THE MORPHOLOGY OF THE CICADIDB.

365

19. The Morphologr of t h e Cicadidx (Homoptera). By J. G MYERS, Sc.D.", Imperial Bureau of Entomology, late 1851 Science Exhibition Scholar.
[Received March 20,1928 : Read April 17, 1928.1

(Text-fi gnres 1-7 5.)

CONTEXTS. Page Introduction ........................................................ 36t5 (1)ICxternal Anatomy ............................................ 365 a. General Characters ............. ............ 365 b. Integument and Coloration .............................. 366 c. Head and its Appendages. d . Thorax and its Bypendages .............................. 380 e . Abdomen and its Appendages ........................... 402 (2) Internal Anatomy ................................. 413 a. Wervous System and Sense-Organs .................... 414 L. Bluccular System .......................................... 426 c . ltespiratory System ....................... d. Alimentary System ....................................... 433 e. Circulatory System _ _ . ._ _ ._ _ ,f. Fat-Body ..................................................... 456 g. I<eproductire System .................................... 456 ... 469 (3) Sound-Organs ......... a. Description of Sound-Organs .......................... 460 71. Historical Reriem .......................................... 467 c . Accessory stridulating apparatus of Tettigacli xe... 472

IXTRODUCTIOS. I n spite of their si/;e and the interest which their loiid."song 1 1 % ~excited in Man from the earliest times, the Cicadas [have been extraordinarily neglected by morphologists. The soundproducing apparatus has been the snhject of considerable study and much controversy, hnt the rest of their structure has remained almost unknown or greatly misunderstood. The following study ~ v a s made largely in the Entomological Laborat,ory of the Rnssey Institution, HRrvarcl University, and I am deeply indebted t o Professor W. M. Wheeler and t o Professor C. T. Brues for help and advice of every kind dnring its course.

(1) EXTERNAL ANATOMY.

a. GENERAL CHARACTERS.

Tile Cicatlas are among the largest insects included not only in Homoptera but in the whole Order. Their shape, with the

* Communicate:l by S. A. NEATE, M.A.,

PROC. ZOOL.S 0 ~ . - 1 9 2 8 ,

D.Sc., F.Z.S.

NO.XXV.

25

366

DR. J. G . JIYERS ON TIIE

I)otiy usunlly short and 5tollt il.11tl tlie wings long or iiioderately so, is rentleretl chracteristic by the normally short but wide head, with proiniiient eyes and conspicuous ocelli, freqnently red in cvlour, n.iid arr:iiiged in a triangle upon t h e crown. The head may be pro(lriced, bnt iiever to sucli an extent as i n t h e FLilgoridztlie oiily other Honioptera newly rivalling tlie larger cicadas in
S17.6.

The fore-legs siiom gre;it swollen spinose feinorn, while t h e wings, tliorigli strong, are usnally tr:tnsparent, but in some cases opaqiie ~ n lriglily coloarerl. The venation is compai,atirely d pyiuiit.ire ant1 siinple, with, i n addition, a n auibient vein cuti,iiig off' a f:iirly wide border. 'I'lie irirtles of all sa.ve one genus posses5 :t coinp1ic;tted souridorgan in the 1st itbtlornin:Ll segment, wliile both sexes ied wit,h t h e most complex chortlotona.1 oigan known, lio~~setl : swelling on e:icIi side of tlie 2nd ~bc1oinin:ri segment. iii t AAil t h e species are piirely pliytophagons, l i \ i n g on tlre sap of pl:ints. The iiyrnplis tlirougllout the fitlnily are snbterriiiiean :tiit1 rioted for. t h e extreme length of the cycle-in one wellc:st,:il)lisheil case leaching seventeen years. Their for,e-legs are liiollified i i i a. c1iar:tcteristic may for fossori:il purposes. 'Ciie eggs are tleposited in t h e stenis of pl:i.iits by means of t h e lx~werfulovipositor carried hy the fernale. The young nymphs 0 1 1 1i:Ltching out, drop a t oiice t,o tlie grountl ancl burrow. 'Vhe :ttiiilts are comparn,tively short-!ived, enclnring at most one ae;\ .on. '['lie song exhibits gi.e:it,coristnnt specific vai.intion n.nd ;1ppreiitly p1aj-s a. part i n the mating preliininnries.

h. INTEGUMENT AXD COLORATIOX. 'rile iritegiinient of cicadas is generdly well-cliitinized. The ciit,icie 1iin.y be ra.rely entirely smooth o r covered with pul~escence or pilosit?'. ' h e 1)ubescence is often fine, close, :ind continuous, :ii11] n~:r,y conipletely obscure t h e colour beneath. Hi1vei.y or gold ll,:i,rIiiiigs foriiietl by p t c l i e s of pubescence w e f i q u e i i t . Pi,tiinosity is less fi,equent t h a n in Fiilgoroit1e:i 01- Mternorri.iyrlcli;~. Lt, is ~rsnally confined t o t h e iiiiriiediilte neigliboui.lioo~-ooften a 1liei.e riiii--of t h e iiioi'e posterior :ibdoniinal spiracles, biit may OccIIr in p t c l i e s elsewliere. Nerertlieless, wnxy inaterid 1ma.y exist n.s it11 extremely tjliiii a n d pra,ctically imperceptible covering < ) ~ t ?the wliole of tlie body-snrface, a.s Fumouze (1888 * fouriti y c) ill 7'ibicen ~7lebria and J17wchTy.s smzguinecc by t h e \ise of cl-ieinical ~ ~ ~ ~ t l i oPilosity is aasoci:ited chiefly wit11 altitnrle. Anlong (ls. tllose X e i r Zealant1 species of ilfelnwpwltu wliieli erijoy ;I, conSiiIeyitbIe i.xlige in altitiidinal distribution, there is a clear correlntiou hetween elevation nncl hairiness. Alpine or subnlpirle species, ho~vever, are rmrrly a.1wa.y~more hairy t h a n closelyi&ted Iowland forms. Tettigarcta crinita Dist., perhaps the lil& ~,rofiisely hairy of linowri cicadas, occurs on &It. Kosciiislto itt :LII elevation of 5000 eet (H. Ashton, in Zitt.).

llOltIHOLOGT O F TKIE CICADID23.

r 1

367

I he colour-pattern of Cica.clas contains usually two elementsan arrangement of (lark coticnlar iiiarkings on ft paler hypotlernia.1 ground-colour. Sonietimes one will predominate ; sometimes t h e other. Net,allic colours would seem t o be zbsent, but, a r e occasionally simula.ted by pubescerice. An interesting study of t h e pigmentntion i n Jfugicicarlu septeirtlecim lias been iiiade by Gortner (191 1 ) *. H e remarks t h a t t h e rapid clinnge froin t h e pale-coloured, almost cre:urry, iiynipli t o Lhe deep black arlult presents one of the most reiiiarka.ble ctinriges i i i colonxtion of which lie is amwe. From his earlier n work on tlre pigmentation of t h e iiieal-worm (Ileiteb~io a o l i t o ~ ) h e h ;t d expected t o find t h a t t h i s c h m g e i n colour wns a n oxitla.tion, induced by t h e action of a.ti oxidizing enzyme, aiiil his expectation was fully realized, altliougli h e fouiirl iiriportant tli fferences. The colour is d u e t o t h e action of n tyrosinase acting 011 some aromatic amino phenol, poilucing, as a i.esnlt of tile action, n hlnck, insoluble pipnieut. Tliat it3 is a n oxitlatiori is shown in tJie following n-nys ( G w t n e r , 1911) :I . I an i m a y jnst, emerged be placed in mnter no colomtion f :ippeni.s ; yet if i t i s reiriovrd :it) an; time before decomposition sets in, t i l e coloration proceeds a t once. 2. If nn iinago just emeiged be kept i n a stream of CO, n o s ; but if i n 0, t h e blackening proceeds norinally.
1 iiere was n o difference i n tlie intensity or rapidity of coioratiori in insects kept duriiig, :&lid after. ecdysis i n sti,ong 1 iglrt, dim liglit, total tlitrkness, o r light filterecl tlirougli blue g 1R ss . On 1). 9 3 Gortner makes tlie iiitorestitig suggestion, inasmncli :IS t h e color,ation is a uniform black over the elitire surface, tliat t h e eiitire new c n t i c n h is forriietl by t h e reaction i)etween t h e oxida+e aiitl the chimiogen, in the s:biiie manner tihiit tlre J:tpnese l:~,cquer formet1 by t h e actJion of laccase on is tlre milky 1:ihex of t h e h e I<hu.sw*i/,icifemc. J.Ire \viiigs are nsnally colourless, b u t in some c:ises opaque aiid i)riiliiintly colonre~l. Gr:ifin von Iincleii (1 901) 11:ts studied t h e vYirig-iiiarlririgs, especi:i.lly i n some of tlie lZat!jpleura spp., which sliom spots :*nd stmxlrs. She finds the m;Lrkings :ire always (Ieterminctl i n position eitber hy tlre position of tlie veiris tlienisel\7es o r of t h e rriiiiiite cross-foltls whicli may be seen i n certain li,vIlts in every cic:idii, tegmen a n d which she believes t o be relics of :L close net-veining. The strongest markings a r e determined j i i position by tlie few cross-veins whicli rein:rin in tlie iiiiago. Eve11 i n tire niost trarisparetit tegmina there a r e frequently dark
I ,

, I

1?~11 refereiiccs are given in a comprehensive bibliography of the f a m i l r now I,t:ing publi.;lied by Meswu. George Routledge & Sons, Ltd. in my book &I the I3iology of Cicadas.

25*

368

D H . J G . MYERS ON T H E .

marks on t h e cross-veins or a.t t h e end-points of long veins. Uiedernimn (1914, pp. 1719-1721) discusses the Grafins results. It is unfortunate t h a t the only two cicada species on which tlie chemists littve worked were, from the pigmentation standpoint, highly anomalous. FLiinouze studied tlie red iiieclicinal Cicada, Ht~echys sccngziinecc, of the Orient. This species is colorired a striking red and black, ant1 possesses in addition a n unpleasant smell, and thus belongs t o the forms coinuiomly regarded as protected by warning coloration. Fumoiize (1888 b) isolated from i t n red pigment- rouge clHueclijs which gires the colour to t h e abdomen, ant1 a yellow hygroinetric sribstance.

APPEKDAGES. The literature on the head :md its appendages is almost

ITS

c . HEAD AND

entirely modern.

After

L :

few references i n c!nssical antiquity

w e find n o more itttention pait1 to this part of the cicada anatoniy

until comparatively recent times. The early post-Renaissance observers were interested chiefly in the sound-organs and worked b a t little o n other parts of the cicada body. Nevertheless, it worild be a mistake to suppose that Aristotle missed anything which could be seen, if not exactly perceived, by the iiieans at liis cornrnand. He believed that cicadas lacked a. inouth, or rather that the iiiouth a n d tongue were uriitecl so as to form a single part, through which, as througli :t root, the flnitls on which they live are sucked up (de part. Animal. lib. iv. 5). .\Ve s l d l return to this conception when discussing feeding1i;tbits. The Egyptians a.lso sw.w the rostrum or ncnleus, aiirl interpreted i t a,s tlie plectruni of the nirisical instrument (Hora-mi lo, V:ileriano Holeani). Pliny (lib. ii. cap. 37) quoted Xigidins t o tlre eIfect that cicadas lack eyes-a st.atenient which is repeated by Monfet (1634),-and is perhaps based on the lack of the waiiness iisnal t o cicadas a.nd other insects in tlie coininon Enrope:rn species, libiceii, plebeicc. !l.l~e modern references will not be discussed in det,n.ilnntil we 1 1 : tlexcribccl tlle parts concernetl. I t will tlieri he easier t o ~ ~ correlate tJie results of previous work. I n h t e r clays some of t,lie most coiitrovertecl qnestions in ext,ernnI insect ariixtoniy have concerned the interpreta.tion of Hernipteroua Jiea,d-strncture. Some member of the family Cica d i t h has very often been tlie type studied, as in tlie case of Smith wit11 his startling lioinologies iri 1892, Marlnt,t, (1895), Meek (1903), and Muir and Kershaw (1911 u, 1911 b, 1912). .Writers on other families of Honioptera have often orientnted tliernseives o n the large Cicaclidx; thus Funlrhoiiser (1917) ant1 Doering rlc22). The literature is therefore biilky, hut,, in addition, t h e f:tct that Elemipteroiis head-structure is remarkably iiiiiforni througliout both suborders, brings in a s exp1anator~of cicnda conditions interpreta.tions based on other forms: arid :dcls to the list such classical contributions as those of Geise

UORPIIOLOGP O F THE CICADIDB.

369

(1583), .Wedcle (1885): and the more recent work of Bagnion and PopoE(1911), and Tower (1914). Tlius it might reasonably be expected that the homologies of the trophi and sclerites have been clearly denionstratecl. B u t this is very far from the case. Moreover, as at least two other contributions (Metcalf, Hussey) on the subject in America alone are now either in the press or in manuscript, the present writer feels t h a t no finality for t h e interpretation here supported can be expected. I n a siibject so much discussed it is, of course, obvious that the structures themselves have now become moderately well known and the discovery of strikingly new morphological evidence correspondingly unlikely. Until the homologies of the Hemipterous nioutli have been finally elucidated by wider and inore comparative embryological studies of this and of all related orders, coupled with an advance in our knowledge of insect morphology in general, with clearer conceptions of the relationships and origin of the head sclerites in the most primitive inandibulate forins, finality cannot be expected in the interpretation of cicada head-structure. I n the following p a p I have therefore, after i~n examination of a11 previously published views and a study of both nymphs and adult cicadas, chiefly of New Zealand species of ~llelan~psulta, especially 111. ciiigulata and X . leptomera, attempted first a simple description of the head and mouth-parts, and then a collation of t h e arguments for and against the various divergent views, the snbsta.ntiation of mliich must lie sub jzbdice at l+mt until the publication of studies now in the press. For the adult condition, as Snodgrass has suggested, it is advanta.geous to study a soft freshly-emerged example, o r even one extracted from the nymphal cuticle just before ecdysis. Since tlie chief controversy concerns the homologies firstly of the two pairs of mouth-set%and secondly of the facial sclerites, the description of tliese will be made especially detailed. Unless otherwise stated tlie particulars concern ,I!felanzpsa,lta leptonzei-a a,clult, of which a head cleaned by Nematode worms more neatly t h n by K O H was fortunately available.
Xouth-parts. Viewed froni the front, the most striking feature is the large, swollen, transversely-striated frons (text-fig. 1, J;..), of which t h e striations correspond with the insertions of the great dilator muscles of the sucking-pump. The upper edge of this sclerite p8sses into the cro\j-n or vertex*, from which i t appears to be separated by a transverse foltl rather than by a distinct suture. En the nymph at the final ecdysis the split does not occur along this fold but more cauda.lly, along two sides of a. wide tria.ngle of which tlie fold in question forins the base. This triangular
d
Following C. F. Baker, we may regard crown as a less committal term than osrtex. I n the Fulgoroids, moreover, the expressioiis are certainly not synonymous, since the vertex invades the region commonly knomn as f r o n t in these insects.

a. Descri$tion. o Head c f

370

DR. J. G . MYERS ON THE

piece, which bears the median ocelliis in tlie adult, is apparently rather a part of the tine frons tlixii of the true vertex. There sceriis n o ieason against considering tlie sizes of the triangle a s t h e arms of the epicranial suture (text-fig. 2, e s . ) . Theie is a fairly tlistirict liiie along t h e arms of tlie Y in the atlnlt head as vien erl internally (testJ-fig. 2). The rorrespoiirliiig sclerite in Membracidz, a s shown by Hniel Brarich *. may also be considered the frons. I n her figure the rpicmnial sutiire is clearly marked, and its arms are formed by the 1:iter:tl edges of the clypeus, she incorrectly Iiomologises t h e plate in qiiestion. Posteiiorly the cron n passas into tlie occiput wjthout furtlirr Text-figure 1.
?C

Dfelampsaltn leptomem Myers.

Face.

f?., o n t ; c., clypens; Z., labrum-epipliaryiis; fs., frontal suture; gs.,geiial fr suture; fp., frontal pit ; oc., rnediaii ocellus; mp.,maxillary plate; Zr., loruni.

sutures. The paired ocelli are on the vertex just behind the arms of t h e I . The conipoontl eyes are large and protruding. Tlie aiitennte comprise, as in all other hncl-ieriorrliynclla, a shaft or peduncle and a whip or flagellum. Tlie peduncle has two segmentg, the fii.st hidden by the horizontal antenna1 shelf, both simple, lacking sense-organs, a.nd considerably thicker tlian t h e first segment of the flagellum. The number of segments of the latter varies within close limits, b u t in Jfelnnzpsalta leptomercc is six. These taper t o the apex, though in Nagicicada septelzdecina especially the apical one i n a fresh conditlion may be somewhat swollen. Hansen (1890) gives five flagellar segments a s normal t o the fnmily, bnt this is an error. Tlie sensillz, t o be described i n a later chapter, a.re confined to tlie flagellum.
Morpliology and Biology of the Meinbracidz of Iiansas, iians. linir. Sc. Bull. xviii. pl. 11. fig. 3 6 , p. 86 ( E 7 ~ g l i a ) .

MORPIiOLOGY O F TJ3E CICADIUB.

371

The lower margin of the frons passes into the clypeus (textfig. 1, c. & text-fig. 6, c.), wit11 well-defined demarcation from tlmt sclerite. The clypeiis is considerably flat,ter, with a, more or less membranous structure wising froni its inner surface and projecting as a narrow groo\7ecl flap beyond its trunca.te apex. This structure--the lnbrurii-epiphar~nx(text-figs. 1 & 2, I.)-is throughout grooved o n its inner sutrfa.ce for the reception of t h e mandibular and maxillary set%, wliich lie basally between it and t h e closely-nppressed groove of t h e Iiypoplrarynx (text-fig. 2, JL).. The distal portion of the labium-epipharyx protects the set= at t h e base of the labium, where t h e dorsal trough of the intter is: shallowest.

MeZnmpsaZta Zeptomera. Occipital view of head, I<OI'I prepariition, trophi removed. bt., body o f , tentorium ; es., epicrailid suture : d t , dorsal arm of tentorium ;
ga., genal apodeme; fa., frontal ayodeme; h., hypopharyns; fp., frontal , plate ; ~ r .lorum (maxillary plates removed) ; Z., labrum-spipharyns ; oc., median ocellus.

On each side of frons and clypeus is a crescentic lorum occupying most of t h e true genal area in cicada (text-fig. 1, I t - . ) . Lateral to this in t u r n is a Ionger plate, which i t largely hides from facial view. This is t h e maxillatry plate (text-fig. 1. nap.), which apically meets t h e free extremity of its fellow t o form with it ant1 t h e labrum a narrow pore through which t h e set= pass into t h e rostra1 furrow. Retweeri lorum and frons is t h e frontal suture (text-fig. 1, fs.), which passes up to t h e crown. Separatir~g lorum from maxi!laqplate is t h e genai suture (gs.), which almost joins the frontal suture posteriorly. The sutures mark t h e external lines of deep flange-like apodemes (text-fig. 4, n ) , t h e first clear interpretation of which

373

DR. J. G . MYERS ON THE

Text-figure 3.

Xelairqmzlta leptomera. As test-fig. 2, but trophi in place.

retractor of mandible ; am., articulation of mandible; ms., maxillary plate; m.,mandible; mx., maxillary seta; t., teiidon of Meek ; ma., maxillary apodeme.
YM.,

Text-figure 4.

MeZampsaZta cingulata (Fabr.).

Exuvial head of ultimate nymph.

a , , frontal apodeme; fp., frontal plate; fr., front; c., clypeus;

t., tendons of

pharynx dilator muscles.

has been given by Muir (1926). The frontal apodeme is narrow near the antenna, but widens apically till it reaches t h e junction of the frons and clypeus, where it forms a strong s t r u t at a point

NORPHOLOGY OF THE CIDADIDB.

373

iiiarked on each side of the face by a deep pit. Surely t>liesepits (text-fig. l,f?.), situated exactly at t h e junction of t h e plates we :),re calliiig J y o n s and cbypeus, are homologous with t h e frontal pits of the cockroiZch and other orthopteroids, and, as such, good landmarks for the recognition of these two sclerites. They may be taken as additional grounds for Mnirs supposition that t h e struts i n que.stion are homologous with the anterior arms of the tentoriuni. These struts flatten out and join in tlie middle line to form t h e posterior portion of a deep, trough-like, strongly-chitinized structure which all previous workers before Muir (1926) appear t o have regarded a.s t h e floor of the pharynx, for which, however, it is only the support, since Muir fonnd i n a nymph just a?.fter ecdysis t h a t the pharynx can be separated complet~elyas a membranous tube lying on the floor of this trough (text-fig. 4: fp.). I have been una,ble to nchieoe this separation myself in nymphs or adults, but I had nolie near ecdysis. This boat-shaped trough is then in its relationship strikingly similar to the fronta.1 plate of tlie tentoriuni in some ortliopteroids, and shoiild certainly be regarded a s liomologous therewith. Smith (J. B., 1892) regards i t as the nientuni! The genal suture forins inteim~llya similar apodeme, t h e .anterior portion of wliich joins its fello~v from t h e opposite side to form a.n anterior support for t h e ventral wa.11 of t h e pharynx. The dorsal a.rm of tli e ten tori II m -n ea rl y always hi t b ert o (Snodgrass ; Muir and Kerdiaw, 19 11 ; Doering, Funkhouser) considered t,he anterior-ariPes froin (Muir sa,ys ?rear>this gem1 apodenie just at its approximation t o t h e frontal apodeme and near t h e base of t h e antenna (text-fig. 2, dt. & text-fig. 3 ) . It joins the body of the tentorium near the niiddle. This latter is a short transverse rod, conspicuous in occipital view, swollen near each end a t the junction with t h e dorsal arms, and a.tt~clied t o t h e apices of the maxillary aporlemes. Thence t h e tentorial body is connected to the outer wa.ll of the liea,d-capsnle by an invagination a t the ill-defined !abial suture of Muir (1926), this inva,gination possibly corresponding with the posterior arm of t h e tentoriuin. A t the base of t h e free apical portion of the maxillary plate on the unclei~sic\eis t h e opening of tlie deep invagination which forms the large maxillary apodeine. (Muir.) The genal apodeme joins it, and t h e two form a pouch i n which lie the mandibular and maxillary set=. The mandible is representecl solely by a long stout seta (textfig. 3 , nz.) of the outer pair and its fairly simple bnse. This base consists essentia.lly of a niodera.tely wide plate bent over a t t h e top, to use Snodgrasss (1921 a ) expression, a,nd connected with tlie ~vvxllof the head-capsule near the upper corner of t h e frons and t h e origin of t h e dorsal a.rm of t h e tentorium. Un this plate a r e inserted the protractor niiiscles, which are attached t o t h e lower portion of t h e face, chiefly of t h e lora. The retractor

374

DR. J. G. M Y E R S ON T f l C

mriscles are iiisertetl at the bulbons portion of the mandible where the seta passes i n t o t,he wide plate, and are attached t o tlie crown behind the ocelli. The innndibular set= are barbed at tlie tip, a i d together loooely enclo\e the tightly-conjoined niit xillary setw within the roPt,rum. The maxilla base is cnrrietl further back into the head-capsule th.m that of the mandible. The protractors attach chiefly t o the loner part of the maxillary plate, while the retractors go t o the vertex near the attachment of t h e in,cndibnldr ones. The m:Lxillai-y setre are fitier and intimately lockell together, forming a water-tight tube which is split with difticultp and has often been described (e. g. Buckton) as one piece, and which lies in the looie channel between the apposed mandibles. The maxillary set= are connecteil by tongues to form this tube, R S seen in cross-section (text-fig. 6 ) in l+icZicina. The central tnbe is bounded on a11 sitles by the inner walls of the maxillary set% (fc.), but the section shows there are three holes enclosed in Text-figure 5.

Bidicina semilata (Wk.). Cross-section of maxillary tube.

.fc.,

food-canal ; s , salivary canal ; Z., lumeii of one maxillary seta. .

the walls of the canal thus formed. Two-one placed excentrically in each seta--are small, and are the lumina of these organs, allowing passage of nerves and tracheE, which stand out as fine Iiairs wlien tlie maxi1l:try setw are cut. The third passage is hrger, and is formed by a deep nenrlv enclosed groove i n one <eta roofer1 over by the apposition of its fellow. Tlie large central canal IS, of course, for the passage of footl, and t h e pnrietsl channel for that of saliva. I n Graphosonza, and npp:rrently in most other I-leteroptera, tlie two canals are more nearly elqnal, the salivary passage being slightly smaller. Tlie set= are capable of considerable extension beyond the tip of the labium (text-fig. 68). Tlie Irypopharynx is a pad-like triangular structure grooved for the reception of the set= and fitting closely against t h e inner surface of the epipharynx. Below t h e terminal portioii of t h e hypoylrarynx is a swelling coctaining the salivary pump, tlescmberl more fully in the account of the alirrientary system.

MORPHOLOGY OF THE CIDGDIUE.

37 5

TIiis terminal part is fiised with t h e cliitinous plate supporting tlie ventral mall of t h e pharynx. Tlie latela1 wing5 of t h e hypopliarynx seem t o be tlilectly continuor~s with tlle lora, a s rioted by Snotlgrass (1921 a ) arid seen ill text-fig. 2 (h., Z/*.)-an extremely ariom:iIoiis conrlition. Possibly t h e term h?JpOp1LcC12/1~~ is t o he restricted t o tlie ternlirlal somewhat spoon-sliapecl portion, as held by Muir and Kerslraw (1911 a, p. 5 ) . T h e lnbiuin forms t h e rostrum 01- sheath for tlie set=. I t i s attaclietl by feebly-chitiriizecl membrane, xiid, a s noted by hleek. s h o w i 5trongei. connections m it11 t h e prothorax tban wit>h tlie liead-a condition (leveloping early i n emblyonic life (Hepnions, 1899). I n its distal portion t h e labium is a tleep, t r o u g h - l i k almost closed tube. Three segments are recogliiznble wit,ll s i p s of IL fourth (text fig. 6). F r o m t h e proxilnal segment t h e cliititiized floor of tlie labial groove becodhes diffeerentintetl 3 s a free chitinizetl roc1 (text-fig 6, v.) f o r t h e inseltion of t h e muscles of t h e labium. Uugnion and Popoff say t h a t i n t h e Pentatomoitl, G~apho,~ol,,a Text-figure 6.

\ ,c

JfeIampsalta Isptomera. Rostrum of iiltirnate nymph. c., clypeiis ; I., labruni-epiphar).ni; gm.,gular meiiibraiie ; v., cliitiiiized rod.

lixeatuna (1911, p. 651), t h e labial canal forms, near the t i p at least, a completely closed tube. Above t h i s (proximally) t h e groove is open to permit t h e labium t o bend away from t h e set= and nllovc. penetration, t h e flexion occurring between t h e first and second labial segments. They nate t h a t , if t h e set= by accident a r e drawn riitirely out of tlie labial trough so as t o escape from t h e closed apical tubular part, t h e b u g is nilable t o r e t u r n them :mrl dies of starvation. W e doubt whether this condition is a t a11 general, and we have shown i n t h e New Zealand Pentatomid, Rhopulinzorpha, t h a t when t h e set= a r e penetrating t o tkie fullest extent, t h e labium is witbdrawn from them completely a n d foldetl back along tlie venter *. I n cicndns, at least in 111. cinpdatcc, t h e apical portion of the labial gioove is n o t shut. T h e shortening of t h e labium t o allow penetration of tlie set= takes place by wrinkling transversely in t h e terminal segments.
iv. pp. 266-266, figs. 1-3 (1921).

The Feeding-habits of Hemiptera, New Zealand Jourri. Sci. R: Tecliii-

376

DH. J. G. MYERS ON THE

'l!lre mechanism of t,he mouth-parts is so linked up with t h e fnnctioning of the sa.livary glands and pharynx t h a t it has been d e d t with under t h e physiology of t h e digestive system. The .chief point t o emphasize is the presence of a distinct salivary canal in t h e wall, a,s it mere, of the much larger food-cha,nnel. This 21:~s been frequently overlooked (e. g. Snodgrass, 1921 u). The set.= can be pushed out by the approximate niuscles with considerable force, even i n alcohol specimens.

13. iVotes o n the Honaologies. Smith (18'32) considered the whole rnandiblo t o be represented by tlie lorum, while both set= were derived from the nia,xilla, as also were t h e maxillary plate am1 t h e labium! He seeins t o have strained his i n t e r p e t a t i o n to accord with cc priori conceptions drawn from a study of the Dipterous condition. I n more detail lie homologized the niaxilla.ry plate with the palpus, t h e rostrum with tlie subgalea.+ga.le:i, and tlie two set= with t h e lacinia and stipes respectively. hlarlatt (1895) considered our frons the clypeus and o u r ln1,rniri-epipliary~ix the labrum or rather pzrt of it. The lorum was considered a " m:Lndibula.r sclerite " bearing the same relation to the nirr.ndib111a.r seta as t h e maxillary plate o r sclerite t o the corresponding sekt-a view still much held in some qimrters, e. y. Comstock. Meek (1903) did a very detailed piece of work, and was one of tlie few cicncla. workers to notice the salivary canal. He considered tile frons t o be correctly Eioniologized. hlarlatt was followed r e p d i n g the " mandihular sclerite " or lorurn. ,Just before Meek, Bentley (1900) considered t h e frons, in our sense, tlie clypens. Berlese (1909) studied both t h e nyrnphal and a.dult heads, but adopted :t terminology which h a s few adherents. Our v e r t e x is cn.llecl tlie post_f'ront,while o w t,riangnI;Lr appendage of t h e frons bea.ring t i i t : median ocellns is t h e tergite of t h e antenna1 or second Iiea.d-somite, of which t h e sternite is coustituted by t h e a.ntenna1 shelf (p. 86, fig. 41). Unfortunately, t h e different sections of Berlese's trementlous work are not always consistent i'nter se, so that, for iiistance, our frons, usually called by Berlese prefronte (p. 86, fig. 41) on fig. 422, is Inbelled clypeus. The adult head is considered more especially on pp. 98-99, figs. 58--60, where the clypeus is so labelled, except a. small a.nd riot really differentiaterl portion which is called Zubmnb, our labrumepiplmrynx not appearing in any o f his figures, having been prolxlbly boiled off in caustic potash. The labium is said t o consist of submenturn, nzentu?n, and a.ppeiida.ges (lobi). The hypopharynx is considered (p. 365) to form the floor of t h e labial gut,terl aiid to be produced backward into t h e head as a peculiar process, wliicli is evidently our text-fig. 6, v. Muir ant1 Kersliaw (1911 a ) considered ndult structure. They recognized mid pointed out t h e importance of our frontal suture,

M O R P H O L O G Y OF THE CICADIDAZ.

377

called " mandibular sutiire," ant1 describer1 tlie niandible articuIntirig by a " true giiiglynius articii1:ttion " a t t h e posterior elid of t h i s suture. The articulation point is taken as a landiiiark between f r o m and clj,peus, m t l oitr present frons is therefore tlesignated cZ!jpezis. A vei'y good :tiit1 clenr account of the rnechnnisni is Gppeiitlecl. Tlie vitriie writers (1911 b ) made embryological studies on a Reduviitl Heteropterou, demonstrating, as Heymons (1899) liacl alrea.dy somewhat indefinitely inclicated, t h a t t h e iiiandibiilxr setx ant1 its :xtual base develop from t h e entire iiianrlible. A t t h e sanie time they confirnied Heymona's demonstration of t h e early tliff'erentiation of t h e ni:txill:t into iriaxillnry plate a r i d seta. A third paper by these aur,hors (191 2 ) considered the nyrnplial head of tJhecicada aricl t h e embryology of tlie Sip'hanta (Fliiticla?) head. The results corifirniecl those previously announced. Snodgrass (1921 a ) gave a very clear arid bea.utifully illiistrated a.ccourit of tlie head of Septe?zdecim, with r2 critical disciission of previous work, iiiaking also it suggestive coniparison with a n Acritliaii l i e d . He considered t h e froris t o be lioiriologized as we liave stxted, a n d called o u r la.~)riiiii-ephipharynxt h e labruiii. Il 1 i e s:div:iry climnel of t,lie mxxillnry set:e was riiissed entirely. l The saine writer (1927), iiifluencecl I)y his new decision tliat tlic second enlftrgeineiit of t h e footl-caii:tl is t h e true pIi:ii*ynx, believes t l i i i t t h e f r o m should be knowii ns the post-clypezis, :rritl tlie clypens, i n o w sense: t>he ante-ch~pe~~cs-a view in which we cannot follow him, b u t which is discussed in tlie account of t h e footl-canal . Mnir (in Zitt., a.iitl in 1926) now believes tliat our frons slioultl I,e so-c:illetl, or perlraps cb~peo-ii.or,s. !L'lie cliitirious piwe supporting tlie posterior l i d f of t h e ph:tryiix or snckiug-primp is corirpnretl with t h e frontal plate of some Orthoptcra., ixiicl is slioxii t o be tentorial rahlier t1ia.n stoiiiot1:wl iii origin. Tlie chief debated points in the literatiire tliiis reviewed concern tlie homology of t h e large, strintetl? facial sclerite, and with i t t h a t of all tlie other pieces of tlie face, anti, secoiidl~-, tlie intet~pretntion of tlie tentorial structtires. If t h e chitinous trough wliicli supports iiiost of tlie srickiiig- pump is, a s illuir suggests, tlie frontxl place of t h e tentorinni-to wliicli i t w a s comp:i.recl also by Snodgr,ass (1921, p. 2),-theii t h e striit coiiiiecting i t witli t h e rest of tlie frontal apodeiiie niarks t h e inoagii:ation of t h e anterior a r m of t h e tentorium. Ant1 as this iiiv;igiriation is generally accepted as inn.i,l;irig t h e boundary between frons n.ricl clypeus, t h e striated facia! sclerite would be t,lre frons. This was t h e T-iew of Sriodprass (1921 cc) arirl of Meek. I n Heteroptern, Bugnion a n d Popoff (1911, 11. 650) woultl seem d s o t o agree mlien they state "plus en arrihre, les bords de 1'6pistorne (clypeus) formrnt tleux petits ailerons (figs. 2, 3, a;.), qui doiineiit attache pm Ieurs amgles posttdiieurs aux branches nnt6rienrs du terltorium. . . ." Btit they believe t h e facial sclerite in question t o be t h e clypeus ( i p i s t o ~ z or c h q x ~ o n ) e

378
I)ecaiisi? i t gives insertion t o t h e pharyngeal dilator iniiscles( line tles ca.r:ict&istiques cln clypeus est quil donne insertion par sa face profontle a a x fasceaux aritk-ieiirs d u clilateur d u pha.rynx. Snor1~r;xss (1921, p. 1) ca.lls the siinie piece tile frons for tly the same reason t h a t t h e French writers named it clypeus. e prominent ridged facial plate (36.) designated t h e fj.ont is IieciLiise of tile att:tchinent of the pli:tryngeal muscles t o it as i n other insects. The clypeiis of t h e two Frencli workers would tiiei.efo1.e a,ppexr t o be t h e front of general morphologists. Siio;igrass, however, believing now t h t the t r u e pharynx is t h e secontl en1;tigerrient i n t h e alinienta.ry canal of cicada (see section 011 digestive system) a n d t h a t his pharynx of 1921 is a Iiioilifiietl niontli-ca.rity, has changed his rien-s on t h e homology of t h e striiitecl sclerite accordingly, and identifies t h e t r u e front with t h e small t r i m g n l a r piece bearing t h e median ocelliis and coiisitleretl by 11s a part of onr frons, which Sriotlgrass iiow coiisitlers largely clypeal, a s did Mnir a n d Kershaw (1911, 1912). I Int, i n t h e rrie;l.iitiiiie illuir has decided t h a t iie was l n i ~ t s l i e n n i liis former v i e w on t h e tentoriuiii, and now s t ~ v a l ~ c t h e views es fortli :it the lieginning of t h i s paragraph, a c c o d i n g t o which t.:e disprrt,ed striated sclei,ite is the frons. Thus, ciuiousiy euongh,-&Luir and Siiotlgrass have in a sense changed places in tlieir interprehtion. We lii~ve seen t h a t Siiodgrass now regirds t h e great siiclringpiiiiip as a inotlifieil p w t of tlie inoritit-cavity rather t l i i i , i i a true l t i m r p x , since is 1i:is no general niiiscul:tr covering, b u t only t h e tlorsnl (iihtors. U n t grnnting, as w e must>,t h a t tliis suclringpiiirip is niidogons in fiuictiori t o that of non-Heinipteroiis itistlcts, tlie loss of tixitsverse muscles and t h e great developinexit s :we per1i:ti)s riot sufficient reasons for declining, , t o r e p r t l i t as also lroinologous i n structiire, ~ t1e:ist i t sliocvs 110 greatel. 1nodific;itioii t h z n tlre trophi thein, selves, if, :LS,211 agree, these also ive1.e tleiivecl from t h e hoinologous p:lvt,s of a inandibuhte insect. Moreover, we limy ask wlrnt coiistitutes a plinryiis other t h n tlie anterior portion of tIie htjoiriotl;enin. llie tleiiiarcatiori between nioutli - cavity a r i d piiarynx, anti between t h i s and mophagus, i s surely more or less ;zrl)i trnry. [f I\Iuii.s present view a.s t o tlie tentorium be correct---:i.nd it \%,oultl be tlifficiilt t o explain t h e coriditiori fount1 i n test-fig. 4 0 1 1 a,np other bnsis,-tlien t h e :interior arms of t h e tentorium i r i t,lre previous sense of Miiir atltl liersha\v (1911, 1912) a.iic1 in tll:\t of Doering (1922) really cori~espond t o t h e dorsal a m i s ill the cochoscli :inti other insects. The likelihood of confiising t,llese two pairs of arms was pointed o u t by Dr. Crampton in
conversation. nfiiir mnkes tire interesting suggestion t1la.t tlie elements of tile t,e,lboriun represent segriienhtl apodemes, a.iising at t h e edge of tire segrneiits composing tlie head. In cicada there a r e four pairs of sllcli inv;igiiiatio~is-frontal, genal, maxillary, and labial.

MORPHOLOGY OF THE CICADIDh?.

379

La6runz-e~~oi23hai?l1zx.--Tht: projecting portion, or the ~ l i o l e of , t h i s terniina.1 sclerite of t h e face 1ia.s been c:illed t h e Z G C ~ Z L by ~L niost of tliose wlio identify tlie Iar,ge, stri:xtetl, swollen pl:i,t,e :is tlje froiis (Snodgrass, 1921 ; Meek, a n d Tower in Heteroptei,:r), a n d epiplia.ryrix by those who name tlie directive sclerite t h e clypeus. &feel< won1d distiii guisli :I. r ntl iinentary 1:I 1 )iwni , d eep1y groovetl iLnd oaerlyiiig tlie epipharynx. Muir ant1 Kerahaw (1911 a , p. 2) s n g g e s t t h a t t h e division between 1:abriim (our clypeus) ant1 epiph:i,rynx is r i o t tlistinct, and therefore t h e piece niiglit he cwlled z~hruin-epioi23har?/ii~ i n Diptera. But witli t h e lioinology as suggested i n t h e present contribution t,he conipountl te1.m might, f o r t,he following reasoxis, lie retainer1 for tlie narrow termina.I piece only. I i t arose from t h e edge of the sclerite mliiclr we f c.dl clypeus, it; woiild be n. simple 1a.bririn : b u t its origiii on tlie iniier surface (text-fig. 2, 1.) of t h e clypeus mid its geiiwal na.t,rire :l,s a grooretl pd, sitve tlist:tlly, are epipli:iryrige:-1.1cli:i~.acters. Y e t tlie epiplmiyiix, iri Insecta generally, :r.rises from t h e iriiier snrf:l.ce of t h e lnliriiin, n o t of t h e clypew (Conistoclc, Inims), tllongll Packard (1898, p. 43) states : The epipli~.ryrixis t h e lintley snrf:i.ce o r phaxyngexl lining of t h e clypeus a.nd labrum, foriiiiiig thz inenibranons roof of t h e month. I 1 these circnm1 staljces we think it ntlvisnble t o retain the compound ntline f o r tJhe t,eiminal sclerite of t h e cicada. face. WitLi regaid to t h e einbryo1ogic:il development of tlie trophi, a moril here rniist be said of Heymonss (1899) cont,iibnt-ion, tlealt \vit,li frilly i n R h t e r section. Cornstoclr (192,5, p. 398) quotetl Heyriions i n support of t h e view t h a t t h e inaridi~~iilni~ sclei.itrs (011r 1ol.a.) a.vc forrrietl from t h e basal part of the rriihlyonic In;lnrlible. while Siio(lgi,ass (1921, p. 8) hrings liini i n 3,s provirjg tile corjt,r:i,ry. It behoves ns, therefore, t o see what Heynioiis re;lJly st,nterl on this qiiest,ion. He (lid definitely pi-ove tliat t,he Inaxilla, split early into a p1:i.te and &a, o r ratlier into two lobes giving rise respectirely t o these strnctiires : b u t coiiceimiiig tlie ~nn.rit]ibles was less t1efinit.e. The following pnsage (Heyrnoris, lie 1899, p. 422) is doubtless Coinstocks witness : 6 DR. (lie Lor;*. bei ilen I-Iomopterou sell~stiindige, tleirtjlicli von {lev Sti1-n itbgegrenzte Sceletstiicke aintl iind da. sic iin derselben RezieIjung z u tlen Ka.iitlibe!n stelin wie die L:3nrinz max. zu (]err ~7ortlerenk x i l l e i i , so kiitliien (lie Lom entsprechenti :L]s Lainiiiz mandibulitres hezeichrieti werden.

DtIt Heynions evidently wished t o reg,%id t h e m:indibiilar Scleyite a s honiologous witli t h e niaxillary plate, though Iiis own elilhryologicaL woi,k and figures indicate no such correspondence. T i i u s (1 899, p. 422) :-Diese S O ~ . Lor& sind in entjwicklnngsgeschielitliclier Hillsiclit keine ganze einheitlichen Bildungen, incleiri sie sO~ol11ituf Restandtheile des Antennensegmentes wie aL1f solelie des Mmdibelsegments zuruckzufnhren sind.
(6

380

UR. J. G. MYERS O N THE

A n d flirther (p. 440) :" Vom anatomisclien Standpunkte lassen sie sicli rleswegen iiiit den L m i i i i a mnxillares vergleichen, weil sie wie tliese d i e TnsertionsflSclie fkr die Protractormuskeln (Mandibnlare Protractoren) e n t h d t e n . Entmickliingsgeschichtlicli hnbe icli dugegen nicht deli Nachweis f iihren lriinnen, dass an cleni A ufbnii tler Lamin= niandibulares sicli anch noch die Extremitiiten tles Manclibularsegnients betlieiligen " (11. 438). " D i e M;i.nc[ibel wiinle derngemiiss :~Isoirri wesentlichen n u r noch clie Inorpliologische Redeutung einer ' Lade ' besitzen."
I n view, therefore, of Muir itnd Kersliaw's work, it niay. be tnken :IS established t h a t the whole of t h e embryonic mandible develops into t h e seta and i t s base. '1'0 siini up t h i s section, we ma.y regard three other points in cic:b(la. head morphology as settled :( 2 ) T h a t t h e sclerite on each side, which we have called tlie l o i ~ i m is : p a r t of t h e genn, ancl is neither a. lateral development , I of t h e clypen.1 region, as M u i r ancl Kerslinw (1911 6, p. 78) heliet7etl, nor t h e sole remains of t h e functionless nia,nclible (biiiith, 1892), nor yet :L iiiandibular sclerite (Meek, Marlntt, and ot,hers). 1 1 1 this interpretatioii Muir, according t o iiix latest paper ( l 9 2 6 ) , agrees witli Hnodgrnss. ( 3 ) Thak t h e iiia,xill~ e represented on each side by n bnsa.1 w pI;Lte miiich becomes inore or less fused with tlie hend-cnpsitle, ancl :I, t1ist:tl setn which witli its fellow forms t h e second or inner p:rir of r o s t d setre. This follows from t h e embryological morlr of 1Ioyinoris :rnd of M u i r and Kershaw, a n d as Sriocigrms (1921 a , p. 8) tins said, iniist be accepted until new embryological evidence is ;rtl~laceclt o t h e contrary. ($1 AII writers, except Smith," writes Xnorlgrass, " agree tl1;t.t t h e labinrii of t h e adult is formed from tlie fused labial :Ippendn.ges of t h e embryo.'' Muir would c d l t h e basal segment i n cic:i,da t h e ?miztzLnz, tlie second, which is somewhat swollen, the pcdpigei. ; aiid tJliethird the ZigrbZa. Rerlese (1909) terms the P:I nie segnieiits srdiiaent~cn~, ?izentrtnz, and lobi, while Marlntt (1895)calls tiiein submentuna, mentuni, 2iqttlcc.
6'

rl. TIIORAX D AK
CX.

ITS

APPENDAGES.
there is coiisiderahle

'7%03'(135.

s in

so many other ChiLracters,

homogeneity in thoracic structure throughout tlie family. A

ii,jticenl,le featnre is t h e greatly eiilargerl mesot11ore.x and esl)ecia.llyi t s notuiii, doubtless associated with t h e very stout a n d powerf ill fore-wings 01- tegminn. I n t h e nymph, 011 t h e other Iinncl, tlie prothorax, carrying t h e niuscnlature of t h e strong fossorid fore-legs, is much t h e largest of t h e thoracic segments ; n,rld srlch is t h e condition in t h e imago of Te'ettigoi-ctcc,where t h e hyprtropliiecl pronoturn (text-fig. 8) overshadows t h e much

MORPIIOLOGY O F T H K CICAD1O.E.

38 I

Text-figure 7.

Meelampsalta nzuta (Fabr.).

Thoracic nota.

p . , pronotum ; ms., mesonoturn ;mt., nietanotuin; ce., cruciform elevation (scutellum).

Text-figure 8.

Zbttigcrrcta crinita Dist.

IIpad and nota shoiving the tremendously enlarged

pronoturn.

reduced mesonotum. I n this primitive form the tegminn are far vveaker, without that inassing of veins on the fore-border 50 characteristic of all other cicadas. I n these latter, however, the PROC. ZOOL. Soc.-I 928, NO. XXVI. 26

382

DR. J. G . MYERS ON THE

pronotum is the second largest of the thoracic segments (textfig. 7 , p . ) . The only worker LO deal especia.lly witli thoracic structure in Cicntlitlze is Taylor (L. H., 1918) who used il'ihicen linnei 8. & C:. (Cicada tihicen n w L.) arid Cryptotynapana epithesia Ilist. l<erlese(1909) gives considera,ble details. Our description and figures are based largely on iWeZanapsalta ci'iagulata and iM. nztcta. Three pairs of small, free, cervical sclerites are present. I n the prothorax the notum is curved down considerably laterally. The surface is mmked with deep grooves associated with muscle insertions, though !raylor would regard them as division marks for a triangular prescatum, a scutum narrow mesally and wider a s it approaches the pleuron, and a scwtellum. -We believe this to be entirely fanciful, as, indeed, Crampton * has indicated in Dissosteira (p. 350) when he says t h a t the sulci of the pronotum divide the surface into areas which have been very incorrectly identified as prescutunz, scutum, etc., although they a r e purely secondary structures having no connection with t h e typica.1 notal subdivisions of the wing- bearing segments. The pleuron of the prothorax is grea.tly reduced and practically fused with the notuni. The fusion concerns chiefly the epimeron, while the episternum is more reduced. The prosternurn is represented by a single plate bearing the well-developed furca of t h e endoskeleton. The plate may be regarded as a fused eusternum and sternellum. The mesothorax, as me have seen, is the largest division of the thorax. The anterior. part of its notum projects beneath the pronotum almost to the occiput, and this part has been distinguished by Taylor as the anterior phragma (prophragnzn of Herlese). The presc.utum is only indefinitely rnarked off by sutures beginning along the outer edges of the dark obconical marks which form a n almost constant feature of cicada markings (text-fig. 7). These correspond to muscle insertions. The greater part of the notum is occupied by the scutum, but t h e scutellum is sharply diEerentiatec1 as the cruciform eZevatio?L of taxonomists (ce). The posterior edge of the scutellum runs down laterally to give insertion t o the axillary cord of the tegnien. The scutellum entirely hides from dorsal view the stronglychitinized post-scutellurn or post-not urn, which forms a shelf beneath it. The post-scutellum is joined t o the epimeron by a ~ ~ a i - r obridge (text-fig. 9, poa.) called by Taylor the postdare, w which is just the prolongation of the katepimeron. On t h e pleuron the pleural suture is distinct, and i n addition there is a suture dividing the episternum into a n upper anepisternum (aes.) and a lower katepisternum. The latter suture is coiitinuons posteriorly across the pleural suture, and there divides the epimerori in the same way, a condition said t o be peculiar to
Soc. Am. xi. pp. 347-366, pl. 32. text-fig. (1918).

* " The Thoracic Sclerites of

the Grasshopper, Dissosteira carolinn," Ann. Ent.

YORPIIOLOGY O F TIIE CICADIDB.

383

t h e Cicadidae. A t the lower extremity of the pleural suture the coxal region shows a distinct posterior meron (me.) and t r u e coxa (c.). I n CryptotympanffiTaylor found that a median portion was marked off on the episternum between what we have called anepisternzcnz and 7catepister.nzcn~ respectively, and this tripartite condition recurs in MeZffimnpsalta (ms.). It will be interesting t o study its occurrence throughout the family. I n the mesosternum of Xelampsaltffia eusternum or basisternite (e.) and a sternellurn o r furcasternite are t o be distinguished. The invagination in the latter t o form the furca presents externally a deep and wide mouth. The trochantiii (text-fig. 9, tn.) is normal. Text-figure 9.

2Kelampsalta sericea (Wk.).

Nesopleuron.

aes., anepisternnm ; aem., anepimeron ; ms., median division of episternuin ; e., eusternnm ; tn., trochantin j c., true coxa ; ks., katepisternum ; k e i ~ .katepimeron; me., meron ; p a . , postalare; sp. 2, second spiracle. ,

The inetathorax is extremely short, especially dorsally, where t h e whole notum is reduced t o a narrow band (mt.) behind t h e scntellum of the mesonotum. The posterior edge r u n s down laterally t o the axillary cord of the hind wing. I n Taylors two species and in Melffinzpsaltffi pleuron shows the a distinct pleural suture dividing off an anterior episternum (es.) and a posterior epimeron (em.), both of which remain undivided further (text-fig. 10). Taylor (p. 2 3 2 ) writes : The epimeron of the female is large and rectangular. In the male n broad lobe, called sometimes the operculum, extends backward from the epimeron and sternum. . . .

We believe that the operculum is purely epimeral. In Xepteizdecim it may be seen distinctly t o curve round the posterior 26*

384

DR. J. G . MYERS ON THE

margin of the sternum, towards its fellow. Moreover, the sternum has a distinct process of its own t o which the folded membrane is attached. W e shall discuss these structures in more detail in connection with the rest of the sound-producing apparatus. The metasternurn shows a eusternum (text-fig. 10, s.) and a sternellum (st.). The trochantin occupies its usual anterior pc!sition (tn.), and must not be confused, as is so often done, with tlie meracanthus (mc.) of Fiebei-, which is a spur arising from the meron. l ' h o m c i c Endoslceleton.--The furca of the prethorax is very strongly developed, the others less so. The chief peculiarity of the endoskeleton is the great development of the mesophragma, giving insertion t o the great lateral muscles of the mesothornx (costccli-dorsccli of Berlese). This pliragma is called by Berlese the preclaaicoln clel metanoto. B u t i t seems t o 11s t o arise Text-figure 10.

mc itfelampsalta muta. Female : metasternum etc.

s , basisternite (eustcrnum) ; ps., pleural suture ; mc., nieracanthus ; es.,epistcrnum ; . em., epimeron ; op., operculum; st., sternellum ; c., coxa; tn., trochantiu.

distinctly from tlie postnotum of the mesothorax, and thus to be a postphragma of that segment. Taylor (p. 230), with views evidently colourerl by Berlese, writes of the post-scutellum or post-notum as *'fused with the anterior phragma of the metathorax," but gives no evidence that the pliragma in question belongs to the metathorax rather than to the mesothorax. We may, of C O U ~ ~ C , the question by admitting that in the last beg analysis, since a phragma is a plate invaginating betwrem segments, i t belongs to neither. Hut Crampton, in his work above cited on Dissosteira, makes the definite statement (11. 354) t h a t the post-scutellum of tlie metathorax bears a pliragnia for t h e attachment of tlie longitudinal muscles arching the notum in the movements of flight ; arid similarly, in an earlier paper *, " The posterior plate , . . . or post-scutellum usually consists of an external region bearing an internal phrsgma." Imms, moreover (1925, p. 47), would appear to hold the same

* G. C. Ciampton, Zoologisch. Anzeigrr, xhr. p. 60 (1914).

MORPHOLOGY O F THE CICAUIVE.

385

opinion. W e therefore name as mesophragma the great plate which cuts off the whole of the fore-part of the body from the rnetathorax and abdomen, leaving a narrow perpendicular slit for the passage of t h e viscera and nerves. ZL'eZatioiLsh~pps.-Tlie tripartite mesothoracic episternum is considered by Taylor a nenropteroicl character. Cramptort h a s :dso usell thoracic structure as a n argument for neuropteroid relatioilships. According to Taylor the threefold division of the Test-figure 11.

Melampsalta muta.

Female abdomen : ventral view; ovipositor largely disengaged. chordotonal organ) ; a.,auditory capsule.

I.-IX., abdominal segments ; 6-9, spiracles ; m., mirror (tympanum of

episternuni is present also in Cicadellide (Jassoidea). The composite mesothoracic epimeron would appear t o occur in no other Homoptera.

p. Legs.
Hansen has made a detailed study of the legs (1890, trans. 1900-1903). The prothoracic legs of the Cicadas are highly characteristic both in the nymphal and imaginal stadia. The femora of t h e first (see later section) are much swollen and strangely modified for fossorinl purposes; those of the second (text-fig. 15) are

386

D A . J. G . MYERS ON THE

enlarged almost as much, and furnished ventrally usually with three %toutspines, directed somewhat obliquely. The trochantins, as we have seen, are distinct, t h e coxre very long, with a considerable distal portion free. The interior angles of articulation are a t a considerable distance from the insects middle plane (Hansen). The trochanters are thick with a very oblique femoral articulation, producing a see-saw movement. Text-figure 12.

Xelamnpsalta wtutu. Male

front view of right fourth spiracle.

m., mirror; a., auditory capsule.

a probably primitive simplicity.

The second and third pairs of legs are unspecialized, showing I n the second pair coxre are moderately short and broad, and situated near t h e middle plane. Their principal movement is rotary (Hansen). There is no t r u e meracanthus, but t,he meron is well developed. I n the third pair the cox%, according to Hansen, are t h e Text-figure 13.

~ r ~ 7 ~ m p s a muta. Male : lateral view of parts of segments vII.-IX. to show Zta position of tenth (and last) spiracle.

sinlplest in the Auchenorrhyncha. They are contiguous in t h e middle plane and articulated in a pagiopodous manner, i. e . only a hinge-movement is possible. The meracanthus is usually well developed. The tarsi are usually three-segmented as in all other Auchenorrliyncha, b u t Cicadidae differ in lacking any empodial formation.

MORPHOLOGY O F THE CICADIDAC.

387

stout, and simple. Those of the The claws are two-equal, nymphs, as first indicated, but not described, by Hanseri (trixnsl. 1901,.p. 151), are very different from those of the adult. I n t h e universal habit of cicadas t o rest habitually in a perpendicular position, whether t h e support be trees, herbs, or Tdxt-figure 14.

jpfelantpsaZta muta. Female. ventral view of base of abdomen. sp. 4, fourth spnacle; S.I., h s t abdominal sternite; S.II., second ditto.

rocks, we may see an explanation of the powerful fore-legs. The attitude is largely a suspensory one. in which the second and third pairs take little part, most of t h e weight fnlling on the prothoracic pails. A similar condition occnrs in certain moths. Text-figure 15.

ZeZampsaZta sericea.

Fore femur, inner view.

The Wings. Chabrier (1822), Amans (1885,1915), andImhof (1901,1905), also Haupt (1913), have devoted considerable study t o cicada wing-structurs from the viewpoint of t h e mechanics of flight. Cornstock and Needham (1898), in their first classic attempt to homologize the wing-veins in all the orders, studied cicada venation in detail and ascertained its condition in the nymphal wing. Their interpretation, with the Tillyarcl modifications, has
y.

388

DR. J. G . MYERS ON THE

been adopted in this paper. Woodworth (1906) dealt more 01less incidentally with cicada venation, ancl drew attention t o the nodal line, miased by Corristock and Nieedhnm (at least in text), but considered at greater length by HorpAth (1913) in a paper whish introduces several divergences from the Comstock-Needham interpretation. Berlese (1909 J pays considerable attention t o t h e axillary sclerites. The first taxonomic division of the family w ~ based, Iinsuccesss fully as events have proved, on the wing-venation, while Iatm workers have used i t for generic separation. Comstocks latest account (1918) is apparently identical with that of Comstock and Neeclhams original coiitribution, while iVlooltons account (1923) ancl figure follow Horvath. Conistocks Text-figure 16.

Melampsalta sericea. Lateral view of male abdomen.

I.-X., abdominal segments j fm., folded membrane.

type was libicen ; ours will be chiefly Melunipsalta muta, with which we lime attempted t o compare other members of the family. Text-figs. 49, 50, and 5 3 are to the same scale, while 48 and 54 are more enlarged.

i. Tracheatiom. A transverse basal trachea exists between the costo-radial and the cubito-anal groups of pre-venational tracheE in the developing wing-base. I n this respect cicadas a l e less primitive than Membracidq Cercopidae, arid some a t least of the Cicadellidz, where the basal connection is absent-a condition t h a t exists nowhere else according to Comstock (1918, p. 275), b u t in Plecoptera and certain Blattids. Cornstock has shown tliat when this transverse basal trachea is present the medial trachea tends t o migrate along it towards the orbit-:ma1 group. This is well shown in cicadas, and reaches an extreme in those cases where the M and Cu are united in a common stalk a considerable distance after entering t h e tegmen (text-figs. 48, 49 ; cf. 50, 55).

MOltPIIOLOGY OF THLE CLCADIUB.

389

The tracheation is of t h e simple type, in which only the principal veins are represented by pre-existent tracheae. The nymphal scheme is much more primitive than the adult venation would lead one to suspect, and, in fact, it approaches very closely t o t h e Comstock-Needham hypothetical type- T h e great reduction of H 1 is a striking difference.

ii. Vencctioiz. In the following interpretation we have followed, in company with Muir, Imms, Alexander, and many other workers, t h e Tillyard modification of the Comstock-Needham system, considering the latter's 1 A to be really Cu 2 .
Text-figure 17

.NelampsaZta muta. Male : lateral \iew of left sound-appaiatus.

t., tymbal; sp. 3, thiid spiracle; op., opercnlum ; a., auditory C R ~ P U I P ; mirror. nz.,

First is a description of the venation in Melanzpsffilfcc mtctffi, and t h e n n discussion of divergences i n other members of the family. T h e orthodox abbreviations need no explanation. I n t h e tegmen (text-fig. 48) there is a conspicuous massing of veins on t h e costal margin ; but t h e elements are distinguishable under magnification. Costa is very thick and coincides with t h e margin of t he wing. S c is fused with R, at lenst a s f a r a s t h e noclrtl furrow, which will engage much of our attention later. R I is apparently obsolete i n t h e imago. The fate of t h e other branches will be apparent from text-fig. 48 (cf. 49, 50). The

390

DR. J G . MYERS OK TUE .

base of M is in close contact, if not actually united, with the common Sc R stem until the arculus is reached. There M splits off to form the anterior arculus and to join Cu, by which latter fusion the posterior arculus, seen in more typical cicadas (e. g. ilibieen, text-fig. 5 5 ) and representing the cross-vein between M and Cu, is obliterated. So far we are in accord with Comstock and Needham, save for the differences in the species described. These writers state that there is a peculiar and unparalleled condition in certain insects, including cicadas, consisting in the fusion of 1 A with the base of Cu, so t h a t it appears a branch of Cu, far separated from the common stem of 2 A and 3 A . Tillyard has given good reasons for. considering their 1 A as Cu 2 , thus making Cu two-branched-a far more probable explanation

Text-figure 18.

Melniirpsalta serieea.

Male : lateral view of right sound-apparatus, with segments softened and stretched to fullest extent.

op., cut base of operculum; c., metacoxa; fm.,space normally covered by folded membrane ; ms., metasternum; mt., metanotnm. Other letters as in text-fig. 17.

of the condition. The anal furrow lies along the conjoined courses of Cu 2 and 1 A. The venation of t h e hind-wing is niore specialized than t h a t of the teglnen, the difference depending, in Comstocks opinion, on the more nearly basal forking of R in the former. From this it results that only R 2 + 3 fuse for a considerable distance with S c ( R 1 being completely lost), while R 44-5 coalesces for a short distance with M. The common S c + R Z + Y stern is in close contact with C, and in ihttigai-cta (text-fig. 53) seems actually fused with it. While M is normally only three-branched, there occur aberrations in which exists a small remnant of cell M 2 , usually crowded out by coalescence of M 2 and M 3 ap in our fignres. This venationnl variation is only one of many which

NORPHOLOGY O F T E E CICADIDB.

39 1.

occur witb frequency in all genera of which I have seen sufficient material for comparison, and such characters should be introduced with great circumspection into taxonomy, even when they appear constant. Pauropsalta, founded by Godirig and Froggatt on such a venational difference, I do not think can stand, and Bergroth (1911) has been critical of some of Distants siniilar characterizations. Applying the Tillyard correction in the hind wing, we find in the adult (text-fig. 54) that Cu branches in a U-shape right at the base of the wing, and the two branches become considera,bly septrated, Cu 1 dividing into Cu 1a and Cu 1b, while Cu 2 runs in the anal furrow with, but really separate from, 1 A. The Text-figure 19.

b -

Mngicitadu septendecim (L.). Male : KOH preparation of mirror and adjacent parts, flattened out. fin., folded membrane; i., inteisegmental membiane between thorax and abdomen ; I., II., abdominal aternites ; m., mirror; p . , process to chordotonal organ ; t., tymbal; w., wing (see p. 466); a.,auditory capsule.

two remaining veins, by this interpretation, are 2 A and 3 A respectively, thus being avoided the Rnomalous attribute of four anal veins, or a two-branched 3 A , required O K ~ the Comstock hypothesis. The anal area is much more developed in t h e hind wing than in the fore, and folds under the corium when at rest.:Finally, both tegmen and hind wing are surrounded by a strong border between the ambient vein (text-fig. 48, an&) and the edge of the wing. Venation in the Fumily.--l\lain features, as described above in Melampsalta, recur throughout the family, save in the subfamilies TettigarctinE and TettigadinE. Net-veining of the tegmen, a character on which Amyot and Serville based the first classification of the family, has almost

392

DR. J. G . MYERS ON THE

certainly arisen independently in several different branches of the family,and occurs to-day in several unrelated genera :-Polyneum, Anycc~nirma,Talai?zgcc, Eenridictya, and others. The condition, moreover, in such a form as Polyneurw, brought about by ercessi\ e regular dichotomy of the longitudinal veins, differs grettly from the network of irregular cells found in the Hemidictyirie forms. Lenz3eju (text-fig. 56) and Cystosonzcc show a more or less complete abkentse of the coriaceous border on the tegmen, the ambient vein coinciding with the margin, which it slightly thickens. Text-figures 20 8 21. :

21.

Magicicada septendecim.

Female : as text-fig. 19. Internal view.

ac., ieitiges of Vogel's rudimentary abdominal coxites I. and 11.; I.-111. abdominal sternites; p., left member of pair of pores of unknown function ; .$., flap (see text).

Pig. 21. Portion of the same preparation as in text-fig. 20, but with flap turned forward.

More fundamental differences occur, as we have mentioned, in Tettigarcta and in the Tettigadinz, which show between them a gradation leading up t o the specialised condition of the dominant and more typical cicadas. Venationally there is much more in conimon between Tibicen and Xeelanzpsabta, which are separated as far as possible under the present classification, than there is between either a nd Tettigarcta. Yet the present scheme places the latter with Melumpsalta in the same subfamily. Y'ettigarcta is a t the very base of the Cicadide, and is extremely

MORPHOLOGY OF THE C1CAI)ID.B.

393

primitive in far more respects than absence of sound-organs, and not least so in venation. The TettigadinEe, of which we have examined Chonosia c m s s i pemais (text-fig. 49) as a type, connects it, venationally only, with the dominant groups, and so may be described first. I n the first place, Chonosia shows a distinct strongly-marked membrane fold along the nodal line, to be discussed later. Secondly, Cu 2 lies, as usual, in the a n d furrow or on the cltwal sutnre, but considerably distant from 1 A, both basally and distally, as pointed out in the characterisation of the subfamily by- Jacobi (1907 c). The strondv-chitinized base of the clavus a t ." the point corresponding with the posterior tuberosity of Amans Text-figure 22.

NIeZnmpaZta Zeptomera. Ultimate female nymph : junction of thorax and

abdomen extended to fullest extent.

Lateral view, right side.

mt., metanotum; sp. 2, second spiracle: ac., region of future auditory capsule; a., region which in male develops future tymbal; c., metacoua.

i s said to be strongly developed in the whole subfamily Tettiga-

din=, and is supposed t o form a plectrum for the stridulating apparatus on the mesonotum (text-fig. 59). I n Chonosia crussip e n i ~ i s do not find that it is much more protuberant than in I other Cicadids, and Mr. W. E. China, from experiments with a softened specimen of a Tettigccdes, expresses grave doubts n s t o whether this structure could be employed as a plectrum. Unfortunately observations in the field are lacking. I n Te'ettiyarcta (text-fig. 50) the whole tegmen is markedly coriaceous, the basal half especially so. The surface is covered with irregular shiny taberculations. The nodal line is extremely distinct, as well on the membrane as o n the veins themselves. The anal furrow is very deep, the vein Cu 2 on its floor rather weak. The ambient vein is normal, but the coriaceous border

394

DR. J. G . XYERS ON TEE

somewhat narrow. The main veins bear long hairs. Costa lies widely separate on the fore border of the tegmen, with a large costal cell between i t and the Sc, R, and M group. There is thus practically none of that nrassing of the main veins on the costal bortlev SO noticeable in other cicadas. The strong longitudinal grooving of the tegmen may counterbalance the general weakneb? due to this lack. This folding is very conspicuous, t h e costal cell and the anal furrow being especially depressed. Cu is peculiar (text-fig. 50j; it forks early, and Cu 1 goes u p to join M at the inner end of the anterior arculus, where the prominent boss on t h e surface of t h e tegmen makes the course of t h e veins difficult t o follow. 1 A is extremely distant from C u 2 and is strongly convex, lying on the brink of the deep anal furrow. Text-figure 23.

Jielampsalta cingulata. Male : KOH preparation to show junction of abdominal tergites arid sternites.

]IT.-V., abdominal segments ; t., tergites ; s., sternites; 5-1, spiracles.

The cross-ridging of the tegmen (see later) is practically obsolete, probably owing to the coriaceous texture. "lie venation on the whole is strong, standing out in bold relief. 'Che hind wing of Tettigumto (text-fig. 53) shows, as usual, more specinli~ationthan the tegmen. The hind wings of t h e Cicadidz as n. whole are mere appendages to the stronger and larger tegniina. Sc and R 2 + 3 a l e fused as far as the distal part of the wing-coupling apparatus (see later), where the common stem forks, apparently into Sc arid R 2 3, so that a large additional cell, Sc, i s formed. I n conjunction with this the cross-vein Y between R 2 + 3 and R 4 + 5 is carried very much nearer the apex of the wing. The cell Sc is squeezed out by vein coalescence i n most other cicadas. A vestige is visible, almost hidden by t h e wing-fastener in the l'zbzcen chloronzei-a figured (text-fig. 51).

MORPHOLOUY OF THE CICADIDE.

395

The anal veins are distinctly three, arising from a n apparently common base and spreading beautifully fanwise in a primitive fashion. Practically the whole surface of t h e hind wing shows innumerable microtrichia, which appear t o be absent on the tegmen. The coriaceous border is as in the tegmen. The axillary membrane in Cicadids is well developed and frequently coloured, when it affords a good taxonomic character. It contains the axillary sclerites, and is bounded posteriorly by t h e axillary cord (text-figs. 50 & 59, az.). The ccbseizce o R1.-Muir f (1923, p. 217) writes : The Text-figure 24.

si. 5
HiH

Metampsalta sericea. Male : external view of left auditory capsule. t., position of tymbal; a., auditory capsule; sp. 5, fifth spiracle.

absence of a distinct free Rl in the adult tegmen is characteristic of most of the living Auchenorrhynchous Hornoptera, but it is found in the Mesozoic Cicadid, Mesogereo~Tillyard. We have seen that the R 1 trachea occurs in the nymphal wing of Cicadi&e, as also in that of Membracide. B u t I believe that it the presence of R 1 in ilfesogereosz is only apparent-that depends on a n interpretation of Mesogereon tegmen different from that given here t o the cicada tegmen. If we compare Tillyards figure (1921, fig. 66) of Mesogereon tegminal venation, not with that of a typical modern cicada, but with t h a t of lettigwcta (fig. SO), which shows an intermediate condition,

396

DK. J. G . MYERS ON TIIE

R in both cases is seen to end in three branches meeting the margin of the tegmen or, rather, t h e coriaceous border. The condition in Mesoyereon has necessarily been interpreted from adult structure only, and the first of these three branches has therefore naturally been called R 1, as indeed would probably the first of t h e three radial branches in Tettigurcta and also in more typical cicadas, did we not know from t h e nyniplial tegmen of t h e latter that the true R 1 is lost in development. Therefore there is no proof that the first branch of R is the adult tegmen of iWesogereon is other than homologous with the first branch of R in modern cicadas, which we know from ontogeny is R 2. The p e s t i o n o Costa.-I have followed in t h e foregoing the f commonly-accepted Comstock-Needham interpretation of t h e anteriormust vein of the tegmen as C, since this view is based on,
Text-figure 25.

As text-fig. 24, but viewed internally. t o chordotonal organ ; a c , auditory , capwle; sp. 4 trachea enterlug from fourth spiracle ; t., shows position of tymbal.
m., mirror (mostly out away) ; p., process

arid seems t o accord best with, t h e evidence from nymphal tracheation in Cioadidze. This vein, however, may not be true C. The recorded nymphal evidence in Auchenorrhyncha as a whole is conflicting, since in some forms the corresponding trachea seems t o arise from t h e alar bridge, and in others from Sc, in which latter case i t is apparently homologous with the humeral. Muir (1923, p. 216), in discussing the whole question, notes wit,h reference to Fulgoroida : B y calling this vein the costa we are faced by t h e fact that, in a large proportion of the fulgorids, t h e costa vein and costa margin do not coincide, but the vein lies considerably within the membrane, leaving a precostal cell o r costal area. This is a condition recognized in no other order of insect,^."

NOEPIIOLOGP 01 TIIE CICADIDB.

397

Similarly. Tillyarcly states :" I t seems, however, very iinliltely t h a t any t r u e costal vein, distinct from tlie anterior bortler of tlie wing, was ever present in any Panorpoid type. seeing that such CL vein is absent Jj-orn almost all known insect wings." (Italics mine.) Conistock has bhown that R 7. is conipletely aborted (1918, p. 272). It would therefore seem thnt if our C is Sc and our R + S c stem thu5 R alone, even then t h o vein we have called

Text-figure 26.

Fidiilicina semilatn (Wk.). Internal view of male right auditory capsule. Arrow represents hair thrust into fourth spiracle, and through trachea which emerges from i t ; II., HI., abdornind segments; s., sternites; tg., tergites ; t.,position of tymbal; a., auditory capsule; sp. 4,fourtli spiracle; sp.5, fiftli spiracle; tT., trachea (dorsal) ariying from it and sending off a thinner rentml branch.

Bc, branching oE, t tile node, t o the fore-border, would not be a R 1, buh rnther :t cross-vein betwetn R and Sc. But here again we are led into difficulties, since, as we shall show below, HorvAth is wrong in considering the vein on t h o fore-border, distttl t o the node, as a continuation of tlie airteriorniost vein proxinial t o t h e node (his and our C). It tends t o be continuous witzhthe sliort vein, Tvliich leaves the R (or R+8c) stem at the node, and wliieli cannot therefore be a cross-vein.

* Proc. Linn. Soc. New South Wales, xlir. p. 649 (1919).

PROC. ZOOL.

SOC-1928.

NO.

XXVII.

27

398

UR. J. G . MYEllS ON THE

HorvBtli believes Comstock and Needham incorrect in stating that C terminates a t the node, its place on the anterior margin being taken by Sc. He claims that our Sc (text-fig. 48) is really a continuation of C past the node. Tiewed dorsally i n Xe1arn;oscclta muta (text-tig. 5 2 ) , C is seen distinctly to end at t h e node. Continning i t s line is a piece of similar width and coloiir, b a t flat o r even hollow, being the edge of the tegmen d o n g the subcosta. Viewed ventrally, as Horvith suggests, this continuous portion anterior to Sc does look vein-like, bnt the convexity is merely the bottom of the grooved edge of the wing. Moreover, i n the nymphal tegmen, t h e costal ti acliea is distinctly shorter than t h a t of Sc, it5 termination being a t a point corresponding with the node of the adnlt tegmen. Text-figure 27.

M&zmpsnlta sericea. Male : back view of mirroir

(9% ) and tymbal-muscles (tm., p . , proceess t o chordotonsl organ ; a., auditory capsule.

The costal area or precostal cell meiitioiied by Muir in tlie quotation above is well developed in some Cicadidze. especially i n some of tlie Platyplenrinz. Thus in Ycmga pzdveiea arid I,,/cria n ~ ~ a d n ~ i s c c w i e(textJ-fig.57) there is a wide ampliation of tile ~~sis fore-borcler outside the co&ta. Ihis p~ecost:~I is as wide as cell (Yangcc) o r wider than ( P y m c ~ ) the true cohtal call. sucil , ampliation is an element in a general secondary platyptery considerably rniwked i n these a n d related genera, and accorvipanied by the tlcvelopinent of pxraoota.
iii. l h e Cyoss-ridging. A striking and peculiar feature of the teginen of Xesogereoiz and of all Gicadidzc except lettiyaiatrc is the presence of numeyous cross-ridges, especially visible in ce;t;Lin lights and interpreted by

MORPHOLOGY OF TIIN: CICAUIDB.

399

Tillynrcl as arcbeclictya.1. In the Palzeontinidat there are apparently no sigiis of this cross-ridging, but the condition of the material is usually not such as t o show such a character. Forbes (W. T. M., 1922) saw thiscross-ridging in an emerging For just a few minutes it cicada of cindeterininerl species. wits perfectly distinct. The arrangement w a s very definite ; t h e n:i.rrow cells were filled with a series of simple evenly-spaced cross-veins, while iii cells B, 1st M 2: and M they formed a double series of cells alternating with each other. On t h e coriaceous border beyond t h e ambient vein they were evenly speed, the regular longitudinal veins each ending opposite the middle of L: marginal cell. Towards the costa there were two veins opposite each definite cell, while opposite cells M 3 and M 4 there were three, and more posteriorly even four. The margin of the hind wing was similar, but the clisc of the wing was not .observed. I n the cell 2 A, instead of cross-veins, there was a Text-figure 28.

>&&mpsuZta serieea. Left tymbal from within to show attachment of muscle. pZ.,terminal plate of tymhal-iiiusde; tn., its tendon ; t., tymbal.

series of closely-spiced parallel longitudinal veins, which remained visible i n the dried wing. Forbes regarded these transient veins as * fugitive blood veins, and suggests they are relics of a net-veining such as occurs i n the henroptera. H e believes t h e different arrangement i n the :ma1 region especially suggestive, a,s it recalls the plaited portion of the wing in the Orthoptem, where there exist numerous ynriillel longitodind veins. This :igrecs with Tillyards origirtai theory in 1916, when he H:I.IV in Xesogereon, o n account of these cross-ridges, a n intermediate stage between B u g e m o n and the PnJ~eohemiptern. Yliotrgli Ez/,gereon is n o longer considered Heinipterous, and the Pa1;eolieiniptera are disposed among recent groups, it still rc,rrinins prolnble that tliis condition is n relic of the net-veining of :I, 1iLlreo:lictyo~)teroidor Protorthopteroid-PalEodictyopteroid ancestor of the Iiemiptera.

27

400

DR. S G. MYERS O N ?'HE .

Gonreau (1843, p. 204) remarked that the cicidas are except i o n ~ ~ l tliat the hind wings do n o t iridesce: Gadd (1908 b, in p. 143) very rightly contradicts him, tliough admitting that the tegniina. are certainly tlie more iriclescent. The Russian a u t h o r notices further tliat tlie iridescent wing-colours of older cica.tlas are predominantly yellowish, while those of younger ones are slipblne. iv. The ATodul Line.
lliis is a very reniark;thle structure, present to some degree in the tegrrreii of all Cicadicl:t, and especially ni:i.rketl in tlie very archn,ic Ykttigamta a.ntl in the proliably primitive ~lf"ogamm,ia,in
( 1

whiclr i d in others the basal portion of the tegirien is t h n s m iterl :is :L more coriaceoins aiitl opaque part, at 1ea.st superfici:illy vesernbling R typically Heteropterous, 6.g. Pentntonioid, coiiditjion. It is least developed in such as form 11sMelccnzpsnltu (text-fig. 581, hut a point overlooked is tha,t tlie breaks in the wing-veins where the nodal line crosses them are xlwa.ys complete, Text-figure 20.

,?l~gicicndnseptsizdecim.

Male : inner view of portions of stcmites I. and 11. showing the miirgs (w.), wliicli have beeu described as an abdominal h c a .

even in those which are most divergent from the 2kttigcwc8a-liiie coiitlition, hrihof has described tlie nsn:~l coiitlition of tlie 1~ot1;d Iiiie i l l 22 gcnera in very great tletnil (1'303), recogrrizing 13 e1eiiients7 mid has enumerated tile various kiritls of joints a n t 1 foitis formwl ill veins :mtl ineinhrarie by its p : i sotnt; of tliesc joints e n l a i p i in tpxt-figs. 51 : I T line hegins on the costxl margin : L t the tljstinct iiotle, where the fore-lmrder, e.g. in i2/leZampsalta ci~yzclnta, ofteri howed N J I ~ , is riiiiriing hetween the twinination of C a.nd the b:as:r.i p u t of Sc, crrt,a the S c + R stem. I t then curves in somew1i:Lt tomartis the .wiiig-h:tse ant1 bre~nks t 2 3 ant1 112 3 4 in siiccesjiori, riinniug A tlieu tiown Cn 1 h to the hind-margin (text-fig. 51). Zri tile doininmit modern cicnclu.s, repimerited o n the one linntl by 2'ih ao(1 0x1 the other by ~l~felampsultcc, iiotial line o n the riieiiitlie bra,ne I i e t m e e i i the veins is very indistinct, bnt is aln tiys visible if t,he insect IDA held a t it certniri angle. In those fornis where t h e J i i X ~ aline, as iii Xogamairc, divirles a Insal cnriaeeous, often ~ I~rigIi tly-coloi~re;l, coriuiii from a distal transparent ~ ~ i e r n b r n n e ,

+-

UOILPIIOLOGY O F iLlE CICADIDrE.

401

it I S , of course, very distinct. I n Choizosia, where sucli . I tleinarc&iou is not present (text-fig. 49), the iiodal line is nevertheless very well market1 on tlir membrane. Finally, i n lettigccrcta the riorldl line is extremely conspicuous, rrossing not only tlir vein<, bnt forming on tlie intervening meinl,mne :I tleep groove or fold. tlie lip5 of which appear a s two fine p m l i e l lines. Iinliof (1. c.) a n d Conistock (1918, fig.) both notice t h t , the break in tlie veins where tlie nodal line crob~es not pal ticipated is in h j the t r a c h e a The nodal line is apparently n o t piesenr in iWmogereo?z, and its absence, coupled with t h a t of apical celli, inakes m e reluctant t o place Xesogereoib i n t h e Cicadidz a s M u i r woirld do. In the Palzontinidz, however, both it and t h e node ale present
Text-figure 30.

X e Z n n p n l t a Zeptonzsra.

Feiiiale : iriiier view o f left half of pygophor, the anal segmerit boiled off.

VII1.-IX., abdoiiiiiial segments; s., steriiite; t., tergite; r., cut edge of right
median process ; Z., left anterior process ; my., middle piece, fused distally.

t o some degree, as evidenced by t h e published figures (Haase, H;~ricllirscli,Oppedieirn), writ1 by t h e examples studied by nie in t h e Musenin of Coinparative Zoologj. One of the latter specimens slio\vs a distinct k)reaIi i n tlie curbature of the costal border of tegnien, with au inwvartlly ciirvetl traiisveise line of tlie m m e shape arid relatioilships as t h e n o d d line of cicadas, and appweiitly clividiiig off a inon8 cori.rceous basal portion. In t h e nymphs of :dl recent Cic:didE examined t h e wing par1 itsrlf shows n T ery clistinct riotla1 line in the apical third, dividing off a, very much thiciter basal t w o thirds. Yhis condition appeitrs strikiiigly Heteropterous. S o nirrclr for the facts. Tlieir significance we believe with HorL 6th atid Mon!tori t o he great. Both tliese writers consider the iioclal line homologous with t h a t sepmating corium from

402

DR. 3. G . 'MYERS O X THE

meinhrfine in t h e Hetei-optera. W e have discussecl its possibly priniitire character throughout the order in a later section tle\oterl to pliylogcny. Here we wonld give reasons for coiisitleriiig it a Iiiglily piiiiiitive clisrncter in the C i c d i d s :( I ) lt reaches its greatest development among adult cicadas i n f'ettiprctci, ~ h i c l iin venation, head-structure, absence of somitl-orc:ms, and in othei cliaracters is t h e most priniitive exi<tiiig cic:icla. (2) I t shovs an intermediate condition in such less primitive f o r n i x :I\ Cl~,ottosaa arid other Tettigadins. ( 3 ) It i b least, developed in those cicadas which are the most higlily evolx ed in other directions, notably in clevelopmeiit of the souritl-oigRiis ancl their accessories. (4) I t i5 liiglily tlevelopetl arid strikingly Heteropteroid in the Iiyinlphs of even tlie most modern cicatlas, and in all of such that we 11:ive exaininetl. ( 5 ) It definitely occurs in t h e most cicada-like (Tillyard) all tlic pi e-Cretzceona fossils-the Pakeontinida. (6) It i x tlefinitely present in aZ1 living cicadas knomn, and tiici.efoi-e iepreserits a definite tendency of the cicada stem.
v. The %iny-cozcpli?ty Apparatus. !rile wing-coupling arrangements are simple in the extreme. A recurved flange on the dsrsal. surface of the median part of tlie cost:il border, liiiicl-wing, hooks in a recurved flange on the ventral surface of tlie hind-border of the tegmen. The hind-wing flange

is strengthened by the conjoined stem of c! Sc fig. 51).

e. ABDOMEX AND
ITS

+ +R 2 + 3 (text-

APPEXDAGXS.

u. Gemwal.

'L'CXA distinct segments itre recogiiizable in the cicada abdomen in botli sexes, and, in addition, posterior. t o t h e tenth. are t w o successive parts of the anal '' segment,'' whicli rnay perliaps be interpreted as segrnent XI.and a telsori respectively, a s Hansen (tixi~sl, 1902) suggests. Segnieiits 1 i ~ n d are modified extremely in tlie service of . II. hearing ancl sound-production in the male, arid of the former function :done in tlie female. They will be considered ill greater detail when we come t o describe the organs concerned. In both sexes, segments IX. and following are devoted largely t o reproductive purposes, ancl will be described in connection with the external genitalia. The tergites of t h e ordinary abdoniinal segments (text-figs. 11, 16) are greatly developed and strongly arched round the ventral surface. The sternites, bearing the 5th to 9th spiracles, a r e commensurately restricted.

MORPHOLOGY O F THE CI CAD ID B.

403

There has been consitlevable discussion with regard to the lateral regions and as t o the presence o r absence of abdominal pleurites. Hltnsen (1902, p. 215) takes a somewhat broad band lying between tergite aiid sternite to be the pleuion. This consists of two pal ts-extrinnlly (dorsally) a conspicuous chitinous plate, nhich, except in swollen abdomens like that of Cpstosoma, IS separated from the tergite by R very narrow, thin, marginal uidmbi:ine j and internally (ventrally) by a distinct, narrow, thin inembiane between plate c~ncl sternite, again except in Cystosoma. Heymoris (1893) gives vrluable embrq ological evidence. R e finds (p. 422) in tlie einbrj o very distinct and Heteropteroid iergitinulsfe, which axe passed on t o the nymph and appear to be foriiied partly of tergite and partly of paratergite. They disappear in the adult, or rather become plates which are sepnrated from tlie sternite by a sutuie, while laterally they reach the slinrp body-edge (p. 424). These correspond to o u r under-arched part of the teigite, a n d are seen in text-fig. 11. Heymons (pp. 377, 4 i 7 ) denies entirelythat special pleurites are present in Text-figure 31.

~ a ~ - i i t e t a , f i n * ~ n(Germ.). J f d e : evteriial view of left lower pygoplioral process osa (aborted genital e t j l e of Muir).

any Hemipterons ahdonien. H e finds that i n embryonic development each sterriite consists of three plates, of which he calls tlie two laterd ones parctstemites. These are represented in the acliilt cicadas at most by the a n tlas Stigma angrenzende Partie of the ventral plate (pp. 377,423). Neither paratergites nor parasternites are pleural in origin, though Verhoeff (1893) citlied them respectively obere Pleuren and unteren Pleuren in various other Heniiptera. Vogel (1923), as we shall see in the description of sound and hearing-organs, recognizes quite a. considerable developnient of the paratorgites ancl pmrsternites-more than we can follow. Doering (1922, pls. 59, 62, p. 558) shows in the abdomen of the Cercopicl, Lepyronia, large and distinct pleura which seem to correspond with most of the lateral and ventral part of t h e tergite. It is possible she lilts used the tern1 only in the topographical sense. Berlese (1909) figures no abdominal pleural structures in a n y insect, and mentions (p. 255) as the acme of complexity LacazeDuthiers epimerite from the tergite and episternite from t h e

404

DE. J. C. NYG1tS OX THE

sternite, evidently corresponding \T itEi Hcymonss paratergite and para s t eriiit e . Comstock (1925, p. 75) states that a n abdominal segment consists typically of a tergurn and a sternum united by lateial coiijunctivze. Sonletiines there are one or two small sclerites in tlir lateral region which are probably rediiced pleura. lmrvis (1925, p. 41) states that the abdominal p h z r a axe meinbrnnous and iisu:iliy without differentiated scleiites. W e are theirfore justified in interpxeting the abdominal elements iu cicada as a strongly-developed over-arclled tergite meeting an entiiely veutrxl steriiite, which beam the spiracle.

Tibic~n chToronzera w k . I\lak : lateral 1 iew ofcntirt. R n d segment. X., tel.t;i :~bdomirialsegnreiit; dm., right mtmbcr of pair of dorsal muscles ;
rm.,right ineniber of Imir of ventml muscles; u., uncus of Ainericair taxonomists.

It is true that there is : i n irregul:ti,ly rect:mgula,t area at the side of each stelmite, market1 illdistinctly by folds ; but the spiracle i s definitely in t h e niaiu body of the sternite (text-figs. 11, 21). A n internal view is seen in text-fig. 2 3 . The cliief endoskeletal structui.es of the nlvdoiiien are the strong pegs fount1 in t h e latero-ventrid portioiiq of the tergites and sbown i i i text-fig;. 23. The structure wliicli gives support t o t h e tyrnbal-muscles in the male, and is regarded by many R S the furc:i. of t h e first or of the second alxlominal segment, we shall see later is not endoske1et:Ll a t all, but merely the modified ventml wall of tlie segment.

B. Ge?Citalia. Whatever. be the nuinher of :r.oclo~iiin:i segments admitted, w e I have seen tliat the l X t h and succeetliiig ones in both sexes are

V0I:PHOLOGY

OF IIIC IICADIDB.

405

gueatlv riiorlirietl eitlier directly o r iiitllrectly for reproductive piu imhe5. The V I I t l i :inti V I I l t l i a l e d s o soniewhat nlteietl. Altliougii. :is n e s1i:ill see later, Xli\totlr tlescribed copulation, I t rein iiiicvl for &.EalpJglii (1687) aud Re<iuiiiur(1740) t o describe i n coricerr~ed. my detail the o i g ~ ~ t i s r , 1lie descriptions wliicli follow :we I i n b e t l on ~lieZunipsrtZtc~ hpp.

i. X a l e Genitalia. L h e ui;ile goiiitnli;~, Cicatlitla: :IS eiiil>ll:isize~l9 7 Kru*lr:~\va.ncl of I Mnir ( I 0 2 2 ) are v ~ i y inch. Tlie VIItli steriiitc? is 1:rrge :ind tii I,otlllcetl k,:I,ckwnrci (t.ext g.35), i t s s1i:ipe being it good tasoilotriic charact,ei,. Whetlier t ha extension represents a development of tlie YITth coxites is not; known. The po rior borclei. is often oinargiiiate i n tlie middle, :lilt1 thus :iffords soine slight evidence of primitive t1u:ility. The Imckwitrd ert,ensioii covers t h e base of tlie V I T l t h sternite (hypnndrium) as a crescentic free flap. Tile V I l I t l i teugite is Ial,ge a n d carved dowii 1i~tero-cetitra.lly
r,

re x t-fig Lire 3 3 .

,Uelampsalta c i n , p l a t a . KO13 preparation of anal segment. i., intersegiiieiital meuibraiie.

so as almost to forni it complete ring. The V I I l t l i sterriite is large i ~ i i ( 1 boat-sliaped i ~ n d constitutes t h e Iiypaiiduiuni (textjig. 35). Tliere is sonic: evideiice--sliglit hilob;i.tiou at tlie tiptlint tile fused coxites enter into t h e composition of t h e Irypandrium. This coriclition is peculiar t o the Cicndidr-e. 111 all other liiiciienorrliyricl~:~ these coxites aie iiicorporatecl into t h e pygoplior (lX,). 111 the trough of t h e liyparitlriuni t h e pygoplioi lies a t rest. Fuiiklionser ( 1917) tigures a siirii1:w plute in &leinbr;acid~, anrl corihitleus i t :ippareiitly t h e steriinin of tlie nirit,li seguieiit. Newell ( I 918) would call i t tlie I X t h sternite in cicada, but there is no evidence for tliis ant1 inucli t o tlie coiitmiy (text-fig. 35). The pygoplior is ;t ~ti~oi:gri~-cliitinize~ or less cylindrical iiiore case suruouiitliiig t h e teriiiiriatioii of t h e g u t and tlie genitalia. r 1 I he wi~ole t h e dorsal and la.teral portions a r e formed froni the of I X t h tergitr. The ventral sui,face is membranous aiitl of unt certiiiii origin. Ulie lateral margin h w s on each side i process, entirely non-articulate, which Muir and Kersliaw homologize

406

DR. J . G . MYERS ON THE

with tlie otherwise entirely lnclring genital styles (text-figs. 31, 35). Singh-Prutbi does iiot agree Tith this hoinology (1925), and me tiunk i t rather doubtful. True genital styles arc present i r i Zettiprvctu (Wuir, i ? ~ Zitt. ; Singti-Pruthi, 1925). 111 repose no portion of the pygoplior extends beyond tlie h y p n t l r i u m terminally, lorit the teq$ spiiie--a dorso-cnnrlal pi ojection of tlie IXtEi tergite-ends tlie c1ors:d line of t h e ahloinen. This teigwl spine. wlricli is the irieclian d o r d spine of Woodworth (1888), varies specifically, a n d is valuable in taxonomy I. T)istnlly t h e pyqoplior bears the anal segment o r a n a l tube, ni:irle a p chietiy of the X t h tergite. The Xtli qegment is produced ventrally into a inore or les3 wide plate, extending i n Text-figure 34.

Tibicen chloromera. Internal view of left half of male pygophor.

or)., left half of basal opening of pygophor; ej., ejaculatory duct; pe.,periandrial thickenings (busal plates of Singh-Fruth) ; ys., Muirs aborted genital i;tyle ; md., redeagus ; t., tendon for left member of pair of powerfnl muscles.

Melanzpsaltrc cinguluta (text-fig. 35) directly cauclad, but in dL nmta ~ i i d relatives directly ventrad. Proxinially to t h i s its again tlie edge of tire anal segment hears two downwardly-directed, stout, curved, non-articulate hooks which I call copulatory c1:rspers (text-6g. 35). These may be cerci-organs generally considered absent from t h e entire order, Hemiptera ; they a r e appendages of the X t h segment, a n d t h u s in cicada t h e most distal, but they have lost all articulation. Cmmptoii (1922, pl. 3 . fig. 9) calls them sargouopods (cerci? ). Newell labels them cerci, hut notes that in no Herniptern are these organs segmented (1918, p. 125). Berlese (1909, fig. 347) applies t h e term cercus

* Tlie term y?Jqophor is preferable to pygofer-a 11) brid often encountered-and still more so t o the plural form of the latter, frequently used. The dipterological term hypopygmm is equivalent.

MORPHOLOGY O F T E E C I C A D I D E .

407

i n tlie female t o t h e epiproct, presently t o he described, and i n the male (fig. 399) to two sinnller lateral pieces at their base. The copulatory hooks protect the wdeagus, which peeps o u t between them, and they thus take t h e place of t h e absent genital styles. In Z'ibicen ant1 related genera the copulatorv hooks a r e often fused i u the middle line t o form the uncus 2 American taxononlists (text-fig. 32, u.).This structure when single, a s in T i b i c e x ch~0?.07?2ei~6, shows a deep o.roove, proximally. in c tudal view, marking t h e line of fusion. bThe base of each copnlatory hook ciirves round and merely meets t h a t of t h e opposite side, forming an ;ipertiire through which the mleagris passes t o the euterioi. This is a much niore specialized Xtli segment t h a n that of ilIdunzpsultcc ; yet tlie zseclengiis is very iiiiich simplerText-figure 35.

...
I

Jrelanzpsnlta cingulata.

ICOII preparation, partially sectioned.

VII.-S., abdominal segments : sp. 10, last spiracle (position only indicated).
(For esplanstioii of other lettering see text-fig. 3 . 4)

possibly by correla.ted reduction. In ilfelanzpsulta cuesda tlie work of t h e u n c i i s is perfomied by a strongly-curved plate which looks very much like one, but is distal to the true copulatory hooks which lie beneath it, and is thus hoinologoue with the directly caudal extension of segment X. already described i n JL cinguluta (text-fig. 3 5 ) . Both this and t h e true copulatory hooks are shown distinctly i u lateral view in text-figs. 33 and 36 of cingzclata and of sericea respectively. The condition i n the latter suggests t h a t t h e second projection may possibly pertain to an X l t h segment. The modification of the X t h segment occurs in no other Hemiptera (Singh-Prutlii). Beyond the anus and t h e X t h segment is a dorsal lamina or epiproct (text-fig. 36, ep.) and a ventral one o r anal style (as.).

403

DR. J. G . MYERS OX THE

The f o r ~ i i e ris t h e tclson of Doering (1922) in the Cercopitize. A t blie h s e of these on each side is LL rouncletl, sometimes shining bl;ic!r p h t e \vlricli Berlese (1909, fig. 399) labels cercus, but mliich is inore coiicitivnbly a podical plate from which tlie cercus has been lost. &lecigrcs.--ln Jfelccnzpscclta this consists of a stout, basal, b i l o h t c imlb, the per%andri?cnz, and t h e e long distal processes surrouiidiug :i,rid more or less p~rallelt o the penis itself. Muir arid Iietsjhaw (1922, p. 205) ca.11 the distrtl portion of t h e tedeagus the p i s i s , but it would seem better to restrict the term adeagus
Test-figure 36.

illelrr~~z~~sulta sericsa. Male : anal segmeiit complete, soft parts stippled. ep., epiproct ; as., anal stj-le ; i, iiiterseginental membrane. .

t,o tire actual sheath. Of the three arms of tlie tedeagus, two are laterd aid the third uiedian ~ i i t lhe;tririg d i s t d g the orifice of the penis. The pei.iniitlriuni arises as a tliickeiiiiig in tlie body-wall between the IXtli a i i d Xth segiiients. These t,hickenin,vs(~~oJiclZo of Yerlese) are cdled busul plates I)y Bingli-Pmtkii (1925, p. 136), a,ud considered of very great inorphological significance. He wou!d homologize tlierrr with certaiii distinct sclerites near the btwes of t h e zdeagus mid the genital styles in such other Horlroptera ~1s Cicatlellide. This Iioiriology would seein to be entirely fancifiil. Singti-Prutlii states that " basnl plates " are

bIORPIlOLOGII O F TIlE C:ICADIDB.

4U9

present in Cicxdina Dist., Gmmiinze Dist., a n d some TibiciniilE, but, :we a.bsent in t h e rest of the latter. I n such forms as l'ibicen t h e t,hickenings of t h e pei~iiinclriui~ir e cri%ainly very strongly a developed (text-fig. 34, p e . ) . b u t i t woultl appear tlia,t no hard-andfast line can I)e drawn. O n this hiisis Sirigh-Prvthi considers t11a.t t h e CicadiLs of t h e snbfnaiily P 1 a . t y p l e n r i n ~ (Gicadinle Dist.), usually believed on other counts t o lie t h e most highly evolved of t h e family, exhibit t h e most primitive conditions. I n spite of its frequent complexity t h e Cicarlid Lcdeagus may be coilsidered relatively primitive. It shows little of t h a t iiira~ginatiori of t h e tlistnl i n t o t h e nroxiriial nortion characteristic of nlanv other Herniptera-e. 9. Fulgoroidea a n d Heteropte1.a. It is, h o x ever, t r u e t h a t Apgm (1887) lln< described quit? a conililicated

Text - fig tire 3 7.

__-

eritloionle in t i l e adeagus of Nc~giciduda so1~tenrlecinz,while in most cicadas the whole a l e a g u b occiipies a soinewhat internal positioii, brought a bout by invagination of t h e > ~ i i t i anreiitbrnne I of tlic pygoplior arid of t h e intersegmental nienihrarie between that and segment X. The chief n-01 keys or1 the genitalia of male Auclienorihj ncha. and incidentally of CicatliriE, have been ~ 2 a I p i g h i (1687), li6nuninr (1740). Dufour (1833), Newel], lIeyi~ions, Kerslmw ant1 1\lTuir (1922), Muii. (nlany p p e r s ) , LRV son, Singli-Pmthi (1924, 1926), Doering (1922). ii. T h e Pevnrde Genitcclici. female genitalia consist essentially of three pairs cf appenclageh, f r q u e n t l y known as valves. 'i'lie5e make up t h e ovipositor
'Clip

410

DR. J. G . JlYERS O N THE

and its sheaths. The central or inner pair is fused for t h e greater part of its length, forming distally a single solid piece (text-fig. 30, nzp.) with sharp atrongly-cbitinizecl apex. Tlie pygnphor is not unlike that of the male iu shape (text-fig. 37, 1X. t.), with a longer and niore tubular anal segment which lacks the armature of that of the male, but is otherwise es~ent~ially similar. The central piece formed by fusion of t h e inner pair acts as guide-piece for the outer o r proximal pair which partially encloses it (text-fig. 30,Z.). The three pieces are welrled into a functionally single weapon by means of a t least t h e e pairs of Text-figures 38-47.

c
38

Veiitrsl

1 iew of ovipositor tip, left half, of qpecies of X e l a v z p a l t a : cingulatu ( 2 specimens), s t ~ e p i t a n s (Kirk.), ci-uentata (F.), Zeptoazera, sericea, mzcta, ,fidiginosu Myers, and ~nzctamrs. s u b a l p i n a (Hnds.) and czctora (Wk.).

interlocking grooves m c l ridges, well sliomn by Marlatt in crosssection (1907, f g 37). Such interlocking, wliile effestually i. preventing 1atera.l movement of individna.1 pieces, allows a.nd indeetl f:i.cilitatee. longitndind sliding of the ventral or outer pair on t h e iniddle piece, the tip of which is sharp and hard. The apices of the ventral or hternl pieces (text-fig. 30, I.) beyond t h e tip of their infolded portion clasping the middle piece are likewise blxclily, heavily chitinized a,nd furnished with a number of oblique cutting-edges. These a r e t h e effective instruments of perforation. Lawson was the first t o figure these in deta,il for a

MORPHOLOGY OF THE C I C A D I D B .

41 1

number of species (1920, pls. 24, 3 6 ) and to claim that they are of considerable value in taxonomy. Nothing is more needed i n cicada systematics to-day tlinn some new character intlepentlent, of male characters, but o u r experiences in il.llampsalta (textfigs. 38-47) do not lead us t o hope for very much fiom these as characters for specific separ ation. While the prosinial portion of the whole ovipositor is protected Iaternlly if riot ventrally by the pygophor, its distal part is ensheathed by the third pair of appendages, which are less strongly chitinized, apically roundetl, and usually haix y (text-figs. 30,37, s.). Attachments of Ovipositor.--The lateral pieces arise from the V I I I t h sternum (text-fig. 37, VIII. s.), which is represented by two triangular sclerrtes largely hidden by the VIIth, which, as shown in text-fig. 11, abuts largely on the base of the ovipositor. The dorsal basal angle (outer b. a. of Muir and Kershaw, 1922, in Text-figure 48.

hmb
IWelainpsaltaiiiuta. Right tegmen. amfi., ambient veil:.

(S.H.-Veuation

for text-figures 48-59. The usual abbreviations of tlie ComstockNeedham notat,ion employed.)

Cercopirlze) of the lateral pieces is joined to t h e posterior basal angle of the I X t h tergite as shown in text-fig. 30. The I X t h sternum, in SO far as it is chitinizecl, consists of alsrge, somewhat elongate, boat-shapetl sclerite or plate situated on each side of the mid-ventral line and homologized with the orthopteroicl valvifer. Proxinmlly this gives rise to the iniddle piece, formed by the fusion of the inner pair of appendages, which are perfectly separate for some dista.nce from their origin. Distally the I X t h sternal sclerites are continued by a imembranous area, shown stippled in text-fig. 30, into the two hairy appendages (s.) forming tlie ovipositor sheat,h. Berlese (190'3: pp. 302,303, fig. 347) would homologize the latter with the prostili of Orthoptera. The vnrious names applied in taxonomic a,nd morphological literature t o tlie t h e e pairs of gonapophyses comprised in the ovipositor arid its sheaths are so confusing t h a t we are impelled t o siibniit a table of thein.

412

DLl. J . G. MYERS O S THE

g.=snw-valves (t,ext-fig.30, Z.), ventral valves (Tillyard, Doering), anterior, outer, o r ventral processes (Muir aiid I<ersliaw).

g.2=niiddle piece (text-fig. 30, mp.), inner valves (Tillyard), niediaii processes (XIuir and Kershaw), dorsnl valves (Uoering).
g.:j=slieatlia (teat-fig. 30, s) dorsal valves (Tillyard), posterior processes ., (Jluir and Kershaw), lateral valves (Doering).

The first writer t o concern himself with the structure of tlir ovipositor w:xs Aristotle, who merely notes : pariimf [CicndEJ in aruis cessaritibus, excavnntes asperitate przeacutn, quain pnrte liabent posteriore. . . . Hist. ccnzm.; Gnza, lib. v. cn,p. 30.
lCZalpiglii (1687) gives a ventral view of t h e ovipositor (ii. p. 38, tab. xi. f . 2) showing some detail, and then proceeds to describe i t as the male genitalia, thus :-

( I n Ctcnclis etiam fceminis, eadern pudendi conforniatio imuiri ventris occnp:Lt : in his penis A binis c0nsts.t assibus, quorum extreniitas, pluribus eminentiis aspera, glnndem B constituit ; custoditnr a u t e m vagina C, q u a e t ipsa in binas aperitur partes non longe ab an0 D.
Yet on an earlier pzge(i. p. 129, fig. 73) he writes: In Cicadis (73) insignis magniti[clinis terebra. extat, qunin olim pro pene, ita cleceptus, delineavi. H u j u s structora mirani NatiirE s:tg;witaterri pa.tefa,cit. Then follows a good a n d detailed description of tlie ovipositor ; but :I. passage o n p. 131 still leaves i t doubtful whether lie really lr ne EJ tl I e in a1e g en ital ia . These latter were considered with admirable detail arid figured with much clearness by Reaumur (1740), whose description of the ovipositor, more rnnsterlg than t h a t of his gi-eibt predecessor. bec:amo the reference :wxount, for subsequent en tonlologists almost to tlie present (lay. Later workers 011 the structure and lioiiiology of tlie female genitalia. of A.ucheno~rh3.nclia general o r of Cica.didse in parin ticulnr are Doyere (1 837a),H y a t t (1896), aiid Heymons, VerhoefY, Kershxw and Muir (1922). iii. JIonzologies of the Genitalia. Korsha.w and Rluir (1!)22) would homologize the geiiital appendages in the two sexes. They firid t h e more general coiirlition i n H.omoptera. to be as follows. (For convenience iu reference we ha.ve coil:i.ted the synonymy) :g.=;niterior goiiapoph! ses =:ippcridages of V I I I . =hypovalvi*, (CYnmpton). =genital plate (Kershaw & Muir). =subgenital plate (Siiigli.Prnt1!i).

MORPHOLOGY O F THE CICADIDB.

413

g.?=median gonapophyses =appendngsn of I X . =adeagnn of male. =iniddle piece of ovipositor.

g.3=posterior goiiapopliyses=Koirocosites of IX. (Tillyard). =genital styles (K. C Muir). 3faIe. =parameres (Singh.Pruthij. =ovipositor sheaths.

I n the cicatlid female all three pairs are well represented : in the rnale g. are apparently I:dcing, though perhaps fused in the hypxndrium, wliile g. are prncticaily aborted save i n Tettigcoctffi. Several writers, notably Doering (1922) and Singh-Prntbi, have questioned this homology on morphological arid embryological grounds, and Muir (1925) has reviewed tlie evidence anew. Text-figure 49.

Chonosia ci-assipennis (Wk.). Tegrnen of aberrant specimen, with abnormal Cu 1 .

Tillyaril, i n his new general work (1926), agrees t h a t a complete homology exists between t h e two sexes. While most writers agree t h a t t h e posterior two pairs of gonapopliyses-tlie adeagns ant1 genital styles of t h e inale arid tlie inner processes arid onter sheaths of tlie female-are honiologous i n t h e two sexes, Keishnw S: Muir a r i d Tillyard would appear t o be alorie i n considering tlie anterior pair SO also. The forrner have brought forward very convincing evidence for their view. As, however, the qnestion can be settled only by renewed embryological work and by comparative studies not only in Heiniptera but in other orders, i t is outside the scope of this paper and need concern i i b no further here.
( 8 ) INTERNAL ANATOMY.

Our alcohol material heing scanty, we have been compelled t o ilse niost of it for the elucidation of the digestive system, and, with reference t o the other organa, t o make what incidental observations we could arid collate previous references.

PROC. ZOOL. S0c.-1928, NO.XXVIII.

28

414

DW. J. G . MYIGRS O X THE

a. NERVOUS SYSTEM. u. Genernl. The chief writers t o concerii themselves with t h e ner:7ous system a n d sense-organs of Cicadidae, either especially or incidentsally,were Meckel (1818) and Dufour (1833), who laid the foiindntion of so marly branches of Hemipterous anatomy, Brandt (1878), Rossi (1879-1880), Binet (1894), Rerlese (1909), Swinton (1877b, 1879- 1880), and Vogel (1922, 1923); Will (1840), Grenacher (1879). Meckel descrihetl in Iibicen pleheiu the brain and 4 TCnoten, the last I a i p r , Dufour (pp. 264-266, separate pagination) described in Cicadu omai L. a cephalic ganglion and two thoracic ganglia. He noticed that the former was produced by a fusion of two hemisplieroid lobes, the fissure sepwating which WRS only snperficial.
Text-figure 50.

R +M

Tettigarcta eriitita. Tegmen. sxillary cord. (N.U.-Text-figs. 48-22 me of tliffererlt magnifications.)

as., axillmy sclerites; ax.,

The anterior part gives h e to an optic nerve on each sitle. This nerve, pyramidal 01- shortly, thickly club-shaped, embi-aces the ocular bulb. Anteriorly ariaes a trifid nerve of wliich the three branches go to the ocelli. The esophageal ring gives rise t o : pair of neives for the muscles moving the hend. The thoracic L ganglia, far from being separate :md distinct as in ,Yepa, are tiearly fused into one, :ts in the Geocorises. With difficulty one traces the light dernarcation of :in interior ganglion. The ensemble of the two gimgli;t forma an oblong body, deeply situated beneath the ni~~scle-niasses which fill the lower wnllof the tliorax. The anterior ganglion gives rise t o four pairs of principal nerves, and t h e posterior to six. The two nerve-cords, towards tlieir origin, are contiguous as if adherent, but in the abdominal cavity they sepnrate before finally dividing. Such in brief is Dufours century-old description, and it is

MORPHOLOGY O F THE CICAUIDE.

415

snrprising bow little from the gross anatomical viewpoint can be added thereto to-day. Text-fig. 70 shows certain rather striking differences in Melanzpsalta seimicea from the condition described by Dufour in Cicada orni. If these be coilfirmed they would go t o show that there is considerable diversity i n the family, and that Melanzpscclta is considerably less specialized than orni in the development of the nervous system. Instead of the t w o thoracic ghnglionic masses distinguished with rlificrilty i n orni, there are i n sericea three, the foremost of which is quite distinct. Dofour makes no melition of a sub-mophageal ganglion, which he probably confused with the brain itself. Remarkably little is known about tlie general nervous system of any Hemiptera, and there is but scanty niaterial for comparison. It would appear, however, that the Cicdicke are Text-figure 51.
CUlL

s?ibicPJL ehloromera.

Wing-conplirig apparatns-part of hind-border of tegnieii and fore-border of hind wing, in dorsal view.

u . 1, downfold of teginen (viewed througli membrane of wing) ; f tuf. 2, upfold of hind wing.

.corisider:thly specialized so far as the nervous system is concerned, although less so in some respects than tlie Cicadellida (Jassoidea), in which, as for instance in Ur~culaaephnla,Cogan (1916, pl. 22. fig. 32) shows the whole central gnnglionic system as practically one elongate mass with a small aperture for the cesophagus. Berlese fin& that i n some Coccids and Heteroptera the subesophageal ganglion is fused with the thoracic mass. Binet (1894) has supplied perhaps the most thorough study of t h e subintestinal nervous system in cicadas (Cicada o m i ) . H e remarks first of all (p. 486) :p:i,rmi tous les Insectes que nous avons 6tudiBs, nous avons trouv4 les cellules nerveuses les plus considerables chez la
$6

28*

416

D R . J. G . MYERS ON TUC

cjgale; on trouve dans l a r4gion abdoniinale des centres rierveux cle cet Insecte deux cellules g4antes niesuiant 150 p snivant leor plus giand diamktre. K n t Binets greatest discovery (pp. 542, 543) concexns t h e abclornirial ganglia, which, though fused w i t h t h e last thoracic mass, a r e nevertheless distinguished in sections by t h e absence of craral lobes, correlatecl with t h e absence of legs i n the corresponding segments. T h e first abdonrinal gqnglion diEers from t h e sncceeding ones in possessing a sv elling of t h e dorsal lobe. This swelling is formed by two very distinct lobes, wbiclr are superadded to t h e dorsal lobe of t h e ganglion. Each of them occupies tlie externo-superior position, and t h e y a r e separated by Text-figure 5 2 .
11

5, c

_.._ - - !

_ _ _ _ _ _7 _
. .. . -

M 3 +.& ;-,p---?<
. . ~

-J-

LWeeln7~7psaZtn ma&.
I$.,

Part of right trgmen, viewed doisally.

node; j., joints or breaks in t h e veins.

a deep indentation where t h e great ganglionic cells accntxuIate. T h e t w o lobes a r e situated i n front of a ganglionic cell of exceptional size, which lies a t t h e level where t h e crural nerves enter t h e central nervow system. A longitudinal section shows tllat each of t h e new lobes is ~ i t ~ u a t e jdu s t behind t h e carresponding nletathoimcic crur:il lohe. The cocnl lobes, as Binet Cnlls them, are intinlately united wit11 t h e dorsal lobe of tile ganglion, which. is essentially motor. They have no ad2ierence t o t h e ventral lobe, froni which t h e y remain distant for 811 their extent. T h u s t h e y differ from t h e crural lobes, which, situated laterally, in relation a t once with tlie sensory (ventral) a n d nlotor

(dorsal) lobe, a r e both sensory a n d motor. T h e vocal lobe 6 pnrAit etre uniyueinerit done d c fonctions inotrices. It will be interesting to study the relations of t h e ailditory nerve, discovered since Binets time, with t h e abdominal ganglion in q u ~ s t i o n . Vogel (1023) describes it as hranching from the abdominal strand, apparently i n t h e I I n d segment. O u r materia] at present fails t o settle this point.

MOILPHOLOGY 06 I'HE CICADIDB.

417

W e have examined the gross anatomy of the nervous system in imsuitably-preserved niaterial of -Ilela.n~palta wuiffi,M.sericea, and Ccirinetc6 for7nosa. The brain is esconced on the second swelling of the tligestive canal, jnst behind the dorsal dilator muscles of this organ. It is consicleiably wider than long (textfig. 70, 6.). The optic nerves are huge and long, and corpora perZic7zculota lacking. Rerlese notices that the protocerebral lobes are small and the others still smaller. The ocellar nerves are as branching from the common stem described by Dufour-their does not occur until just below the ocelli themselves. Rerlese figures the brain of l'ibicen plebeia (text-fig. 697), a section of the procerebral mass i n the snme insect (text-fig. 694), and also a section of its whole brain (Tav. .I-1, ii.). Text-figure 53.

Tettigareta crinita. Riglit liiiid wing.

The esophageal connectives (text-fig. 70, c.) are stout arid rather long-in iVekcntpsa~ta less so than in plebeic6, ns figured by I3erlese. The subcesophageal ganglion is rounded, joined by long, stout, well-separated cords t o the 6rst thoracic ganglionic mass, which lies largely i n the prothorax. This has two short, very stout connectives to the second thoracic mass, which is much longer than broad and shows signs of two-fold origin. It lies wholly within the mesothorax and. in fact, does not reach the niesophragma. Posteriorly it passes into a superficially single cord which splits into two a s i t enters the abdomen. From t h e sbantlpoint of gross anatomy the abdominal ganglia fused in the second mass are not t o be distinguished. The sympathetic nervous system has not been studied in Cicadidz nor related families.

E y e s a?id Ocelli. The compound eyes have been studied i n Cicada ormi L. by Will (1840) and i n Cicada. gyossa. (?) by Grenacher (1879). According t o the former, each eye i n OWL^ has 11,600 ommatidia (p. 11). Grenacher (p. 96) finds the eyes are of the eucone type, i. e., each visual element coiitains a true crystalline cone, in front

F.

4lQ

DR. J G. MYERS ON THE .

of which are t h e nuclei of t h e cone-cells. The retinnla is formed from eight distinct elements, and the crystalline cone5 are very well developed. The compound eyes iu Grenacllers opinion approximate most closely t o the Hymenopterous type and differ sehr betrachlich from t!he Heteiopterous-a fact which need not occasion surprise when one considers how largely Cicadidz appear t o depend upon their eyes. Their habit of dodging like n squirrel behind a branch, keeping it ever between theinselves and the observer, is very amusing, and is shared also by certain Heteroptera, notably mirids. That t h e compound eyes are the cbief organs involved is suggested by the absence of ocelli in &firid%.
(

Text-figure 54.

s c +R

cu-

Nelampsalta mutn.

liight hind wing.

The ocelli are three in number and markedly red in colour. Berlese (1909, p. 672, fig. 840) has studied in some detail those of Tibicem pbheia, while Link in the same year (pp, 354-356, text-fig. 0, Taf. 24. fig. 2 7 ) gives a more thorough-going account in Cicuda concivzna L. (= ? Cicadatro atra ( O h . ). The most striking feature i s t h e presence of pigment-cells between t h e sense-cells-a characteristic of Auchenorrhyncha. Link shows t h a t these red pigment-cells are derived from the corneagen layer. Berlese (1909, p. 672), however, disagrees. The pigment is in t h e form of red granules, filling the cells. Link erroneously considered the . Gattung Cicada as t h e only one in the Humoptera furnished with three ocelli. W e have seen that the ocellar nerves are bound in a common stem t o near t h e ocelli. Link goes further and finds that the branch then proceeding t o the median ocellus is really double, whence he speculates whether t h condition is secondary, or denotes fusion from a n original pair. Hymenoptera and

MORPHOLOGY OF T H E CICADIDB.

41s

Odonata-notably insects with well-developed visual powersshare this arrangement with the cicaclas. Link repeats the well-known liypotheses that the function of the ocelli may be connected with the need for sudden inovements-since they are well developed in niost jumping insects-or with orientation, as suggested by their position. I n an insect like a cicada, with the head and prothorax almost immovably fixed to the rest of the body, the ocelli would certainly seem t o complete t h e visual equipment 60 far as direction i s concerned. The field of the great conipound eyes is largely lateral j the paired ocelli look directly upward, while the median one is often on an eminence which directs i t forward. Whatever be the respective functions of this rather complicated viaual apparatus, n o one who lias collected cicadas will question the ability of the insects to use it very effectively. Text-figure 55.

SC+R

r 4

Tibieen chloromera. Base of right tegmen.

y . Auditory Organs. Tho supposed absence of auditory organs in the CicaclidE was

long one of the chief arguments against any sexual significance in the song. To Swinton (1877 b, 1879, 1880) must be given t h e credit for discovering a n organ to fill this lack. Unfortunately Swinton wrote in such an involved style, euphuistic, highly allusive, and abounding i n long irrelevant discussions, t h a t he effectually disguised t h e solid scientific contribution often contained in his writings. I n reading him one had always t h e impression of vapourising dilettantism ; one met never without surprise every additional citation indicating t h a t he knew thoroughly t h e literature of his subject. One does not expect an epocbmaking discovery among the amiable ramblings of a n undated volume with such a title as ' Insect Variety: its propagation and distribution.'

420

DR. J. G . MYEBS ON THE

It is true t h a t Swinton reported very briefly his investigations also in a scientific periodical (1879) ; hut, for all that, they were completely ignored by all subsequent workers until practically the present day. The masterly and detailed description of cicada auditory organs by Vogel in 1923 involved their discovery anew, and it was not until Vogels work was ell on t h e way to cornpletion t h a t he found Swintons long previous account. The first detailed descriptions of cicada sound-producing organs had niade knowir the mirror -an extremely thin and delicate, beautifully iridescent membrane closing the cavity of the sound-apparatus posteriorly. I n spite of a t least some experiments t o indicate the contrary, almost every observer since and including Reaumur lms assumed that the mirrors act as resonators. This will be found in every modern account of cicada sound-organs. Swintons contribution consisted in his recognition of tlie mirror as the t,ympanum of an auditory organ.
Text-figure 56.

Lem6ejafatiZoqua (St%l). Apical half of right tegmen. amb., ambient vein.

Swinton was purely a n amateur, and could not have elucidated t h e microscopic structure of the chorclontnl organ itself, even had h e seen it. H e described, however, the auditory nerve, t h e thickened line on the tympanum, the spatulate process at its lateral extremity, and tlie external appearance of the auditory capsule. To Vogel alone we owe a complete account of this remarkable organ and its accessory structures. The mirror is seen a t M (text-figs. 13. 17, 18, 19, 20, 27). It is a n extremely fine skin, according t o Vogel, only 5 p. thick in the middle, in two species of Ne2umpsultn studied by him. As to its origin, that is so wrapped up with the development of the sound-organs that we must consider them together when the l ~ t t e are debcribed. For r the present we content ourselves with a brief account of t h e sound-organ a s it occurs. I n the lateral part of each tympanum is ,z dark thickened line (text-fig. 19,p.) which leads csntiewards but never reaches the middle. From the broadened lateral end of tlie thickening arises a spatulate chitinous process, running

DlORPIlOLOGY O F TIIE C I C A U I D B .

483

obliquely into the auditory capsule, where i t serves as an attachment surface for the sense-organ. r, l h e auditory capsule is a more or less hemispherical swelling (a.) on the uentro-lateral portion of the I I n d abdomiiial tergite (paratergite, Vogel). The lumen contains I-iEmolymph and blood corpuncles, the last often accumulated in the angles. Where the capsule joins t h e body-wall, there are outgrowths forming surrounding chitinous rings and making the capsule larger and more nearly spherical (text-figs. 2 5 , 26). Only on the middle of the inner surface of the capsule there remains a larger oval opening, wliicli is covered by thin membrane (according t o Vogel, the wall of the tracheal sac, as he names our rneseriteric sac). Vogel calls this opening the feizestra ovaZis, from analogy with Text-figure 57.

Iyena wmclagascariensis (Dist.).

Ilasal part of left tegmpn.

am., ampliate precostal area.

the vertebrate ear. On the ventral side the ca.psule coniinunicates with the abdominal cavity by a slit overlaid by the mesenteric sac. This slit leads into one of the grooves formed by the sternum of the I I n d abdominal segment (text-fig. 26). Tlie actual sense-organ itself is st,retched in the cavity of t h e auditory capsule, between two spring-like chitinous pieces. One is the spatuhte process from the typanuro, while the other is a n invagination of the outer wall of the capsule, where it a.ppears as a slit in external view (text-figs. 16, 17, 24). The spatulate precess lies in the same plane as the tympanum, from which it arises. l h e sense-organ sliows all the structure of a chordotonal organ in so far as its sense-cells terminate in the usual rod-like bodies or scolopales, and are stretclied by m e m s of fibril1a.r differentiated hypoderm cells between two points of the cut,icle. It is distinguished from all hitherto-described chordotonal organs by the enormous number of sense-cells or scolopophores. Schwabe
- 3

422

DX. J. G. MYERS ON THE

found upwards of a 100 in the tympannl sense-organs of Locustids and Gryllids ; t h e cicada, Me2nnipsaZtu coriuriu, studied by Vogd has about 1500. But t h e most important distinction shown in cicadas is that the proximal fastening point is differentiated into rz tympanum whose swinging is registered by the sense apparatus. Usually it is the distal one which is thus modified. Vogel speculates as to whether the scolopophores, in their tremendous numbers, are capable of sound-analysis in a manner analogous t o that of the fibres of Corti. I n i%feZun~psnltu coriaria they vary in length from .2-.31 mm., AS compared with *04-.5 mm. i n Man (Helmholtz). H e decides that there would probably be Text-figure 58.

Me?aelampsalta mzcta. Portion of right teamen t o show chief elemelits of nodal line-

only x small range. If it, were demonstrated i t would go farzto explain the great effect of certain sounds on cicadas and their indifference to others, as in Fabres classical experiment with the festal cannon, which the cicadas ignored entirely. \ F o r t h e histology of the sense-organ itself we must refer to Vogels very able paper (1933). There reiliain several accessory strnctures to mention. The first is n simple dorso-ventral muscle which acts as a temor tympnlzi. The second is Vogels great tracheal sac, which he describes as abutting on the inner side of the sense-orgnn and of the tympanum and acting as a pressure-equaliser, maintaining air -pressure

MORPHOLOGY O F THE CI CAD ID B.

423

equilibrium on both sides of the tympanum with which he claims that its wall is fused. This tracheal sac is .pparently t h e inesenteric sac which we hope t o deal with in n later section. Tt, has been the subject of endless controversy, so c contending that i t is part of the digestive tract nnd others th3t it is part of the tracheal system. W e believe that we have ctnfirmed the work of those who regard i t as mesenteric in origin, nd we have certainly found no trace of a connection with the tracheal Text-figure 59.

Chonosia crassipennis.

Part of notuni and base of tegmeii of male to sliow the stridulatory orpan.

sa., stridulating area; p., proiiotum; m., iiiesonotum ; e., costa; cs., costal sclerite ; pt., posterior tuberosity of tegmeu ; ax., axillary cord of tegmeii.

system, nor do we believe t h a t its wall fuses with the auditory tympanum or mirror. The whole evidence is discussed in t h e section on the alimentary system. W i t h regard to the auditory nerve, Swinton (1S79, p. 81) traced it from a thoracic ganglionic niasa to the abdomen and round the tymbal muscle, after which his description becomes obscure. Tlie acoustic nerve is said then to form a c ganglion t h a t enters a groove.

424

DR. J. G . XYERS ON T H E

Accorcling to Vogel t h e auditory nerve arises in the I I n d abdorninal segment from the two great ventral nerve-strands, and rises, running parallel with the body-wall, i n a chitinous groove dorsally t o the sense-organ, where its fibres r u n one into the base of each sense-cell. In my poorly-preserved material I found a distinct nerve emerging on each side of the last thoracic gnnglionic. mass near the end, and running yarallel t o the abdominal cord. I followed i t nearly t o ft chitinous ridge leecling up to t h e auditory capsule, and suspect t h a t it is t h e Text-figure 60.

a
iarinetafor~izosa. Ventral view of dimentnry system, entirely scbematic.
a?.,msophagus ; sp., suspensory ligameiit of meseliteric snc; fc., filter-chamber (stonlac111; ni., ascending intestine (mid) ; di., descendillg intestine (bind) j m., niesenteric sac ; rs., musciilar rectal aac ; a ,anus. .

auditory nerve. i t seems very iniprobnble t h t it should arise from the abdoniina.1 strands, as Vogel seems t o state. I the primitive ilettigarcta, which lacks all trace of soundn o r p u s , I have been unable t o find, in the two females available, any signs of an audit,ory ca.psule o r of a tympanum. But the evidenue is not conclusive, since these specimens were rather badly attacked by ants before they came into my possession, and were especially damaged on the basal abdominal segments, which .are, moreover, thiclrly covered with long hair. If, however, suck

M O R P H O L O U Y OF TIIE CICADIDAS.

425

a n organ is present, it is decidedly extremely reduced. I have since been able t o examine a i d e of the sdnie species (T. cyiraita Dist,.), thanks to the great lcincliiess of Mrb. F. Muir, and have found a very slight swelling on the rentro-lateral angle of the IInd tergite, b u t no external evidence that this is a n auditory CxpS~k.

2. Other Sere/tse-O~*qans. The antennze bear a number of sense organs or, rather, sensillq usually considered (Redese, 1909) :is olfactory, although 1 nm not aware of any experiinerital evidence 111 snpport of this view. Hansen (1890) \\a\ the first t o describe these. The secontl (and of course also the first) segment of t h e antenrial peduncle lacks these organs entirely j but o n t h o untlerside of
Text-figure 61.

Carineta furmosa. Male : dorsal view of anterior portion of meseiiteric sac atid related parts.
m p . , Malpighian tubules.

(Other lettering

as

in test-tig. 60.)

tlie first and second segments of the fl:igelluirr there is a large nnniber (text-fig. 69). Each is a pit wit11 x blurit conical spike at tlie bottom. Hnnsen distinguished two differeilt kinds-some fairly large, wide, and sli:tl:ow, arid others snialler and deep, the spikes hardly projecting f r o m the orifice. A few of the ~ m e l l e r kind recur on the succeeding segments of the fhgellnm. i7Idanzpsulta sericea and Magicicacla sepLeudecim v e r e exnminecl by 116. There the sensilk are very n ~ ~ i r i e r o u ~ o n tlie nnder ; all surface, most on the first flagellar segnient, but niany also on the succeeding ones. Less tliference 1)etween t h e two kinds tliari mentioned by Hansen was seen, and tlie type wit11 projecting spike was found on other segments than the first. especially in ill. seyicea.

426

DR. J. G . MYERS ON THE

iii

Uerlese (1909) discusses and figures the antenna1 sensills

Tibicen plebeia (p. 620, fig. 745).

The tip of the rostrum in botli nymphs and adults is furnished with nunierous liairs wliich conceivably subserve a sensory function. A t lenst such an Iiypothesis i s more probable than the one held as recently as 1839 by Natlianiel Potter, who believed tliat moisture was absorbed through them or, rather, exhalations of vegetable barks.

6.

MUSCULAR SYSTEM.

A very detailed account of the musculature of Tibicerz plebeia is given by Berlese (1909, Cap. 8, figs. 452, 464, 466, 481, 496498) and compn.red with that of other types. The complexity of the trunk muscles in general is not dissiniilar to that attained in Lepidoptern. The muscles of the mouth-parts have been referred t o in t h e account of these organs, RS also those of the auditory and sountlprocjucing apparatus in corresponding places. For the general trunk muscles we refer to Berlese, a s cited above. The recta.1 sac is strorigly mt~scula.r, expelling the waste liquid lvitli colisiderable force a s it thin jet. File tyrnb:rl iriuscles form the largest and strongest pair of muscles in the cicada body.
c. RESPIRATORY SYSTEM.
01.

Spiracles.

There has been much confusion concerning the number, position, arid terminology of t h e spiracles, especially in CicaclidE, but also in Hemiptera. i r i general. Yufour (1883, p. 258) recognized in Cicada o v ~ t isix pairs situated v e n t d l y at tire intel.nd side of a, deep longitudinal fold. fze described these ostioles as lacking peritrernes and as placed or1 sKla11, ronirdetl, whitish, somewhat salient spots. p;&nrd * a t t r i h t e s t o Hemiptern arid Orthoptera two pairs of thoracic spiracles present on t h e t w o anterior segments. 80 far :ts we know, no oiie receiitly has suggested t1ia.t the first, spiyacles belong to the protllol-ax. Lalidois (1867, 1874) considered spiracle ( 3 ) to be a Schrillst,ignia,, the efficient instruiment in the song of the cicada The t,rne souncl-orgnris he considered functionless. Schiodte, in 1870, iiidicat,ed ten pairs of spiracles in all Herniptern-three thoracic and seven abdominal, the former lying or1 the hinda: etlges of their corresponding segments. Ha.ndlirsch, in 1899, from a. study of the relative size of the spiracles and of the branching of the trachea decided that,
1 On the Distribution and Primitive Number of Spiracles in Insects, Amer. Nat. viii. pp. 531-63L (1874).

YOXPYOLOOY OF THE CICADIDAL

427

the third spiracle in Hemipterawns really the first abdominal. H e found it in some cases on the hind border of the metathorax, in others on the intersegmental membrane, and io others on t h e first abdominal segment. The first abdominal segment he found was greatly reduced, and had been overlooked by t n xonomists, who had practically universally callcd the second the first. Heymons, i n 1899, observed that in the nymphs of mnny species of Heteroptera t h e first spiracle is on the meso-, the second on the metathorax, and the third on the first n btlominal segment, but in the colirse of development there is R movement Text-figure 62.

Carinetaformosa.

Female : latero-dorsal view ; the stomach sliglltly displaced.

.e., entrance of Malpigliiau tubules into wall of stomach atjunction with mesenteric SRC j tr., position of tracheal knot viewed from within. (Other lettering as in text-fig. 61.)

forward, so that the first lies between the pro- and inesothora.x, and the third on the 1netathora.x. The first segment in the abdomen to show a stigma distinctly on its surface is the true second. Hansen (1902, pp. 214-216, transl.) p v e the first cleta.iled description of cicada spiracles, thoiigh Handlirsch had studied the nymphal e m v i e of one form. Hansens results may be summarized as follows :1. The first spiracle lies between pro- and met;othorax in the articular membmne. It is almost perpendicular. 2. Occupies a similar position between meso- and nietathorax.

428

DR. J. G . MYERS OW TWC

3. The first piir of abdominal spiracles lies on the ventral side of the body close to the lateral margins iii a de1)ression at the base of tlie :ibdorrien, surroumded by 3olitl chitin \a p r t of t h e nietasterri~ini), which, particu1:dy in the rnale, is very tliick arid of considerable breaclth. They are trarisvelse like ti10 thoiacic spirxles, b u t somewhat diorter tliaii these. . . 4. The second abdominal spiracle is on tlie underside of the body, facing forwnrdb mtl towards the insects mitldle plane. 5-10. .. E:rtch of the third t o eighth wbtloniinal spiracles lies i n the stercite itself, a little helrind its frout margin. . . Spiracles 4-10 inclusive are considerably smxller thail t h e others, and of different structure ; they are entirely open, with an oval o r almost circular orifice, tlie peritrenie being a solid ling, wliicli is also furnished with a great number of hairs directed towmls the centre of the spiracle (Hnnsen, 1. c . p. 216). The condition in Afelctnapsulta nzuta is :I,\ folloivs :We number the spiracies consecutively in one series, owing t o the dispute as to whether number 3 is thoracic or abdomirial. 1 . As described by Nansen. The trunk enteririg from it is very laige, but the spiracle itself difficult to see. It is very slitlike. 2 . (Text-@. 9, 22, sp. 2.) This is more roundetl t h i n (a) and lies i i i tlie intersegmental membrane between mesotliorax and nietathoiax, just below base of tegiiien. 3 . (Text-figs 16, sp. 3, 17, 18. 22.) This is just beneath outside m g l e of the operculuni. 4. (Text-figs. 19, 20, 22, sp. 4.) This lies in close association v i t h the auditory capsnle, and is seen only in facial view in t h e u n t i eated specinien. Hansens deecription is good, except tli;tt tlie spiincle is really on segment 1. (text-figs. 19-21 ; we latei). 5-9. As tlescrilrierl by Hanseu. These are shown in text-fig. 11. 10. This is difficult to find. It lies in t h e intersagmental nienibixrie in the angle between VIIItfi tergite slid VIIItli sternite (text-figs. 13, 35). I n ZhopJw saccacta the 5th spiracles arc covered with a white pruinose nisteriiil, u hicli, with their i,xised rims, renders theiii highly conspicuous. W e are now concerned with the question as to whether t h e 3rd spiracle is thoracic or abdominal, as w e have seen it lias mucli iiioi e in coninion with the preceding (thoracic) ones tlian with the succeeding (abclominal). Yet if we accept tlie evidence of Hanclliiscli and of Heymons, confirmed and accepted by Dogs in iVepa (1 908), by Hegeniarin in C0ri.m (1910), by Hoppe in Xotonecta (1911), and by Wefelsclieid (1912) in Plea, i t would appear tliab i t is abdominal in . . or~igin. Moreopeel., the latter origin is almost a logical necessity unless we are prepared t o admit a prothoracic spiracle. I-Iansen, who was the first to point out t h a t spiracle 3 in Cicadidrc differed

MOBPHOLOGY O F PHIL CICADIDB.

439

from the thoracic type, and who also expressly stated that i t is surrounded by solid chitin of the inetasternum, yet continued t o call it first ccbdonainccl. Crampton * subscribes t o t h e theory of a forward migration of spiracles as described by Heymons. Snodgmss (1921, p. 403) refers t o the spiracles lying just before t h e tymbals in a male cicada as those of the first abdominal segment. These are our sp. 3 (text-fig. 17). Kershnw and Muir (1922, pp. 202, 206) refer t o t h e last spiracle a s t h e eighth abdominal. Vogel(1923) accepts Heymonss results and considers spiracle 3 as now definitely placed on the metathorax. H e was the first to
, I

lext-figure 63.

Carineta f w m o s a . hcinale : ventro-lateral v ~ e w show tracheal knot on to surface of mesenteric sac. (Lettering as in text-figs. 61 and 62.)

indicate that spiracle 4, in front of the auditory capsule, is topographically now attached t o abdominal segment I., as shown in our text-figs. 19, 20, and 21, but by no means obvious unless a KOH preparation is made. Finally, Mammen (1912) agrees t h a t spiracle 3 is t o be considered first abdominal. To avoid confusion i t is best t o number the spiracles consecutively in one series. Then and then only shall we know whether 6 metathoracic spiracle is meant to imply 2 or 3. The most detailed description of t h e spiracles themselves has been supplied by Mammen (1912). H e studied the nymphal exuviz of Cicada gigantea (= 12 Pomponicc g . Dist.) and the adult of a Plutypleura sp. from giautschou (probably P.kaempferi (Fabr.)).
(

* Ann. Ent. SOC.Amer. xi. p. 356 (1918).

PROC. ZOOL. S0~.-1928, NO.XXIX.

20

430

DR. J . G . MYERS ON TEE

I n the former he found the thoracic spiracles pleural, and tlie abdominal he considered t o lie on the border of sternite and parasternite, in Complicated folds. I n essentials the closing apptratns is similar t o t h a t in adult Platyplewa. In the latter the first spiracle showed very interesting specializations. The niesothorax is anteriorly produced far forusrcl into the prothorax, so t h a t the projectir~ghind margin of the latter covers a more or less enclosed spwe (Luftraum) in which lies t h e spiracle in question. There is a n intricate closing airangenient with a special muscle, and, in addition, a complicated system of hairs and bristles t o prevent entrance of foreign bodies. This spiracle belongs to the type known as (Visierstigmen, i. e., t h e spiracle has come t o lie in an insinking of the intersegmental membrane. and this depression has taken up an oblique position so that t h e stigma does not lie directly beneath the opening of the pit. Text-figure 64.

Carineta,forinosa. Tracheal knot of surface of mesenteric sac immediately beneath left third spiracle. External view.

The second stigma lies between meso- and inetathorax and belongs t o Mamniens L Deckstigma type, in which the foreborder of tlie spiracle forms a kind of lid which fits over the opening right to t h e hind edge, t o which i t may be tightly appresed by the contraction of a special closing muscle. Since this spiracle lies perpendicularly beneath the gap between two segments, the risk of entrance by foreign matter is great, but is obviated by the presence of numerous hairs and a chitinous plate which projects over the stigmatic openiiig fiom the fore edge of the met at 1.1orax. The abdominal stigmata in this Plcctypleum lie laterally on the fore-border of each sternite, so that each is slightly overlaid by the posterior edge of t h e preceding segment. The first abdominal spiracle is situated under the hind edge of the metathorax in r2 complicated fold. I the abdoniinal segments, in n general, tlie integument sinks in to form a Eollom, at the bottom of which lies the stigma1 opening. The hind wall of the liollow

MORPHOLOGY O F THE CICADIDrE.

431

evaginates meclially as a. short, plump cone for t h e att,achment of t h e closing muscles. Further details should be studied i n Mammens elaborate paper with the aid of his excellent figures.

8. Trachea
The general tracheal system has received no comprehensive treatment since the work of Dufour (1833), who distinguished in Cicada orni two kinds of trachea-tubular and utricular. The first, lie said, were distributed especially t,o the abdoniinal viscera and were very fine. Those ramifying among the digestive organs were enveloped i n an adipose sheath ; those in association with the genital system brilliant and shining. The thoracic cavity was said to have two moderately large trunks distributing branches as far as the salivary glands and the rest of the head. Text-figure 65.

Carinetaformosa. Male : dorsal view of enlargement of esophagus, posterior to true sucking-pump, showing five pairs of dilator muscles a i d their position in relation to supra-cesophageal gaiiglion (.. s)

Dnfonr describes the iitricular trachea as, i n general, small bull=, some globular, others of diverse shapes, more or less grouped, especially on the lining membrane of the abdomen and on the snrface of the thoracic muscle masses. Independent of these sinall bnllz there is a large conical titride on each side of the mesotholacic cavity. Dufour believed that, in genei~al, very active insects show the greatest development of utricular trachez. H e was therefore surprised to find so few in Cicadidre, with their well-developed iriotor apparatus, but explains the deficiency by ascribing to these insects des habitudes shdentaires. A thorouglrgoing study of the whole tracheal system is much needetl, i n ~ i e wof the controversy as to whether the large abdominal sac is tracheal or mesenteric. We h a \ e referred t o 29*

432

DR. J. G . MYERS O N THE

this before and shall not discuss it l l o ~ save to mention t h a t we , deal with it at length with the digestive system of which we think i t a part, and to state that, in the examples examined we were able to establish no other connection with the tracheal system than that afforded by a. stout, short trachea from the third spiracle, running a t once t o the immediately adjacent snrface of the sac and there, without penetrating, forming a tiacirieal knot (text-figs. 63, 64) which breaks u p into several bmnclies raiiiifyirig over t h e wall of the sac in question. This is Text-figure 66.

carineta fornzosa. Entirely schematic longitudiiial section of posterior part of pharynx and anterior portion of esophagus. mss., dilators OF second swelling of food-canal; msp., dilators of pharyngeal sucking-pump ; fp., frontal plate of tentorium ; CB., cesoph;bgns.

the stigma, nevertheless, with which Snodgrass (1921 b ) and Vogel (1923, p. 207) believe the sac to obtain its only communication with the outside. Vogel (1. c . ) claims that the usual communicating longitudinal and transverse trachea3 are absent in cicadas, at least in the third to eighth abdominal segments. Nevertheless, in Fidicina (textfig. 26) we are able to find a wide transverse trunk running in the tergal arch between the fifth spiracles (third abdominal), and betweeii the inenibers of the same pzir a finer ventral trachea,

II1ORPHOLOGY O F THE C I C A U I U B .

433

From the immediately preceding spiracle :I large trachea likewise arises, but branching soon, it fails t o form a transverse trunk (text-figs. 25, 26). The tracheze of the seventh pair of abdominal spiracles are said t o be devoted wholly to the service of the niycetornes, the elements of which are bound together into the seniblance of an organ by their smaller branchlets. That all the spiracles, apparently without exception, are, however, linked up by a longitudinal trunk on each side is obvious from a n exaniination of the nymphal extivia The connection between spiracles 1 and 2 is a particularly stout trachea. To sun1 up, we know t h a t there are transverse trachea: between the spiracles o f certain pairs, and longitudinal ones connecting the spiracles of each side. There is also a conical air-bladder on each side of the mesothoracic cavity. Further elucidation of the tracheal system must await the acquisition of fresh material for dissection.
d.
ALIMENTARY SYSTEitI.

Bistorical Review. The widespread classical belief t h a t cicadas lived only on dew n a s based probablyon the absence of any striking signs of injury t o the plants frequented by these insects, and partly n o doubt on the lack of anything wliicli the unskilled obser ver could recognke as a n effective mouth. The first name associated with an actual dissection of the cicada digestive system is one which every schoolboy knows. Meckel, in 1808, gave a surprisingly detailed description and figcire of conditions in Fiihicenplebeaa. H e stated t h a t the whole canal was a t least ten times t h e length of the body. H e distinguished esophagus, stoinach, einen aus diesem entspringenden uric1 in ihn znruckkehienden langen Kanal, a long thin and a short tliiclc intestine. The mesenteric sac is shown i n its correct relationship, biit the true intestine runs o ut of t h e esophagusprobably he ruptured the thinner part of the stomach. The salivary glands were explainer1 thus :- Es sind offenbar Organe, die rine Fenchtigkeit absondern, welche cler Cigale das Anbshren des I I o l ~ e s erleichtert. (p. 3.) The next worker was Ranidohr (1809, l S l l ) , who largely followed Meckel so far a s cicada anatomy was concerned, but appears to have discovered the suspensoiy ligament of the anterior projection of the mesenteric sac. This, however, seemed t o liini SO noma ma lo us that lie regarded i t conime un tissu accidental e t contre nature. (Dnfour, 1825, p. 161.) Marcd de Seiies (1813) atlded little save the statement that the NIaIplghian titbiiles entered an enlarged part of the cesopliagns. W i t h his great countryman Dufour, however, i t was far otherwise. This observer lair1 the folindation for a knowledge of t h e internal anatomy not only in Heiniptera, but also in several other
a.

434

DR. J. G . MYERS ON THE

insect orders. I n 1825 he described the main parts of t h e digestive system with great clarity, and clainied t h a t les Cicadnires sont jusqu& ce jour les seuls insectes ou ce ligament subpensenr [of the niesenteric sac] existe. (p. 161.) H e explained this latter ss due largely t o the habitual vertical position of these insects. I n 1833 he published an amended account of the digestive system in Cicucln orni. Perhaps the one point on which he was inistaken lay in supposing t h a t the (intestiniform t u b e in ascending opened into the cavity of the stoniach rather than coursed merely in its walls. The fact of the latter condition was established by Doyere in 1839 in a paper criticising Dufours work and claiming, further, t h a t the latter was incorrect in describing four Malpighian tubules since there were only two. To this Dufour (1839) replied, accepting very gracefully the correction concerning the filter-chamber, but, re-assertirg h i s belief that there are four Malpighian tubules, i n which cf cowee he is abundantly confirmed. Text-figure 67.

Melampsaltp leptomera. Salivayy pump. a., lateral view of wliole organ from left; b., ventral view of entrance of sa1ivar.y duct.

Lubbock (1859) described the internal gland of Coccids. Later workers on cicaclfin splanchnology have been Schindler (1878), Nassonow (1899), Quaintance (19021, Gxdd (1902, 1910), Ha.rgitt (1903, 1923), Lirent (1911 a, 1911 6, 1912), Berlese (1900), Kershaw (1913,1914.), Hickernell (1920, 1923)*Snodgrase (1921 6). Gxdds paper (1910) is of great value i n t h a t it gives conipa.rative data on eight species in the genera ITibicem7 &ado, Ck&itm, :ind Me~cmpscdtc6. Burnett (1851) clfi.imed t h a t in Jfagicicccdn septendecim thcre was n o trace of the digestive syst,em in t h e m d e . La,ter American entomologists, even as recent as J. B. Smith, believed that t h e gut, was more or less atrophied in this species. Finally, t,llere are those who, like Cnrus and Graber, interested priniarily in the sound-producing apparatus, took the mesenteric sac for an air-bladder of the tracheal system. Appa.rently the only one who has hdcl t o this belief R.fter dissecting the d i m e n h r y tract is Snodgrass (1921 6). All t h e other workers cited above have given an interpretation, wit,h which, i n essentials, t h e account set out i n the following p g e s will be found to agree.

MORPHOLOGY O F THE CICADIDB.

435

p. Gewerul View. It will conduce to tlie clarity of the succeeding account if me prefix it with a sketch of the salient points of the whole digestive tract (text-fig. 60). FunctionalIy the first pnrt of the alimentary canal consists in cicadas, as in all other Hemiptera, of an extremely fine passage between the closely-apposed mouth-set%, which are themselves enclosed in the trough-like rostrum. Soon after entering the head the food passage widens into a capacious and powerful sucking-pump, a dilated portion of the pharynx, which is succeeded by a smaller but muscular swelling j u s t below the entrance t o the thorax. A t the posterior extremity of the thorax t h e esophagus enters t h e pecnliar complex known first by Lubbock R S the internal gland (in Coccirls), but now more generally termed
Text-figure 68.

Mggicicada septendecim.

Stj lets iiiserted in branch (from Quaintaace, 1902).

the filter-chamber. The essence of this arrangement consists in the zig-zagging of the nxending part of tlie mid-intestine in close association with the attached portions of the Malpighian tubules in the actual walls of the anterior part of the stomach. By this construction it is believed that the watery constituents of the plant sap (Licent), o r these and the surplus sugars (Berlese), I)"\" directly by dialysis through the lining of the stomach and t l 1 P malls of the intestine, and thus are carried direct by the rectum t o the exterior j while the more nutritious elements are \elected t o pass by the more circuitous digestive route. The f o r e p ' t of the intestine thus forms o long and very intricately coiled loop which appears finally to end in t h e stomach, as indeed l h f o u r oiigiiinlly believed it t o do. The position, already complicated by the serpentine twistings of the intestines a n d

436

UR. J. G . MYERS ON THE

Ma!pighian tubules, is further obscured in cicada by the extrclordinary development of a huge, thin- walled, mesenteric sac, occupying most of t h e abdominal cavity in the male and considered by Graber, Snodgrass, n.nd many others as a n air-sac of the tracheal system. It is, however, a part of t h e stoimch in direct cominuuication with the cavity of the interior part of t h e latter (not merely of the filter-chamber), a,nd gives rise at its own posterior extremit,y t o the ascending mid-intestine, which, after many convolutions, enters the main stomach-wall near the junction of this main stomach with the mesenteric sac. After a. tortuous course within the wa,lls this tube receives the Malpighian tubules which have acconipanied i t within t h e investmerit of the stomach, then joins the posterior intestine at the anterior end of the stomach, which it leaves n e w tlie entrance of t h e esopha,gus, a.nd proceeds nfter much winding to t h e musculnr rectn.1 sac which opens to the exterior on segment X.
r t

1ext,-figure 69.

_ _ -= Q

_--_

--

--\\
'\

:::,q) a");
-__--_
x..__-r

.--_

._ _ _ --

Jfa.qi&ada septeiadecinz. Sensillre of first nutairrid flagellar segment

The salivary glr7.11dsnppear R S bunches of lobules filling the greater part of t h e head-cavity. Their deferent cannls unite to form a single duct carrying the salivary secretion t o the salivary pump or syringe, wliicli forces i t down tlie sinaIIer of tbe two maxillary channels into the pierced plant-tissue. The juice from the latter is sucked rip the 1;Lrger 01' food-c:mal formed by the close apposition of tlie maxillnry settr.
y. Horpholoyg. i. Xozcth and its Appendages. The food-passage begins with the mouth, but i t is largely all acaclemic question where tlie moiith lies. Rugnion and Popoff (191 1) consider it t o be at the tip of t h e rostium: xnd functionally the entrance of the tube between t h e apposed maxillary set= must be t h e mouth, even when the stylets are sunk in tissue fatbeyond the tip of the labium (test-fig. 68). Embryologically the opening of the storiiotlmni corresporids, liowevei., to the egress pore of t h e stylets from the head-cspsule before they enter the labial trough a,t all. AS shown by Snodgrass, this is the functional

MORPHOLOGY O F THE CICAUIDB.

437

relic of a very wide orifice whiclr extends between the apposed edges of the epipliarynx and hypopharynx. Into this niouthpore opens a short,, narrow tnbe, found by Meek t o be strongly chitinized, which is the beginning of the pharynx ; and to this same spot the efferent duct of the salivary pump, opening actually at tlie tip of the hypopharyiix, delivers the salivary juices. ii. Pharynx. Foilowing the narrow, chitinized tube is a n extremely wide chamber lying in the boat-shaped chitinous trough formed largely Text-figure 70.

MeZampsaZta s e i - i c e a .
oc., coininon

Male : dorsal view of chief gmglia of central nervous system; drawn in one plaue.

stmi o f ocellar nerves ; op., optic nerve; 6., supra-esophageal ganglion; c., tcsopl~ageal connective; sy., sub-esophageal g m g l i o i i ; tg., gaiiglionic masses of thorax.

by the frontal plate of the tentoriuin. This is the suckiiig-pump. A t rest tlie dorsal wall, wliicii is strongly invaginnted, touches t h e floor. which is fused both in old nymphs and in adults to the chitinons sribstaiice of the frontal plate, though Muir says it may be separated in a nymph just after ecdysis, before the chitin has ha2 dened. Thus tlie whole of this plmiyngenl chamber, despite the commonly-accepted opinion, is niembraiious except for a

438

DR. J. G. MYERS ON THE

chitinous median longitudinal portion of the deeply invaginated dorsal wall. Here are inserted the very numerous and feathery muscles which are attached t o the inner surface of the swollen frons, whose striations correspond with the attachments of these muscles (text-fig, 66). I n t h e nymph of Neelffimpsalta&g&ta, and probably in others, the front is much more swollen and smoother than in t h e adult, hut t h e striations are marked with liries of bristles which are used as a cleming apparatus for t h e Text-figure 71.
- . .. .. . . . . . . -.

Carineta formas.%. Male: dorsal view of dorsal vessel.

I.-IX., abdominal segments; ao., aorta.

fossorial fore-limbs. Snodgrass considers this sucking-pump a part of the mouth-cavity and not of t h e trne pharynx (1927). H e is led to this riew chiefly by his belief that a true pharynx is always a muscle-covered organ, whereas the dilation under discussion is membranous and chitinous, its only muscles being the dilators of the roof. He finds t h a t t h e dilators of the mouthcavity are mostly attached t o t h e clypeus in other insects. This mould make t h e striated facial place, called by us the from, really

MORPHOLOGY O F THE CICADIDB.

439

tho clypeus. Meek describes the pharynx o r sucking-pump as formed by two troughs, one lying within the other-a very a p t illustration. The sucking-pump leads backward into a narrow tube, which, just above the tentorium, dilates once more t o form a strongly muscular sac, shown by Snodgrass (1921 b, fig. 6), but apparently Text-figure 72.

Curineta fomzosa. Male reproductive organs, semi-dorsal vien-.

t., right testis; wd.,riglit w s deferelis; ws., left vesicnla seminalis; ej., ejaculatory
duct ; ced., zdeagus; ac., right accessory gland.

missed by Meek. I t was then considered by Snodgrass as a second bulb, but he now (1927) believes it to be the true pharynx. It is provided with numerous dilators of which the distribution is somewhat obscure (text-figs. 65, 66). Thus Snodgrass in ilfugicicadn septendecina found the first dorsal dilator muscie to go to the neighbourlioorl of the mer1ia.n ocellus, the second t o t h e dorsal arms of the tentorium, the third t o the lateral parts of the

440

DR. J.

a.

MYERS ON T'UE

vertex, and the fourth to the occipital margin. In addition he saw two pairs of h t e r a l iniiscles proceeding to t h e posterior arm of t h e tentorium. The brain mas found to lie on the surface of the organ before the first dilators. This was taken as evidence that severd anterior dilator rnuscles of biting-insects are here suppressed, since these ordinarily arise from between the bands of cii-cular muscles, t h e absence of wEiich in the sucking-pump ma,kes Snodgrass deny its identity with the pharynx, extending soiiie distance in front of the ganglion. B u t in the Paraguayan Cicri7zetrcformosa (text-fig. 65) studied by us, while t h e diatribution of the dilator muscles on this second enlargement agreed in the main with Xnodgrass's notes on septeizdecinz, yet the brain Text-figure 7 3 .

ow,,

Curineta forntosa. F m a i e reproductire orpins, ventral t-iew. right ovary; o., left oviduct ; ac., left accessory gltnid; co,, commoii oviduct; sp., spermattiece ; acs., basal portion of left sperinatliecnl gland.

was ensconced on the surface of the swelling distinctly behind the roots of a group of a t least two pairs of dorsal dilators. The posterior dorsal dilator niascles are in close relation with the anterior extremity of the aorta.

iii. aSsop7~uy.u.~. Behind the second enlargement, and just before entering t h e bhorax, the food-canal, now the cesopliagus, makes a sudtlen downward bend 50 that the subct?sopliageal ganglion is directly anterior t o it. Throngh the thorax the cesophagus, tightly enclosed between the great, lateral muscle masses, shon s little differentiation, although h o d g r a s s (1921 b ) describes and tignres

DIOHPIlOLOGY O F THE CICADID&

441

in Xugicicada s e p t e d e c i n z a transversely-corrLigatec1 distensible crop or proventricnlus. Such a structure. if present a t all, is very much less developed in Carineta fornzosu. A t the entrance into the stomach ih the very effective ccsophaSeal valve. Following Licent (1913) and Kershaw, we regnrd t h e remainder of the alimentaiy tract to the pyloric valve as mcsenteric in origin. IIickemell apparently does not accept these as landmai.ks, since (1920, p. 226) he states t h a t in cicadas i t is difficult to recognize the boundaries of t8hefore, mid, and hind guts. I n stating t h a t the whole of the canal between the miophngeal valves and the pyloric valves is mesenteric in origin, 1 do not wish t o claim that this portion is derived entirely from endoderm, but rather t h a t it cloes not form part of the embryonic invaginations, stomoclzeal arid proctodzeal. Kershaw (1913, Text-figure 74.

Carinetafommosa. Female : part of reproductive system, ventral view. (Lettering as in text-fig. 73.)

p. 176) considers that in i%pha?Ztffi acutu at least a part of this tiact niay be ectodermal in origin. The whole of the region thus delimited is lined with bare epithelium rather than with chitin.
iv. Stomach and Eilter-chamber. A t the junction of thorax and abdomen, or a little before, tile cesopllngus enters the stomach, the entrance being guarded by a r:tlve. I n Carineta part of the stomaah may lie even in the tlioyacic cavity before the mesophragma. Tlie stomach (text-figs. 60-63) appears as a bilobed sac, called by Hickernell the anterior crop and by Kershaw the pouch or filter-chamber. The esophagus enters the anterior lobe slightly beneath and behind its anterior end (text-fig. 62). The first lobe is opaque, apparently niusctilar, and with an irregular surface,

442

DR. J. G . MYERS O N THE

while the second is rather thin-walled and spherical. By t h e position of the esophageal valves this stomach must be considered part of the mesenteron, as Licent, Kershaw, and Snodgrass seem to agree. I n some of the other Auchenorrhyncha, and especially the Cicadellidz (Jassoidea), the anterior chamber forms a much more distinct diverticulum of t h e stomach, and is there called by Licent the poche. It would seem to be homologous with the huge anterior diverticulum which Kershaw found t o extend into the cephalic horn of lantern-flies and other Pulgoroids, b u t these insects differ from most of the other families of the Auchenorrhyncha in lacking n filter-chamber altogether. The condition of the stomach in the Paraguayan Cchrineta fornaosu is not quite like that of any hitherto described species, but resembles nearest the figure (pl. 4 fig. 5) given by Kershaw . (1914) of a n undetermined Trinidad form. It may be best comprehended by reference to text-figs. 60-63. Although it Text-figure 75.

Carilzetaformosa. As text-fig. 74, but dorsal view. up., apical pouch of spermatheca j ms., muvculal ploamal part of spermatheca.

curves collar-like in a very intimate manner partially around the anterior portion of the mesenteric sac, it does not communicate with the cavity of the latter until near its own extremity closely adjacent to the spot where it receives the mid-intestine on its way from the posterior end of t h e mesenteric sac. The course of the mid- or ascending intestine in t h e thin wall of t h e second lobe of the stomach is relatively simple. and nray be followed with ease from the outside ventral aspect. The four Malpighian tubes enter near the same spot, on the wall of the passage between the stomach and the mesenteric sac, but they are dorsal in position and run almost directly into the thick mall of tlle first part of the stomach, where they are immediately lost to siglit. The course of the intestine and Malpighian tubules within the wall of the filter-chamber or first stomach-lobe is difficult to follow. Their coils are extremely thin-walled. Here it is t h a t the chief arrangements of the filter-chamber function. The loops and zigzags of the intestine within the chamber are,

MORPHOLOGY O F THE CICADIDLE.

443

however, much less coinplicated and numerous in Cicadidte t h a n in CercopidE, where the peak of evolution in this matter is reached. Both Licent and Kershaw confirm this independently. In Curinetu formosa the intestine keeps rather closely t o the anterior and ventral wall of the stomach, passing, however, dorsad of, and anterior to, the entrance of the esophagus, and then curving round, receiving the Malpighian tubules and passing out of t h e stomach-wall as the hind or descending intestine. The Malpighian tubules follow a more dorsal course. The condition in Cicadidze is somewhat peculiar. They
li montrent une structure de lepith6lium des lacets intestinaux toute differente B la fois de celle des Cercopidae e t de celle des autres insectes du groupe. . . . Les lacets [folds of t h e midintestine within the filter-chamber] b cause de leur dkveloppement, peuvent bien filter lenii, mais leur Bpithdium a lair beaucoup moins specialis6 dans cette fonction que celui de toutes les autres cicadines B poche. (Licent, 1912, p. 88.)

W e have considered throughout, as dissection seems to show and as Licent asserts, that the intestine and the Malpighian tubules in the filter-chamber are actually in the wall of t h e stomach between the epithelium of its lumen and the external muscular investment. B u t Kershaw (19 14, p. 66) explains t h e condition differently, as follows : T h e zigzags of the various parts of the gut do not pierce the walls of the pouch in order t,o enter or leave it, b u t are merely enclosed between its exterior wall, the peritoneal membrane, muscular and connective tissues sealing u p their entrances and exits and the narrow gap or slit-like opening between the exterior walls of the pouch. Hickernell (1920) expressed somewhat the same view, but t h e distinction is largely a merely verbal one, since the peritoneal membrane of the stomach (pouch) is, after all, a p a r t of its investment. The striking feature anatomically is t h a t the membrane separating the very thin-walled intestinal and Malpighian filter coils from the lumen of the stomach is extremely thin. Muir (1923, p. 213) credits all the Auchenorrhyncha except t h e Cicadellidze and the Fnlgoroidea, and all the Sternorrhyncha u~ithout, exception, with a filter-chamber, and draws phylogenetic conclusions therefrom, on the assumption that such a structure probably did not arise nioie than once. But J i c e n t h a s shown that, though the filter-cha mhcr niay be physiologically similar throughout these groups, i t differs greatly morphologically. The dirergence between Auclienorrhynclioiis and Sternorrhyrichous conditions is especially great, as Licent remarks (1912, p. 29) : I1 est absoluinent impossible, nu point de vile anatomique classimiler a aucun clegre, B ces formations des Coccidae, la poche rles homoptbres superieurs dans laquelle lcesophage ni le rectum ne sont absolument pas engagBs.

444

DR. J G . MYERS ON THE .

Similarly, Hickernell (1923, p. 214) says :(Berlese (1909) describes the digestive organs of certain 1 scale insects. I 1 these insects the rectum is large and extends anteriorly a5 f a r as the esophagus. This results in :I knitting together of rectum a n d cesophagus, and also cames the intestine t o describe a complete circuit of the abdominal cavity before it finally joins the rectum. This condition, while resembling it superficially, is entirely diferent from the arrangement found in the adult cicada, since in the latt8erthe relatively enormous enlargement affects the midgut, while the rectum is small. Hickernell (1923) found that the filter-chamber is typically developed, even in the newly-hatched nymphs, a result to be expected from the similarity in feeding-habits throughout t h e life-history ; so this condition evidently arises in the embryo. v. The Mesenteric Suc. This is one of the most disputed organs of cicadan anatomy. By far the greater part of t h e abdominal cavity, often in both sexes, but especially in the male, is taken u p by a huge thinwalled sac (text-fig. 60, n ~ . o which t h e greatly wrinkled )f anterior portion is wedged in the V between t h e two large tymbal muscles, and is produced into a horn extending into t h e thorax and suspencled from the esophagus by a tendinous band. This is the sac which Graber (1876, pl. i. fig. 5, Be.) and Snodgrass (1921) claim as part of the tracheal system, while Vogel (1923) believes it accessory t o t h e auditory organ; but in t h e sonth American Carineta jormosa I find that not only has it no connection with the tracheal system, but t h a t it also opens wiclely definitely into t h e anterior part of t h e posterior portion of the stomach, thus confirming t h e views of Dufour, DoyBre, Nassonow, Gadd, Licent, Kershaw, Hargitt, and Hickernell. Licent calls it le seynae7at c o n i p e of t h e dilation ventriculuire ; Kershaw refers t o i t as a sac of the midgut j while Hickernell names it posterior crop. Soorlgrass (1921, p. 403) considered t h a t this sac, as a part of the tracheal systeni, received its air-supply directly tbroug2i the first abdominnl spiracles (sp. 3), which lie just before t,he tympana, but Hickernell (1923, p. 220) writes :-6 It is easy to be deceived as t o t h e continuity of the lumen of the posterior crop with the exterior through these first abclominnl spiracles. I n gross dissections there is only the most delicate epithelial membrane limiting this abdominal sac in the region of these spiracles. . . I a specimen is allowed f t o become dry, the portion of the wall which is in front of the spiracular opening may easily rupture, :tnd then there is an external opening in fact. Sections show that the spiracle opens into a very small chamber the walls of which break up almost immediately into a number of tracheal tubes [textfigs. 67, 681 which distribute themselves over t h e external surface of the posterior crop. I ani therefore still inclined to

M O R P I I O L O G P O F llIE CICADIDE:.

445

question any interpretation which gives this organ a respiratory fnnction. It certainly hecoiiies modified ill later life, but it is n.t all times a, part of t h e digestive system. This condition, which is easily observed in sections, inalres uniiecesssry t h e postillation of any secondarily derived function on the part of 11. The rnethorl of gross dissection, then, is inadequate n the conditions found. Hickernell further notes t h a t the interior of a tracheal sac should hn.ve a. chitinous lining, while sections demonstrate conclnsively t h a t the present orgnri is lined with epithelium, although in the rriiildle portion the walls become so extremely thin that the cells may be cuboid or even squamous instea.cl of columnar. O w gross dissections (text-figs. 63, 64) and shose of Lucas (1887) confirm Hickernells results concerning the relation between the third spiracle and the adjacent wall of tlie ineseiiteric sac. These relations are so close tha.t we suppose i t within the boiiucls of possibility for the sac thus t o gain an opening with the exterior similar t,o t h a t described by Suodgrass. Rut t h e fact would nevertheless rernairi that the connections with the rest of the digestive system a,re primary ancl funtlninental, and t h e communication of the mesenteron with the exterior through tlie body-wdl direct wonlcl constitute a condition so anoma.lous that the proof required to estitblish it would he almost as rigorous as that tlerna,ntletl by Huxley for the presence of a unicorn trotting down the Strand. Wit,h regard to the extent of the sac, Hickernell found in extreme cases tha.t it went back as far as the V I t h segment. I n Cffirineta,in the male at least, itasdevelopment is greictei- still. J7Tickeriiellascertained that the size of the sac increases not only with t h e life of t h e adult insect, especiallg with decrease in size of the fat-body, but also throughout ontogeng. I n the nyniph (1 923) i t is much smaller and considerably folded, with a tortuous liinen. I n the a d u lt male, as noted first by Hickernell, its walls may. closely a.pproach the actual integument on all sides, lenving extremely little space in the main part of t h e abdcrnen for the orgams of circulation ancl reproduction, a n d for the profuse coils of t>he intestines a.ncl Malpighim tubules, all of which, as Snorlgrnss puts it, are in places janiined in a space no thicker than R piece of paper. Hickernell is inclined to hoinologise this pea.$ sacculax diverticulum with the prolongation described by Kers1~a.w Fulgoroids and st~atedin his earlier work t o arise iii from the cesophagus. But, a,lthough this latter has also been shown now to he of mesenteric origin, s e t both Kershnw (1914) a,ncl Licelit (1912) agree in homologismg it with t,he poche or filter-chnmher of the columate Suchenorrhyncha. There is certainly no doubt that the two organs are morphologically disfiimilar in origin, though both mesenteric, bnt wbether they may not serve the same function is another question. Licent shows in a very instructive manner that throughout the PROC. ZOOL. S0~.-1928, NO.XXX. 3Q

446

DR. J. G . MYERS ON THE

Auchenorrhyncha" the meaenteric sac (flegment coniqzce)increases in size with reduction of t h e filter-chamber. Thus the former is large in Cicadidz, where the filter-chamber is somewhat reduced, while in Typhlocybids, where t h e latter is altogether absent, the sac becomes simply enormous. vi. The fxicl-intestine. The part of the alimentary tract now to be considered is csllecl by Licent the mediintestin, of which the free portion forms t h e boucle mediantestinale or the anse rnecliintestinale. Snodgrass names it '' the tubular continuation of the stomach," Kershaw simply " midgut," and Hickernell " ascending intestine." Ainong the coils of intestines and Malpighian tubules coating t h e exterior of the mesenteric sac the portions pertaining t o the midintestine may be distinguished macroscopically from those of t h e Malpighian vessels by their larger calibre and smoother surface, and froin those of the hind-intestine by their colour, as remarked by Dufour in 1833. Similarly, Nassonow found it crammed with yellowish granules. A t the posterior extremity of the mesenteric sac is a knot of yellow tubes consisting of the first portions of t h e mid-intestine. This may be unravelled and the continuity of the mesenteric sac and the mid-intestine may be demonstrated, but the lumen of the latter is so minute t h a t t h e actual communication can be confirmed only by sectioning, as has been done by Hickernell, Licent, and Gndd, while Kershaw shows the same passage in his figures (1914, pl. 4 fig. 5). The numerous and . comp1ic:tted yellow folds of t h e mid-intestine disentangle themselves eventually froin the intermingled fat and Malpighian tubes, and the distal portion enters the filter-chamber, as described in the account of that organ. Licent found that the mid-intestinal loop in the Cercopicla showed a constriction n e w the end of its proximal third. Gacld (1902) believed that this constituted a n actual stoppage, and regarded, therefore, the whole loop as two csca joined a t their distal extremities. B u t Licent demonstrated histologically that while the two portions were different in structure and apparently also in function, yet the lumen was never entirely blocked at t h e constriction as stated to be t h e case by Grtcld. Nassonow had earlier found the lumen continuous in the Cicadida, and Licent discovered far less histological differentation between the two parts in this family. Such may be regarded as a more primitive condition than t h a t prevailing in t h e Cercopida and most other coluinate huchenorrhyncha. The mid-intestine is joined by the Malpighian tubules just before the pyloric valve, which i n most of the Auchenorrhyncha is marked by a n externally visible swelling, but which in Cicadids seems not to have been noticed.
Licent persists in calling the Auchenorrhgncha " H o m o p t h e s supdrieurs." I f means anything a t all phylogenetically i t connotes specialization, and the Sternorrhyrichn are very much the most highly specialized of all Hornoptera, albeit it IS in many respects a specialization by reduction.
" height "

MORPHOLOGY O F THE CICADIDB.

447

vii. find-intestine and Rectum. The hind-intestine as it issues from the filter-chamber is grey, and easily distinguishable, even in its most complicated coils, from the bright yellow mid-intestine. I n Curineta formosu its distal extremity enters the strongly muscular, pear-shaped, rectal SAC nearer the *stalk of the latter than in Magicicudc6 septendecwn, as figured by Snodgrass and by Hickernell (see our textfig. 60). The rectum is confined to segments V I I I . to X. (or XI.) and lies very close t o the dorsal integument. As Licent remarks :- L a forte musculature de cette poche expiique surabondamment ldnergie avec laquelle les homoptkres etudiks en ces pages projettent les goutelettes pleines quils evacuent par lanus en si grand abondance.; Nassonow describes the little twist made by the hind-intestine as it enters the rectal pouch. I n this twist are incorporated t h e distal exti emities of t h e Malpighian tubules. viii. Salivary Glu7zd.s a d Pump. The first detailed description of t h e salivary glands in Cicadida is apparently t h a t of Dufour (1825), amended in 1833. H i s account is reinnrkably accurate. &[any wiiters have been fascinated by the salivary pump of Heniiptera, and since this organ shows little variation in essentials, their descriptions ai e largely applicable to Cicadidze. In 1910 Bugnion and Popoff studied in detail the salivary glands of iiuinerous representatives of the order, and concluded with a description of conditions in Cicudu o m i . This has been in p i r t confirmed by Snodgrass (1921) in ilIugicicada septendecim, and by ine in CcLrinetu formosu and in several New Zealand ,pecks of Il.leelampsulta. The glands form whitish digitiform lobules clustered in oiinches and filling most of the head-cavity between t h e muscles, bnd extend also into the prothorax. There is an anterior pair orming the chief gland which supplies the principal canal, and L posterior pair-the accessory or aberrant gland-joined t o t h e irst by a short thick cord on each side. A third gland pair, bonsisting of a long filiform organ on each side with a small pheroid body i n its course, joins the principal canal near its . . brigin. Thcre a l e thus, on each side, three canals meeting at he same place on the deep face of each principal gland. The principal gland exhibits two kinds of digitations-larger nes to the number of sixteen or twenty and white in colour, n d smaller ones more or less transparent. Among the former, 1 Carineta we found several lobules (acini) similar in structure 3 the white ones, b u t deep black in coloui-. These were irregud y distributed among the others. The principal canal unites, after a very short course, with that f the opposite side, and the common t r u n k runs t o the salivary 30*

445

DR. J. C . MYERS O N l H E

ptiiiip. I n Nelanzpsffiltcc leptomera (text-fig. 67) its entrance into the antero-ventral wall of tlie panip-ciianiber is flanked by t n o short, stout, curved transparent horns. I n Cicccdu o m i the debouchure is similar in position antl marked by a rounded swelling. In Xc~gicicffidcc, wliile Meek shows it simiiar here also, Snodgrass (1921 a,p. 6) states that the colnnion clnct appears t o open :it the very tip of the terminal point of the hypopharynx. This is surely an error. Gacltl (1909, fig. 2) shows in Cicadatm at7u a coiiclition identical with that in Xelffimpsaltcc leptonzera. The aberrant gland contains a score of lobules sirnilm t o those of the principal gland, but a little larger, more transpment, and easier to isolate. The cord joining i t tQthe principal gland has an opaqiie covering, formed largely of trachez (Bugnion & Popoff). The filiform organ is a very fine and delicate tube, in o ~ n i about 20 mrn. lorig. Beginning with a free, rounded, blind extremity it ninkes many turnings behind the liead and then joins the principal gland at its origin. Coniposed of the two distinct segments, ir, shows in its distal part a relatively wide tube with thick walls, furnished internally with cells badly delimited and difficult t o distiiiguish (B. & P.). The proximal segment, niuch shorter, has a chitinous canal surrounded by n thin investment like that of the collecting canals in general. A t the junction of the two segments is a small rosette containing some atrophied glandular lobnles. Bugnion and Popoff consicler that this peculiar nberrffintgland, discovered by Ihfoui- in 1825, takes the phce of the great abdominal salivary gland of Fulgora. The salivary pump is a tiny but tough chitinous cylinder or elongate bell of transpxrent material lying beneath the tro1lgh-like frontal plate. The piston or plunger (text-fig. 67), s t i k t e d deeply but sparingly in both I?felanzpsalta leptoinerffi a n d Cicada orni (B. c P.), is darker in colour. The pump b anteriorly * continues into a narrow heavily-chitinized tube Posteriorly the shaft of the opelling near the mouth-pore. p h q r e r expands into two branches, each serving for the insertion of it \vide and powerful protractor muscle. Miiir antl Kerbhaw (1911, p. 79) have shown that the salivary p 1 1 p is a diflerentixted part of t h e common salivary duct, from 111 which it arises in tlie embryo. ,IS regards the histological structure, Bugnion xncl Popoff found tll:Lt the glnntlnlar cligitations show an external cuticle supporting t r , ~ ~ larid~in ~ , interior a greyish cyt oplasni without distinct ~ - the cell-liniits, and a branched nuclens. There is, t o all appearances, a single nucleus or acinus in spite of its relatively large dimensions. Tho presence of these ramified nuclei gives to fresh pieces in saline :I peculiar aqpect ; the teiminal nuclear branches being a lit,tle tlilatetl, one sees in each lobule a great number of small ,wp]liiigs joined to one another by straight britlges. Finally, ?r the hypothetical positioii, \.ritll t h e head of the iiiscct poriect. Pelllaps
(6

crp~lalad woald be n clearer term.

XOR P HOL OGY OF THE CIlCADIDA5.

449

above the great ramified nucleus there are small multiple nuclei situated a t the surface of t h e acirius arid pertaining to t h e small flat cells just beneath the cnticle and concerned in its production. Although minified nuclei of the kind described in Cicada are characteristic of insect salivary glands in general, yet the Gymnoceiate Heteroptera, studied by Bugnion and PopoE, show only simple oval or rounded nuclei in the salivary acini. This is a great divergence within the same ortler, and one would be inlerested to see how widely it extends. Bugnion ancl Popofs remarks on the pliylogenetk development of the Heinipterous salivary glands ai-e to be largely discounted by the fact that they regard the highly-specialized Aphidid= as lower forms, stnd seek to derive the coiitlition in Ciccldu, Ftdgoru, arid the Heteroptera from that in Aphides. The latter may have retained a primitive condition in this respect, but there must be no a priori assumption t h a t such is the case. Gadtls work on Yibicen plebeia, Cicada o m i , Cicadatru utm, G. hyulina, arid Xelccmpsaltn mo?ataiou seems t o agree largely with that of Bugnion and Popoff.

ix. iVcdpigluiaiz iltcbules.

In the Cicdicle the relationships nncl even the number of the Alnlp~gliiantubules are more obscured, firstly by their extreme lengtli, secondly by their participation in tlie riiesenteric complex kriowii RS the filter-chamber, ant1 thirdly by the inextricable iiiirigling of their distal extremities i n the final knot of t h e hindiiitestiiie just as it enters bhe rectum. Froin this point the tiiluules zigzag over the surface of the rnesenteric sac until they enter the wall of the stomach near the entrance of the midintestine. The distal portion of the tnbules is much thinner t h a n tlie proximal, as shown by Licent in Czcudoc orni (1912, fig. 28). I n the proxiiml portion. just after emergence froin the filtercliaiiiher, tlie nnri.ow first part is succeeded by R very short, smooth, rather swollen section, apparently hon-rologous with a more noticeiLble swelling, greatly developed in Cercopidre and responsible f o r the secreticn of the viscid mate1 id, which gives tenacity to their foam or cuckoo-spit. The reinninder of t h e tubule is varicose. The i\Zalpigliian tubalw are four in number, b n t after serpentining in the filter-chamber, and just hefore joining t h e mid-intestine near its junction with the hind-intestine, two, nccoiding t o Licent aiid to Pilassonow, reinain free, while the other t w o fuse into a ureter-a very unusual condition found b y Cndtl and by Licent to occur also in the Cercopida R u t in S q ~ t m d e c i m Snotlgrais (1921 b ) found that both pairs fuse, so , t l i a t only two ureters join the intestine instead of one uieter and two tubules; while Keyshaw (1914, p 4. fig. 5) shows the same ! . condition very clearly in an undetermined Trinidad cicada.

450

DR. J C . MYERS ON THE .

The histology of the Malpighian tubules is considered at great length by Licerit (1912). H i s masterly work should be consulted by all interested i n the digestive system of Hornoptera. A n interesting feature is mentioned by Hickernell (1920, p. 236) : A t the point where t h e tubules enter the internal gland there i s . a n abrupt change in the character of their walls. They become thin, the nuclei decrease in size and are less chromatic. They retain their niembranous character until they finally empty into the intestine near the anterior margin of the internal gland, at the junction of the ascending a c d descending intestine. A t t h e same time their lumen increases greatly i n size-all of which is highly significant, in view of t h e supposed function of the filter-chamber. x. Othey AnrLexed Glands. The excrementitious glands of Dufour (1833, footnote, p. 225) are almost certainly the mycetomes, which house supposedly symbiotic fungi. The liquid squirted from the anus, and derived by Dufour from these glands, is obviously ordinary digestive waste, expelled by the contraction of the muscular rectal pouch.

8 . Physiology. One still frequently sees it stated in entomological works that t h e rostrum is inserted into t h e tissues of prey, or food-p!ant, as the case may be. This is a n entirely false assumption (see Myers, 1921 c). The rostriim o r labium serves solely a,s a sheath, or in some cases (Psyllids) as x manipulator, of the set=. W h e n the latter are deeply inserted, the labium must be shortened or else bent back a t one or more of the articulations. The former method occurs i n cicada, where the setre a r e not so excessively longer than the rostrum. The mandibular setre apparently retain tlie whole setal organ in position i n the tissues, while t h e maxillary set% form the connecting channel. I n the salivary pump the retractor muscles of the piston draw back the organ,. causing saliva to enter the pump-cylinder from the duct-opening in the ventral wall. Thereupon the miiscles relax, and the piston tends to return t o its former position by virtue of the elasticity of the strongly chitinous wall. This forces the saliva down the smaller of the two niaxillary channels, and thus into t h e foodtissue. Snodgrass (1921 a, p. 6) apparently missed this smaller passage, and was thus at a loss t o understand that further pumping did not counteract the upward flow of tlie food-liquid. The function of the saliva remains still problematical; it ma,y be solvent or digestive, or both. Meek (1903), following Plateau, considers the second function predominant. Bugnion and Popoff (1908) consider that a diastase is present which dissolves cellulose

MORPHOLOGY O F THE CICADIDB.

451

cell-walls and begins t h e digestion of &arch grains. The combined effect, by whatever method, is doubtless a readier flow of sap. The reaction of the salivary secretion, according t o Plateau, .Bugnion and Popoff, and Kershaw (Cercopicls), is alkaline. The sap ascends the larger of the two maxillary canals, under suction by t h e great pharyngeal pump, exercised by t h e action of it,s dorsal dilator muscles. Snodgrass finds t h a t on relaxation of these muscles the anterior end of the pump closes first, thus ensuring t h e retention of the liquid ; and doubtless t h e second pharyngeal swelling, with its powerful dilators, acts in conjunction. The return of the dorsal wall of t h e pump would seem to take place by the elasticity of the chitinous supports and possibly by the chitinons dorsal rod, and not, as Muir (1926) suggests, by the reverse action of the dilator muscles, which would be hardly possible. So much is fairly clear. W e have t h e sap, more o r less altered, in the cesophagus. It enters the filter-chamber through t h e esopltageal valve, and the problematical begins. Berlese has suggested for the Coccid filter-chamber-the internal gland of Lubbock-that by its means the wat,er and the excess of sugars. in the sap may soak through the wcclls of t h e coil into the rectum and thus take a short cut t o the anus, while the more nutritious matters follow, as usual, the whole course of the alimentary mnnl. Kershaw (1913, p. 177) notes t h a t feeding the insects (in t h i s case the Flatid, Xiphanta) on colonred liquids tends to confirm this theory, since the contents of the long loop of t h e midgut are very faintly, if at all, tinted, whilst, die rectum is heavily coloured. But neither of these examples deals with the t r u e filter-chamber of the Auchenorrhyncha, as exemplified in Cercopidae and in a less highly-developed state in C i c d i d s . H e r e Licent (1912) has done some careful work, leading him t o adopt Berleses theory for these forms also, though he found t h a t in Cercopidw a t least there mas no evidence of sugars in the liquid from the rectum; a.ncl thus he confines the action of t h e filter-chamber to the eliniination of surplus water. Yet it is well known that the liquid excrement or honey-dew of most other Auchenorrhyncha is rich in sugars. Rates describes t h a t of a n Amazon cicada as sweetish (1863, p. 227). According t o Licent the astonishing thing is that the folds of the mesenteron, in the wa.lls of the filter-chamber, originally absorbent, reverse the action of their dialysing power and excrete t h e same water concurrently with the Midpighian tubules, with whose enclosed portions he found them in CercopidE t o agree in the structure of the epithelium. Licent stresses the fact t n a t the sap of plants is relatively innutritious-it must be taken in bulk. There is thus of necessity a rapid flow, showing itself, among other WR.YS, in the copions evacuation of watery material from the anus in most of these forms-an evacuation a t times so profuse i n the case of the Cercopiclae and Cicadida as to lead to t h e belief in rain-trees, which water t h e ground beneath them.

452

DII. J. G . XI'XRS

ON THE

Liceiit. believes that digestion m~oalrl he difficult were there riot some filtering device by wliiah the bulk of aqueous constituents conld be very qiiickly separated from t h e nutritious remainder. To resume, t h e sap entering froin t h e cesopliagu:; is separa.ted by osinosis i n the fi1t8er-cliamber into its: watery constituents, wliich, perltaps :~ccoiiip~nietl excess sugars, pass thimigli t h e by of wa.11~ t h e intestinal coils a.nd straight o u t by t h e hinrl-iiitestine, arid i n t o t,he more nutritious matters which pass along t h e lumen of tlie s t o m x h to entei. tlie long and tortuous loop formed b y t>lie iriesenteric sac niirl by t h e inirl-intestine itself. This brings us t o tlie function of the huge a n d enigmatical inesenteric sac. Note first t h a t iti lies posterior to t h e filter-chninher, n . n d i s t h u s concerned, if our theory of t h e latter be true, only with tlie more concentrated niitritious portion of t h e s ~ p . EIistologically, botli Licent and Hickernell found i t t o be lined wit,h digestive epitlieliuin, cha.ra.cterized hy '' nctivit6 skcrktrice e t absorbante," app'irentlv not less i n Cercopidp, wliere it is of niorlera.te size, t l i a r i in Cicadidx!, where it is very large, ant1 in TpphiocybidE, wliere i t is simply trenientlous. W e r e it largest in Cica.didq one woiild perhaps explain such developmerit as due to a function accessory t o t h e auditory tyiiipannm (mirror) as claimetl by Vogel. This woiild t h e n he a case slightly andogous with that of t h e Teleost swim-bladder, though in the Cicaditlx! t h e sac need not rrecesswily have lost so coinpletelj- its dimeittary capcity. Hiit its still greater evolnt'ion in t,he a.ppnrent,ly silent Typhocybids, i n mhicli, moreover, no auditory organ has yet been described, 1,enclers a, pradorninoritly gastrononiic explann tion more logical. Th:it, tlie mesenteric sac may 11 receive : i i r a n d t h u s act secoiirla,rily t o ninintain normal :I pressure on tlie inner snrface of the auditory t y m p n u n i is rendered pi,ohable by t h e fact t h a t i n Cercopid nymphs, accortling t o l i e r s h a w (1914, p. 6 5 ) ,
"the air, mhicli d l sucking insects doubtless imbibe in qiimtity n.long with tlie liquid food, appears t,o pass through t h e alimentary canill nncl he utilizetl i n forming tlie air-bubbles conted .rvit,h mucinoirl which a r e eniitted front t h e %nus and form tlie frotli iit mliicli the nynipli lives. After examining this C'ercopirl, 1 ail1 the inore inclinecl t o believe that . . . . t h e ' food-reservoir ' i l l t h e head of FZuta functions in p a r t a s an air-sepa,ra.tor t o rid the liquid food of snperahundaiit a i r before i t passes t hroiigli t.he a1i 1x1eritary caii a I. "

Earlier (1913, p. 179) t,ha same writer, tliscussing t h i s foodreservoir after stating t h a t its chief fiinction, 011 account of its secretive activity, seems t o b e digestive, says :It also collects a. quantity of a,ir, separateti from t h e food imbibed ; there i s always soyl1.e air, often ( especiailp just a f t e r tile moult to adult) a very Inige amount."
'(

NIORPHOLOGY O F THE CI CADID B.

453

Tliis food-reservoir of the Fnlgoroidea is not lromologous wit21 the mesenteric bac of the Cicadidae and Cicaclellirl~, though botli perixirl to the midgut. But physiologically, as in superficial appearance, they may have much in common. Granting t h a t there is some superabundant a i r to be separated from the food-liquid, i t r i d supposing that this is not accomplished by the filter-chamber, then the air rnay be segregated, arid perliaps even stored in t h e cicaclas niesenteric sac, atnd serve secondarily a s an adjunct to tlie auditory organ. In t h e silent nymph the sac is much less developed and considerably folded-the air may n o t be stored. I n dissecting cicadas under wa,ter, one sornetiixes finds the sac t o be full of gas, proba.bly a.ir. H u t such a n enticing theory is to a large extent opposed by the still greater developnient of the sac in the Typlilocybids and other Cicadellids. Le:tving the mesenteric sac, we find that the n e s t section of the digestive tract---the mid-intestinal loop-is, according to Licent, botli absorbent and excretory. The proximal portion, u p to t h e constriction so well mar.lred in Cercopidz but obscure in Cicndirla, is almost entirely absorbent ancl digestive, while the part from there to the filter-chamber is excretory, like that within t h e latter organ. It also stores waste matter from the proximal portion. This would explain Hickernells(l923, p. 218)observation t h a t in i V ~ , g i c i c ~ d septendecinz some sections of the ascending c~ intestine (mid-intestinal loop) showed a n epithelial lining of enorInous cells filled with granules, while iii others the walls were thin. The Xalpighinn tubules, in so f::.r a s they enter into the filterchamber, have been already discussed. The smooth, short, proximal portion (outside t h e filter-chamber), which in Cercopidz secretes the albuminous viscid material for imparting teiincity to the foam? has not been shown to be secretory in cicatla. !rhe familiar varicose, more distal stretches perform their nsua.1 excretory functions. Tlic swollen excretory rectnm expels the liquid waste in t h e foicible iiiaunei so often rernarketl in the CicadidE, particularly in Er1rope:in species. It is extraortlinnry that l h r g i t t (1923, p. 210) slioulcl actrinlly btate, with regaid t o septendeciin, that the adults leave n o s i w s of excretory wastes, such as clefecx.tive prodncts, anti deduce from t h i s t,li:it very little feeding takes .place in tlie iinagind star. This brings us t o the subject of tlie next section.
c.

Peediiig-habits.

Two feature5 of cic:ida life, a k m e all others, appealed to t h e

imagination of the ancients. One was the lour1 ant1 piercing song, and the other concerned the mysterious feetling-habits. Few popular beliefs have taken so long t o die a s that which rtttributes t o the cicada a diet of dew. Even Aristotle (Hist. Anim. lib. iv. cap. 7) saw nothing improbable in this alinientnni uiium

454

DR. J. G . NYERS ON THE

e t peculiare. Pliny, of course, followed him, while Cardano, t h e physician of Milan, probed logically the nouiishing properties of dew, fiom the fact that it was the source of t h e manna nhich sustained the Israelites in the wilderness ! Hales. t h e English plnut pliysiologist (1 727), believed in t h e valuable properties of dew, which was absorbed directly by the plant. To the question of manna produced by cicada-feecling punctures we shall return l a t e r ; but here we are impelled t o quote from Donovan, who & a s responsible for the charming English popular name applied to a Chinese cicada, namely Bea-locust. Dono>nn writes that cicadas had been observed t o fly among ash-trees, t o r e many holes in them, and when the manna had oozed out, return and carry it off. The earliest note I have, however, of cicadas feeding on something if not more substantial, at least more nutritious, than t h e dew of heaven, is an obbervation made by Tancred Robinson in 1683 1684, but published in 1714-1716 : Coming near Cnpua, I ohservd a Species of Ash, or O m u s , on the trunk whereof many Saccharin Concretions were visible. This proved the true iWccn~a,h a t issues out thro the Incisions t made in this tree h j the Iriliabitants of Ccdahricc. Swarms of Cicuclccs were sucking the Body and Boughs, and perhaps by wonriding them made way for fresh Manna. (p. 474.)

R8:iumur somewhat later (1740), describing the rostrum, says (p. x) :L c Cette trompe apprend que la cigale nest pas faite pour vivre uniquement do roshe ;

and instances a n observation of one of his correbpondents that a cicada, burldenly seized on a tree-trunk, often withdrew its mouth-parts from the bark only with great difficulty. But the clew theory migrated t o America and held sway until compar~tively recent years. Potter (1839) claimed t h a t CicadidE live solely on the exhalation from vegetable barks- the insensible perspiration of the vegetable skin. Hildreth (1830) saw the same species (septendecim) with its proboscis inserted in the bark of a tree, and a drop of liquid exuded when the mouth-parts were withdrawn. Jaeger (1 854, p. 102) retained the old fetish- they suck with their snorits only the dew of leaves. It is possible, although not proven, that we may have to recognize a difference, especially quantitative, in t h e feeding-habits of different species. Thus, in Europe, Yibicen plebeia, though wellknown to feed (Fabre), appears t o do so less frequently t h a n Cicada orni, which was doubtless t h e subject of Tancred Robinsons observation, quoted above. I n Australia, Froggatt (1 903, p. 339) remarks t h a t Cyclochila australasice! does not appear to feed in the adult stage. I n North America it is usually t h e

MORPHOLOGY O F THE CICADIDB.

455

periodical cicada which is given the palm for abstinence. W i t h regard t o this species, there were stray observa.tions by Riley, by Davis, arid by March (1589), but it was Quaintance (1902) who proved that i t fed commonly, and sometimes in such numbers that the sap from the punctures was streaming down the trunks and branches of the trees. Townsend (1892) saw Tihicen rnonteauma (Dist.) a.lmost certainly feeding on Yucca aizgustilfblic~ Gravely (1915) had a captive specimen of Lem,uriana apicalis (Germ.) which fed on the sap of a piece of the tree on which it was caught. During feeding it emitted from time to time a jet of colourless liquid with considerable force from its hinder end. This squirting of liquid surely is nn indirect proof of feeding. W e have observed it in iMelarnpsnitu c i ~ ~ y ~ ~ l a t a . fluid disThe eharged is probably different from t h a t emitted by the adult during the final ecdysis-a kind of moulting fluid (Davis, 1922, in Tibicen auletes ; Krumbach, 1917, in Y. plebeia). The feeding evacuations are specially numerous in CYicadaorni (Krumbach, 1917) ; in PZatypZeura capitnta (01.) (Pacilopsaltria subfiufu Dist.) (Biscoe, 1896) in India ; andEhave been described in an Amazon species (Bates, 1863, p. 227). I Australia, n Psaltoda marens (Germ.) feeds continually on the trunks of Angophora lanceolata, emitting meanwhile a spray from the anus (Froggatt, 1903, p. 340). The most copious rain-makers are, however, reported from Mexico. Dr. W. M. Mann found that in one new irrigation district a main ditch was being constructed with a n angle t o avoid a rain-making tree which was considered almost as valuable as a small irrigation system, and far cheaper. An undetermined species of cicada. was thickly perched on this tree, busily producing the rain. An exactly similar tree with cicadas is recorded, lilrewise from Mexico, by Krieger (1904). This writer found t h a t the evacuation WRS performed with a chorus effect, much like t h a t observed in the song of the other species.
Beginnt eine der Cioaden ihren Tropfen z i i schleudern, so ist dies das Zeicheri fur die ganze Gesellschaft, eingleiches zu tun, untl in wenigen Sekunden ist die Ersclieinung des Baumes, der nach Belieben regnen Iasst, fertig.

I n the case of some species, at least, it would appear t h a t no very great restriction in feeding-habits is shown. Thus o r n i , which seems in some districts t o prefer ashes and olives, in others haunts pines, and Siwinton (1908, p. 380) found it in Spain feeding on especially resinous mushroom-topped pines. He was a t liberty t o pick as many of the intoxicated bridegrooms off the sticky trees as I pleased, for they had drunk t h e spirit of turpentine, which is a poison to Man, long and deep.

456

DR. J. 0. MYERS ON THE

e. CIRCULATORY SYSTEM. The t1ors:tl vessel or heart stretclies practically the entire length ot the abdomen with a conipartinent in nearly every segiiient and the usual lateral muscular wings (test-fig. 71). The first o r most anterior of these compartments is largest and almost circnhr. It is situated in the Ilncl segment, and sends f o i ~ ~ a rthe aorta, which reaches thus from the I I n d segment to (I the Iiead, p s s i n g between the great lateral thoracic muscle-masses. Thence i t threads the folnmen between the esophageal connectives, and lends above tlie cesoplragus to the posterior smelling of the pharynx, where it eiitls in a n open mouth in close relation with the bases of the posterior dorsal dilator muscle of this orwn. h e aorta is colourless o r white. while the heart, itself is dnrkcoloured-at least in alcohol material of Cari?~utcc jomzosa.

f . FAT-BODY.
1.he fat-bo?ly, representing a tissue rather than an organ, mas first considered in Cicadidae by Dufour (1825), who described it a.s glnmcoiis-green in coloiir and most beautiful near t h e end of the abdominal c:tvity. I t is cliffuse and plentiful in nymphs and in adults of both sexes, especi:dly in the abdomen. It is permeated by trachete ant1 trncheoles, a n d biricls in sheets the intestine and Malpighian tiihes. The niesenteric S:LC has a. c a p c i t y for adhering to it wherever contact, occws, and sepmxtion is extremely difficult mit31iouti*npturing tlie thin walls of the foriiier. The function of the fat-body is probleni,ztical. Comstock considers it i s tlevotetl primarily t o tlie storage of nutriment and secoritla.rily t o excretion. It plays a n interesting rdle in the cicnt1:i.s arid otlier Homoptera in snppplying housing t o the supposeetlly symbiotic Saccharomyces. According to American observers, infection by the fungus, Jlcissospolol.a cicadiiza, which C:LII jes such :I. heavy mortality among males of septeiadacim, increases with tlie retluctioii in t h e fat-body during adult life. Is i t pssible tliat there is ail antn.gonisni between the Sacclmroniyces a r i d hlre ilfwsospora spores ? Mrilc lins already suggested a bactericidal fnnction for the symbionts.
r .

g. REPRODUCTIVE SYSTEM. The so-called " genit,nlia '' or appendages of the reproductive systeni have been tre:itetl with the chitinoris skeleton generally i n :t previous section. This leaves 11s with tlie gonads themselvesorgans with which few observers h v e concerned themselves. As i r i so iiiany other feutures of internal anatomy in the Hemiptera, we t u r n to Dufaur for the Grst thoroughgoing study of these 0 rgan s. Our oirn observations have been mnde chiefly upon the neotropical CaTineta fomaosrc.

YOItPI-IOLOGY O F THE L I C A D I D X .

457

a. Jlale Beproductive O i g an s . Apparentl) the first account is t h a t of Meckel (1808, pp. 5-7, figg. 2-5). H e states that the median swollen ejnculatoiy duct appears to be glnndular ; its inner surface is befet with small openings. Tlie figures illustrating conditions in Tibiceiz ( l e t t i gonia) plebeia sliow the testes, their ducts, one pair of thick and long accessoiy glands, the ejaculatory duct, and the adeagus. Dufour (1833, p. 186, sep. pag.) deals with Cicada o&. He describes the geneial relationships with detail and accuracy. r , I i v o nccessoiy glands were found, and considered a s seminal vesicles. The ejaculatory duct is desciibecl as swollen nt the base. Ueilese (1909, p. 856) copied Dnfours figure, hut added little t o the FIench authors account, which iemains the best extant. J n Cayineta forrnosa the two testes are ioumhly spherical, but at the same time diffuse (text-fig. 72, t.). %he vas deferens from each is extremely long and complexly coiled (wd.). A n apparently equally long and veiy much thicker accessory gland of iinknown function opens with it into the base of the conimon ejarulatoly duct; b u t just before this fusion the vas deferens and accessoiy glanti together form a small swelling which niay be considered a vesicula seminalis (vs.). The proximal portion of the ejaeulatorj- duct is inordinately swollen and furnished with thick walls. Berlese names it ampolla eiaculatrice, while Apgar (1887) mistook it for the single testicle. Between this organ zinc1 the base of t h e Edeagus the ejaculatory duct (textfigs. 34, ej., 72), while still stroiig, is of sinaller calibre. It enters the :deagns by the basal foramen. The greatly-developed accessory glands WOlild seen1 t o be nzesadenia, i. e. formed froin the mesoderm, but observations on development are Iacliing t o establish this. Unlike tlie external genitalia, the gonads, especially in the male, :we apparently very similar througlioiit the family.

Eewac.de Z?epodzcctive O r y a m Meckel (1808) dealt with tlie feiiiale as with the male gonads in lihicen plebeia. H e was the first t o describe the single long, thick, unpaired, accessory gland opening near tlie apertuie of tlie vagina a t tlie base of the ovipositor. He states that thi4 gland nrisps f r o m a swelling whicli give5 iise t o two tiibes,one the iiecliof the spei-nmtheca and the o?ller the conimon oviduct. Tlie latter branches into two pairs of canals--one those of the prirecl accessory glands and the other the ovidncts. Tlie paired glands seem to contain the same matter as tlie impaired. Dufonr gzve a detailed description i i i 1825, recognizing two lorlg vessels secreting sebaceous liuniour. The whole account IV\V&S modified in 1833, and, among other matters, three of these vessels were shown. They a i e described as being half a s long a s the insect, membianous, senii-diaphanous. Their positions ni e
(

a.

458

DR. J. 0. WYERS OX THE

as observed by Meckel. I n the figure (188) the spermatheca is labelled as the reservoir of t h e sebaceous gland. DoyAre (1837 6) studied the parts immediately adjacent to the base of the ovipositor in great detail, in ~ ~ c u d a nzunnifera (Fabr.). (=Fidicinu 172.). H e described what he believed t o occur during copulation, stating t h a t the penis proper finds its way into the diverticulum, which h e called poche copulutrice. H e discovered the small vesicles on the neck of t h e latter and t h e long fillform tubes which Gadd (1910) thought had escaped all previous eyes. Doyhre noticed t h a t these vesicles contained a fatty substance of a yellow colonr, evidently secreted by the hairlike tubes. H e found the tubes themselves to be five to ten times as long a s the body, and confirmed t h e presence in this third species of the three stouter accessory glands of Meckel and of Dufour (1833). Holmgren (1899) used Cicadellids, Cercopids, and one Fulgorid in his investigations of the female reproductive system, but made a comparison with DoyBres and with Dufours results. The chief point of difference appeared t o be t h e lack of the paired accessory glands opening a t the junction of the oviducts. Herineguy (1904, p. 166) remarks that the poche copulatrice receives the Semen in Cicadidae, and this is then stored in the receptacula seminis, of which t h e cicadas possess two (p. 169). This is surely a n error, unless t h e latter name is applied to the small tesicles on the neck of the bursa. Berlese (1 909) fails to mention t h e unpaired accessory gland, but otherwise follows Dufour closely. Gadd (1910) studied the female reproductive system in Tibicem plebeia. Cicnda orni, Cicudatru querulu, C. atru, C. hyalinu, Melan~psaltu udustu, and &f. nzontam-seven species in all. I n all he found the unpaired gland first described by Meckel, and noticed t h a t in plebeiu and orni it is very long, reaching to 20 mm. in the former, mhile in Cicudutru i t is reduced, and in iMelurnpsulta much shorter than the oviduct. This may therefore explain my own failure t o find it in Nelanzpsultu and in Carinetu. Similarly, the efferent canal of the 1eceptaculum seminis in both Cicudutru and Melunzpsulta (Cicudettcc) is proportionately shorter than in orni and in plebeiu. Gndd claims t h e discovery of the paired filiforin glands opening into vesicles a t the entrance of t h e receptuc u k m senzinis. W e have seen, however, that these were accurately described by Doyere in 1837. Gadds receptaculuna senzinis is evidently our apical pouch (text-fig. 75, up.). I n Curineta there are, as in other species known, two large ovoid ovaries (text-fig. 73, ow.). These contain in a European species 70-80 ineroistic ovarioles, according t o Berlese. The number in Curineta seems t o be of the same order. The oviduct (text-figs. 73, 0.) bends shortly and runs t o join its fellow and form a stout, short, cominon oviduct or vagina (text-figs. 73-75, co.). This bends over very sharply near its beginning, and just after the union of t h e oviducts receives

MORPHOLOGY O F THE CICADIDX.

459

a long, winding, accessory gland from each side. These appear homologous with the very similar structures in the male, but are shorter. They can, of course, be considered neither spermathecal nor colleterial glands, and their function remains for the present unknown. Near the external opening t h e vagina receives a large sac-like spermatheca. Comstock (1925, p. 160) remarks t h a t a bursa copulatrix is said t o be wanting in Hymenoptera, Diptera, Heteroptera, and Homoptera, except t h e Cicadas. I am not clear t o which organ in the cicadas Comstock would apply this term. There is only one large sac-like appendage t o t h e vagina in this and in other recorded species. This appendage, the speinatlieca, shows, however, a differentiation very marked, into a relatively soft-walled apical pouch (ap.) and a very stout and muscular portion proximally (ms.). Possibly the latter subserves t h e function of bursa copulatrix, but we are here calling the whole organ a spermatheca. A t each side of t h e entrance of the spermatheca (or bursa, Comstock) is a tiny thin-walled sac receiving a long, much coiled, thin, and thread-like white gland discovered in other species by Gadd in 1910 (text-figs. 73-75, acs.). These glands, which are cut short in our figures, are t o be considered as spermathecal glands. Finally, as t o the large, very long, unpaired gland opening into the I agina near its orifice, as discovered by Dufour and confirmed by Gadd (1910), b u t omitted by Berlese (1909, fig. 1132) from his copy of Dufaurs figure, I have found traces neither in Cariszeta formosa nor in Melarnpsalta scutelluris ; b u t the condition and quantity of my material hardly warrant the assumption that it is therefore absent in these species. Gadd found it very short in Nelcc?npsaltaadusta. I n Melampsulta scutellccris t h e filiform sperniathecal glands are thicker and somewhat shorter than in Carineta. The vesicle on each side, into which each opens, is larger than in the other genus, translucent and bright yellowish-green instead of white as in Carineta.

(3) SOUND-ORGANS. W e now come t o those parts of cicada anatomy which, above nll others, have enabled these insects, in more senses than one, t o make a noise in the world. W e have here to deal with t h e most complicated sound-producing organ in the animal kingdom-and one which usurps such a large part of the cicada economy as t o remind us forcibly of Platos story t h a t these insects were once men who gave up their whole lives to song, neither eating nor drinking, and singing, died. The following comprises, firstly, a description of these organs as they occur in Melampsaltu, notes on their mechanism and a discussion of the further specialisation exhibited by the Platypleurinz, and a few notes on t h e condition of t h e corresponding

460

DR. J.

a.

MYERS ON TIIE

parts in females and nymphs ; secondly, a. historical review of our knowledge of c i c d a sound-organs ; and, thirdly, an account of the accessory strirlulating apparatus of t h e subfamily Tettigadinze. It has been necessary to depart from historical sequence i n this arr,iiigeiiieiit, since a description of the organs is a pre-requisite to : just evaluation of enrlier contributions to their morphology. L

DEACRIPTI~N OF SOUXD-ORGANS. (Text-figs. 10, 14, 16-23, 27-29.) Unless otherwise stated, the description is Lased on New Zealnritl species of ilfelasnpsalta, chiefly sericecc and nztcta. A n effoit has been made to stabilize the terniinology, 1. The opcrczda are two large plates on the iinclersicle of t h e body, extending posteriorly from the inetathorax and covering a large ventral cavity. I n Afelonapsultffithey are short, in Lenabejo (fatdoqua) and in lettigccrcta entirely absent, but in some species of the Oriental Dundubiini-in many respects the most liighly specialiLecl of all czicadas-they are extraordinarily elongated and attain nearly the apex of t h e abdomen. Text-fig. 10, if nothing else, shows that these organs are extensions of the metathoracic epiniera, a h statecl by Carlet. Mayer (1877) arid Prochnow (1907-8) consider them episteriaites. Comstock calls them outgrowths of the sternella, but in septendecina they may be seen distinctly t o curve round t h e metasternum, which itself bears a projection (apophysis of Carlet) for the insertion of the folded membrane. Tlie opercula are the volets of Reauiiiur, Carlet, and F,ibre. They have been inexcusably mistaken for the agents of souiicl by several tLxonomic writers ; thus Francis Walker calls them the drums, and ;Lpplies the term opercula to the lateral coverings of the tynibals in P l a t y p l e u r i n ~ . Projecting over the base of each operculuni is a nieracanthus which BBanniur (cheville) believed to function as a lock t o preveiit the operculum from moving too far from the plane of the abdomen. But the operciilum is immovably attached to the thorax, ant1 the space between i t and t h e abdomen, which in some species is varied during singing, is altered by lifting the nbtlonien from the operculum. The effect, of course, is the same. The opeicula are moderately developed in the female. 2. Tlie ccczlity-the chzcr.ch or la gleaso of tlie Proverigal peasant -1ir.s beneath tlie opercula. It is bounded anteriorly o n each sitle hy n yellowish folded m e ? ~ z ~ ~ a nirich is apparently t h e ~ze, intersrgniental mcriibrane between the thorax arid abdomen, :lnd posteriorly by a naiwor, which has already been described as the tyinpniini of tlie chordotonal organ lying in the lateral wall of each I l n d ab~loniiiinlsegment. The mirror has usually been consi,lered ( e . g. by nllayer, 1877) as the intersegniental menibrane between the Is-t and Ilncl abdominal segments. but Vogel, as we s b d l sep later, sees in it the PlPzcralhcczct of segment I. Apart from the opercula, tlie mirrors are the only organs of this
a.

NORPHOLOGY OF THE CICADIDB.

461

complex which are present in the female. The folded membrane is probably only of articulatory importance ; but Carlet describes a sniall tensor muscle, and believes the membrane is an accessory vibratory structure, a character as5ignetl also, by assnnipt,ion, t o the mirrors until they were recognized as t h e tympana of the auditory organs. Vogel describes also (1923, Abb. 5 ) a tensor muscle of the mirror, but I have been unable t o find a n y trace of this in Nelanzpscclta (text-fig. 27). 3. The spiracles associated, in our opinion only topographically, in that of other observers (Cnras, Lanclois, Swinton) functionally, with the sound-organs are tlie third and fourth pair (first and secoiitl abdominal). Their position has already been described in section (2) c. of tlie present paper. W e have seen also t h a t the t,achecd sac of many observers is mesenteric in origin and has no communication yet proven with t h e spiracles. Some writers, concerned only with the sound-organs and dissecting probably dried material, have not even recognized a sac, but speak of the third spiracle as opening directly into the body-cavity (Carlet, 1877, p. 20, in Tibicen plebeia). Vogel claims that the mall of the sac is f u s e d with the tyiripnnuni, the two together forming a wall only 0.5 p. thick ! 4. W e have reserved till litst the effective instrument of sound. All the foregoing stroctnres are mostly protective and accessory. The sides of the 1st abdominal segment show as strongly-chitinizecl, paletergal modifications two convex, 0 ~ 3 . 1 ~ colouretl p h t e s in a, franie even more stoutly chitinized. In dfelc~nap~alta surface of each tynzbal shows usually a posterior the smoother portion with one strong rib a t its edge arid a n imterior with strong, more or less parallel ribs, slightly oblique but, not far from perpentliculnr t o t h e long :]xis of the cicada body. These ribs are darker in colour than tlie rest of the t y m b d . I n JiCcbgicicacZcc there is ft greater development of parallel ridges and less smooth space. The tgrnbal of l'ihicen chZoro?iaera is more like that of Xeelanzpsalta. 5. Dissection is necessary to lay ha.re the musculature which pnts these organs in niotion (text-fig. 28). It is convenient t o cut off the abdomen a t about the IIIrd segment. I n ca~~clal view t h e mirrors w e then very conspicuous, and the other orga.ns may be seen throiigh them. Prominent among the latter is a large chitinons V, standing on the sternal surface mid reaching its ~ r i n s p almost to the tlorsum. TIiis is the triangle Bcccilleux of u R&:Lumurand the e~~tngastre Audouin and Carlet. The whole of or pitrt of this structure has been hoinologisecl incorrectly as the fiwcn of an a,bdominal segment, interpreted either as I o r 1 . . 1 We shall consider its true nature in t h e seqneel. Parallel but dorsal to the a,rmsof the V, and inserted in aviclge in the ba.sal portion of each, are two large muscles-the lmgest single muscles in the body. These are the tymbaI-muscles. Each brindle enrls distally in a chitinous plate-XehneizpZatte of Togel, 31 PROC. ZOOL. Xoc.-l928, No. XXXI.

462

DR. J G . JIYCRS ON TEE .

plapzce caTtalaginezLse of RBaumur, kpiclhe of Audouiii, d i s p e t e r n b i d of Carlet,-from which R narrow tendon extends t o its attachment in the doisal portion of the inner face of tlie tyiiihl at the main ridge of the miootfler portion (test-fig. 28). 2'?~7iLbcil-covei-s. --We have described the complete boundpioducing apparatus as it appeals in its simplest foini. I n t h e Vlntyplenririre, t o which subfamily belongs Tibiceiz plebeia, the species most often studied by European woikers, tlie tymbalz are coiiipletely hidden from view by a forward lateral expansioii of the second abtlominnl tei gite, which forms on each side a n accessory cavity of which the inner mall is forniecl partly by the tymbal. These second cavities-cellules of Ri.auniur, cavernes of Dug&+are niucli narrower but considerably deeper than the main one bonndecl by the opercula, folclecl membranes, and mirrors. I n the subfamily Cicadinw (Gsminse Diat.) the tyiiibal covers are incon~plete. The essential elements of the sound-producing apparatus are the tyiiibals and tlie tymbal-muscles. The experiments of Laurentus (Zanotti, 1731, p. SO), of Lepori (1869), of Mayer (1877), of Solier (1837), of Goiireau (1838), of Medici (1847), of Lucas (1887), to mentioii only N few, all attest the fact that the destruction of actual or supposed accessory parts-opercula, folded membranes, mirrors, " tracheal " sac-and the stoppage of the adjacent spiracles have but the slightest, if any, quantitative effect on the production of sound in the living cicada. Landois's theory that the third spiracles were tlie instruments *lf sound never had the slightest experimental foundation, but, on t h e other hand, went against all previous evidence. The more reasonable belief of Carus and others that the spiracles acted as accessories by supplying the air t o the " tracheal" sac was based 0 1 observations made on l'ibicerz plebeicc while singing. 1 This species, like several of the New Zealand me lamp salt^, modulates the sound by alternately widening and narrowing the entrance t o the main cavity by raising or lowering the abdomen. This rhythmied movement was mistaken by Carus for a respiratory one. I n those species, like Jlelffinysnltcc nzuta in New Zealand, which do not thus manipulate tlie opening while singing the song is nniform and monotonous, though the volume o f sound is no

smaller.
W e nre reduced, then, to the tymbals and their muscles. The former are convex, and if they are pulled inward and released by nianipulation of the appropriate muscles in a moist specimen, they will regain their former convexity by virtue of the elasticity which resides especially in the strongly-bowed ribs. This process is accompanied by a short click like that emitted by a tin call similarly indented-and these clicks, rapidly repeated, constitute the song of t h e cicada. No further arrangement is required. B u t the simplicity of this explanation, first clearly demonstrated, though not discovered, by RBaumur, has been repugnant t o several investigators, and Graber would see in the sound a result

SIORPIIOLOGY O F THE CICADIUB.

463

of friction between the rib5 of the tyinbal during the alternate moverrierits of contraction a n d relaxation. Hingston 11922) carries this idea still further-believes that the click elicited by inniiipulntion of t h e apparatus in a dead cicada results from t h e tapping of a tooth boine by one of the ribs on a bar which corresponds with another thickening. Since the tymbal sti ucture iircessery for this method is fay froni general in t h e family, and i i i ~ y even coilfined to t h e one species studied by Hingston, be thi5 esplnnntion, even if partially true, is too limited to apply to cicada music as a whole. So far as we can bee, in n deacl specimen of libiceiz c/~~o~o7irercr inovenlent of the tyinbal is just a simple the buckling on a line near the middles of the chitinous bais, and allnost parallel with the long axis of the body. Ciilet attempted (1877) to measure t h e vibrations of the tyinbal by fixing a fine ghss thread with wax t o this organ and plaririg the other end of the fibre in coiitact with a revolving cylinder fnrnislied with a smoked surface. The experiment was riot 1 ery successful, largely because a captive cicada never sings pioperly, b u t eniits an alarm cry like that of a bird under similar cii curiistances. The same observer (1876 b ) observed that tlie tymbal-muscles contract aimultaneously. It is needless to state that the sound-organs are confined entirely to the males. Apparently Amyot (1836) was the first t o make the statement that certain unspecified exotic forms had the sonnd-organs almost as well developed in the male as in the females, and that presumably these females sang too. Although this seems to he entirely incorrect, it has been repeated in several reputable works, even as recently as 1909 (Berlese). Girard (18613)states the belief very dogmatically. Ilouaologies of the ASo~nd-07-gans.--The interpretations offered towards the solution of the homology of the structures involved i n tlie sound-producing apparatus are confusedly various. The following points zwe, however; agreed upon by almost all workers :-

a. The tynibais are modified portions of the first abdominal tergit e. 6. The tymbal-covers in the PlatypleurinE are outgrowths of the I I n d segment of the abdomen. c. The folded nienibrane is the modified interseginental membrane between the mietasternum and the ventral part of tlie abtlornixial base.
It niay be further taken as established, in spite of assertions to the contrary, t h a t t h e following homology is substantially correct :cc.

The operculs are productions of the metathoracic epimera 31

only.

464

L)R. 3 . G . XYERS
b L

0s THE

b. The mesenteric (so-called part of the alimentary tract.

tracheal) sac is primarily a

The most important of t h e remaining questions concern t h e origin of the furca-like structure snpport,ing the tymbal-muscles, ant1 of t h e mirrors or auditory tynipxna fastened along one of its edges. I n the solution of these problems it is possible t h a t a dissection of Tettigarctcc, which is devoid of sound-organs, wonld be of assistance. Scarcity of materid makes this impossible a t present, h u t an external examination shows the following. conclitions in t h e adult mnle :--The opercnla are conrpletely lacking ; there are n o t even t h e rudiments seen in t h e females of other species. The 1st) abdominal segment is greatly reduced, but shows 1:r.ter.nlly a slightly swollen area, free from t h e long Iinirs which thickly clothe tlie rest of this region, and furriislied with faint ridges. Were n o t nearly ,211 t h e other ch:tracters of Y e t t i gaisctcc a.ppxrently highly primitive, one would be inclined t,o see i n tliis structure the last vestiges of tynibals lost in the history of the race. There are no signs of a mirror. Herlese considers tile, chitinous V in question t o be t h e furca of t h e second abtloininal sternite (p. 707) or of the first (fig. 633) -an inconsistency not unusual i n t h e 1tali:~n workers gre2.t boolc. Conistock (1 925) agrees with the first honiology, while Ininis (1925) states t h a t t h e tynibal-muscles arise from tlie mesofurca. Tho two authors, however, who have given the matter the most a.ttent,ion a,re Carlet (1877) arid Vogel (1923). The former sees in t h e structnre in qnestiori an endoskeletal developinent of the 1st abdorninal segme:it: and calls i t the entoqccstre, while Vogel likewise ascribes it to segment I., :I. conclusion which, as we shall see, there can be no disputing. All previous workers Iiave studied the honiologies of t h e soiindorgans only in aclnlt males, with an occasiona.1glance at a female, Init we have attempted t o follow a developmentnl series, beginning witii a feirinle nymph, as likely t o contain t h e least rudiments of tlie :Lpp:tratiis, then studying t h e male nymph, the atlnlt female, a.nd then the adult male. This has been done witli those species Q \\-liidi w e liave sut>henticniat,erial of the pre-imaginal stages-n:i.iriely JIelrcmpsrcltrc Ieptomera, iW. serierc and iK. ciit,plccta, and ilItcyicicatlcc septendecinb. 1. Tlie female ultimate nymph shows t h e first sternite greatly reduced-iii fact :dmost obsolete. 1nfern:~lly there is a power.fii1 furcn from the metasternam, but iiot tlie slightest trace of enclos1ielot:tI structure in a.hc1omina.l segments I. nnd T I . (text-fig. 22). 2. The male ultimate nymph shows snr~prisiiigiy little dif-ference from the feninle condition. There i d , however, a raised dome-like area of smooth, white, very thin skin where tlie tynibal \vill lie on the 1st :rbdoniinnl segment. This is so laige as t o suggest t h a t the arliilt orgmi owes its structure and strength la.rgely t o excessive folding of this area. The rudiment of the hearing-capsule is more swolleii than i n t h e female. Tlie

NORPHOLOGY O F THE CICADIDB.

465

boundary between segments I. am1 1 .is rather indistinct, and, 1 a p r t from the a.nditory capsule, very membranous. Them is n o ,iiLoYe trace of ccbdonainal fiwcce here than in the female of correspoiiding a g e , while those of t h e t h o r a x are distinct a n d well developed. 3 . ln t h e adult female (text-fig. 20) t h e nymphal condit>ions liave been iiiucli modified, especially i n nbcloiiiiiial segments I. imtl 11. T h e chitiiiized part of tlie 1st steniite (test-fig. 14) is very sniall. The inii.ror has come i n t o being, b u t i s much snialleit1i:tii i n t h e rnxle. Its front border is formed by a narrow chitiiiizetl piece. Its hind-border arises distinctly from t h e 1 s t steriiite :i,s a chitinous s t r i p which, as seen from within, runs into t h e 2nd sternite a n d seems t o be continuous with t h e posterior margin of a flap folded back on t h e 2nd sterriite (text-fig. 20,3.), along tlie intersegmentrtl line. The lower margin of the mirror ched t o t h e posterior edge of t h i s fiap. W e shall see later whetlier the flap itself belongs really t o I. or II.,b u t t h e fact is assured t h a t t h e mirror lies wholly i n segment I., a n d consists of n tliinniiig of t h e sternite itself, as Vogel (1923) was appwently t h e first t o recognize. Xoceover, t h e line separating I. from 1 . 1 p:i.sses between the 4 t h spiracle ant1 t h e anclitory capsule, and the fornier inust therefore be considered as topographically belonging t o I. There is n o s p i ~ x a l en.ctually situated OH seginent 11. If I-leyinons and others ~ h believe t h e r e is a foi.n-a,rd o shift of syira.cles (luring development in t h i s region a.re correct, theti spiracle 3 belongs t o n.hrlomina.1 segment I., b u t lias moved on t,o t h e posterior edge of the meta.tlior:ax, while spiracle 4 lias made R siniilnib shift from 11. t o I. If, o n tlie other hand (see section (2) c. of this paper), these workers a r e wrong, a n d spiracle 3 k)elongs truly t o t h e tliorn.x, as its structure mould seeiii t o snggest, we are faced, wlietlier 4 belongs primitively t o .I. 01' II., with t h e ariomalous conclusion t h a t a spiracle occiipying : ~ n intermediate position i n tlie series has been lost elltirely. W e prefer the former explanation. Vogel believes t h t t h e line sepal-itting t h e m e s d border of tlie mirror from t,he chitinized niesal part of sternum 1 is Ironiologoiis with Ifeyiiions's eiiibryonic division between sternite and parasternite. !C]IIIS t h e iiiirimrs woulcl correspond with the pamsteriiites of segiiieiit I. Tile ratlidity of this division as :i.pplietl to :~.dult st1,nctiire is, liowever, very questionable (see section (1)e . of t h i s paper), arid we prefer t o reg:trcl t h e auditory t,ynipnnum merely : , a, t1iff'erenti:ttetI p a r t of sternite I. without conimitting onr]S selves t o its homology with a more or less liypothetic~~.l parasttwin.1 segregate. There is no sign of endoslreletal structure in tile female abiloniiiia,l segments I. rid 11. Turning now t o tlie a.tlult male (text-fig. In), we can understand t h e position better by coinparision mitli t h e feniale. The small stroi~gly-cliitinizecl 1st sternite of t h e feiiinle is here nierely bent

466

DK. J. G . MYERS ON THE

more slmrply upward on each side, and continued into the forerim of tlie mirror i n accordance with t h e much greater derelopineiit of this organ. I n the male this fore-border is inuclr wider, stronger, and more wing-like (text-fig. 29)?arid extends, as in the felil:~le, t o the neiglibonrliood of stigma 4 ancl of the lower corner of the tymhal-i. e. to tergite I. Thus this winglike fore-border is really part of the outer body-wall of tlie 1st segment, and, in fact, merely a differentiated part of sternite I. There is nothing really endoskeletal about it-yet i t is an arm of the IT wliich has been called the fnrca, of the I I n d sternite. A glaiice at text-fig. 29 will show that it belongs entirely to segment I.-tlie line of demarcation is clear. Anterior to this diEerentia.tec1 part of sternite I. on each side lies tlie folded membrane, ancl posteriorly the mirror. I we accept the paraf ste~nite coiiception, then this wing is part of parasternit'e 1.along with the mirror. All t1ia.t has happened is tha.t the entire lntero-ventral portions of sternite 1. have been stxorigly constJricted--pashecl dorso-mesa.lly, so as t o leave a resonatory c:3.vity between what, is really the ventral surface of segment I. a.nd the opercnla, which lie in the normal plane of the venter. No encloskeleta.1 structure lias been produced since t h e whole of segnieirt 1. lias participated i n t h e change. The tergite has supplied the tymbal on each side, while the sternite has differentiatetl in each lateral region into a strong chitinous anterior portion serving as a support for the greatly thinned posterior part or mirror. Text-fig. 89 shows that there is a slight inward buckling near t h e base of t h e two sternal wings i n septendecim. The two parallel ridges formed thus, one on each side, tend t80 approximate each other. They are the nearest a,pp~o:~ccli endot.o skeletal strncture found in this segment, and t,hey serve as t h e fixetl insertion of the tymbal-muscles. I n Tibicen c h l o ~ o n ~ e r a these ridges a.re better developed and practically meet, forming a strong caxina perforniing the same function. The hind edge of the mirror is formed, as in the female, by a stron~ly-chitinizet1 flap a,pparently from the 1st sternite, folding over within the nnterior edge of t.he 2nd sternite, thus making a strong reinforced rim. Whether this flap itself (text-fig. 20), cnrving ba.cliward and inward over the edge of the 2nd sternite, belongs t o segment 1 . or t o segment I. is difficult t o decide. 1 The 1st segment sepmates easily from the IIiid along the intersegnicnta.l line in KOH preparations, and the tear always continues right along this fold, thus separating the flap in qiicstion from segment 11. Rut along the fold there is n o sign of n ten1 suture"text-fig. 21), ancl, moreover, the posterior sternal wings (text-fig. 29) a.re not continuous with tlie flap, as viewed from within. A t the lateral or dorsal extremity of the fold, on the other hand, the chitin is continuous. This questlion, however, does not aEect tlie conclusion tha.t the wliole of the sound-producing apparatus, together with the tympn nil. of tlie chordotonal organs, belongs t o abdominal segnient I.

MOR PHOLOGY O F THE CICADTDB.

467

To the homologies already tabulated a t tlie beginning of this subsection we should therefore like to adtl the following :( r . The wings of the so-calleclficrc~(Eerlese, 1909, figs. 433, 882) supporting the tynibal-muscles a t their base are not endoskeletal, but are rlifkrentiatecl anterior parts of the 1st sternite. b. Tlie mirrors or tympana are differentiated posterior portions of the 1st stprnite. c. Tlie whole of the sonncl-organs pertains t o segment I.

b. RISTOI~ICAI, REVIEK. ihefirst Irnown observation on tlie methorl of sound-production in the Oicadiclze-that of llesiod, about the eighth centiuy before Christ--we have seen t o be :iccurate so far a s i t goes. The tettix pours forth f r o m under his w i n g s his shrill song. Arist>otlesmuch later and more detailed account, a n extensioii, so far as tlie scientific equipment, of the time would allow, of Hesiocls observation, w a s the first ibttempt at a formal explanaL tion. We must exaniine i t : little closer. The relevant passages (Hist. h i m . , lib. iv. cap. 9 ; De part. anim., lib. iii. cap. 161, tilough sliort, have given considerable trouble t o the corrimentators The following iriterpretn.tion is based on the various editions and comnientn.ries cited in the bibliography, and o n the exegetical notes of Lnriclois (1 867). The explanation involves Aristotles conception of respiration, ~vliicli he regarded as essentially a process of cooling. Vertebrates accompli,cli tliis result by taking air into the body and expelling it, a,gain. Insects do not breathe in tliis sense, but achieve the same end by tlie a.gitation of the air enclosed within their bodies. By this nioveirient the body fluids :we brought into contact a t tlie hyposorrzn with tlie thin menihrnne which there sepa.rates them froin the cooler outer air. Ogle and Landois tliinlr that the hyposonia is tlie waist-line o r septum between thorax arid abdomen in its veritrnl part. I n Bies, bees, and similar buzzing insects this sound 1s produced by tlie vibration OF the membrane at the h?yposonzn,brought about L i attritu s p ~ i t z ~ s . I n those tettiges which sing, the liyposomatic insinliing is rnucli more greatly developed, but the song is proclncerl in the same way. Landois compares the agitation of t h e innate spirit (Ogle) mith the nioveinent we might impart t o the air in our chest by moving our dinphragm up and down while keeping the nostrils :tnd lips closed. \Ye do not think Ogles above tmnslation for rrve6pa is well-advised, for Aristotle apparently meant the term t o connote merely enclosed xir mithout any metaphysical iniplications. The movement was conceived as produced by relative motions of the internal organs. Snch was the explanation which held sway until after t h e Renaissaiice. It possessed the merit of distinguishing clearly between the song of the cicadas a.ncl the striclnlation of Orthopteroids; and it was as definite a s the biological concepts and scientific equipment of the time would allow. The absence of

468

DR. J .

a.

JIYEHS ON THE

even t h e riidirnents of knowledge concerning muscular action on tlie one liniitl and respiration on tlie other, i n invertelurntes, inacle i t impossible to distinguish i n prirjciple tlie song of t h e cicadas and t h e biizzing of bees a n d flies. The two other main t.heories of cicada souncl-prodiictiol1i atdmnced i n classical times a r e superficia.1 i n comparison. On t h e one hand, illeleager, Apollonides (or Philippus j, a n d other writers i:i tlie Greek Anthology, by confusion with Orthopteroid st,ridulation, spoke of such methods a s b e a t i n g i t s belly with i t s wings, striking its feet with its wings, or its body. On the other hand, was t h e idea, apparently Egyptian, thougli mentioned also I)y Bia.nor i n tlie Greek Anthology, that t h e rostrum, ritbhirig on t h e under surface, was tlie plectra1 instrunlent of sound (Hoi~:i~pollo, Valeriano). But these superficial notions conipetecl little with t h e Ai,istotelixii liypotheses, which was repe:ttecl by inediEval :md later writers down t o hldrovandi (1618) a n d Moufet (1634), a n d constituted, from tlre snhola.stic viewpoint, a perfect explanation, requiring oiily abstention from renewed investigation t o ensure its iiitlefiiiite perpetuation. T h e oiie glimmer of experimentnlism -the observstjion of Albertus Ilagiius (published 1405) t h a t cicadas could sing f o i n considemble period a f t e r decapitatiunservetl only t o confirin tlie theoi,y. Tlie fiist, t o dispnte, i n the opening of tlie seventeenth century, t h e a.ntliority d i e per. tant :mni n.vea c e h t o il vero, were Cnsserius a n d Gnlileo. G:xlileos statement is a clear expression of tlie nioderri spirit a.nd a clean sweep of tradition, b u t a t i t s pnblica.tion t,he effective organs of cicada. sound had been already cliscovered liy his contemporary. Gialio Casserio (Casserius), erstwhile tloinestic in the liouselioltl of Fa.hricius xb Ayuapendente and Ister liis successor in t h e Chair of Anatomy a n d Pliysics at Padna, piihlislietl i n 1600 a detailed illustrstecl acccunt of several species, figuring t h e t,ynIhals ( ~ m m d m t 7 i m bmcteales), mirrors ($I/~/L~CI~L~), folded rrienihrane ( n z e ~ i ~ b i ~ i n Zutem,), and tymhalce muscles, wliich l a t t e r he described as cid mouendas menzbmnas efecti. The relatioris of t h e muscles a r e clearly sliomn, save t h a t tlie tei~niirinldisc and tendon :+re omitted aiid tlre muscle drawn as insertetl directly on t h e tynilr):ai by a slightly narrowed apex. There is, however, some mistnke i n his experiment, since he reimrks concerning t h e folclerl nieinbiane (nzessibrcc7icc Zutea) L tlisnipta smztcs perit. Considering t h e t i m e at which he wrote, a n d tlie more t1i:i.n a centmys interval which passed before his studies received R worthy suppIernentjt t h e contributioii of Casserius is highly nore~vorthy. H i s work, publishetl i n N. geiiera.1 ana.toinica1 t,reatise, seems largely t o ha^ been overloolied b y his successors. lhe next account, t.ha.t of t h e Spaiiish botanist Giiilio Pontedera (1718), f:i.lls considera.lily sliort of t h e standard set, by Casseiius so long hefore. It is t r c e t h a t h e proinises a fuller a n d illuati~ated description, b u t this apparently never appeai.ed. He indeed

MORPHOLOGY OF THE CICADIDA3-

469

recognized t h e tpmbals, and obssrved t h a t t h e sound ceased when they were broken. His outlook was predominantly traditional, however, and he appeared stiil t o believe that the tymbals were moved by the air enclosed in t h e body, but his accouiit is so obscur-e t h a t one follows his thinking wlth difficulty. Giovairibatista Felici is credited by Medici (1847. p. 142) with publishmg the first figures of the sound-organs (1724), but we have seen t h a t another Italian, Cxsserio, lias no fewer t h a n 124 years priority. Rnt if that honour go to Casserio, neveitheless the standard of Felicis woik makes liiiii a worthy second. Felicis investigations were said to have been made and communicated in 1717, seven years before publication. Felici held n generally correct view a5 to the relntione of all tlie parts ; he discovered t h e terminal plates of the tymbal-muscles, ossetti o cartilagimi dzcre, and the teiidons connecting them with the inner face of the tymbsl. H e knew no previous theories save those of antiquity, and these he criticised soundly, characterising t h e Aristotelian conception as a well-thought-out hypothesis(Ma quanto ( o Dio !). agevolmente singannano coloro, che per investignre le cagioni degli effetti natuiali. non hanrio a h a rnigliore hcorta che la propria fantasia ! (1724, p. 67.) By experiment he showed that the mirror and other necessary parts-one by one-could be eliminated as sound-prod ucers, until finally he could demonstrate this r61e in the t j nibals alone. Felici figures and names five diffeieirt species of true cicatlas. Znnotti (1731) records the investigations of Pozzi (Putius) and Laurenti, carried on some ten years previously. Those of Lxnrenti were devoted to the soulid-oigans. Zanotti refeis t o Felicis \voile, and i d s to show t h a t Laurenti adds anything t o the latters contribution. The tynibals are desciibed, a n d the now almost usual experiment performed to show that their extirpation renders the arimal mute. The tymbnl-muscles are n o t , IioweT er, mentioned. Swammerdani (1737, p. 504) knew no cicadas in nature : we niay therefore excuse his iepetition of the Aristotelian explanation. Four times a t lenst now lixd the true sountl-organs been independently discovered arid their niechanism moi e or less cleaily desciibed. It is significant that the pooiest account of the four was that of the investigator who placed the most emphasis on the teaching of the classics. To the extent that these isolated experimenters broke away from traditionalism, t o t h a t extent also did they tend to ignore the work of their contemporaries and immetlinte predecessors. In no one were both tendencies more clearly displayed tlian in the fifth independent discoverer of cicada sound-organs. B u t it was clue t o the unfailing accuracy and patience of his observations, to the clarity and detail of his illustrations, and to the lucidity and charm of hi5 literary style t h a t RCauniur (1740) not only escaped the oblivion which has

470

DR. J.

a.

MYERS ON THE

largely swa.llowed the work of his forerunners, but supplied t h e sccoiiiit ~vhichis the basis of nearly all subsequent ciccumte clescript,ions of these organs. Even Fabre considered t h a t the '' riiaster" had cleared up most of the mystery, leaving only a few suppleiiientary facts for his tlisciples to glean. F'aLbre is right---tkiose who, like Landois (1867, 1874), loudly proclaim that., instead of aclcling a, little t o the work of their predecessors, they linve reaped an entirely new harvest, are found usually to have stored nothing but chaff. Re'aumur's account, reniar1i:able for its iincomproinising clenial of any other inotor agent than t h a t f urnishetl by t h e tyiiibnl-muscles, is too wellkriown, :it least by imitation ant1 abridgment, to require further
comment.

Later workers who, while adhering t o tlie essence of the Ileauiiiarian interpretntion, have in many cases adder1 greatly to our detailed knowledge of cicada. sound-organs are Carlet (1876 ( I , 1876 h, 1877, 1879), who lias perhaps given the best generd account ; Fitbre ; H:(.swell (1887) ; Lucas (1887) j Mayer (1 877) ; Lepori (1869); Goureau (1837, 1838); Targioiii (1865) ; Meilici

(1847) : Lntreille (1822) ; Micldleniiss (1886); Hingston (1922); Dugi~ (1838) ; Graber (1872, 1876, 1877) ; N. Potter (1839); Lloyd Morgan (1880, 1886). Lepori and Goureau sliowed that the stigninta liad nothing to do witli tlie sound, since it was not altered when these were blocketl. C:r.rns (1829) einplia.sisec1 very strongly a n dlegecl respiratory factor nieritionetl by C'hahrier (1820), but which Cariis clninied t h a t R6auniur mid liis disciples hat1 overlooked. He obeet~ved (l'ibicen plebeia in Italy. Now plebeia is one of those species, like Meluiiqxalta scutellaris in New Zealnnd, which '' play " their instriiinents with considerable ingenuity, modifying the sound by alternately lifting the iLbdomen from and appressing it to t h e opercula, ~ n c tlh u s varying the size of the entrance t o the sonnclorgan cavit#y. A t its gre;i.test divergence the abdomen is slightly curved tlorsail of t h e usua.1 outline of tlie b u d p , and appears swollen. The whole process looks very like a swelling a,nd deflating of the abdomen itself. Hingston (1922) describes i t as a n nctual operation of this liintl-the abdoinen in a Hinia1:iyan species was seen to distend and collapse in accortlance with increase mid diminution of pitcli in the song. Yet a hole cut in t h e side of the abdomen a.nd of tile mesenteric sac failed to stop the song. Chrus, however, believes that, the movements were distinctly respiratory, and acconipanied tlie alternate expiration and inspiration of a i r through the third spiracles. H e regarded this process as per se essential t o sound-production, and knew nothing of the alternate increase and decrease of tlie space between operculn arid venter. I n this interpretation he was followed by Biirmeister (1833) ; van Hasselt (1882) Milde (1886) ; arid, in independent agreement, hy Soiier (1837). We

MORPlIOLOGY O F THE CICADIDE:.

47 1

have seen that Lepori, Goureau, a n d Hingston have supplied refuta1;iori based on experiment, while we have interpreted w i t h Fabre, Swintoii (1880, p. 2 2 2 ) , a n d others t h e eEect, of t h e moremerits i n another way. Swiriton (especially 1908) would seem t o believe, nevertheless, t h a t t h e movements a r e respii,atory. There remain four theories which are now little inore t h a n historico-scientific curicsities. Griffini (1897) writes (p. 570) of cicadas : L e loro elitre hanno le nervaturn bxsali vesicolose, ossia alte : tlilxtarsi ed a depriinersi alternntivariierite. L This alternate swelling a n d deflating of t h e nt)i-vatura produces cicada song ! Bartram (1894) claims t h a t the males of Xagicicada septemkcirn niake a noise by a tremendous motion of two air-bladders under their wings. H i s countryman Hildretli (1826), by a happy thought, extends t h i s hypothesis t o expluiri also t h e origin of the Scottish national iiiusical instrument. H e refers t o septendecinz as screaming with their air-blarlders o r bagpipes placed u n d e r their wings, a n d suspects t h a t t h e inventor of bagpipes received t h e idea from some insect of this kind -surely a Southron origin for t h e pipes. Possibly Hildreth was influenced by t h e question as t o tlie mnsica.1 properties of either instrument. ltiisel(1749 I suffered with R e a u n ~ u rwhose work h e knew well, , t h e tiisadvantage of working with (lead material. B u t while t h e French naturalists cicadas mere preserved i n alcohol, t h e Nukernberg observer liad apparently only dried specimens. He found tha.t t h e tendon from t h e terniirial plate of t h e tymbalmuscle was riot attached to t h e inner face of tlie tynibal, b u t actetl, he tlionglit, as a plectrum, eliciting sound by rubbing its t i p across t h e ridges of t h e Iattei,. l!lie fourth schism from tlie now ovthoclox interpretation of R4:~uriiurwas initiated Iiy Lanrlois (1867, 1872, 1874). T u r n i n g from his investigations on t h e buzzing of bees and flies, he attempted n siniilar explann.tion for cicadas- described complicated g Stinimbiinder, bortlering t h e tllird spiracles, wliich he m i n e d Bchrillstignien a n d n n n o u r c r d as t h e sole instrument of sound. The tyrnbnl-muscles he waved airily aside Die Trommelmuslrel ist s t a r k chitinisirt und wurde von alteren Forschern einfxcli als Cliitinstibchen gecleutet. Die Muskelstructur desselben karin nach cler microscopiechen t Untersuchung clurclians niclit zweifelliw~f sein. Wegeii seiner stxrlien Chitinisirung k a n n dieses Tiibchen nicht contrahirt werden. (1873, p. 348.) H e calls Aristotle t o t h e support of his theory, a n d concludes tri~iinplii~ntl y So koniiiit man oft durch genaue mikroscopisclie Stndien wietler auf das zuriick, W A S die Viilker vor Jahrtausericlen richtig geahnt uncl benannt haben. (1867, p. 158.)

472

THE NORPEOLOCY O F TEE CICADIDB.

Landois's view, severely arid effectively criticised by his countrymail Qraher (1872), by Lloyd Morgan, and by Haswell, t o inention only :t few, was neveitheless accepted by Darwin (1871, p. 330), by Huxley (1878, p. 377\, and by a large number of text-books.
c. ACCESSORY STRIDULATIXG APPARATUS TETTIGADINB. OF

Jacobi (1907 c) was the first t o signalibe in t,he faniily Oicaditire the existence of a. second sound-producing apparatus of a, far leas anomalous and complicated character than the one we have just studied. The subfamily TettigaclinE, poor in genera and species, is essentially a neotropical group of fairly large, robubt forms, with extremely well-developed tyrnbals, richly ribbed. On the lateral angle o f the mesonotuni is a n inconspicuous oral area witliout hair, but with a series of strize running pardlel t o the long axis of the body, and thus across the oval area, t o which they are confined. I n Tettigctdes chilensis A. 6 S., according to Jacobi, the ruts are not perpendicular to hlie surface, but inclined t o forin short scales. W e have been able to exa.n:ine only Choncsia crffissipemzis (Walk.), in which we have counted some 30 of these' ridges (text fig. 59). Jacobi states tliat there are between 15 and 35 in all species of T'ettigacles and ClzoTiosia studied, but only about 6 in B'abms. \Vhile in other cicadas the hind-base of the clavus forms approxiniately a right angle (posterior tuherosity of Ama.ris), in Tettigades a n d CJ10~2osicc, according to Jacobi, i t is d r a w n out, while in Babras, it is extended inucli further into a rounded lobe. This is the plectra1 portion of the stridulatory n p p x t i i s . But n o one has yet recorded observations in the field to est;t,blisli these a.ssumptioiis, and we confess tliat in 2honosia crnssi2ie7s7zis the anal lobe (text-fig. 55, p t . ) seems little more cleveloperl than in orctina.ry cicadas, though apparently capable, nerei~tlieless, rubbing on tlie stridulatory surface (sa.). of This apparatus is equally developed in both sexes, and is espla,iiied by Jacobi as a S ~ l i r e c l ~ m i t t e lp i' n s t such enemies a~ as birds, the abdoiniiia.l sonnd-organs being devoted to sexual purposes. We should like field-observations. I r i t w o of our largest New Zealand cicadas, Bfela~~zpsalta cingirlata and iW, sti*epita.izs, there is a loud wing-clicking protlncerl by botli sexes and additional t o tlie niale's song. It results from a rapitl h t e r a l movement of the wings from the roof-like resting positiorl to one a t a n acute angle with the body; but the movement is so quick t h a t it is in~possible t80 be sure whether tlie noise is prodwed by friction between teginiiia anrl hind-wings on each side, or between one or both pairs and the body. I f the h t t e r , then the developlilent of the stridulating areas on the mesonotum of t h e Tettigadiria: is only a further step in tlie same direction.
&'