrella) spp.
,
2(1)	Paramedial plates of posterior scutellum very close to each other (subgenus *Anagrus* Haliday, 1833 s. str.)............................3

2'	Paramedial plates on posterior scutellum widely separated (subgenus *Paranagrus* Perkins, 1905) ......................................... ..........................................................A. unilinearis Soyka, 1950
,
3(2)	Clava with 6 longitudinal sensilla; forewing (Figure 2) 4.6-4.9 x as long as wide, with longest marginal cilia subequal to or slightly longer than maximum forewing width (*stethynioides* species group) ...................................................................A. stethynioides sp. n.

3'	Clava with 3 or 5 longitudinal sensilla; forewing more than 6.0 x as long as wide, with longest marginal cilia much longer than maximum forewing width .......................................................4
,
4(3)	Clava with 3 longitudinal sensilla (*atomus* species group)...............5

4'	Clava with 5 longitudinal sensilla (*incarnatus* species group) ........7
,
5(4)	F4 with 1 longitudinal sensillum .....................................................6

5'	F4 without longitudinal sensilla .............A. ustulatus Haliday, 1833
,
6(5)	Bare area on forewing disc long and narrow (along posterior margin) ................................................A. frequens Perkins, 1905

6'	Bare area on forewing disc short (subapical) ................................... ...........................................................A. atomus (Linnaeus, 1767)
,
7(4)	F2 subglobular, about as long as F1 ................................................ .....................................................A. brasiliensis Triapitsyn, 1997

7'	F2 cylindrical, usually much longer than F1 ....................................8
,
8(7)	Mesoscutum with a pair of adnotaular setae .................................10

8'	Mesoscutum without adnotaular setae ............................................9
,
9(8)	F2 usually with 1 longitudinal sensillum; if without, then F3 or F4 with 2 sensory ridges .................A. gonzalezae Triapitsyn, 1997

9'	F2 always without longitudinal sensilla; F3 and F4 each with only 1 sensory ridge ........................A. breviphragma Soyka, 1955
,
10(8)	F2 with a longitudinal sensillum ....................A. yawi Fullaway, 1944

10'	F2 without longitudinal sensilla ......................................................11
,
11(10)	Forewing disc with one incomplete row of at most 10 setae beyond venation (as in Figure 5) ...................................................12

11'	Forewing disc with at least with one complete row of more than 10 setae beyond venation or with several rows of setae...........13
,
12(11)	Clava white; F3 with a longitudinal sensillum ....................................... ............................................................A. ogloblini Triapitsyn, 1999

12’	Clava brown; F3 without longitudinal sensilla ....................................... .....................................................................A. urichi Pickles, 1932
,
13(11)	Forewing disc with only one complete row of setae beyond venation and without additional complete or incomplete rows, at most with a few scattered setae at apex........................................14

13'	Forewing disc with more than one (complete or incomplete) row of setae beyond venation, with or without bare area at broadest part ..15
,
14(13)	F3 without longitudinal sensilla, much shorter than F4 ............................ .................................................................A. lineolus Triapitsyn, 1999

14’	F3 with a longitudinal sensillum, about as long as F4 .............................. ..........................................A. oahuensis Triapitsyn & Beardsley, 2000
,
15(13)	Broadest part of forewing disc with a more or less defined bare area near posterior margin ........................................................................16

15’	Broadest part of forewing disc more or less uniformly setose, without bare area near posterior margin..............................................18
,
16(15)	Ovipositor very long and strongly exserted beyond apex of gaster, ratio of total ovipositor length to length of its exserted part 3.5-3.7:1; ovipositor length/protibia length ratio about 3.7:1 ................................... ...............................................................A. armatus (Ashmead, 1887)

16’	Ovipositor not very long and only slightly exserted beyond apex of gaster, ratio of total ovipositor length to length of its exserted part greater than 6.0:1; ovipositor length/protibia length ratio about 2.2-2.5:1 .........................................................................................17
,
17(16)	F2 and F3 each much shorter than F4 or F5; F3 and F5 without longitudinal sensilla .....................................A. raygilli Triapitsyn, 2000

17’	F2 and F3 each about as long as F4 or F5; F3 and F5 with a longitudinal sensillum .................................A. empoascae Dozier, 1932
,
18(15)	Body uniformly dark brown or black; F5 with 2 longitudinal sensilla (rare, known from southern Argentina only) ........................................... .............................A. subfuscus Foerster, 1847 sensu Debauche, 1948

18’	Body color different; F5 usually with 1, sometimes without longitudinal sensilla (common) .............................................................................19
,
19(18)	Body color more or less uniformly light brown to brown; F3 without longitudinal sensilla ...............................A. flaveolus Waterhouse, 1913

19’	Body color contrastingly black and yellow; F3 with a longitudinal sensillum ...................................................A. nigriventris Girault,1911

Anagrus (Paranagrus) unilinearis Soyka, 1950

USA, Louisiana, Orleans Parish, New Orleans, 13.x.1934, H.L. Dozier, 2

(“Sweeping grass in back yard”) [USNM].

Distribution: Southern Palaearctic and Afrotropical (South Africa), southern Nearctic (Louisiana, USA) and Neotropical (Mexico and Trinidad).

Triapitsyn (1997) gave the background information about the distribution of this species and also provided an illustration of the forewing. It is for the first time recorded here from the United States and the Nearctic region.

Anagrus (Anagrus) stethynioides Triapitsyn sp. n.

HOLOTYPE

on point, GUATEMALA, Sacatepequez, Chylla (ca. 34 km from Guatemala City) lat 14°38’N, long 90°42’W, el 6877 ft, 26.x.1998, M. Hoddle (on avocado) [USNM]. PARATYPES: ARGENTINA, Misiones, Loreto, 23.viii.2000, P. Fidalgo, 1

on point (“YPT in Ruinas Jesuíticas”) [IMLA]. GUATEMALA, same data as holotype, 1

on slide [UCRC]. MEXICO: Nuevo León, San Juan, Río San Juan, 14.vii.1983, A. González H., 1

on slide. Yucatán, Chichén Itza, 27.vii.1984, G. Gordh, 1

on point [UCRC]. TRINIDAD, Curepe, CIBC Lab. grounds, 26.ii-13.iii.1974, F.D. Bennett, 1

on slide [CNCI]. USA, Texas, Uvalde Co., Speir Ranch, 3 mi. NW Uvalde, 6.v.1977, T. Eichlin, M. Wasbauer, 1

on slide [CSCA].

Female (holotype and paratypes): Coloration: General color yellow to light brown with the following parts darker (brown): transverse trabecula, gena, mouthparts, flagellum (clava dark brown), anterior half of mesoscutum, axillae, propodeum, legs, lateral spots on gastral terga, and tip of ovipositor sheaths. Eyes and ocelli dirty pink. Wings hyaline. Head about as wide as mesosoma; stemmaticum present. Mandible tridentate. Antenna (Figure 1) with scape about 3.0 x as long as wide; F1 oval, F2 and F3 very short, cylindrical, either subequal or F2 slightly longer than F3; F4-F6 subequal in length and width; longitudinal sensilla on F4 (2 or 3), F5 (2 or 3) and F6 (2 or 3); clava about as long as combined length of F5 and F6 or a little longer, with six longitudinal sensilla. Mesosoma: Mesoscutum with adnotaular setae. Posterior scutellum somewhat longer than in most known Anagrus species, 0.65-0.70 x as long as wide; each paramedial plate 1.40-1.45 x as long as wide. Forewing (Figure 2) 4.6-4.9 x as long as wide; longest marginal cilia subequal to or only slightly longer than maximum wing width; marginal vein usually without a microchaeta between proximal and distal macrochaetae. Forewing blade narrows just beyond venation before broadening considerably; disc rather densely setose beyond narrowing, leaving two indistinct bare spots, one in the middle and the other near posterior margin. Lengths of distal and proximal macrochaetae in ratio 1.2-1.5:1. Hind wing blade bare except for rows of microtrichia along margins. Metasoma: Ovipositor barely reaches mesophragma anteriorly and posteriorly slightly exserted beyond its apex. Ratio of total ovipositor length to length of its exserted part 7.0-9.0:1. External plates of ovipositor each with two setae in distal half. Ovipositor length/protibia length ratio 2.5-2.6:1.

Measurements (n=3): Body (taken from dry specimens): 708 (583-812). Antenna: scape (including radicle): 96 (91-102); pedicel: 45 (40-47); F1: 18 (15-22); F2: 24 (19-29); F3: 24 (17-32); F4: 54 (44-62); F5: 54 (44-66); F6: 53 (44-60); clava: 111 (93-124). Forewing: 579 (482-683)/124 (99-150); proximal macrochaeta: 57 (52-66); distal macrochaeta: 79 (63-95); longest marginal cilia: 154 (102-193). Hind wing: 552 (462-646)/28 (25-31). Ovipositor: 317 (255-365).

Male: Similar to female in color. Antenna typical of the genus, flagellomeres subequal except F1 slightly shorter. Forewing (Figure 3) 4.5 x as long as wide; marginal vein with a microchaeta between proximal and distal macrochaetae; bare spots on disc more distinct than in female. Genitalia (not illustrated because the single available specimen is poorly mounted) elongated, appear to be similar in shape to male genitalia in some typical representatives of the “incarnatus” species group, as described and illustrated by Chiappini and Mazzoni (2000); digitus with two hooked spines.

Measurements (n=1): Body: 637. Antenna: 740. Forewing: 573/128; longest marginal cilia: 168. Genitalia: 156.

Diagnosis: This species is easily distinguished from all other described species of Anagrus in having the clava with six sensory ridges. It is the sole described member of the “stethynioides” species group of Anagrus s. str. as defined above.

Etymology: The name refers to the superficial resemblance of the new species to the members of Stethynium.

Anagrus (Anagrus) frequens Perkins, 1905

Distribution: Australasian, Afrotropical, Oriental and eastern Palaearctic; Neotropical (Colombia, Ecuador, Mexico, St. Kitts, Trinidad); Oceania, Hawaiian Islands (introduced).

St. Kitts is a new distribution record for this widespread species, an egg parasitoid of various planthoppers (Triapitsyn and Beardsley 2000). It is very likely that A. frequens was long ago accidentally introduced into the Neotropics from Australia or the Orient with the sugarcane crop.

Anagrus (Anagrus) breviphragma Soyka, 1955

COLOMBIA, Bogotá, 22.ix.1975, Lazardo, Posada, 4

(“ex. Dalbulus maidis”) [USNM].

Distribution: Palaearctic and Neotropical (Argentina, Brazil, Colombia, Guadeloupe, Guyana).

Hosts: See Moratorio and Chiappini (1995), Triapitsyn (1997, 1999), and Virla (2001).

Colombia is a new country record for A. breviphragma, a common egg parasitoid of the corn leafhopper, Dalbulus maidis (De Long in Wolcott, 1923), in South America (Virla 2001).

Anagrus (Anagrus) yawi Fullaway, 1944

ARGENTINA, Buenos Aires, Buenos Aires, ix.1999, S.G. Bado, 3

, 9

(“Ex. eggs of Tenthecoris bicolor Scott (Miridae) on Crinum zeylandicum and C. asiaticum (Amaryllidaceae), Jardín Botánico Lucien Haumann, FAUBA”) [IMLA, UCRC]. HAITI, Port-au-Prince, 30.xii.1929, H.L. Dozier, 1

, (“ex. sweetpotato foliage infested with Protalebra”) [USNM]. HONDURAS, Fco. Morazán, San Antonio del Oriente, El Zamorano, 30.ix.1987, G. Evans, 1

(“on Cupressus ?lusitanica”) [FSCA].

Distribution: Nearctic (Mexico and USA) and Neotropical (Argentina, Haiti, Honduras, Mexico); introduced into Hawaii (USA).

Hosts: The mirid bug Tenthecoris bicolor Scott, 1886 is an additional host record to those indicated by Triapitsyn (1997).

This species was redescribed and illustrated by Triapitsyn and Beardsley (2000) based on the type specimens from Mexico and additional material from Hawaii.

Anagrus (Anagrus) urichi Pickles, 1932

BRAZIL: Distrito Federal, Brasilia, 16.ix.1991, Ileana, 7

, 3

on eight slides (ex. “Deois flavopicta (Cercopidae) on Brachiaria decumbens” [palisade grass]) [FSCA, UCRC]. Goiás, Campinaçu: lat 13°52’S, long 48°23’W, 19-25.ii.1996, 3

[UCRC]. GUYANA, Demerara, 11.viii.1944, H.C. James, 4

on points (“from soil contg. stages of Tomaspis flavilatera”) [USNM].

Female: Coloration: body and appendages brown to dark brown except for the following parts lighter (pale to light brown): scape, pedicel, posterior scutellum, and legs. Wings slightly infuscated; venation brown. Head as wide as mesosoma; stemmaticum present. Mandible tridentate. Antenna (Figure 4) slender; scape with faint sculpturing (longitudinal and transverse on opposite sides), without distinct cross-ridges; F1 subglobular, F2 and F3 subequal in length, each markedly shorter than following segments; F4 slightly longer than F5 but a little shorter than F6; clava slightly longer than combined length of F5 and F6; longitudinal sensilla only on F4 (2), F5 (2), F6 (2 or 3), and clava (5). Mesosoma: Mesoscutum with adnotaular setae; mesoscutum, axillae, anterior scutellum and paramedical plates of posterior scutellum with weak cellulate sculpturing. Forewing (Figure 5) shorter than body, slightly curved, 9.0-9.5 x as long as wide; disc with one incomplete, irregular median row of six to ten setae in distal half. Lengths of distal and proximal macrochaetae in ratio 1.3-1.5:1. Hind wing blade with one row of microtrichia beyond venation. Metasoma about as long as mesosoma or slightly longer. Ovipositor not overlapping mesophragma anteriorly and posteriorly notably exserted beyond apex of gaster. Ratio of total ovipositor length to length of its exserted part 4.9-5.3:1. External plates of ovipositor each usually with two, in some specimens with three, distal setae. Ovipositor length/protibia length ratio 2.5-2.6:1.

Measurements (n=3, specimens from Brasilia, Brazil): Body: 649 (628-683). Antenna: scape (including radicle): 89 (80-95); pedicel: 45 (44-47); F1: 18 (18-19); F2: 47 (44-51); F3: 47 (45-49); F4: 55 (51-58); F5: 53 (51-55); F6: 57 (55-59); clava: 115 (111-115). Forewing: 505 (485-533)/55 (51-58); proximal macrochaeta: 51 (48-55); distal macrochaeta: 71 (66-73); longest marginal cilia: 167 (160-172). Hind wing: 490 (482-507)/18 (17-19). Ovipositor: 307 (292-321).

Male: Similar to female in coloration and other morphological features except for sexually dimorphic characters. Unlike in female, scape with distinct cross-ridges, otherwise antenna (Figure 6) somewhat unusual for genus: F1 subglobular, either without longitudinal sensilla or with one short longitudinal sensillum, much shorter than other flagellomeres; F2 and apparently F3 with two, F4-F11 with four longitudinal sensilla each. Forewing 9.1-9.3 x as long as wide, with one incomplete median row of seven to eight setae in distal half of blade. Genitalia (Figure 7) very unusual both for “incarnatus” species group and genus: genital capsule subtriangular, aedeagus very short, and digitus without spine.

Measurements (n=2, specimens from Brasilia, Brazil): Body: 540-637. Forewing: 464-514/51-55. Genitalia: 68-69.

Diagnosis: This species belongs to the “incarnatus” species group of Anagrus s. str. It differs from all other described species in that group by the unique combination of antennal characters and the distinctive forewing that has only one incomplete median row of setae in the distal half of disc. The closest species to A. urichi appears to be A. lineolus Triapitsyn, 1999 which has the female antenna with only one longitudinal sensillum on F4 and F5 and also a complete median row of setae on the forewing disc. The male genitalia in A. lineolus are also triangularly shaped, but the genital capsule is much narrower and the aedeagus much longer (Triapitsyn 1999).

Hosts: De Santis (1979) recorded A. urichi from Brazil and Trinidad from the following froghopper (Cercopidae) hosts: Aeneolamia varia saccharina (Distant, 1909) (as Tomaspis saccharina Distant, 1909 and Aeneolamia varia (Fabricius, 1787)), Mahanarva fimbriolata (Stål, 1854), and Tomaspis sp., without, however, referring to any specimens. The material examined in this paper indicates two new records of the froghopper hosts of A. urichi: Deois flavopicta (Stål, 1854) in Brazil and Aeneolamia flavilatera (Urich, 1914) in Guyana. The latter host association may be traced to the work of James (1947). Additional froghopper hosts were indicated by Guagliumi (1962, 1971) for “Anagrus sp.” in Brazil and Venezuela; it is quite likely that those unidentified species could be A. urichi. However, it is impossible to verify that without studying the voucher specimens, if such exist.

Triapitsyn (1997) gave the background information about the status of the type material of this species, which is apparently lost. The original description by Pickles (1932) is very poor as he indicated mainly the generic characters of Anagrus; however, there is no doubt that the two series of specimens studied here belong to the species described as A. urichi, based on a thorough comparison with his illustrations, the description, and also on the fact that all of them were reared from the eggs of Cercopidae. The specimens from Guyana are all on points and are shriveled, and therefore their body length (316-366) is much smaller than in the specimens measured above; the clava is pale, possibly due to fading. Therefore, the above redescription is made from the slide-mounted series of specimens from Brazil.

Anagrus (Anagrus) lineolus Triapitsyn, 1999

MEXICO, Chiapas: Ocozocoautla, “El Kikapu”, 15-20.vii.1984, G. Gordh, 1

. PERU, Madre de Dios, Manu National Park, lat 11°55’S, long 77°18’W, 1-30.ix.1986, K. Person, 1

[UCRC]. USA, Florida: Manatee Co., Bradenton: 8.vi.1990, 1

(on watermelon) [FSCA]; 20.vi.1990, 1

[UCRC]. Osceola Co., Canoe Creek, Turnpike Plaza, 21.vi.1991, F.D. Bennett, 1

(“Bemisia tabaci material on Emilia sonchifolia”) [FSCA].

Distribution: Southern Nearctic (Florida, USA) and Neotropical (Argentina, Brazil, Mexico, Peru).

It is likely that A. lineolus is present throughout the tropics and subtropics of the Neotropical region, where it appears to be rather common in bulk samples. It is for the first time recorded here outside of the type localities in Argentina and Brazil.

Anagrus (Anagrus) oahuensis Triapitsyn & Beardsley, 2000

MEXICO, Sinaloa: 11 mi. N La Concha, 25.x.1982, J.T. Huber, 1

. 12 mi. N Mazatlán, 25.x.1982, J.T. Huber, 2

. Yucatán, Chichén Itza, 27.vii.1984, G. Gordh, 2

[UCRC].

Triapitsyn and Beardsley (2000) described this species from the Island of Oahu, Hawaii and supposed that it might be a native species there. The discovery of A. oahuensis in Mexico makes that assumption almost certainly incorrect because it is more likely that it was accidentally introduced from there to Oahu Island.

Anagrus (Anagrus) armatus (Ashmead, 1887)

HOLOTYPE

on slide, labeled: 1.”Litus armatus Ashm., Jacksonville, Fla. Type”; 2.”Anagrus armatus (Ashmead), E, USNM Type No. 13810" [USNM]. In the original description, Ashmead (1887) did not designate the type(s). According to Girault (1911), who slide-mounted many of Ashmead’s types including probably this specimen, the type designation had been written in Ashmead’s handwriting on the original label.

While visiting the USNM in 2000, I found the second specimen that likely belongs to the type series of Ashmead’s Litus armatus and thus might represent a paratype, labeled as follows: 1.”Litus armatus Ashm.”; 2.”E. Fla Ashmead.” [USNM]. This point-mounted female (length of body 512) is in a much better condition than the holotype and, very importantly, it has a complete set of wings that are missing on the holotype.

Other material examined: PUERTO RICO [USA], Lajas, 22.xi.1935, H.L. Dozier, 1

on slide (“Sweeping grass and small weed at small pond edge along border of Lake Cartagena”) [USNM].

Female: Coloration: General color light brown; head (except face), flagellum, anterior half of mesoscutum, basal terga of gaster, and sheaths of ovipositor darker (brown). Head: Width 164; stemmaticum present. Mandible tridentate. Antenna (Figure 8) with F1 almost cylindrical, F2 longest of funicle segments; F3-F6 progressively shorter than preceding segment; clava slightly shorter than combined length of F5 and F6; longitudinal sensilla on F3 (1), F4 (1), F5 (2), F6 (2), clava (5). Mesosoma: Mesoscutum with adnotaular setae. Forewing (Figure 9) slightly shorter than body, narrow, 10.5 x as long as wide; disc with one complete median row of setae and two additional incomplete rows of setae in the distal half, leaving narrow bare area along posterior margin. Lengths of distal and proximal macrochaetae in ratio 1.5:1. Metasoma: Ovipositor anteriorly broadly rounded, very much alike as in many species of Omyomymar Schauff, 1983, strongly overlapping mesophragma, and posteriorly markedly exserted beyond apex of gaster. Ratio of total ovipositor length to length of its exserted part 3.5-3.7:1. External plates of ovipositor apparently each with two distal setae. Ovipositor length/protibia length ratio 3.7:1.

Measurements: Holotype: Body: 551. Head: 114. Antenna: scape (including radicle): 91; pedicel: 46; F1: 22; F2: 72; F3: 67; F4: 65; F5: 63; F6: 57; clava: 110. Mesosoma: 0.171. Legs: [given as femur, tibia, tarsus]: fore (pro-): 129, 118, 167; middle (meso-): 114, 190, 163; hind (meta-): 123, 217, 160. Metasoma: 0.266; ovipositor: 437.

Non-type

, Puerto Rico (parts of antennae missing): Body: 637. Forewing: 610/58; proximal macrochaeta: 55; distal macrochaeta: 85; longest marginal cilia: 197. Hind wing: 573/18; longest marginal cilia: 161. Ovipositor: 500.

Diagnosis: This species belongs to the “incarnatus” species group of the subgenus Anagrus s. str. and is probably related to the Nearctic A. delicatus Dozier, 1936, which lacks a longitudinal sensillum on F3 of the female antenna.

Distribution: Southern Nearctic (Florida, USA) and Neotropical (Puerto Rico).

Numerous previously published records of this species, other than of the original Ashmead’s material from Florida, USA, are incorrect (Chiappini et al. 1996). Thus, the female from Puerto Rico is the only non-Ashmead’s specimen of A. armatus known to date. I was thus wrong to assume that A. armatus does not occur there although the previous record of this species from Puerto Rico cited by De Santis and Fidalgo (1994) was almost certainly incorrect due to a misidentification (Triapitsyn 1997).

Anagrus (Anagrus) raygilli Triapitsyn, 2000

DOMINICAN REPUBLIC, Bani, Villa Sombrero, lat 18°16’N, long 70°21’W, 19.iv.2000, M. Hoddle, 3

(on avocado). MEXICO, Baja California Sur, Las Barracas (~30 km E Santiago, lat 23°28’20"N, long 109°27’10"W), P. DeBach: 16.xii.1984, 1

; 14.iv.1985, 1

. Michoacán, Uruapán, 18-20.x.1999, M. Hoddle, 6

(on avocado). USA, Florida, Manatee Co., Bradenton, 20.vi.1990, 1

[UCRC]. VENEZUELA, Carabobo, Samán Mocho, 3.viii.1982, Kinido Gómez, 1

(on cassava) [CNCI].

Distribution: Southern Nearctic (Mexico and Florida, USA) and Neotropical (Guatemala, Dominican Republic, Mexico, Venezuela).

This species appears to be present throughout the native range of avocado plant following its leafhopper host(s); the only one known, I. minuenda, is a minor pest of avocado in Mexico.

Anagrus (Anagrus) empoascae Dozier, 1932

USA: Delaware, New Castle Co., Newark, 3.viii.1927, H. L. Dozier, 1

(“on window”) [USNM]. Florida: Alachua Co., Gainesville: 17.v.1990, F.D. Bennett, 2

(“on Cucumis melo”); 6.x.1990, F.D. Bennett, 1

(“on Euphorbia heterophylla”); 1991, Stonaker, G. Evans, 1

(“on Lycopersicon lycopersicum”). Orange Co., Turkey Creek, 12.vi.1990, F.D. Bennett, 1

“on Solanum americanum”. Palm Beach Co., West Palm Beach, 13.vi.1992, F.D. Bennett, 1

(“on Chamaesyce hyssopifolia”) [FSCA]. Louisiana, Webster Parish, Buckhorn, 27.vii.1937, H.L. Dozier, numerous

and

on three slides (“sweeping from Helianthus infested with Empoasca erigeron De Long”). South Carolina, Richland Co., Columbia, 10.ix.1916, A.H. Beyer, 2

, 1

(“on alfalfa. Webster No. 12196, ex. Holticus citri eggs” [USNM]. VENEZUELA, Lara, El Taque, 24.iv.1991, F.D. Bennett, 1

(“on Chamaesyce hyssopifolia”) [FSCA].

Distribution: Nearctic (Mexico and USA) and Neotropical (probably throughout, except South); Hawaiian Islands.

Hosts: See Triapitsyn (1997). Most likely, this species can parasitize eggs of many Empoasca spp. including Empoasca erigeron De Long, 1931 mentioned above. According to Beyer (1921), eggs of the mirid bug Halticus bractatus (Say, 1832) were parasitized by Anagrus nigriventris Girault, 1911; however, examination of his voucher specimens from Columbia, South Carolina, has shown that this record was incorrect due to a misidentification of A. empoascae.

This common New World species was redescribed and illustrated by Triapitsyn (1997) based on the type series from Haiti.

Anagrus (Anagrus) flaveolus Waterhouse, 1913

BELIZE, Tower Hill, iv.1966, numerous

and

(“ex. Saccharosydne rostifrons eggs”). CUBA, Baragua, 17.ii.1930, L.D. Christenson, 2

(“ex. eggs of Saccharosydne saccharivora (Westwood)”). HAITI: Damien, 25.i.1930, H.L. Dozier, 1

(“ex. Peregrinus maidis on corn”); Port-au-Prince, 9.viii.1930, H.L. Dozier, numerous

and

(“ex. eggs of Saccharosydne saccharivora on sugar cane”). JAMAICA: vii.1959, F.D. Bennett, numerous

and

on two slides (“ex. eggs of Saccharosydne saccharivora”); 4.viii.1966, 2

, 2

(“ex. eggs of Saccharosydne saccharivora (Westwood)”) [USNM]. USA: Arizona, Santa Cruz Co., 7 mi. S Patagonia, 27.viii.1979, J. LaSalle, 1

. California, Orange Co., Huntington Beach, 23.xii.1981, H. Andersen, 2

[UCRC]. Florida, Hendry Co., Clewiston, 3.iv.1961, J.R. Gifford, 3

, 1

(“ex. eggs of Saccharosydne saccharivora”). VENEZUELA, Aragua, El Consejo, 5.viii.1950, P. Guagliumi, 1

(“ex. eggs of Saccharosydne saccharivora (Westwood)”) [USNM].

Distribution: Nearctic (Mexico and southern USA) and Neotropical (probably throughout, except temperate South).

This common, mainly Neotropical species is an economically important biological control agent in Argentina (Triapitsyn 1997) and, according to the above host records, elsewhere in the New World as an egg parasitoid of the leafhopper and planthopper pests on corn and sugarcane.

Anagrus (Anagrus) nigriventris Girault, 1911

BRAZIL, Rio de Janeiro, nr. Ipomena, 25.v.1991, F.D. Bennett, 1

(“on Sonchus sp.”) [FSCA].

Distribution: Nearctic and Neotropical (probably throughout the New World from Canada to Argentina and Chile); Hawaiian Islands.

This common species, especially in North America, is for the first time recorded from Brazil.

Acknowledgment

I thank Gregory A Evans (FSCA), Patricio Fidalgo (IMLA), Raymond J Gill (CSCA), John T Huber (CNCI) and Michael E Schauff (USNM) for the loans of specimens and access to the respective collections. Vladimir V Berezovskiy (UCRC) made line drawings and provided other technical assistance.