4 S y stematics and Taxonom y 1 . Intr oduc t ion S y stematics ma y be defined as the stud y of the kinds and diversit y of or g anisms and th e relationships amon g them. Taxonom y , the theor y and practice of identif y in g , describin g , namin g , and classif y in g or g anisms, is an inte g ral part of s y stematics. Classification is the ar - rangement o f organ i sms i nto groups ( t ax a , s i ngu l a r t axo n )ont h e b as i so f t h e i rre l at i ons hip s. I t f o ll ows t h at id ent ifi cat i on can ta k ep l ace on l ya f terac l ass ifi cat i on h as b een esta bli s h e d .It s h ou ld b e emp h as i ze d t h at not a ll aut h ors a d opt t h ese d e fi n i t i ons. Taxonom yi so f ten use d a s as y non y mofs y stematics (as defined above), while classification is sometimes used rathe r loosel y (and incorrectl y )asas y non y m of identification . S ystemat i cs i sanact i v i ty t h at i mp i nges on most ot h er areas o fbi o l og i ca l en d eavor. Yet , i ts i mportance (an dfi sca l support f or i t) seem to h ave di m i n i s h e di n recent years. To some extent, t hi sma yb et h e f au l to f s y stemat i sts w h o ten d to wor ki n i so l at i on, o f ten f ocus i n g o n some small and obscure g roup of or g anisms. This ma y be especiall y true of entomolo g ica l s y stematists who, faced with the enormous diversit y of the Insecta, tend to be seen as “counters o fb r i st l es,” “measurers o fh ea d w id t h ”an dp er f ormers o f ot h er act i v i t i es o fli tt l e re l evance to t h e outs id ewor ld .In f act, as Dan k s (1988) e l egant l ypo i nte d out, not hi n g cou ld b e f urt h er f rom t h e trut h .S y stemat i cs h as p l a y e d ,an d cont i nues to p l a y ,ama j or role in fundamental evolutionar y and ecolo g ical studies, for example faunistic surve y s, z oo g eo g raphic work, life-histor y investi g ations and studies of associations between insect s an d ot h er organ i sms. In app li e d entomo l ogy goo d systemat i cwor ki st h e b as i s f or d ec i s i on s on t h e management o f pests. In d ee d , Dan k s (1988) prov id e d examp l es o f pest-management pro j ects i nw hi c hi na d equate or f au l t y s y stemat i cs resu l te di n f a il ure, somet i mes w i t hg reat economic and social cost ( and see Section 2 ). The taxonom y of insects, like that of most other g roups of livin g or g anisms, continue s t o be based primarily on external structure, though limited use has also been made (some- ti mes o f necess i ty, espec i a ll y b etween spec i es) o f p h ys i o l og i ca l , d eve l opmenta l , b e h av i ora l, an d cytogenet i c d ata. Mo l ecu l ar bi o l og i ca l ana l yses o f pro bl ems i n i nsect systemat i cs h ave increased exponentiall y over the past two decades (Caterino et al ., 2000). These anal y ses , principall y usin g mtDNA sequences, have principall y focused on the resolution of rela- t ionships at lower taxonomic levels, for example, among subspecies, species and species groups. Mo l ecu l ar p h y l ogenet i c stu di es o fhi g h er i nsect taxa (e.g., re l at i ons hi ps amon g 91 92 CHAPTER 4 o rders), thou g h far fewer, have nevertheless g enerated important, sometimes even contro- v ersial, conclusions (see Cha p ter 2 for exam p les). T he purpose of this chapter is to provide a short introduction to the s y stematics o f i nsects, i nc l u di ng some o f t h e tec h n i ca l terms app li e db ywor k ers i nt h ese fi e ld s, as a b as is f or Chapters 5 –10 inclusive, which deal with individual insect orders. 2 . Namin g and Describin g Insects Fo r a v a r i ety o f reasons b ut most o b v i ous l yt h e enormous di vers i ty w i t hi nt h ec l ass Insecta an d econom i c cons id erat i ons, i nsect taxonom i sts usua lly wor k w i t hi n f a i r ly narro w boundaries. Onl y b y doin g this can the y acquire the necessar y familiarit y with a particular g roup (includin g knowled g e of the relevant literature) to determine whether the speci - m en t h ey are exam i n i ng h as b een d escr ib e d an d name d or may b e new to sc i ence. Even a f ter a part i cu l ar group h as b een c h osen f or stu d y, t h ere are typ i ca ll y super i mpose dbi o - g eo g rap hi c constra i nts, t h at i s, taxonom i sts restr i ct t h e i r stu di es to part i cu l ar g eo g rap hi c re g ions . M an y frequentl y encountered insects, especiall y pests, have a “common name” b y w hich they are known. The name may refer to a particular species (e.g., house fly) or to a l arger group (e.g., scorp i on fli es) an d re fl ects a c h aracter i st i c f eature o f t h e i nsect’ s appearance or h a bi ts. Un f ortunate l y, i nsects o f w id e l y diff erent groups may h ave s i m il a r habits (e. g ., so-called “leaf miners” ma y be larvae of Diptera, Lepidoptera, or H y menoptera) o r the same common name ma y refer to different species of insects in different parts of the w o rld. Thus, to avoid possible confusion, each insect species, like all other organisms bot h f oss il an d extant, i sg i venaun i que l at i n i ze dbi nom i a l (two-part) name, a system i ntro d uce d b yL i nnaeus i nt h e ear l y 1700s. In t h e Lat i n name, w hi c hi sa l ways i ta li c i ze d ,t h e fi rst wor d denotes the g enus, the second the species (e. g . , Mu s ca d ome s tica f or the house fl y ). Rarel y , the name has three parts, the third indicatin g the subspecies. (It should be noted, however , that some national entomolo g ical societies such as those of the United States and Canad a p u bli s hli sts o f t he a pprove d common names f or spec i es i nor d er to a ll ow t h e i r use, ye t a v o id poss ibl em i sun d erstan di ng.) Spec i es are norma lly di st i n g u i s h e d on t h e b as i so f a sma ll num b er o fk e yf eatures ( character s ) that exist in a s p ecific c haracter stat e i n each species (e. g ., “number of tarsa l se g ments” is a character, and “five tarsal se g ments” is a character state). Thus, a taxonomist will b ase t h e d escr i pt i on o f a new spec i es on t h ec h aracters a l rea d y esta bli s h e df or ot h e r spec i es i nt h e same group to f ac ili tate compar i son w i t h t h em. Care f u l co ll ect i on an d curat i o n ( preparat i on, preservat i on, an d ma i ntenance) o f spec i mens are cr i t i ca l to taxonom y to ensure that potentiall y important characters (which ma y be minute and delicate) are not dama g ed. The specimens must be properl y labeled with the date and place of collection (preferabl y us i ng map coor di nates) an d t h eco ll ector’s name. To f ac ili tate proper ma i ntenance, as we ll as access ibili ty f or f urt h er stu di es, spec i mens are usua ll ysu b m i tte d to a centra l repos i tory, t h e name o f w hi c hi s i nc l u d e di nt h epu bli s h e dd escr i pt i on o f t h e spec i es, to b ecome par t o f the reference collection . T he s p ecimens whose descri p tion leads to the establishment of a new s p ecies form t h e t ype ser i e s , one only of which becomes the standard reference specimen, th e h olo- t y p e ,t h eot h ers i nt h e ser i es b e i ng parat y pes. T h e name g i ven to a new spec i es must f o ll ow t h eru l es an d un i versa l nomenc l atura l system l a id d own b yt h e Internat i ona l Com- m ission on Zoolo g ical Nomenclature (published in the International Code of Zoolo g ica l 93 S Y S TEM A TI CS A ND TA XONOM Y N omenclature). The species-specific part of the name ma y be a g enuine Latin word, as in t he dra g onfl y Hemicordulia fl ava (from the Latin “flavus” meanin gy ellow, referrin g to the extensive y ellow coloration on the bod y ), or ma y be a latinized form of a word, for example, a name o f a person or p l ace, as i nt h e d amse lfl y N eosticta f raseri, name df or t h e Austra li a n amateur o d onato l og i st, F. C. Fraser. Somet i mes, aut h ors s h ow remar k a bl e i mag i nat i on i n nam i n g a spec i es, ma ki n g stu dy o f t h e d er i vat i on o fi nsect names (“entomo l o gi ca l et y mo l- o gy ”?) a fascinatin g sub j ect in its own ri g ht. Take, for example, the Australian kat y di d Kawanaphila lexcen i R entz 1993 (in Rentz, 1993), the g eneric name of which is derived f rom t h ea b or i g i na l wor d “ k awana” mean i ng fl ower, a re f erence to t h e f act t h at a ll k now n spec i es f requent fl owers, w hil et h e spec i es i s name di n h onor o f Ben Lexcen, d es i gner o f th e Amer i cas Cup c h a ll enge r A u s tra l ia II ,in whi c h t h e k ee li ss i m il ar to a structure ( t he sub g enital plate) on the female kat y did! Similarl y , the damselfl y Pseudagrion jedda W atso n and Theischin g er 1991 (in Watson et al. , 1991 ) receives its name from the 1955 film J edda , parts of which were shot in Katherine Gorge, Northern Territory, Australia, the t ype localit y (p l ace o f co ll ect i on o f t h e h o l otype) f or t h e spec i es! In pu bli cat i ons, a spec i es’ name w h en fi rst ment i one di sg i ven i n f u ll ,an d may b e f o ll owe db yt h e name o f t h eor i g i na ld escr ib er (authorit y ), which ma y be abbreviated, and sometimes the y ear the description was pub- lished as in the two precedin g examples. In some cases, the name of the authorit y (and d ate) a pp ears in p arentheses as, for exam p le, in the termite Por otermes adamsoni rr ( Froggatt , 1897), s h ow i ng t h at t h e spec i es was d escr ib e dfi rst un d er a diff erent genus, su b sequent l y s h own to b e i ncorrect. In t hi s examp l e, Froggatt or i g i na ll yp l ace d t h e spec i es i nt h e genu s Calotermes. A s noted above, most s p ecies are described on the basis of their structure, es p e- ciall y external characters. However, on occasion such “morphospecies” are not equiva- l ent to bi o l og i ca l spec i es (repro d uct i ve l y i so l ate d popu l at i ons); t h at i s, groups t h at can - not b e diff erent i ate d structura ll y may nevert h e l ess b e true bi o l og i ca l spec i es an d are sa id t o b e“s ibli n g spec i es.” Suc h spec i es h ave b een d etecte dby avar i et y o f means, i nc l u d - in g their different host preferences (e. g ., some mosquitoes), matin g behavior (courtship son g s in some kat y dids), and c y to g enetics (kar y ot y pes of some black flies). The reco g - n i t i on o f s ibli ng spec i es an d t h e i r h ost spec ifi c i ty are cr i t i ca ll y i mportant i n bi o l og i ca l contro l programs. For examp l e, i nt h e contro l o f pr i c kl y pear ( O puntia spp.) b y cater- p ill ars o f C acto bl astis ( see C h apter 24, Sect i on 2.3), i t i snow b e li eve d t h at t h e“s l ow ” start made b y the insects ma y have been due to introduction of the “wron g ” siblin g species which failed to establish themselves, not an unsuitable climate as su gg ested earlier (McFadyen, 1985) . If a new spec i es i ssu ffi c i ent l y diff erent t h at i t cannot b e ass i gne d to an ex i st i ng genus, a ne w g enus i s propose d , f o ll ow i ng t h e same cons id erat i ons as f or spec i es w i t h respect t o name, authorit y , and date as, for example, An ax L each 181 5 , and this species is then de n oted as t h e t ype spec i es for this g enus. Since 1930, it has been a requirement for a t ype species to be selected for any new genus. For genera described before this time an d l ac ki ng a type spec i es, t h eCo d e spec ifi es h ow t h e type spec i es s h ou ld b e d eterm i ne d . W i t hi n a genus, espec i a ll y one w i t h many spec i es, t h ere may b ec l ear l y d e fi ne d group s o f spec i es, an d eac hg roup ma yb e gi ven i ts own su bg ener i c name p l ace d parent h et i ca lly after the g enus; e. g . , Aedes (Chaetocruiomyia) s pp. for a species g roup of mosquitoe s endemic to Australia. Each taxon above the g enus level will also have its authorit y an d d ate, an df or eac hf am il y( b ut not f or taxa hi g h er t h an t hi s) t h ere i sa ty pe g enu s ,w hi c hb y d e fi n i t i on must h ave a name t h at i s i ncorporate di nto t h e f am il y name (e.g. , Ap is i nt h e b e e f am ily Ap id ae). 94 CHAPTER 4 3. Classificatio n Bi o l o gi ca l s y stems o f c l ass ifi cat i on are hi erarc hi ca l ,t h at i s, t h e l ar g est taxa are su bdi - v ided into successivel y smaller taxa. Thus, each taxon has a particular level (rank) within the s y stem. Groups of the same rank are said to belon g to the same taxonomic cate g or y ,to whi c h a part i cu l ar name i sg i ven. Some o f t h ese categor i es are o bli gatory (cap i ta li ze din t h e examp l e b e l ow), w hil eot h ers are opt i ona l .Tos h ow t h e hi erarc hi ca l arrangement an d to i ntro d uce t h e names o f t h evar i ous categor i es, l et us ta k e as an examp l et h ec l ass ifi cat i on o f the hone y bee , Apis mellifer a : KIN G D O MAn i ma li a PHYL U M U n i ram i a S u b p hyl um Hexapo d a C LA SS Insecta S ubclass Pter yg ota I n f rac l ass Neo p ter a D i v i s i on O li goneoptera ORDER H y menopter a S uborder A p ocrit a S uperfamil y Apoidea FAMILY A pid a e S u bf am il yAp i na e Tr ibe Ap i n i S ubtribe — G ENU S A pi s S ubgenus — S PE C IE S A pis me ll i f era S u b spec i e s [ In zoo l ogy, t h esu b spec i es i st h e l owest category cons id ere d va lid ; i n b otany, var i ety, f orm, an d su bf orm are recogn i ze d (an d g i ven l at i n i ze d names). ] C l ass ifi cat i on, t h en, i s a means o f more e ffi c i ent ly stor i n g (an d retr i ev i n g ) i n f ormat i on about or g anisms. In other words, it is not necessar y to describe all of the characteristics o f a species each time that species is referred to. For example, as standard practice, a lar g e proport i on o f entomo l og i ca l researc h art i c l es i nc l u d e i nt h e i rt i t l es, a f ter t h e name o f t he spec i es b e i ng stu di e d ,t h e f am il y, (super f am il y), an d or d er to w hi c h t h e spec i es b e l ongs. In t hi swa y , a rea d er can i mme di ate ly g a i n some i ns igh t i nto t h e nature o f t h e i nsect b e i n g studied, even thou g hheorshema y not be familiar with the species. Related to this last point, c lassification is also important in that it enables predictions to be made about incompletel y stu di e d organ i sms. For examp l e, organ i sms are a l most a l ways c l ass ifi e dfi rst on t h e b as is of t h e i r externa l structure. However, once an organ i sm h as b een ass i gne d to a part i cu l a r taxon us i n g structura l cr i ter i a, i tma y t h en b e poss ibl etopre di ct, i n g enera l terms, i ts h a bi t s ( includin g life histor y ), internal features, and ph y siolo gy , on the basis of what is know n c oncernin g other, better studied, members of the taxon . A classification may be either artificial or natural. It is possible, for example, to ar - range organ i sms i n groups accor di ng to t h e i r h a bi tat or t h e i r econom i c i mportance. Suc h cl ass ifi cat i ons may even b e hi erarc hi ca li nt h e i r arrangement. Art ifi c i a l c l ass ifi cat i ons are usuall y desi g ned so that or g anisms belon g in g to different taxa within the s y stem can b e 95 S Y S TEM A TI CS A ND TA XONOM Y separated on the basis of sin g le characters. As a result, such schemes have extremel y re- stricted value and, usuall y , can be used onl y for the purpose for which the y were initiall y d esi g ned. More importantl y , artificial classifications provide no indication of the “true” or “natura l ”re l at i ons hip so f t h e const i tuent s p ec i es. A l most a ll mo d ern c l ass ifi cat i ons are natura l ,t h at i s, t h ey i n di cate t h ea ffi n i ty ( d egree o f s i m il ar i t y ) b etween t h eor g an i sms w i t hi nt h ec l ass ifi cat i on. Or g an i sms p l ace di nt h e sam e t axon (showin g the g reatest affinit y ) are said to form a natural g roup. There is, however, considerable controvers y amon g s y stematists over the meanin g of “de g ree of similarit y ,” “natura l group,” an d “natura l c l ass ifi cat i on.” Essent i a ll y systemat i sts f a ll i nto t h ree ma j or groups, accor di ng to t h e i r i nterpretat i on o f t h ea b ove terms. T h ese are t h ep h y l et i c i sts , cladists, and pheneticists. To the cladistic group, led by Hennig (see Hennig, 196 5 , 1966, 1981), belon g those s y stematists who base classification entirel y o n genealogy , the recenc y of common ancestr y . Critical to the modus operandi of cladists are the distinction betwee n primitive and advanced homologous characters (so-called “character polarit y ” ) and the recogn i t i on o f si ster g roup s (see b e l ow f or f urt h er di scuss i on o f t h ese terms). Among t h e var i ous ways use db yc l a di sts to ass i gn c h aracter po l ar i ty are pa l eonto l ogy, ontogeny, an d out g roup comparison. In theor y , the stud y of fossils should clearl y show when a character fi rst appears, makin g the separation of primitive and advanced characters an eas y task . However, the fossil record is typically discontinuous and preservation imperfect so tha t v i ta l c h aracters are m i ss i ng. T h e id ea t h at “ontogeny recap i tu l ates p h y l ogeny,” suggeste dby Haeckel in 18 66 , proposes that an organism’s development will reflect its evolution, givin g clues therefore as to which of its features are primitive and which are advanced. Onto g en y has b een relativel y little used b y cladists, however, perhaps because in development evolutionar y steps are compressed, omitted, or masked. Out g roup comparison, which is the metho d most use d , i s a compar i son o f c h aracter states i nt h e group un d er stu d yw i t h t h ose i n i ncreas i ng l y di stant s i ster groups. T h ec h aracter state common to t h e l argest s i ster groups is g enera lly ta k en to b et h epr i m i t i ve con di t i on. T hi s met h o d requ i res, o f course, some prev i ous knowled g eofa g roup’s ph y lo g en y and has been criticized because of its circularit y .As a result of their studies, cladists usuall y express their results in the form of a cladogram . Beginning in the 1950s, some taxonomists, dissatisfied with the perceived subjec - ti ve approac h to c l ass ifi cat i on, b egan to d ev i se sc h emes b ase d on t h e num b er o f commo n c h aracters among organ i sms, regar dl ess o f w h et h er t h ese were pr i m i t i ve or a d vance d .T he pheneticists (ori g inall y known as numerical taxonomists), led b y Sokal and Sneath (se e Sokal and Sneath, 1963; Sneath and Sokal, 1973), have as their ma j or principles: (1) th e more characters studied the better; (2) all characters are of equal weight; and (3) the greate r th e proport i on o f s i m il ar c h aracters, t h ec l oser are two groups re l ate d .P h enet i c i sts usua ll y present t h e resu l ts o f t h e i r ana l yses a s p h eno g ram s or scatter d ia g rams . Ph y leticists such as Simpson (19 6 1) and Ma y r (19 6 9, 1981) ma y be considered a s formin g a “middle-of-the-road” g roup, emplo y in g both cladistic and phenetic information on which to base their classifications. The p ro p ortions of cladistic and p henetic information u se d may vary s i gn ifi cant l y d epen di ng, f or examp l e, on t h e extent o f t h e f oss il recor d ; in ot h er wor d s, i n contrast to t h ec l a di st i can d p h enet i c met h o d s, t h ep h y l et i c system d oes no t f o ll ow a set o f care f u lly esta bli s h e d ru l es . An implicit point of natural classifications, re g ardless of how the y are derived, is tha t t he y are based on g enealo gy (i.e., relationship b y descent). In other words, the y show evo- l ut i onary re l at i ons hi ps among taxa. T h us, t h e k ey step i n any natura l c l ass ifi cat i on i st he d eterm i nat i on o f h omo l o gy (w h et h er f eatures common to groups were d er i ve df rom t h e sam e f eature i nt h e most recent common ancestor o f t h e g roups). S i m il ar, b ut non- h omo l o g ous, 9 6 CHAPTER 4 f eatures are said to sho w homoplas y (analo gy ) and are the result of eithe r p arallelis m ( the f eatures had a distant, common ancestor ) or convergence ( the features are derived fro m e ntirel y unrelated ancestral conditions). Once homolo gy is established, it is then a matte r of d eterm i n i ng w h et h er t h ec h aracter states un d er cons id erat i on are a d vance d ( d er i ve d ) o rpr i m i t i ve (ancestra l ) ( ap omor ph ies o r pl esiomor ph ies, respect i ve l y). Bot h comparat i ve m orp h o l o gy o f extant f orms an d t h e f oss il recor dh ave b een use d extens i ve ly i n suc hd eter - m inations. Apomorphies shared b y taxa are said to b e s ynapomorphies, while those uni q u e toata x o n a r e desc ri bed as a utapomorphies. N either auta p omor p hies nor p lesiomor p hies ca n s h ow re l at i ons hi ps b etween groups. Broa dl y spea ki ng, t h e greater t h e num b er o f synapo - m orp hi es, t h ec l oser w ill b et h ere l at i ons hi p b etween taxa. Eac h taxon, regar dl ess o f ran k, will h aveas i ster group— i ts c l osest re l at i ve—so t h at t h e d eve l opment o f c l ass ifi cat i ons and ph y lo g enies is the establishment of successivel y lar g er sister g roups, often depicte d as a branchin g dia g ram known as a phylogenetic tree ( see the section on Ph y lo g en y an d Classification under each order for exam p les). An ancestor and all of its descendants form a m onop hyl etic group; w h en some o f t h e d escen d ants are l ac ki ng, t h e rema i n i ng d escen d ants are sa id to b e parap hyl etic. Groups d er i ve df rom more t h an one ancestor are sa id to be polyphyletic . It must be emphasized that the actual ancestor of two taxa is rarel y known, thou g h its g eneral features (the so-called “ g round plan”) will be defined b y the plesiomor- p hic characters of its descendants. The ter m st em group re f e r stoco ll ect i o n so ff oss il st h at h ave some p l es i omorp hi cc h aracters o f a more recent group; t h ey may b ec l ose to, b ut are n ot di rect l y on, t h e group’s li ne o fd escent . As the followin g section (and comparison of the current with previous editions of this book) shows, ideas on relationships amon g insect g roups chan g e with time, sometimes quite si g nificantl y . Thou g h partl y related to the acquisition of new knowled g e, it is also becaus e taxonom i sts diff er i nt h e i r ana l ys i san di nterpretat i on o fd ata, or use diff erent d ata sets o n whi c h to b ase t h e i r conc l us i ons . 3 .1. The Histor y of Insect Classification W il son an d Doner ( l 937) h ave f u ll y d ocumente d t h e many sc h emes t h at h ave b een d ev i se df or t h ec l ass ifi cat i on o fi nsects, an di t i s f rom t h e i r account t h at t h e f o ll ow i ng s h or t hi story i sma i n l y comp il e d . (Papers mar k e d w i t h an aster i s k are c i te df rom W il son an d D oner’s review.) Onl y the ma j or developments (i.e., those that have had a direct bearin g o n m odern schemes) have been included, thou g h it should be realized that a g ood man y mor e systems have been proposed. Insect systemat i cs may b e cons id ere d to h ave b egun w i t h t h ewor k o f Ar i stot l e, w h o, according to Kirby and Spence (181 5 –1826),* included the Entoma as a subdivision of the Anaima (invertebrates). Within the Entoma Aristotle placed the Arthropoda (excludin g Crustacea), Echinodermata, and Annelida. Authors who have examined Aristotle’s writin gs differ in their conclusions regarding this author’s classification of the insects, but it doe s appear c l ear t h at Ar i stot l e rea li ze d t h at t h ere were b ot h w i nge d an d w i ng l ess i nsects an d t h at t h ey h a d two b as i c types o f mout h parts, name l y, c h ew i ng an d suc ki ng . Amaz i n gly , i t was not f or a l most anot h er 2000 y ears t h at f urt h er ser i ous attempts to c lassif y insects were made. Aldrovanus (1 6 02)* divided the so-called “insects” into terres- trial and aquatic forms and subdivided these accordin g to the number of le g s the y possessed an d on t h e presence or a b sence an d t h e nature o f t h ew i ngs. In A ld rovanus’ c l ass ifi cat i on t h e term “ i nsect” encompasse d ot h er art h ropo d s, anne lid s, an d some mo ll us k s. T h ewor k o f Swammerdam (1 66 9)* is of particular interest because it represents the first attempt 97 S Y S TEM A TI CS A ND TA XONOM Y t o classif y insects accordin g to the de g ree of chan g e that the y under g o durin g develop- ment. Althou g h Swammerdam’s concept of development was inaccurate, he distin g uishe d clearl y between ametabolous, hemimetabolous, and holometabolous insects. A more elab- orate sc h eme o f c l ass ifi cat i on, st ill b ase d pr i mar il yont h e d egree o f metamorp h os i s b ut a l s o i ncorporat i ng suc hf eatures as num b er o fl egs, presence or a b sence o f w i ngs, an dh a bi tat , w a s t hat of Ra y and Willu g hb y (170 5 ).* Ra y was the first naturalist to form a concept o f a “species,” a term that was to take on more si g nificance followin g the introduction, b y L innaeus, of the binomial s y stem some 30 y ears later. Between 1735 and 1758, Linnaeus* gra d ua ll y i mprove d on hi s system f or t h ec l ass ifi cat i on o fi nsects, b ase d ent i re l yon f eature s o f t h ew i ngs. L i nnaeus recogn i ze d seven or d ers o f “ i nsects,” name l y, t h e Aptera, Neuroptera, Co l eoptera, Hem i ptera, Lep id optera, D i ptera, an d Hymenoptera. O f t h e seven, t h e fi rst f ou r orders each contained a hetero g eneous g roup of insects (and other arthropods) that toda y are separated into man y different orders. The Diptera, Lepidoptera, and H y menoptera hav e remained, however, more or less as Linnaeus envisaged them more than 200 years ago. Like ear li er aut h ors, L i nnaeus i nc l u d e di nt h e Aptera (w i ng l ess f orms) sp id ers, woo dli ce, myr i - apo d s, an d some non-art h ropo d an an i ma l s. He f a il e d a l so to di st i ngu i s hb etween pr i m i t i ve ly and secondaril y win g less insect g roups. S urprisin g l y , perhaps, up to this time no one had made a serious attempt to classif y insects on the basis of their mouthparts. However, the Danish entomologist Fabricius, wh o wa sa s tu d ent o f L i nnaeus, pro d uce d severa l “c ib ar i an” or “max ill ary” systems f or c l ass ifi- cation during the period 177 5 –1798.* The primary subdivision was into forms with bitin g mouthparts and forms with suckin g mouthparts. Like Linnaeus, however, Fabricius included av a riet y of non-insectan arthropods in his s y stem and, furthermore, based his s y stems on a sin g le anatomical feature. De Geer (1778),* w h oa l so stu di e d un d er L i nnaeus, a pp ears to h ave b een one o f t h e ear li est systemat i sts to rea li ze t h e i mportance o f us i ng a com bi nat i on o ff eatures as a b as i s f or c l ass ifi cat i on. Suc h an approac h was use dby t h e Frenc h entomo l o gi st Latre ill e, w h o, d urin g the period 179 6 –1831,* g raduall y produced what he considered to be a natural arran g ement of the Insecta. In 1810 Latreille separated the Crustacea and Arachnida from th e “Insecta,” i nw hi c hh e i nc l u d e d st ill t h e Myr i apo d a. T h e l atter group was not g i ven c l ass status until 182 5 . In the final version of his system Latreille distinguished 12 insect orders. Th eL i nnaean or d er Aptera was sp li t i nto t h eor d ers T h ysanura, Paras i ta ( = A nop l ura), and Siphonaptera, althou g h Latreille did not appreciate that the first g roup was primitivel y w in g less, while the other two were secondaril y so. The order Coleoptera of Linnaeus wa s subdivided into Coleo p tera ( sensu str i ct o ) , Derma p tera, and Ortho p tera. The Phi p hi p tera ( = Streps i ptera), b e li eve d to b ere l ate d to t h eD i ptera i nw hi c h or d er t h ey h a db een i nc l u d e d, w ere separate d as a di st i nct group b y Latre ill e. T h e Frenc h man was a l so among t h e ear li es t s y stematists to appreciate the hetero g eneit y of the Linnaean order Neuroptera, splittin g t he g roup into three tribes, the Subulicarnes ( = m odern Odonata and E p hemero p tera), Plani p ennes ( = m odern Pleco p tera, Iso p tera, Meco p tera, and neuro p teroid insect s 1 ) an d Pli c i pennes ( = mo d ern Tr i c h optera). Dur i n g t h e fi rst h a lf o f t h e 19t h centur y a l ar g e num b er o f s y stemat i sts pro d uce d t heir version of how insects should be classified. A ma j orit y ar g ued, like Latreille, that t he win g s (presence or absence, number, and nature) were the primar y feature on which a classification should be established. Yet others, such as Leach (1815)* and von Siebold 1 Insects t h at are i nc l u d e di nt h emo d ern or d ers Neuroptera, Me g a l optera, an d Rap hidi optera. 98 CHAPTER 4 ( 1848),* considered that the nature of metamorphosis was the first-order character, with w in g s, mouthparts, etc. of secondar y importance. If nothin g else, the use of metamorphosis as a separatin g character drew further attention to the hetero g eneit y of the neuropteroid group, w hi c h conta i ne db ot hh em i -an dh o l ometa b o l ous f orms. In d ee d , i n hi sc l ass ifi cat i on v o n Si e b o ld a d opte d Er i c h son’s (1839)* arrangement i nw hi c h t h e term i tes, psoc id s, em bi - id s, ma yfli es, d ra g on fli es, an dd amse lfli es were remove df rom t h e Neuroptera an d p l ace d to g ether as the suborder Pseudoneuroptera in the order Orthoptera . T he foundations of modern s y stems of classification were laid b y Brauer (1885),* who appears to h ave b een great l y i n fl uence db yt h epr i nc i p l es o f comparat i ve anatomy an d pa- l eonto l ogy esta bli s h e db yt h e Frenc h zoo l og i st Cuv i er, an db yt h ewor k o f Darw i n. Brauer di v id e d t h e Insecta i nto two su b c l asses, t h e Apterygogenea, conta i n i ng t h epr i m i t i ve l y w in g less Th y sanura and Collembola, the latter havin g been g iven ordinal status b y Lub - bock (1873),* and the Pter yg o g enea, containin g 16 orders, in which he placed the win g ed and secondarily wingless forms. Three major divisions were established in the Pterygo - genea: (1) Menognat h a ameta b o l aan dh em i meta b o l a( i nsects w i t hbi t i ng mout h parts i n b ot hj uven il ean d a d u l t stages, or mout h parts atrop hi e di nt h ea d u l tan d w i t h no or part i a l m etamorphosis) containin g the orders Dermaptera, Ephemerida, Odonata, Plecoptera, Or- thoptera (includin g Embioptera), Corrodentia (which included the termites, psocids, an d l ice), and Thysanoptera; (2) Menorhyncha (insects with sucking mouthparts in both the juve - nil ean d a d u l t stages), conta i n i ng t h eor d er R h ync h ota ( = H em i ptera); an d (3) Menognat ha m eta b o l aan d Metagnat h a meta b o l a( i nsects h av i ng a comp l ete metamorp h os i s, an d w i t h bitin g mouthparts in the j uvenile sta g e and bitin g , suckin g , or atrophied mouthparts in the adult), containin g the neuropteroid insects, and the orders Panorpatae ( = M eco p tera), Tri - c hoptera, Lepidoptera, Diptera, Siphonaptera, Coleoptera, and H y menoptera. Thus, Brauer apprec i ate d t h e h eterogene i ty o f t h e “Neuroptera” an d correct l y separate d t h eP l ecoptera , Od onata, an d Ep h emer id a f rom t h e neuroptero id s, Mecoptera, an d Tr i c h optera. He f a il e d, h owever, to reco g n i ze t h e h etero g ene i t y o f t h eor d ers Ort h optera an d Corro d ent i a. B etween 1885 and 1900, a number of modifications to Brauer’s s y stem were su g- g ested. Most of these were concerned solel y with the subdivision or a gg re g ation of order s accor di ng to t h e aut h or’s v i ewsont h ea ffi n i ty o f t h e groups. T h ere were, h owever, two proposa l st h at h ave a more di rect b ear i ng on mo d ern systems. In 1888 Lang* propose d t h at t h e terms Apterygota an d Pterygota b esu b st i tute df or Apterygogenea an d Pterygogenea, respectivel y . Sharp (1899) refocused attention on the importance of metamorphosis, but , c laimin g that the terms Ametabola, Hemimetabola, and Holometabola were not sufficientl y definite for taxonomic purposes, proposed new terms describing whether the wings de- v e l ope di nterna ll y or externa ll y. H i s arrangement was as f o ll ows: Apterygota (pr i m i t i ve l y wi ng l ess f orms); Anapterygota (secon d ar il yw i ng l ess f orms); Exopterygota ( f orms i nw hi c h the win g s develop externall y ); Endopter yg ota (forms in which the win g s develop internall y ). S harp was criticized for g roupin g to g ether the secondaril y win g less orders (Mallopha g a , A no p lura, Si p hona p tera), as these contained both hemi- and holometabolous forms, and the term Anapterygota was di scar d e d .T h e terms Exopterygota an d En d opterygota were w id e ly accepte d , h owever, an db ecame synonymous w i t h Hem i meta b o l aan d Ho l ometa b o l a, re - s pect i ve ly . It was not unt il t h ewor k o f Crampton an d Mart y nov i nt h e 1920s (see b e l ow) that it was realized that these terms had no ph y lo g enetic si g nificance but were merel y de - s criptive, indicatin g “ g rades of or g anization.” Sharp reco g nized 21 orders of insects. Hi s s ystem i mprove d on Brauer’s ma i n l y i nt h esp li tt i ng o f t h e Corro d ent i aan d Ort h optera, t h ere b yg i v i ng or di na l status to t h e Isoptera, Em bi optera, Psocoptera, Ma ll op h aga, an d Si p h uncu l ata . 99 S Y S TEM A TI CS A ND TA XONOM Y T ow a rd the end of the 19th centur y the full force of Darwin’s ideas on evolution an d t he importance and usefulness of fossils be g an to make themselves felt in insect classifi - cation. Gone was the old idea that evolution was a sin g le pro g ressive series of events, and i n i ts p l ace came t h e apprec i at i on t h at evo l ut i on was a process o fb ranc hi ng. T h us, i nsec t c l ass ifi cat i on entere d ,att h e b eg i nn i ng o f t h e 20t h century, t h ep h y l ogenet i cp h ase o fi t s d evelopment, althou g h Haeckel (18 66 )* had been the first to use a ph y lo g enetic tree t o indicate the relationships of the Insecta. Unfortunatel y his ideas on g enealo gy were incor- rect. Most recent s y stems have been influenced to some de g ree b y the work of an Austrian pa l eoentomo l og i st, Han dli rsc h ,w h ocr i t i c i ze d ear li er wor k ers f or t h e i r one-s id e d systems, i nw hi c h as i ng l ec h aracter was use df or separat i on o f t h ema j or su bdi v i s i ons. Anot h e r f a il ure o f t h e 19t h century aut h ors was, h ec l a i me d ,t h e i r i na bili ty to di st i ngu i s hb etween parallel and conver g ent evolution of similar features. Finall y , he pointed out that almost n o one had taken into account fossil evidence. Handlirsch’s first scheme, p roduced in 1903, w as, at the time, regarded as revolutionary. He raised the Collembola, Campodeoidea ( = D i p l ura), an d T h ysanura eac h to t h e l eve l o f c l ass. (Pr i or to t hi st h eD i p l ura h a db ee n cons id ere d usua ll yasasu b or d er o f t h eT h ysanura.) He a l so ra i se d t h e Pterygogenea o f B rauer to the level of class and arran g ed the 28 orders of win g ed insects in 11 subclasses . His second scheme, published in 1908, was identical with the first except for some sli g ht changes in the names of orders. In 1925 Handlirsch published his modified views on insect c l ass ifi cat i on. In t hi ssc h eme h ere i ntro d uce d Brauer’s two su b c l asses, Apterygogenea an d Pterygogenea. In t h e f ormer group h ep l ace d t h eor d ers T h ysanura, Co ll em b o l a, D i p l ura , and the recentl y discovered Protura. In the Pter yg o g enea he listed 29 orders (includin g th e Z oraptera, first described in 1913) arran g ed in 11 superorders (his former subclasses). Th e most si g nificant point in Handlirsch’s work was his reco g nition of the hetero g eneous natur e o f t h e Ort h optera, t h e contents o f w hi c hh esp li t i nto or d ers an d regroupe d w i t h ot h er or d ers i n two superor d ers, Ort h optera (conta i n i ng t h eor d ers Sa l tator i a, P h asm id a, Dermaptera , D i p l o gl ossata, an d T hy sanoptera) an d B l attae f orm i a (conta i n i n g t h eB l attar i ae, Manto d ea , I soptera, Zoraptera, Corrodentia, Mallopha g a, and Siphunculata). He did not appreciate , h owever, the orthopteroid nature of the Plecoptera and placed the g roup in a superorder o fi ts own. Han dli rsc h was a l so i n error i n regar di ng t h e Corro d ent i a, Ma ll op h aga, an d S i p h uncu l ata as ort h optero id groups. T h ey are un d ou b te dl y more c l ose l yre l ate d to t h e Hem i ptera. Han dli rsc h ’s arrangement was strong l ycr i t i c i ze db yB¨orner (1904), w h osa id that it did not express the true phylogenetic relationships of the Insecta. B¨ orner consid- ¨ ered that fossil win g s did not have much value in insect s y stematics, and, in an y case , t here were far too few fossils for paleontology to have much bearing on classification. Comparat i ve anatom i ca l stu di es o f recent f orms, B¨orner argue d ,wou ld g i ve a more ac- curate p i cture. B¨orner, w h ose system was w id e l y accepte d , arrange d t h e19or d ers o f w in g ed insects that he reco g nized in five sections. Three of these correspond with th e “paleopteran orders,” “orthopteroid orders,” and “hemipteroid orders” reco g nized toda y . I n other words, B orner correctly assigned the Corrodentia, Mallophaga, and Siphuncu- ¨ l ata w i t h t h e Hem i ptera. T h e two rema i n i ng sect i ons conta i ne d t h een d opterygote or d ers, th oug h B¨orner’s id eas on t h e i ra ffi n i t i es were to b es h own b yT ill yar d (see b e l ow) to b e i ncorrect . C omstock (1918, and earlier), an American entomolo g ist, supported Brauer’s arran g e- ment as a result of his comparative studies of the win g venation of livin g insects. Comstock w a s th e fi rst person to ma k e extens i ve use o f w i ng venat i on i n d eterm i n i ng a ffi n i t i es. He emp h as i ze d , h owever, t h at c l ass ifi cat i ons s h ou ld b e b ase d on many c h aracters an d not w i ng s a l one . 100 CHAPTER 4 D urin g a period of more than 20 y ears, be g innin g in 1917, Till y ard expounded hi s v iews on insect ph y lo g en y , stemmin g from his extensive research into the fossil insect s o f Australia and North America. Althou g h he made important contributions concernin g t h eor i g i nan d re l at i ons hi ps o f many i nsect or d ers, T ill yar d ’s (1918–1920) wor k on t he e n d opterygotes i s part i cu l ar l ywe ll k nown. In t hi swor kh es h owe d t h at t h e Hymenopter a an d Co l eoptera (w i t h t h e Streps i ptera) f orm two rat h er di st i nct or d ers, on ly di stant ly re l ate d to the other endopter yg ote g roups which collectivel y formed the panorpoid complex. Withi n the com p lex, the Meco p tera, Tricho p tera, Le p ido p tera, Di p tera, and Si p hona p tera form a w e ll d e fi ne d group, w i t h t h e neuroptero id or d ers c l ear l y di st i nct f rom t h em. In f act, as note d i n Chapter 2, Hinton (19 5 8) made a strong case for excluding these orders entirely from the panorpo id comp l ex an d p l ac i ng t h em c l oser to t h eCo l eoptera. While Till y ard was concentratin g on the ph y lo g en y of the endopter yg otes, hi s A merican contemporar y , Crampton, was directin g his efforts toward solution of the prob - l ems of exopterygote relationships, especially the position of the Zoraptera, Embioptera , Gry ll o bl att id ae, an d Dermaptera. Fo ll ow i ng hi s anatom i ca l stu d yont h enew l y di scovere d wi nge d zorapteran Zorot y pus h u bb ar d i, Crampton (1920) conc l u d e d t h at t h e Zoraptera were related to the orthopteroid orders, and he placed them in a g roup (superorder Panisoptera) that also contained the Isoptera, Blattida, and Mantida. However, the followin gy ear Cram p ton revised his views and transferred the Zora p tera to the p socoid (hemi p teroid) su - peror d er, a f ter cons id erat i on o f t h e i rw i ng venat i on. In 1922 Crampton p l ace d t h e Zorapter a i nt h eor d er Psocoptera an d suggeste d t h at i twas f rom psoco idlik e ancestors t h at t h emo d er n hemipteroid orders evolved. Ori g inall y , Crampton (191 5 ) had placed the Gr y lloblattidae in a separate order, Notoptera, in the orthopteroid g roup. Five y ears later he concluded that th e g r y lloblattids were closer to the Orthoptera ( s ensu str i ct o ) than the blattoid g roups and made t h e Gry ll o bl atto d eaasu b or d er o f t h e Ort h optera. T h emo d ern v i ew i st h at t h e gry ll o bl att ids are pro b a bl y surv i vors o f t h e protot h opteran stoc kf rom w hi c hb ot h t h e ort h opteran an d bl atto id li nes d eve l ope d . Crampton cons id ere d t h at t h ec l osest re l at i ves o f t h eEm bi opter a w ere the Plecoptera, placin g the two g roups in the superorder Panplecoptera. In his earl y schemes Crampton also placed the Dermaptera in the Panplecoptera. He later chan g ed this vi ew an di nc l u d e d t h em i nt h e ort h optero id superor d er,att h e same t i me po i nt i ng out t h at t h eD i p l og l ossata (Hem i mer id a) are paras i t i c Dermaptera . A l most s i mu l taneous l y i n 1924 Crampton an d t h e Russ i an pa l eoentomo l og i st Martynov proposed an apparentl y natural division of the win g ed insects on the basis of th e abilit y to flex the win g s horizontall y over the bod y when at rest. In the Paleoptera ( = P aleopterygot a = Archipterygota) are the orders Ephemeroptera and Odonata whos e m em b ers d o not possess a w i ng- f o ldi ng mec h an i sm. It must b e emp h as i ze d , h owever, t h at t h etwoor d ers are on l y very di stant l yre l ate d t h roug h t h e i rpa l eo di ctyopteran ancestry. The remainin g orders, whose members are able to fold their win g s over the bod y , are p laced in the Neo p tera ( = N eopter yg ota). The latter contains three natural subdivisions, the P olyneoptera (orthopteroid orders), Paraneoptera (hemipteroid orders), and Oligoneoptera ( en d opterygote or d ers) . E ven recent l y, v i gorous d e b ate h as cont i nue d over t h e taxonom i c ran k o f ,an d natur e of t h eevo l ut i onar y re l at i ons hi ps amon g , h exapo dg roups (see C h apter 1, Sect i on 3.3.1 [ apter yg otes], and Chapter 2, Section 3.2 [pter yg otes] for a fuller discussion). For exam- ple, most authors consider the Collembola and Protura to be sister g roups and sometimes un i te t h em i nt h ec l ass E llip ura ( = Para i nsecta). However, t h e p os i t i on o f t h eD ipl ura is l ess c l ear; Kr i stensen (1991) p l ace d t h em c l ose to t h eE lli pura pr i nc i pa ll yont h e b as i s [...]... 239) 44 Very small species; body soft and weakly sclerotized; tarsi two- or three-jointed .45 107 SYSTEMATICS AND TAXONOMY 108 CHAPTER 4 45 46 47 48 49 50 51 52 53 54 55 56 Usually much larger species; tarsi usually with more than three joints, or, if not, body is hard and heavily sclerotized and cerci are absent .46 Cerci absent Book lice PSOCOPTERA (Page 199) Cerci unjointed,... 276) 22 Tarsi five-jointed; if rarely three- or four-jointed, hind wings are smaller than front ones and wings lie flat over body; no cerci 23 Tarsi two-, three-, or four-jointed; veins and crossveins not numerous 26 105 SYSTEMATICS AND TAXONOMY 106 CHAPTER 4 23 Prothorax small or only moderately long (In Mantispidae prothorax is very long, but front legs are strongly raptorial.) 24 Prothorax very... ending in forceps 42 42 Abdomen strongly constricted at base; prothorax fused with mesothorax Ants, etc HYMENOPTERA (Page 330) Abdomen not strongly constricted at base; broadly joined to thorax 43 43 Head not elongated ventrally 44 Head elongated ventrally forming a rostrum, at tip of which are mandibulate mouthparts Scorpionflies MECOPTERA (Page 239) 44 Very small species; body... Taxonomy, Freeman, San Francisco Tillyard, R J., 1918–1920, The panorpoid complex A study of the holometabolous insects with special reference to the sub-classes Panorpoidea and Neuropteroidea, Proc Linn Soc N.S.W 43 :265–2 84, 395 40 8,626–657; 44 :533–718; 45 :2 14 217 Wagner, W H., Jr., 1969, The construction of a classification, in: Systematic Biology, National Academy of Science, Washington, D.C., Publ No... CSIRO, Melbourne 111 SYSTEMATICS AND TAXONOMY 112 CHAPTER 4 Martynov, A V., 19 24, The interpretation of the wing venation and tracheation of the Odonata and Agnatha, Russ Ent Obozl: 18: 145 –1 74 [Original in Russian Translated, with an introductory note, by F M Carpenter in Psyche 37(1930): 245 –280.] Mayr, E., 1969, Principles of Systematic Zoology, McGraw-Hill, New York Mayr, E., 1981, Biological classification:... wing; tarsi three-, four-, or five-jointed 27 Hind wings smaller than fore wings; wings held at rest folded back against abdomen; radius and media not fusing; tarsi two- or three-jointed .28 27 Tarsi apparently four-jointed; cerci usually minute; wings with a transverse preformed suture near the base; social species, living in colonies Termites ISOPTERA (Page 163) Tarsi three-jointed, front... exserted; tarsi five-jointed GRYLLOBLATTODEA* (Page 173) R Cerci short; no ovipositor; tarsi four-jointed; social forms, living in colonies Termites ISOPTERA (Page 163) Tarsi five-jointed; body usually very slender and long Stick insects PHASMIDA (Page 179) T Tarsi two- or three-jointed; body not elongate 53 Front tarsi with first joint swollen, containing a silk-spinning gland, producing... to the Hemiptera-Homoptera, Psyche 29:23 41 Crampton, G C., 19 24, The phylogeny and classification of insects, Pomona J Entomol Zool 16:33 47 Cranston, P S.,Gullan, P J., and Taylor, R W., 1991, Principles and practice of systematics, in: The 1nsects of Australia, 2nd ed., Vol I (CSIRO, ed.), Melbourne University Press, Carlton, Victoria Danks, H V., 1988, Systematics in support of entomology, Annu... (Page 1 84) R Hind legs not enlarged for jumping; wing pads, if present, in normal position 47 Prothorax much longer than mesothorax; front legs fitted for grasping prey Mantids DICTYOPTERA, Suborder MANTODEA (Page 161) Prothorax not greatly lengthened .48 Cerci present; antennae usually with more than 15 subdivisions, often multiply subdivided 49 No cerci; body often hard-shelled;... (The Austrosaginae, Zaprochilinae and Phasmodinae), CSIRO, Melbourne Ross, H H., 19 74, Biological Systematics, Addison-Wesley, Reading, MA Sharov, A G., 1966, Basic Arthropodan Stock, Pergamon Press, Elmsford, NY Sharp, D., 1899, Some points in the classification of the Insecta Hexapoda, Proc 4th Int Congr Zool., pp 246 – 249 Simpson, G G., 1961, Principles of Animal Taxonomy, Columbia University Press, . most s p ecies are described on the basis of their structure, es p e- ciall y external characters. However, on occasion such “morphospecies” are not equiva- l ent to bi o l og i ca l spec i es (repro d uct i ve l y i so l ate d popu l at i ons);. recenc y of common ancestr y . Critical to the modus operandi of cladists are the distinction betwee n primitive and advanced homologous characters (so-called “character polarit y ” ) and the recogn i t i on. of the characteristics o f a species each time that species is referred to. For example, as standard practice, a lar g e proport i on o f entomo l og i ca l researc h art i c l es i nc l u d e i nt h e i rt i t l es,