HOLARCTIC TRIBES OF THE ICHNEUMON-FLIES OF THESUBFAMILY ICHNEUMONINAE (PIMPLINAE).
By R. A. CusHMAN and S. A. Rohwer,Of the Bureau of Entomology, United States Department of Agriculture.
INTRODUCTION.This paper, which is a joint contribution of the Branch of Decid-uous Fruit Insect Investigations and the Branch of Forest Insects ofthe Bureau of Entomology, is the result of study extending over aperiod of several years, and embodies the opinion of the authors as tothe relationship and number of tribes of the Ichneumoninae (Pim-plinae Authors) as represented in the Holarctic region. As originallyplanned, we had hoped to prepare a joint paper on the entire group,but since, because of interruptions by other work, such a paperis found impracticable we have considered it advisable to presenta tribal synopsis as a basis for subsequent revision of the minorgroups.The change of the subfamily name is necessary because the typeof the genus Ichneumon is Ichneumon manifestator Linnaeus ^ and hasas synonyms the isogenotypic genera Pimpla Fabricius and EphialtesGravenhorst (not Schrank).^ The subfamily name Pimplinae musttherefore be suppressed in favor of Ichneumoninae and the subfamilyIchneumoninae of authors will be Joppinae after the name of the oldestincluded genus. HISTORICAL.The beginning of the classification of the Ichneumoninae dates ofcourse from Linnaeus, but since the writers preceding Gravenhorst
1 Morice and Currant, (Trans. Ent. Soc. Lond., 1914, p. 388) contend that the type of the genus Ichneumonis Ichneumon persuasorius Linnaeus, a species given by Lamarck (1801) as an example for the genus Ichneu-mon. We can not agree with them in this nor do we believe that the rulings in the Code of International Zoo-logical Nomenclature will uphold the acceptance of the Lamarckian examples as tj^pe designation. The codespecifically says "The meaning of the expression 'select a type' is to be rigidly construed. Mention of aspecies as an illustration or example of a genus does not constitute a selection of a type." Lamarck andmost of the older -nTiters, including most of Latreille's works, gave examples only as an illustration of thogenus and not as a type designation.
? See Cushman and Rohwer, Proc. Ent. Soc. Wash., vol. 20, 1919, p. 186.Proceedings U. S. National Museum. Vol. 57?No. 2315, 379
380 PROCEEDINGS OF THJE NATIONAL MUSEUM. vol. 57.had done little more than describe genera and species they contributedonly slightly to the taxonomy of the group and it is not necessary todiscuss their work in detail.From the beginning the classification of the Ichneumoninae hasbeen rendered unsatisfactory and difficult by the exclusive employ-ment of characters of such nature that they apply to but one sex,or are conspicuous only in more or less extreme types, while absentor inconspicuous in other genera obviously closely allied. In addi-tion all recent classifications have followed very closely that ofFoerster which was based almost entirely on Holarctic genera, andthe attempt to adapt this arrangement to the genera of the world hasadded much to the confusion.Gravenhorst.?In 1829 Gravenhorst^ published the first realattempt at the classification of the Ichneumonidae. He divided thefamily into thirteen genera. Most of these he subdivided into anumber of families or subgenera to which he gave names, and somefew of these he further divided into unnamed groups which he calledsections. AU of Gravenhorst's subgenera and many of his sectionshave since been raised to generic rank.Of the genera treated by Gravenhorst as subgenera that are hereplaced in the subfamily Ichneumoninae, seven, Glypta, Lissonota,PolysiyTiinda, Clistopyga, Pimpla, EpJiialtes, and Khyssa, he consid-ered as subgenera of Pimpla. As subgenera of Pimpla Gravenhorstalso included ScTiizopyga and (TracJiyderma) = Tylocomnus, both nowplaced in the subfamily Tryphoninae. Xorides, Xylonomus, Odon-tomerus, and the Cryptine genus EcMhrus constituted his genusXorides. Coleocentrus and Arotes were the unnamed subgenera IVand VI respectively of his genus Banchus. Pliytodietus he placedas subgenus VIII of Cryptus. Acoenites formed a genus by itselfwithout subdivisions.Gravenhorst's key is in the form of a chart classifying the insectsdown to subgenera, which are bracketed into their genera. Thecharacters employed are mostly superficial, indefinite, or unisexualand have largely persisted up to the present for the major divisionswithin the group.Holmgren.?Holmgren^ separated the Ichneumonidae into fivefamilies corresponding to the usual five subfamilies. He made nokey to these families, but gave a rather long description of each,mentioning nearly all parts.
1 J. L. C. Gravenhorst, Ichneumonologia Europaea, Vratislaviae, Sumtibus Auctoris, 1829, vol. 1, pp.i-xxxiii, 1-827; vol. 2, pp. 1-939; vol. 3, pp. 1-1097.
' A. E. Holmgren, Forsok till Uppstallning oeh Beskrlfning af de i Sverige Funne Tryphonider, KonigligaSvenska Vetenskaps-Akademiens Handlingar, 1855, pp. 93-394, 2 pis.
No. 2315. TRIBES OF ICENEUMONINAE?CUSHMAN AND ROHWER. 381In 1859 ^ he published his first synopsis of his family Pimplariae.This he divided into two main sections which he called Pimplariaeand Xorides, the latter corresponding to the tribe Xoridini of Ashmead,and the former including genera since divided into the tribes Acoeni-tini, Lissonotini, and Pimplini. His main divisions are based onpractically the same characters as those used by Gravenhorst, whilethose of the smaller divisions were new and still largely persist inthe more recent keys. The generic descriptions are full and detailed,and the Swedish species are listed under each genus. In the follow-ing year he published a larger work ^ in which the generic key islargely reprinted from the earlier paper, but each Swedish species isdiscussed in considerable detail. His two main sections are herecalled subfamilies.Cresson.?The fii*st American writer to take more than a generalinterest in the Hymenoptera was Cresson, who in 1887 publishedhis Synopsis.^ For this work Cresson claims little originality, con-fessing to having compiled his keys from the writings of previousauthors. He did, however, a valuable work in marshaling theknown North American species and added much to the knowledgeof the group in America,Cresson's key to the Pimplinae is much easier to use than mostothers. He, however, made no attempt to divide the subfamilyinto tribes nor to express by his key the relationship of the generato each other. His specific keys, based largely on color, are useful,although too much reliance must not be placed on characters of thissort.Foerster.?A few years after Holmgren had published his synopsisFoerster* produced his system of classification of the Ichneu-monidae. In tliis work he divided the group into 34 coordinatefamilies, 4 of which, the Pimploidae, Lissonotoidae, Acoenitoidae andXoridoidae, together with Ashmead's tribe Labenini, constitute thefive tribes into which Ashmead divided the subfamily Pimplinae.The Xoridoidae represent Plolmgren's section II; the Acoenitoidae,section I, division 1 ; the Pimploidae, section I, division 2, phalanges1 and 2; and the Lissonitoidae, section I, division 2, phalanx 3.
? A. E. Holmgren, Conspectus Generum Pimplarlarum Suecia, Ofversigt af Konigliga Svenska Veten-skaps-Akadeniiens Forhandlingar, vol. 6, 1859. pp. 121-132.2 A. E. Holmgren, Forsok till Uppstallning och Beskrifning af Sverlgos Ichneumonider, MonographiaPimplarlarum Sueciae, Konigliga Svenska Vetenskaps-Akademiens Handlingar, vol. 3, No. 10, 1860, pp.1-76.3 E . T. Cresson, Synopsis of the Families and Genera of the Hymenoptera of America North of Mexico,1887, supplementary volume of Trans. Amer. Ent. Soc, pp. i-vi, 1-350.< Arnold Foerster, Synopsis der Familien und GattungenderIchneumonen,Verh.nat.hist. Ver.preuss.Rheinl., vol. 25, 1868, pp. 142, 162-170.
382 PROCEEDINGS OF THE NATIONAL MUSEUM. vol.57.For his larger divisions Foerster used many of the old Graven-horstian characters supported by those of Holmgi-en, but addedmany new ones to define his much finer subdivisions. He tabulatedand named many new genera, most of them without further diag-nosis and without including any species or designating types. This,together with the extremely minute differences which Foerster con-sidered of generic value, makes the determination of his genera verydifficult. Indeed, many of them stand to-day without includedspecies. The difficulties in tliis respect are perhaps less in theIchneumoninae than in almost any other gi'oup, although even thereone must exercise considerable liberality in the interpretation ofcharacters to satisfactorily place a species in its genus.Practically all authors since Foerster have followed him veryclosely . Especially is this true of Ashmead and Schmiedeknechtwhose keys are largely translations or adaptations of Foerster, withnew genera and new characters interpolated occasionally.Foerster's work, left unfinished as it was, has thi-own much lighton the classification of the Ichneumonoidea, but because of his un-supported use of unisexual and variable characters it has also addedmuch to the confusion of this difficult group.Thompson.?In his treatment of the Ichneumoninae in his Opus-culaEntomologica Thomson^ followed largely the System of Holmgren.He published no tribal nor generic tables, but his keys to speciesabound in new and useful characters. It is indeed unfortunate thatThomson cUd not apply his clear insight to an attempt to clarifythe classification of the Ichneumonidae as a whole. No otherworker has appreciated as did he the extent of variation in the groupor the little dependence that can be placed in the superficial charac-ters used for the separation of the larger groups.His contributions concerning the Icluieumoninae are scatteredthrough several fascicles of his Opuscula Entomologica and con-sist largely of keys to Swedish species and observations on thosespecies.Davis.?In presenting his review of North America TiyphoninaeDavis ^ gives, without gi'ouping them into subfamilies, a synopsisof the tribes of the Ichneumonidae. This synopsis follows veryclosely (so closely in fact that up to couplet 10 it is a translation)Foerster's key of the natural families of Ichneumonidae, and has itsuse mainly in being its author's interpretation of Foerster and ingiving a definition of the tribes he treats.
>C. G.Thomson, Opuscula Entomologica, Lund, fascicles 5, 8, 9, 12, 13, 19, and 21,1873-1896.* G. C. Davis, A Review of the Ichneumonid subfamily Tryphoninae, 1897, Trans. Amer, Ent. Soc,vol. 24, pp. 193-348.
No. 2315. TRIBES OF ICHNEUMONINAE?CUSHMAN AND ROHWER. 383Ashmead.?Ashmead^ in Ms treatment of the Iclineumoninaegroups together, and considers as tribes, Foerster's families Acoeni-toidae, Lissonotoidae, Pimploidae, and Xoridoidae and adds the tribeLabinini, a group not represented in Europe. In his method oftreatment and in his choice of characters Ashmead follows very closelythe work of Foerster, and in the main his paper is a translation ofFoerster with the addition of new and a subsequently describedgenera. Many of the characters are taken only from the female,which makes it impossible to satisfactorily place males, and thevenation is used extensively. The shape and presence or absenceof the areolet is used repeatedly as a primary character and muchvalue is attached to the angulation of the discocubitus, the presenceor absence of a ramulus, the position of the nervulus, and the pointof fractm-e of the nervellus. In fact the entire classification isfounded on an insufficient and superficial study of a few types. Thecharacters offered will not apply to all of the species which wereplaced in the various genera as arranged in Ashmead's collectionor that of the United States National Museum as it was arrangedby him. Unsatisfactory as his classification is, it has been usefulbecause it brought together and gave some characters for the numer-ous genera described up to 1900. It must be remembered, however,that Ashmead endeavored to include all the described genera, and in anumber of cases was forced to use only the descriptions which arefrequently insufficient and offer only characters that are often ofquestionable value.Schmiedeknecht.?The treatment of the subfamily Ichneumo-ninae as given by Schmiedeknecht^ in the Genera Insectorum adds butlittle information which will aid in the satisfactory classification ofthese insects. The work is fomided, in great part, on that ofFoerster and Ashmead, and is a conservative adaptation of theirwork with the recently described genera included. There are, how-ever, some transfers of genera and in some places certain groupswhich Ashmead treated as genera are treated as subgenera, yet manyof the mistakes made by Ashmead are copied and the same kind ofcharacters are used. It is, however, a useful work and if it showsbut little origmality we can perhaps excuse the author because of thedifficulty of the group, the area covered, and the lack of representa-tives of many of the genera.
1 W. H. Ashmead, Classification of the Ichneumon Flies, or the Superfamily Ichneumonoidea, Proc.U. S. Nat. Mus., vol. 23, 1900, pp. 1-220.* otto Schmiedeknecht, Subfamily Pimpilinae, Gen. Ins., fasc. 62., 1907, pp. 1-120, pis. 1-2.
384 PROCEEDINGS OF THE NATIONAL MUSEUM. vol. 57.Morley.?Claude Morley in his recent papers ^ dealing with theinsects grouped together as the subfamily Ichneumoninae has offereda number of new suggestions in arrangement and expressed a doubtthat all the members are closely related, as the following quotation(la, p. xv) will show: "That the Lissonotides have any closerelationship with the typical Pimplides I do not for a moment believe
;
the Acoenitides, as at present grouped, are very heterogeneous ; andthe Banchides are admittedly aberrant, wherever placed; whilethe Xoridides, though related to some extent in their thoracicstructure with Rhyssa, appear worthy of ranking as a distinct sub-family."In 1908 (la) Morley adds the tribe Banchides to his subfamilyPimplinaeandin 1913 (1 6) he raises the genus Rhyssa &nd allies to tribalrank (in 1908 he still had this group in the Pimplini) and makes atribe, Ecthromorphides, for the genera Lissopimpla and EctliromorpJia.This last tribe is an arbitrary grouping on two variable venationalcharacters and the lengthening of the malar space, the latter sovariable as to be of doubtful generic value.In his definition of the subfamily Pimplinae (la, p. xvi) he makesuse of a secondary sexual character and adds in a qualifying way anextremely variable specific character. The key to the tribes (la,p. 1) makes use of some of the usual characters and one is at a loss toknow how the Theronini can be placed in the Pimplides as he definesthem. It would seem that Morley has done but little more than offera rearrangement of names, for when he has given additional charactersthey are usually of such nature as to be subject to individual varia-tion or are unisexual and should not be used, unsupported, as primecharacters of genera or higher groups.It must not, however, be implied that we would belittle the workof Morley, because with all its shortcomings it is very useful andclears up many obscure points about the species which are repre-sented in the British Museum by type material, and gives useful keysto distinguish the material in that museum.Viereck.?In the recent synopsis of the genera of Ichneu-mon flies of Connecticut, Viereck,^ does away with subfamily divi-
1 (a) Claude Morley, Ichneumonologia Britannica III. The Ichneumons of Great Britain, etc., Pim-plinae, 1908, H. and W. Brown, London, England, pp. i-xvi, 1-328.(6) Claude Morley, A revision of the Ichneumonidae Based on the collection of the British Museum, part2, Tribes Rhyssides and Echthromorphides, 1913, London, pp. i-vi, 1-48.(c) Claude Morley, Idem., part 3, tribe Pimplides, 1914, pp. i-viii, 1-122.(<?) Claude Morley, Idem., part 4, tribe Banchides, 1915, pp. ix-x, 135-151.2 H. L. Viereck, The Hymenoptera, or Wasp-Uke Insects, of Connecticut, 1917, Bull. 22, Geol.and Nat.Hist. Survey Conn., pp. 243-326.
No. 2315. TRIBES OF ICHNEUMONINAE?CUSHMAN AND ROHWER. 385sions/ and, after removing the closely allied genus Banchus on a char-acter subject to specific variation, he separates the genera belonging tothe Ichneumoninae as here defined from the other genera of the Ichneu-monidae m much the same unsatisfactory manner as that of Ashmead,and in many places only on secondary sexual characters. There are,however, a few new characters, and the work has contributed some-thing to render the Ichneumoninae of the limited region somewhatbetter understood. CLASSIFICATION.In presenting this revision of the subfamily Ichneumoninae we donot wish to be construed as expressing an opinion that it is a naturalgroup. The family Ichneumonidae is a group composed of elementsshowing remarkable differences but at the same time extreme homo-geneity. So true is the latter that the grouping into five universallyrecognized subfamilies is, in our opinion, untenable. The publishedkeys for the separation of these five subfamilies leave the placing of aspecies in its proper subfamily ahnost entirely to the imagination orexperience of the worker. On the other hand, the strict interpreta-tion of such characters as these keys offer frequently leads even theexperienced taxonomist to entirely misplace an insect; and disagree-ment among workers as to the allegiance of certain genera or groupsof genera is very frequent. The Plectiscini, Banchini, and Panisciniare notable as bones of contention, while all of the subfamilies, notablythe Tryphoninae and Ophioninae, are aggregations of groups not ataU closely allied but thrown together on such superficial charactersas compression or depression of abdomen, long or short ovipositor,possession or lack of sternauli, shape, presence or absence of theareolet, etc. Anyone who has studied the insects of this group inan at all intensive manner knows that such characters as theseare valueless as used in the keys, and he also knows that in de-termining species he places them in the subfamilies without regardto the characters of the keys but entirely from his knowledge ofgenera.The subfamily Ichneumoninae, as treated here, is practically thesame as the subfamily Pimplinae in the sense of Holmgren, Cresson,Ashmead, and others, with a genus here and there rejected as notreferable to the subfamily. We have adopted this restriction of thesubfamily as a matter of convenience and in accordance with the
J Since the above was written Viereck has (Proc. Biol. Soc. Wash., vol. 31, 1918, p. 69) elevated all of thetribes of the Ichneumonoidea to subfamily rank for the very naive reason, "to avoid confusion in referringto groups of genera by word of mouth." In Ent. News, vol. 31, 1920, p. 16, he makes a family for the generaLabena Cresson, Apechoneura Kreichbaumer, and Psiloparia, new genus. The value of this grouping hasbeen discussed by Cushman, Proc. Ent. Soc. Wash., vol. 22, no. 4, 1920.144.382?20?Proc.N.M.vol.57 25
386 PROCEEDINGS OF THE NATIONAL MUSEUM. vol. 57.project as originally conceived, which was to revise the NorthAmerican Pimplinae.It seems very probable that the Ichneumoninae as here treated iscomposed of a number of phylogenetic branches, associated by moreor less similarity of hal)itus and superficial structure, and forming agroup which, because of its discordant elements, is so closely re-lated to groups in other subfamilies as to make definition of it as aunit extremely difficult if not impossible. The same is true, togreater or less extent, of the other four subfamilies. Because ofthis, it does not seem advisable to offer any other definition of thelimits of the subfamily Ichneumoninae than to say that this tribalsynopsis is based on the PimpHnae as limited by Ashmead and thegenera described since 1900. The student unfamiliar with theseinsects will therefore necessarily have to refer to the unsatisfactorydefinitions offered by Ashmead, Cresson, and others.In spite of the probable diversity of origin of the tribes, constantdiagnostic characters are very rare, and our keys to the tribes in-clude the most available characters. Not all of these characterstaken singly are of tribal value; in fact such characters are very rarein the Ichneumonidae. It is rather an assemblage of characters,which taken together form the peculiar structure and habitus thatdistinguish the members of one tribe from those of another.The remarks in regard to the subfamihes set forth above areequally appUcable to the usual five tribes of the Ichneumoninae.None of the existing keys to these tribes is usable by the beginner inthe taxonomic study of these insects unless he has access to an ex-tensive named collection for comparison. This is due partly to theattempt of the authors of the keys to place all the genera in thefive tribes originally proposed by Ashmead; partly to the use ofextreme or unisexual characters unsupported by characters applyingto the other sex, or of those variable within a genus, or the positivestatement of characters that do not apply to all species placed in agiven tribe; and partly to a too superficial study of the species, or,when careful study has been made, a laissez-faire policy of notattempting a revision of the keys. In other words, in order to runa given species to its genus one must be able from the knowledgegained only by long experience to tell at once the tribe to which thespecies should be referred. The tribal keys, being unusable by thebeginner and unnecessary to the experienced, are useless, or worse(so far as the beginner is concerned), misleading.We are convinced that the entire biology and the charactei-s dis-played by all stages are of importance as indicating relationship ordivergence and that whatever of this sort of information is availableshould be taken advantage of as a guide to classification. In pro-
No. 2315. TRIBES OP ICHNEUMONINAE?CUSHMAN AND ROHWER. 387posing the following arrangement of tribes we have therefore takeninto consideration what we know of larval structure, nature of thehost, and relation of parasite to host. We have found that thestructure of the ovipositor and of the terminal segments in thefemale have been developed similarly in genera which have similarlarval structures and host relations. These characters in the femaleare of such nature that they can be easily seen and expressed briejflyand positively. We have therefore considered it advisable topresent a key based entirely on the female, making use of sexual andsecondary sexual characters. We recommend the use of this key tobeginners and others unfamiliar with the group and believe that lessdifficulty will be experienced in its use than in the use of the keyin which the secondary sexual characters are eliminated. Thislast mentioned key, while not entirely satisfactory, serves to dis-tinguish the tribes which we believe to be represented among thegenera studied and sets forth the bestcharacters applicable to both sexes.There are three new terms which areused in this paper, and the authors ex-pect to make use of these terms infuture works. They may be defined asfollows
:
A perpendicular rbcrveUus (fig. 1,/, g)is one in which the anterior end is op-posite the posterior end, that is, one in fig. i.-nervelli: o,6andc,reclivous;which a line drawn touching both the ^ and ?,inclivous;/and g,?EnPE^-
, . . , DICXJLAE.anterior and posterior ends is at rightangles to the longitudinal axis of the submediallan cell. Otherauthors have called this a continuous or interstitial nervellus.An inclivous nervdlus (fig. 1, d, e) is one in which the anterior endis nearer the base of the wing than is the posterior end. This hasheretofore been spoken of as an antefurcal nervellus.A reclivous nervellus (fig. 1, a,h, c) is one in which the posteriorend is nearer the base of the wing than is the anterior end. Thishas heretofore been spoken of as a postfurcal nervellus.We have substituted the terms inclivous and reclivous for theambiguous and unsatisfactory terms antefurcal and postfurcal; andhave adopted the less used term perpendicular because we believethat it is less confusing and more in keeping with the terms incliv-ous and reclivous. The terms interstitial, antefurcal, and post-furcal are used in the usual sense for such veins as the nervulus andrecurrents and in this sense are easily understood and in keepingwith the exact meaning of the words.
388 PROCEEDINGS OF THE NATIONAL MUSEUM. VOL. 07.
KEY TO TKIBE3 BASED ON PEMALES.
1. Ovipositor vnih a dorsal notch a short distance back from apex; (internal parasitesof Lepidopterous larvae), fig. 2 2.
kFig. 2.?AncES of ovipositors: a, Glypta simplicipes Cresson; 6, Lampronota Americana Cresson;c Arenetra nigkita Walsh; d, Meniscus scxttellaris Cresson; e, Cylloceria lugubris Cresson;/,' Lampronota frigida Cresson; g, Lissonota verberans Gravenhorst; Ii, Amersibia prionoxy-STI EOnWER.Ovipositor without such a notch2. Tergites without oblique furrows Lissonotini.Tergites with oblique furrows extending from basal middle to near apicalmargin - Glyptini.3. Tarsal claws pectinate; apex of ovipositor spear-head like; (parasites of Lepidop-terous larvae), fig. 3 4.
FiQ. 3.?Apices of ovreosiTORs: o, Toxophoeide3 albomaeginata (Ceesson); &, Phytodietus burgessiCresson. Hind tarsal claw: c, Phytodietus buegessi Cresson.Tarsal claws simple or with a large basal tooth or lobe, fig. 4 -5
Fig. 4.?Hind tarsal claws: a, Itoplectis conquis-itoe (Say); 6, Ichneumon irritatoe Fabricius. Fig. 5.?Apex of female ab-domen OF TOXOPHoroidesalbomarginata (Cresson)(A=HYPOPyGIDIUM.)
4. Tergites 1-4 with oblique and apical transverse furrows and strongl> sculptured;scutellum margined laterally; hypopygidium heavilj chitinized and extendingto or beyond apex of abdomen, fig. 5 Lycorini.Tergites \nthout furrows and polished; scutellum not margined; hypopygidiumneither especially heavily chitiuizcd nor prominent Phytodietini.
No. 2315. TRIBBS0FIGHNEVM0NINAE?CU8HMAN AND ROHWER. 389
5. Ovipositor short, never more than half as long as abdomen, compressed, with a dis-tinct swelling below at or about the middle, beyond which it tapers to a veryacute point; clypeus strongly convex, rounded at or most truncate at apex, nevermedially impressed or inflexed; last tarsal joints swollen, claws and onychiavery large, all claws with basal tooth; face narrow, convergent below; mandiblesnarrow at apex, upper tooth much the longer; areolet only rarely defined; (ex-ternal pai'asites on spiders), fig. 6 , Polysphinctini.
Fig. 6.- -Apices of ovrPosiTORs: a.PoLYSPHiNCTA texana Cresson; b, Hymengepimecis wiLTn (Cres-soN); Manbible: c, Hymenokpimecis wiltu (Cresson).Ovipositor either short or long, but never formed as above; clypeus most frequentlyimpressed and emarginate medially, occasionally inflexed and truncate orrounded at apex; apical tarsal joints rai'ely swollen or with large claws andonychia; mandibles either broad and bidentate at apex with equal teeth oracute and edentate, in the latter case rarely with a small inner tooth 6.6. Ovipositor never nearly as long as body, cylindrical, or nearly, occasionally de-pressed or decurved at apex; claws simple, without a basal lobe or tooth, occa-sionally (Itoplectis) with claws of front tarsi lobed or (Apechthis) all or front andmiddle claws lobed, in the last genus the ovipositor is decurved at apex; notaulieither absent to obsolete or very deep and pit-like anteriorly, where they aieset off by sharp carina that runs back along the margin of the lateral lobe; areoletalways present; nervellus always strongly reclivous with the discoidella at ornear the upper end ; clypeus broadly truncate or arcuate at apex, rarely with adistinct median notch-like emargination 7.Ovipostor compressed, or if cylindrical it is very long and slender or upcurved ; allclaws either with or without basal lobes or teeth; notauli strong, rarely weak, orentirely wanting, but never defined as above 8.7. Dorsal margin of lance straight to apex; propodeal spiracle slit-like, the surroundingcarina prominent, separated from anterior margin of propodeum by less than itslength; notauli subparallel, terminating abruptly posteriorly; polished, withabdomen impunctate; species usually largely bright ferruginous or yellowish;(secondary parasites), fig. 7 Theroniini.
Fig. 7.?Apex of ovepositobOF Theronia fulvescens Fig. 8.?Apices of ovipositors: a, Itolplectis conquis-Cresson. itor (Say); 6, Apechthis picticornis (Cresson).Dorsal margin of lance either decurved near apex or it is flattened at apex; propodealspiracle usually round to long oval, rarely slit-like, and usually separated fromanterior margin of propodeum by at least its length; notauli when strong com-plete and convergent posteriorly; species usually black or blackish with abdo-men distinctly punctured, seldom both pale and with abdomen polished im-punctate; (internal parasites of Lepidopterous pupae), fig. 8 Ephialtini.
390 PROCEEDINGS OF THE NATIONAL MUSEUM. VOL. 57.
8. Hypogidium very large, voraeriform, acute at apex, very heavily cliitinized;clypeus broadly truncate at apex, frequently eharply inflexed and with a moreor less distinct median tooth; labrum exserted, (parasites on wood-boringlarvae), fig. 9 Acoenitini.
10
FIG.39.?APEXJOF ABDOMEN OF FEMALE OF COLEOCENTRUS OCCIDENTALIS CrESSON.Hypopygidium not as above, usually retracted far from apex of abdomen, veryrarely (CHstopyga) reaching apex 9.Occipital carina obsolete or interrupted dorsally; mesoscutum and scutellumtransversely rugose throughout; apical tergite greatly lengthened; (externalparasites on wood-boring larvae) Bhyssini.Occipital carina complete; mesoscutum and scutellum not transversely rugose, atmost the scutum is rugulose; apical tergite only rarely greatly lengthened. 10.
, Abdomen inserted above, frequently far above, the hind coxae; first tergitenarrow throughout; head transverse; occiput narrow, barely concave; templesshort and strongly convexly sloping; eyes emarginate within; propodeumnearly straight and horizontal from base to insertion of abdomen; hind coxae,long, slender and nearly uniform in diameter, fig. 10 Labenini.
Fig. 10.?AreoletOF Labena obal-LATOR (Say). Fig. 11.?Areolets: a, Tromatobia rufovariata(Cresson); 6, Itoplectis coNQxnsrroB (Say);c. Epiubus albokicta (Cresson).
11. Not agreeing entii'ely with above, fig. 11 11Abdomen sessile (not distinctly tapering from spiracles to base and with promi-nent anterior lateral angles), very rarely {Perithous) clavate and slightly com-pressed at apex; areolet usually defined; claws rarely without basal tooth; (ex-ternal parasites on lepidopterous, coleopterous, and hymenopteroua larvae andpupae, or in spider egg-sacs), fig. 12 Ichmeumonini.
Fig. 12.?Sessile hrst tergite ofPerithouspleubalis Ckesson. Fig. 13.?Petioatel first tergite of Xor-IDES YUK0NENSI3 (ROHWER).Abdomen petiolate (tapering from spiracles to base, and without prominent anteriorlateral angles), clavate to subcylindrical and more or less compressed apically;areolet usually wanting; claws without basal tooth; temples broad; (externalparasites on wood-boring larvae), fig. 13 12.
No. 2315. TRIBES OF IGHNEUMONINAE?GUSHMAN AND ROHWER. 391
12. Mandibles edentate at apex, rarely with a email entodoreal tooth; legs slender,fig. 14 Xoridini.
Fig. 14.?Mandible of Poemenia Americana (Cres.sok).Mandibles bidentate at apex, the teeth subequal in length ; legs stout . Odonfommni.KEY TO TRIBES.1. Abdomen inserted above, frequently far above, the hind coxae, first tergite narrowthroughout; head transverse; occiput narrow, completely margined, barelyconcave; temples short and strongly convexly sloping; eyes emarginate within;propodeum nearly straight and horizontal from base to insertion of abdomen;hind coxae long, slender and nearly uniform in diameter; thoracic dorsum notat all transversely rugose Labenini.Not agreeing entii'ely with above 2
.
2. Mandibles edentate or with a much shorter entodorsal tooth; first tergite petiolate,spiracles before middle; areolet usually wanting; thorax depressed, mesopleuradistinctly longer than high; head subquadrate; notauli complete or nearly bo.Xoridini.Mandibles bidentate apically, teeth subequal or upper tooth longer 3.3. Occipital carina wanting or interrupted medially; mesoscutum and scutellumtransversely rugose tliroughout; abdomen inserted rather high or propodeum,occasionally far above insertion of hind coxae; first tergite with spiracles beforemiddle and shorter than or subequal to second, which is parallel-sided. Rhyssini.Occipital carina complete; mesocutum and scutellum not transversely rugose, atmost the mesoscutum partially mgoluse 4.4. Abdomen distinctly compressed in apical third or half, (deeper than broad).Acoenitini.Abdomen not distinctly compressed 5.5. Abdomen petiolate; head subcubical, swollen below antennae, not, or scarcely,narrowing behind eyes; eyes small and placed well forward, cepbalo-candadlength of posterior orbits longer than or subequal to that of eye; thorax andpropodeum depressed, the latter very long dorsally, short posteriorly; legs,especially the femora, stout; areolet wanting Odontomerini.Not entirely as above, though rarely agreeing with one or two ch aracters 6.6. Tergites, at least 2-4, with oblique furrows which converge anteriorly until theyapproximate in the dorsal middle 7.Tergites without such furrows 8.7. Tergites 1-5 in male, 1-4 in female, with apical transverse impressions whichtogether with oblique impressions set off a median, transverse, sub-triangulararea; malar furrow present; first tergite with dorsal carinae short; scutellumcarinate laterally to apex; intercubitus nearly or quite twice as long secondabscissa of cubitus; nervellus strongly inclivous Lycorini.First tergite without either oblique or transverse impressions, and with dorsalcarinae extending beyond middle; other tergites usually without transverseapical furrows; ' scutellum not carinate laterally; intercubitus not nearlytwice as long as second abscissa of cubitus; nervellus reclivous, perpendicular,or slightly inclivous Glyptini.
? None of the North American Glyptini have the transverse furrows, but the South American genusZaglypto7norpha Viereck has them on tergites 2-5. This genus, however, has none of the other charactersof the Lycorini.
392 PROCEEDINGS OF THE VATIOUAL MUSEUM. vol. 57.
8. Tergitea beyond first without either furrows, depressions, or elevated aieas; dorsalcarinae of first tergite defined at most only very briefly at base (in difficultspecies the spiracles of first tergite are very close to the base), mesoscutumanteriorly usually with a cuneiform pale spot on each side 9.Tergites beyond first with more or less distinct elevated areas, depressions, or fur-rows or combinations of or all of these factors; dorsal carinae of first tergitedistinct and setting off of a distinct basal concave area (in the very rare difficultspecies the spiracle of the first tergite is far from the base) 10.9. Propodeum entirely without carinae; claws strongly ctirved, with few (about 6)very long, closely set teeth; entire body smooth, at most very minutely punc-tate Phytodietini,Propodeum usually with at least an apical transverse carina, rarely without carinae;claws long, weakly curved and if pectinate the teeth are smaller, more numer-ous, or sparsely set; at least the thorax dorsally and propodeum distinctlysculptured Lissonolini.10. Propodeal spiracle slit-like, the surrounding carina prominent, separated from theanterior margin of the propodeum by less than its length; notauli subparallelending abruptly posteriorly; body smooth and shining, mostly bright ferru-ginous or yellow; propodeal carinae very strong and high Theroniini.Propodeal spiracle round or elongate the surrounding carinae not prominent,removed from the anterior margin of the propodeum by at least its length;notauli obsolete or converging posteriorly; usually sculptured and dark colored,occasionally ferruginous or polished, but rarely both; propodeal carinae obso-lete or weak, at least not veiy high and strong 11.11. Notauli weak or absent; or if very strong and complete they are deep and pitlikeanteriorly and set off by a sharp carina that runs back along the lateral marginof the mesoscutum ;i head set very close to prescutum; mesopleirral furrowstraight or curved but not angulate opposite the punctiform fovea. . . Ephialtini.Notauli usually deep, at least anteriorly; the anterior margin of the mesoscutumdistinctly trilobed; head, by reason of the longer pronotum, set off from theprescutum; mesopleural furrow angulate opposite punctiform fovea 12.12. Notauli strongly impressed throughout, prescutum very prominent (if notauliare not strongly impressed, as in Hymenoepimeds, the prescutum is neverthe-less very prominent and the other characters are especially well marked);temples fiat or slightly convex, sloping to the strong occipital carina; faceconverging below and at least as long as wide at clypeus, the latter convex orslightly flattened, usually rounded at apex and with a reflexed margin, rarely{Hymenoepimeds) very weakly, broadly emarginate, never medially impressedor inflexed; mandibles narrow at apex, upper tooth distinctly the longer;Bcutellum elevated and compressed from the sides; areolet very rarely defined.Polysphinctini.Notauli rarely complete, weakly impressed posteriorly, prescutum not especiallyprominent (if complete and prescutum prominent, as in Clistopyga, the insectdiffers radically in other characters) ; temples usually strongly rounded ; veryrarely flat, less sharply sloping; face usually wider than long; clypeus usuallymedially impressed and emarginate at apex, sometimes inflexed and truncateor very weakly emarginate ; teeth of mandibles subequal in length ; scutellumbroad, convex, or flattened; areolet usually complete, occasionally wanting orincomplete Ichneumionini.
? None of the Holarctic genera have the notauli strong, the genera in which they are strong being princi-pally oriental.
No 2315. TRIBESOFICHNEVMONINAE?CVSHMAN ANDROHWER. 393
Tribe LISSONOTINI.As here defined this tribe includes most of the genera placedthere by Ashmead and other writers. Of the Nearctic and Pale-arctic genera Eyho'phanes Foei-ster and Phytodietus Gravenhorst areexcluded. HyhopJianes, we agree with Thomson, is a Tryphoninebelonging in the subtribe Thymaridina, tribe Mesoleptini; Phyto-dietus forms the new tribe Phytodietini; while Phidias, unknown tous except from descriptions and Vollenhoven's figure, will verylikely not run here, and probably should be referred to anothersubfamily.The group is very homogeneous, and v/hen once understood iseasily recognized. It is very closely allied, especially throughArenetra Holmgren, to the Banchini as represented by ExetastesGravenhoi"st and its nearest allies. The males of some of the Lisso-notine genera are likely to be confused with males of the Tryphoninesand apparently certain portions of that very heterogeneous sub-family are rather closely related to the present tribe. Within thesubfamily as here treated its closest relative is the Glyptini, theabdominal structure being the only real difference, and these twotribes form a group not at all closely related to the rest of thesubfamily. GLYPTINI, new tribe.The tribe Glyptini is founded for the genus Glypta Gravenhorstand its allies, Teleutaea Foerster, Dihlastomorpha Foerster, andGonoblasta Foerster. Ctenochira Foerster, and Hoplitophrys Foersterare unknown to us, but apparently belong here. All of these generahave heretofore been referred to the tribe Ichneumonini.The remarks above concerning the affinities of the Lissonotiniapply in large part to the Glyptini.LYCORINI, new tiibe.As here defined this tribe includes of described genera only LycorinaHolmgren and Toxophoroides Cresson. These genera have hereto-fore been placed in the tribe Ichneimionim, to which they are perhapsmore closely allied than to the Glyptini, with which the structure ofthe tergites superficially allies them. The real affinities of the tribeare very obscure. PHYTODIETINI, new tribe.The only genus known to us that is referrable here is PhytodietusGravenhorst, heretofore placed in the Lissonotini. Although super-ficially resembling the Lissonotini it is doubtful if it is closely alliedto that tribe. It may be that it has some affinity with the Ly-cormi, and the ovipositor suggests the possibility that they mayhave had a common origin with the Ciyptinae.
394 PROCEEDINGS OF THE NATIONAL MUSEUM. vol. 57.THEEONHNI, new tribe.Heretofore the genera of this tribe have been placed in theIchneumonini. As here defined but two genera, Theronia Holmgrenand Neotheronia Krieger, occur in the Holarctic fauna. In thetropical regions certain other allied genera occm*.In the form of the ovipositor and the secondarily parasitic habitthe Theroniini are very distinct, though probably more closely alliedto the following tribe, Ephialtini, than to any of the other tribes.EPHIALTINI, new tribe.The type genus of this tribe is Ephialtes Schrank {=-PimplaAuthors and Pimplidea Viereck)^ while the other genera are ItopledisFoerster and ApechtMs Foerster, in addition to several tropical andoriental genera, such as Xantliopimpla Saussure, EcMhromorphaHolmgren, and AllotJieronia Ashmead.Except in superficial facies the Ephialtini are very distantly relatedto the Ichneumonini, to which the genera have almost universallybeen referred. POLYSPINCTINI, new tribe.This tribe is erected for the genera Polyspincta Gravenhorst,Acrodaciyla Haliday, Colpomeria Holmgren, Zatypota Foerster, andHymenoepimecis Viereck, all heretofore assigned to the Ichneumonini.Their very peculiar habits ally them much more closely than toany of the other tribes of the Ichneumoninae to certain of the Try-phoninae, such as Monohlastus Holm^gren and Polyhlastus Hartig.We believe that the facies and biological aihnities exhibited by thesetwo groups is of much greater importance as indication of relation-ship than are the superficial characters of form of abdomen andlength of ovipositor.The following six tribes we believe form the true Ichneumoninae.All are externally parasitic and each is related by more or less inter-mediate genera or by common characters of structure and hatitus toone or more of the others. Tribe LABENINI.Ashmead was the first to recognize this group as a tribe, but inhis classification he included also the Ophionine genus NonnusCresson. As defined here the tribe includes, of described genera,only Labena Cresson and Grotea Cresson.EKYSSINI, new tribe.The only author to teat this group as of tribal rank is Morley,other authors having placed the genera comprising it in the Ichneu-
1 See Cushman and Rohwer, Proc. Ent. Soc. Wash., vol. 20, (1918) 1919, p. 186.
No. 2315. TRIBES OF WHNEUMONINAE?CUSHMAN AND ROHWER. 395
monini. The only described genera occurring in this region whichbelong to this tribe are Rhyssa Gravenhorst and Megarhyssa Ashmead,but in other parts of the world others occur, and all the genera treatedby Morley in his tribe Rhyssides appear to belong to the Rhyssini ashere defined.Through Apechoneura Kriechbaumer it is related to the Labenini,and, through certain genera of the Xoridini and Ichneumonini tothose tribes. Tribe ACOENITINI.As here restricted this tribe embraces only those genera which, inthe female, have the hypopygidium very long vomeriform andpolished. In the Holarctic fauna this includes, of the genera whichwe have examined, only Arotes Gravenhorst, Coleocenfrus Graven-horst, Acoenites Latreille, Phaenolohvs Foerster, and MesoclistusFoerster.Of the genera placed in this tribe we have not had opportunity toexamine Asthenomeris Foerster. The type of Asthenomeris has neverbeen described; but according to Schmiedeknecht the genus is inter-mediate between the Acoenitini and Banchini. Crypturus, synony-mous with Endurus Rondani, was transferred by Schmiedeknechtto the Tryphoninae, where it forms his subtribe Endurina of thetribe Mesoleptini. This treatment of the genus seems to us thelogical one.Leptobates Gravenhorst and Procinctus Foerster we place with theBanchini; Apltanorhoptrum Foerster with the Tryphonini, where itis closely allied to Stilhops Foerster, removed thence from the Ich-neumonini ; and Collyria Schi0dte to the Mesoleptini, where it wouldform a distinct subtribe.In biological habits and ovipositor and clypeal charactei-s, togetherwith somewhat similar general form, this tribe is most closely alliedto the Rhyssini. Tribe XOKIDINI.The tribe Xoridini of previous classifications is a very hetero-geneous group. As here restricted it indues of the Holarctic faunathe genera Deuieroxorides Viereck, Xorides Latreille sensu latiori( = Xylonomus Gravenhorst), and Foe/n^ma Holmgren (= CaUiclisisFoerster)
.
The genera Echthrus Gravenhorst, Nyxeophilus Foerster, Helcos-tizus Foerster, Xylopliruridea Viereck, (= Cryptoideus Ashmead),and Xylophrurus Foerster we exclude entirely from the subfamily,placing them in the Cryptinae.^ Odontomerus Gravenhorst andAplomerus Provancher are removed to form the allied tribe Odon-tomerini. Helcosiizidea Rohwer, originally placed by its author in
1 See Cushman, Proc. U. S. Nat. Mus., vol. 55, 1919, p. 536.
396 PROCEEDINGS OF THE NATIONAL MUSEUM. vol, 57.the Xoridini, we are agreed is Campopligine and allied to PyracmonHolmgren.Through Deuteroxorides the Xoridini are related to the Rhyssiniand through Xcyrides to the Odontomerini.ODONTOMEBINI, new tribe.Erected for the genera Odontomerus Gravenhorst and AplomerusProvancher, this tribe is most closely related to the Xoridini, especi-ally to the genus Xorides Latreille.Tribe ICHNEUMONINl.From this tribe as treated by Ashmead we have withdrawn thegenera constituting the tribes Rhyssini, Lycorini, Glyptini, Poly-sphinctini, Theroniini, and Ephialtini. Of the Nearctic and Pale-arctic genera that remain we have had no opportunity to examinespecimens of the following: Trociocerus Woldstedt, AtractogasterKJriechbaumer, OpisorTiyssa Kriechbaumer, Idiogramma Foerster,Tromera Foerster, Eremochila Foerster, and Panteles Foerster.Stilhops Foerster, Dyspetes Foerster, ScMzopyga Gravenhorst andPolyspinctomorpJia Ashmead, are in our opinion Tryphonine, thefirst belonging to the Tryphonini, the second to the Mesoleptini, andthe third to the Exochini; while PolyspinctomorpJia is Mesoleptineand synonymous with Neliopisihus Thomson.^Through Pseudorhyssa Merrill this tribe is rather closely alhed tothe Rhyssini.
"Cushman, Proc. U. S. Nat. Mus., vol. 56, 1919, p. 378.