Digitized by the Internet Archive in 2009 with funding from University of Toronto http://www.archive.org/details/n1annalsentomolo03ento , ss epee Acts > BON Ouiccslh ae ce eras raecw eaare 7 Namber Toe a oH; he Entomological Society of America — rs "EprroRiaL BOARD tm WM. W ‘Managing Editor, -. OHIO. Publications of the Entomological Society of America. Annals; \V ols. dcand Th complete caelr ss 5.) t's Gi ik SoS A a Ra ee ve ee $3.00 Anne Vols: Tand- th, Parts 1/2 aud) a veaehy sii. Uitte sila ow brtleroe ey gett) Annis Wolsvk and Tle Part i3. caches ye Sek ACL Sea ley v4 OE, cobain 50: REPRINTS Proceedings. of. first three FAGCHD ER Constitution, By-Laws and List of é WEED ETS ert i eee ON a a aad ois 2 Renta Bie ard Wa Ree Y AA 5) * WHeeLrr, WM, M.—Polymorphism of Ants....0. 620. ee ee od .30 Osporn, Hertert—The Habits of Insects as a Factor in Classification... ./. . .20°- Severin, H. H. awp Severin, H. C—-Anatomical and Histological Studies of the Female Reproductive Organs of the American-Saw fly, Cimbex © Americatiay Deagh i os Paeisd en che Nicci dl. wc adeteice a PER Ue Se Ske aay OB. sts Fert, BE. P—Some Problems in Nomenclature... . 0.0... 050. eee esses ie eae i . Hanmorar, A.G.—On the Nervous System of the Larva of Corydalis cornuta L «25 Brapiey,.J..C.—A Case of Gregarious Sleeping Habits among Aculeate Eby ICROPLERA ae KG oy 5 he ae DES oe ible eo cac saben Harada Vier tae ate Bate Seales Pa 10. ont Davis, J. J--Notes on the Lile History of the Leafy PEPE of the Box-, elder Aphid, Chaitophorus necundinis "Bhoa os. 4 Re Pe ae ataatp he aie ae 10° HAMBLETON, J, C.—fhe Genus wana with a, Review of the North and o Middle American Species..... +. gaat Sach Lahrbis avy oe Mak dit Gates 120. ey GiRAULT, A. A.—Biological Notes on Colorado Potato Beetle. : 2.2.2... aad ENROL hae of Giraurt, Ao A—A Monographic Catalogue of the Mymarid Genus Raphie 2005 4 eee Severin, H. H.’anp Severity, H. C. ied i's Organs of Heprodacy on OF fate EAT GSS WEY Pes ce AE mi SIE A Nc nvm Pere. UM gm ey eS Ee Pla one aay, ‘Smits, O. P—A Preliminary Study of the Aranex. Theraphosae of California Ay fia Davis, dy J,—-Studies.on Aphididae... 0.2 eee eee te eee MERE RES ES an | Ler Rizey, W. A.—Muscle Attachment of Insects........:. Dia aTUR RS OME TS 1 are ho a NeEpHam, J. G.—Critical Notes on or Classification of the Corduliinae be er iehe OGOn aa) iE CRG Pe ee AR LS Vee eo a ctcloabs vel ate bk oideue erate fe hen kOe “Howarp, L. OA Key. to the spebiea’ of. ‘Prospaltelle with Table of Hosts EC ALIS Dy and Descriptions of Four New Siiecieg ities. tec. sos ee ee eee Bs Ae ae Hoop, a §, aan New pPpcries of vende roa Bs Bs Sa arch AD. 12 ” Address : ae ANNALS ENTOMOLOGICAL SOCIETY OF AMERICA, plone! Building, 0.8. U., actwnespeier. ths Ohio | y | Pe I ib i 1 iV v, A ne - oo A is ’ - , i) i} “ - , ' 7 ry . . " ° ‘ u : : aaa ua : ' . t i ay : . U i + i U i ty ‘ ’ 1 « ‘ ' 4 ¥ s 7 : ri pA? = , : : ; | : ‘ i 1 ty f ‘ . z . 4 kp 1 ‘ , : f ri Hi : j A ' ; . ‘ Y : , ; io , 3 : i'y 4 : 4 t x ‘ + 1 j : i 1 thy m v4 ‘ ae i ; ' 1 7 = ' 7 A i (cae f . x ( ; r ( Py , : n , : z ‘ it i ‘ i ‘ t 1 U 4 7 . ' : ; A 4 PS ry poe ° are t i 1 F 1 ae : ‘ l ‘ ) ‘i : mY 1 F r | } r i 1 i j " fy aa i u in ar | a 5 ; 7 +) x i i 1 i ; a t rf v1 } 7 | : Di , . ‘ i, 7 ’ ; ; a ms ; ; bi ’ : “oe a ‘ 1 ) pe t 7 — ; ‘ f i 1 q : : = | ‘ 5 . ¥ ah 7 « if - J t ; : ‘ ; : i 4 ‘ x beat ' 7, \ = a" i h ‘ pers = i i Ae on i ' : : mY ont . ~ i 7 y is 7 ices ; WM. H. EDWARDS Plate I. ANNALS OF The Entomological Society of America Volume II] WAAR GH. 1940 Number | MINUTES OF THE BOSTON MEETING. The Fourth Annual Meeting of the Entomological Society of America was called to order by the President at 10:30 A. M., December the 30th, 1909, in the buildings of the Harvard Medical School in Brookline. The President announced the deaths of the following Fellows and Members: William Henry Edwards, Honorary Fellow, April 4, 1909. Mark Vernon Slingerland, Fellow, March 11, 1909. Braxton Honoré Guilbeau, January 16, 1909. William Brodie, August 6, 1909. H. M.S. Seib, August 28, 1908. The minutes of the last meeting were accepted as printed in the ANNALS. The President announced the appointment of the following committees: Committee on Resolutions: Messrs. Satterthwait & Brues. Commuttee on Nominations: Messrs. Gillette, C. W. Johnson and Burgess. Auditing Committee: Messrs. Field, Johnson and Sanderson. Committee to Draft Suitable Resolutions Concerning the Death of Mr. Edwards: Messrs. Newcomb, Wheeler and Field. Committee to Draft Suitable Resolutions Concerning the Death of Professor Slingerland: Messrs. Comstock, MacGillivray and Riley. The following report from the Executive Committee was read by the Secretary-Treasurer, and adopted, the constitutional amendments to be brought up at the next annual meeting. 2 Annals Entomological Society of America [Vol. III, REPORT OF THE EXECUTIVE COMMITTEE, December 30, 1909. The following six persons were elected members in June: Ei} W.. Berger: E. H. Smart, W. A. Thomas, We diSairo; sepals (Coollaye, W. O. Strong. Ten more members were elected at the meeting of the com- mittee yesterday: Miss E. A. Newell, Mr. T. C. Barber, Mr. A. C. Lewis, Mr. W. V. Reed, Mr. W. W. Chase, Dr. C. G. Hewitt, Mr. S. F. Blumenfeld, Mr. J. W. Hungate, Mr PS Ee onminthe Mr. W. H. Shideler. The following resignations have been accepted and member- ships terminated: Rev. G. Birkmann, Mr iG. L, Pollard. Mais F Grothe Miss A. M. Fielde, Mr. G. H. Chadwick, Mr. F. M. Needham, Mr. M. Rothke, J Pe Baumbercer, C..E. Brown: PCockburn, A. Ellsworth, E. Gerstenhorn, Be @. Greene J.C. Huctennin, a Des lavas, Rope Wea, J. M. Rankin, We D2 Richardson, C. Stevenson, W. L. Tower, AS Je Meier, A. F. Winn. The Executive Committee proposes to the Society for its con- sideration the following amendments to the constitution to be brought up at the next annual meeting. To amend Section I of Art. IV, by striking out the words ‘‘and a Secretary-Treasurer’’ and inserting in their place the words “a Secretary and a Treasurer; but these last two offices may be held by the same person, so that the section will read: Art. IV. Officers. The officers of this Society shall be a President, two Vice-Presidents, a Secretary and a Treasurer; but these two last offices may be held by the same person.” Igo] Minutes of the Boston Meeting iS) orpamend Arts TV. Sec. 3; to read: “Sec. 3. Councilor to the American Association. The President and preceding Past-President shall represent the Society upon the Council of the American Association for the Advancement of Science.” To amend Art. IV, Sec. 2, by striking out the word ‘‘who” and inserting after the words ‘‘additional members”’ the words ‘‘five of whom”’ and by inserting after the word ‘‘Society”’ the follow- ing: ‘‘and the sixth shall be, ex officio, the Managing Editor.” So that the Section will read: ‘“Art. IV, Sec. 2. The business of the Society not otherwise provided for shall be in the hands of an Executive Committee, consisting of the officers named in Section I, and of six additional members, five of whom shall be elected from the Fellows by the Society, and the sixth shall be ex officio the Managing Editor. Four members of the committee shall constitute a quorum.” The committee farther recommends an amendment to the pro- posed amendment to Art .V, by inserting at the end thereof the following: ‘‘Their term of office shall commence with the first of June following their election.” During the year a memorial drawn up by Mr. W. C. Wood regarding the tariff on insects and signed by the President and Secretary-Treasurer, was sent to the Honorable Sereno E. Payne. No action by Congress resulted. CLASSIFIED EXPENDITURES INGiiGeS programs, Clem (SOCIeby) ae «risa sess 2 ne ee oes oe $ 10.25 SieatiOMamyen (SOCIEUY eines fai due eta ans Soars ue 2-2 avi $ 24.61 PNT GM SUEULU Seer way she ee eas ek Mey , moth with wings spread; 7, same with wings closed as at rest. Fi igs. b.b., ori- eel ps ae ‘after Chittenden, Farmers’ Bulletin, U. S. Dept. Agriculture, No. £55) D: Fig. a Pediculoides ventricosus, gravid 2. Greatly enlarged. Redrawn from Brucker. PEATE TV. Illustrating various forms of the dermatitis lesions caused by attacks of Pediculoides ventricosus. From Photographs by Doctor Schamberg. PLATE V. Fig. 1. Lesions caused by bites of Pediculotdes ventricosus, in experiment of Dr. Joseph Goldberger. From drawing by F. H. Wilder. Courtesy of Public Health and Marine Hospital Service. Fig. 2. Showing reduction in yield of wheat attacked by jointworm, Isosoma tritici Fitch. The tube at left contains yield from 100 heads from unin- fested straws: tube at right contains yield from 100 heads from infested straws. Photographed by W. J. ~Phil llips, Bureau of Entomology. Bis. 3. Showing attack of joint-worm, /sosoma tritict, in field. Note the enlarged and distorted stems. From photograph by Geo. I. Reev es, Bureau of Entomology. MYRMELEONIDZ FROM AUSTRALIA. By NatTHANn BANKs. The ant-lion flies of Australia have been mostly described by Walker and Gerstaecker. In collections sent to me by Mr. Dodd, from Kuranda and Port Darwin, and by Mr. Perkins, from several parts of Queensland, are a number of the described species and several that are new. These are described in this paper. The Australian Myrmeleondiz. are more like the European forms than like those of the United States. Two genera, Myrme- leon and Acanthaclisis, they have that occur also with us, but these genera are generally distributed throughout the world. The new general are more allied to the European than to our genera. There is practically no similarity to the South African forms. The eight genera known to me can be tabulated as follows: 1. No spurs, one cross-vein, before radial sector in hind wings; wings broad Chrysoleon ISMN OME wea a op ee ao am doar assaf oC eG oRa sous g022 2 2. sOnercrossvemu betore sadialssector ims hincdiwineSnen eee ene 5 Several cross-veins before radial sector in hind wings; anal ends before fork olmadialisecton_pronotmmeb.oaderauhane Once een teen 3 3. In middle of apical half of each w ing is a nearly straight line formed by the bent longitudinal veins and very short cross-veins; anal vein of hind wings runs into lower cubitus, legs short and stout. .Acanthaclisis INoMStchMiline sot ybentaveinssand Short) Cross=veinse seen tani eeeeeeeere 4 4. In both wings the upper cubitus and the median unite long before margin (OL Wings sa WwitleS Mera IDLOACi6 i oni ek ners ae eee Callistoleon Upper cubitus and median not united, wings more narrow. .Myrmeleon 5. Basal joint of tarsi longer than apical joint; anal ends far out on wings, beyonduorigin, of fish fork of tradial (sectors) san ees Protoplectron Basal joint of tarsi not longer than apical; anal vein ends nearer base of wing, at or before origin of first fork of the radial sector.......... 6 6 Spursyash lone eaSsinsh tO umELaAnsalu|GinitSet@Oeeunehnn sl a nee Distoleon SOURS lonely Ionvesie qaeia iheSE jor OE WATS, 5.54555 00da55c50 45a oe 7 Wings broad in stigmal area, in middle of this area is a straight line as in “Acanthaclisis formed by bent longitudinal veins and very short cross- ~] WSLS cee i hue ae aca he aOR GEN LS, el say tguaee Sic oRC CNIS eee Glenurus Wings more narrow; no such line formed by bent veins and short cross-veins Formicaleo Glenurus pulchellus Ramb. Mid-Queensland (Perkins) also New South Wales (Froggatt). Glenurus dissolutus Gerst. Port Darwin, 12 May (Dodd). Glenurus stigmatus, n. sp. Face pale, a large interantennal black spot, a transverse pale band above this, and the vertex brown; antenne black; pronotum with a broad black median stripe, containing a pale median stripe in fore part, sides pale, streaked and dotted with black; rest of thorax black, with a few small pale spots on the lobes; abdomen black; legs black, tibia I and II with basal and preapical bands pale, sometimes also a median spot above; hind tibie pale behind, except tip, and in front at 4o Igro] Myrmeleonitde from Australia Al base and tip pale; base of first tarsal joint pale; legs with short black hair and longer, erect bristles; spurs rather longer than first tarsal joint. Wings hyaline, with many small dots, mostly on forkings of veins, and in apical part, larger ones along radius and cubitus, forming rather oblique lines at ends of anal veins, and on gradate series in fore-wings; stigma a large, black spot in both wings, very prominent; wings of usual shape and much like G. dissolutus. Expanse 80 mm., abdomen long 20 mm. From Kuranda, Queensland, Australia, March, (Dodd). _ Protoplectron costatus n. sp. Face pale, black in antennal region, vertex pale, a transverse black line, and behind it a row of spots, a small trapeze in the middle and two spots each side toward eye; antennz pale brown, short, with large knob; pronotum brown, darkest behind on sides, paler in middle, with some long white hair, and also on the pleura; abdomen brown, with extremely short white hair; legs pale, femora dark above, tibie black at base and tip, tarsi black in middle, pale on most of first and fifth joints, with much long white hair, and some black bristles; tarsal joint I much longer than IV, one-half,as long as tibiz, spurs as long as first joint. Wings’ hyaline, narrow, acute at tip, and fore wings rather falcate; venation mostly dark, but some patches of cross- veins wholly yellow, and long streaks of pale in the longitudinal veins; stigma brown at base. In fore wings the lower cubitus runs parallel to anal vein for a long distance; but one series of costal cells in either wing; eight branches of radial sector before stigma in fore-wing. Expanse 48 mm., abdomen 18 mm. From Port Darwin, Australia, 10 April (Dodd). Protoplectron pallidum n. sp. Lower part of face yellowish, upper part and the front black; vertex with a broad transverse pale band from eye to eye, divided on the median line; antenne yellowish brown, the second joint with a darker ring; pronotum pale yellowish; thorax dark brown on sides, ‘pale through the middle, and a pale spot above base of each wing; pleura mostly pale or ight brown; abdomen black, legs pale, rather reddish on femora and tarsi, no bands or marks, clothed with long white hair and black bristles. Antenne rather longer than head plus thorax; pronotum once and a fourth longer than broad, only slightly narrowed in front; abdomen short, not near as long as wings. Legs rather short; tarsi very slender, basal joint more than one half as long as tibia, the spurs of the same length, joints 2, 3, and 4 very short, 5th joint more than one-half as long as the first, on its basal part beneath are several short, curved spines forming a comb. Wings yellowish hyaline, unspotted; stigma scarcely visible; venation yellowish, or brownish yellow, not dotted. Wings moderately slender, acute at tips, a double costal series of cells in fore wing, single series in hind wing; about 7 cross-veins before origin of radial sector in fore wing, only one in hind wing; in fore wing the first branch of 42 Annals Entomological Society of America [Vol. III, the radial sector arises long before end of the anal vein, which is nearly one-half way out on the hind margin; apical portions of both wings have the longitudinal veins very close together. Expanse 58 to 68 mm., abdomen 20 mm. Several specimens from Port Darwin, Australia, 8 April to 4 May, (Dodd). Callistoleon n. gen. Wings rather broad; anal veins of fore-wings ending long before origin of the first branch of the radial sector; posterior branch of cubitus running obliquely down toward anal vein, two series of cells between it and anal vein; about seven cross-veins before origin of radial sector in fore-wings, some crossed; a single series of costal cells; the anterior branch of cubitus and the median vein unite before tip, and from the anastomosis the veins radiate somewhat, to the margin, more strongly so in the hind wings. In the hind wings about four cross-veins before radial sector; the posterior branch of cubitus running obliquely down to margin, before origin of radial sector; head rather broad; antenne moderately long, pronotum broader than long; tarsal joints short, first much shorter than fifth; spurs a little longer than basal joint. Type—Myrmeleon erythrocephalum Leach. Callistoleon erythrocephalum Leach. Mid-Queensland (Perkins). Distoleon n. gen. Wings rather narrow; seven cross-veins before radial sector in fore wings, one in hind wings; anal ends before origin of first branch of radial sector; a single costal series. In hind wings anal ends soon after origin of lower cubitus, not parallel thereto. Spurs as long as first four tarsal joints, fifth joint much longer than first; tarsus I about as long as tibia 1; pronotum broad. Type—D. verticalis; also includes Formicaleo bistrigatus. Distoleon verticalis n. sp. Head pale, a black band below and one above base of antenne, another band above this, narrowly interrupted in the middle or broken into four spots, and on the vertex four spots forming a trapeze in the middle and two each side toward the eye; antenne brown, annulate with pale; pronotum pale, darker on margin, rest of thorax grayish brown, abdomen blackish, a pale spot above in middle of each segment, legs pale, some black dots at bases of bristles, black and white bristles. Wings hyaline, a dark dot on stigma, one on hind margin one-third the way out at end of anal vein, and two at end of cubitus and median veins, and other smaller spots in apical part of wing, and at forkings of longitudinal veins, these veins black and white in long streaks, some cross-veins all black, others all yellow, others black and yellow. Hind wings with two dots at end of cubitus and less distinctly on apical portion. Antenne long; legs short, spurs. Tgto] Myrmeleonde from Australia 43 heavy, curved, as long as first four joints, fifth joint twice as long as first. Pronotum broader than long, not narrowed in front. Wings rather long acute; one costal series, cells in region of gradate series are long and narrow. Expanse 68 mm., abdomen long 25 mm. Mid-Queensland, Australia, (Perkins). Distoleon bistrigatus Ramb. (M. striola Walk.) Port Darwin 20 April, 28 August; Kuranda, March; Mid- Queensland. Chrysoleon n. gen. No spurs; legs slender, first tarsal joint as long as second plus third, hardly as long as the fifth; pronotum slender. Fore wings broad, hind wings narrower and longer than fore wings; costals of fore wings frequently forked; three cross-veins before radial sector; anal ends a little beyond origin of first branch of medial sector; a double series of anal cells; the lower cubitus running obliquely down to the hind margin. In hind wings one cross-vein before radial sector; anal ending beyond origin of first branch of radial sector; lower cubitus running down to margin, upper cubitus uniting with median before tip of wing. Chrysoleon punctatum n. sp. Head yellow, a faint dark mark each side on vertex; antenne yellowish, darker at tip; pronotum pale, with some scattered brown dots; rest of thorax mostly pale; abdomen pale, a black stripe each side; legs pale, minute brown dots at base of bristles, not distinct on hind tibie. Wings hyaline, outer apical part of fore wing with small brown clouds at forks of veins, and one under the stigma, also in hind wing, but fewer in the posterior apical part. Venation pale, frequently dotted with brown in fore wings, especially along cubitus. Antenne short, not reaching beyond middle of thorax; pronotum slender, narrowed in front, this and the rest of thorax with long white hairs, abdomen with short white hair; legs with some long white hairs and black bristles. Fore wings broad below stigma, barely acute at tips, cells of wing small and irregular; radial sector seven branched before stigma; costal area broad, the cells narrow. Expanse forewings 43 mm., hind wings 47 mm., abdomen long 13 mm. From Port Darwin, Australia 20 April (Dodd.) Formicaleo dispersus n. sp. On the same general plan as F. septus and F. bistrigatus, but no dark stripes in front wings instead of which are about twenty small dark spots in the area of the stripe of F. septus toward the hind margin, and in hind wings is a stripe as in F. septus but its apical part broken up into small spots. The head is yellow, a large black spot on front and upper face; antenne pale on basal part, brown 44 Annals Entomological Society of America [Vol. III, toward tip; pronotum pale yellowish as also thorax, abdomen rather darker on sides and near tip. Legs pale, unmarked, black bristles; spurs as long as joint I, 5th joint as long as rest together. Wings of general shape of F. septus, and similar in venation, seven cross-veins before origin of radial sector in fore wings, one in hind wings; from end of anal vein there is no vein bending up and outward asin F. bistrigatus, in this respect like F. septus. Expanse 54 mm.; abdomen long 17 mm. From Port Darwin, Australia, 13 March (Dodd). Formicaleo marginalis n. sp. Pale yellowish; a broad black stripe through antennal region from eye to eye; vertex with two black submedian spots, and a median spot behind them; antenne pale brown; pronotum with two dark spots on front margin and a stripe behind each of them to the hind border; thorax striped with dark, a submedian pair in front reaching to middle of mesothorax, a lateral one just above wing, and a short pair on metathorax, and a dark spot over base of wings; pleura with some black spots, mostly on lower parts; abdomen blackish, with short white hair; legs pale, with black bristles, and short black hair. Wings unmarked, except at posterior margin near tip of both pairs where the margin is infuscate for about one-third the way from tip to base; venation yellow, subcosta interruptedly black in both wings, and some dark dots on cubitus in fore wings. Venation similar to F.. bistrigatus, a vein bending upward and outward from end of anal vein in fore wings, seven cross-veins before radial sector in fore wings, one in hind wings. Pronotum rather slender, narrowed in front; legs short, tarsal joint I about as long as II plus III, much shorter than V, spurs a little longer than first joint. Expanse 43 mm., abdomen long 13 mm. From Kuranda, Queensland, Australia, March, (Dodd). Formicaleo septus Gerst. Port Darwin, latter part March through April; also at Kuran- da, March. Myrmeleon pictifrons Gerst. Port Darwin; common, 1o to 21 April; and also 3 September; Mid-Queensland (Perkins). Myrmeleon uniseriatus Gerst. | Port Darwin, 9 September; Mid-Queensland. Myrmeleon croceicollis Gerst. Kuranda, March. Acanthaclisis fundatus Walk. Kuranda, March (Dodd); Mid-Queensland, (Perkins). Acanthaclisis subtendeus Walk. From Kuranda, March (Dodd). THE ANATOMY OF THE LARVA OF CECIDOMYIA RESINICOLOIDES WILLIAMS. By Francis X. WILLIAMS. In the January number of Entomological News, Vol. XX, 1-8, 1909, I described this midge in its egg, mature larval, pupal and imaginal instars, and gave a brief biological report on it. The curious habit of the larva in living in a mass of exuding resin might be expected to have brought about some adaptive modi- fications of its structure, especially perhaps of its tracheal system. It is for this reason particularly that the attempt has been made to study the larval structure in detail. As much of the anatomy of the larva is described here as could be worked out in a limited time, and although over a hundred larvae were examined and dissected, the results are far from complete. Considerable diffi- culty was encountered in tracing out the various systems of the larva, and the writer has deemed it best to omit the doubtful things and to declare only those facts which were made manifest over and over again by the examination of a large number of specimens. | The work embodied in this paper was done in the entomological laboratory of Stanford University. EXTERNAL ANATOMY OF THE LARVA. Head, Pinel Pies: 1 and 2: The head is very small and imperfectly developed. Its brownish-black chitinized portion consists of a broad irregular ring from which a pair of prong-like rods project well into the supernumerary segment. These rods diverge posteriorly, are more heavily chitinized than the ring, and have muscles attached to them that run from the body wall. The antennae are short, conical, obscurely two-jointed processes placed well above the opening of the mouth. Ratzeburg and L. Defour believe these processes are palpi, but Osten Sacken as well as Laboulbene and Perris, from the position of these organs, consider them rudimentary antennae. I incline to the latter view since they are situated latero-dorsally on the anterior part of the ring and well above the mouth opening. This latter con- sists of a soft, fleshy swelling taken by Ratzeburg for the labium. Breastbone, Pl. I, Fig. 3. Ventrally on the posterior half of segment I, and situated in a sort of fold is the “‘breastbone,”’ 45 46 Annals Entomological Society of America [Vol. III, also known as the “‘sternal spatula”? or ‘‘anchor process,” an organ said to be peculiar to the Cecidomyidae. This is a brownish chitinized process varying somewhat in shape, but usually widest at the extremities, slightly constricted at or near its middle and at the caudal end, and somewhat incised anteriorly where a small portion of it projects beyond a transverse fold. The caudal extremity is concealed under the posterior fold of segment I. The anterior incised portion of the spatula is somewhat scooped out, and the more heavily chitinized central axis sends out a ridge on either side of this depression. Otherwise the whole piece may be quite uniformly chitinized. A lateral view of this organ shows that it follows the curved outline of that part of the seg- ment in which it lies. The use or the homology of the ‘“‘breastbone”’ has been a sub- ject of some speculation. Giard (8) says that some of these Cecidomyid larvae have the power of leaping, using their well developed spatula and terminal corneus papillae for this process, and in this wise: the larva bends itself almost into a circle, and hooking together the breastbone and papillae breaks the con- nection thus formed and the spring occurs. In the species re- ferred to by Giard however the spatula is well muscled, strongly chitinized, projects well externally, and the corneus papillae are developed. The advantage gained by this faculty of leaping 1s, according to Giard, the assurance of the dissemination of the species, which in turn insures an abundant food-supply and therefore the successful perpetuation of the species. The larvae of Diplosts lott and jacobeae, cited as examples by him as having this leaping power, are gregarious and would not leave enough food for their descendants if they were not thus motile. This theory does not seem adequate inasmuch as the adults look after the food-supply for their progeny, and being active and winged, would have no difficulty in finding another suitable plant on which to oviposit. The motile habits of the larva brought into play just before pupation would indicate a search on the part of the insect for a proper place for pupation, the pupa being formed underground and probably at no great distance from the abandoned food-plant. In C. resinicoloides the spatula is not well developed, projects but slightly anteriorly, and the posterior corneus papillae are absent, also as the insect lives both in the larval and pupal stages in pitch, the breastbone could hardly be used as a locomotor organ. In fact it does not appear to be con- Tgto] Anatomy of Cecidomyta restnicoloides 47 nected with any muscles. Owing to the medium in which the larva lives, it is possible that it has lost this power of leaping by the degeneration of the organs so well developed in the saltatorial Cecidomyid larvae. Locomotion in the larva of C. restnicoloides is chiefly, if not wholly, accomplished by the extension and con- traction of its spiny segments, with the possible assistance of the pseudopod-like processes. Another theory advanced to explain the use of the breastbone is that it is used as a perforator (17). It could thus be used in some gall-forming species which pupate outside their gall, to force an exit therefrom. L.O. Howard (11) refers to a view pro- posed by Enock and sustained by Marchal according to which the spatula is used to reverse the position of the larva in the puparium. In the case of C. restnicoloides this last theory might apply, but it seems more probable that the spatula is used in making the cocoon, in pushing out a space in the resin, and form- ing the thin, convex cocoon cap. Inasmuch as, according to Giard, the spatula is only fully developed in the last moult, it is quite evident that this organ is intimately associated with the prepupal actions of the larva. Osten Sacken says (14): ‘‘If the supplementary (14th) segment be considered as a part of the head, this breastbone might be taken for the mentum, in analogy to the horny mentum of the larvae of the T7pulariae.”’ Pseudopods. ‘These are rounded, somewhat nipple-like pro- cesses,-most prominent on the thoracic segments and usually in double pairs on the segments. Segment I has a pair on each anterior side of the spatula and bordered anteriorly by the imaginal leg discs; segments 2 and 3 have each a more prominent double pair, transversely arranged in the middle of the segments and likewise bordered anteriorly by the leg discs, those of the wings being lateral; segment 4 has an inconspicuous pair of pseu- dopods; segment 10 some indistinct protuberances; while on segment 11 are a pair of rather widely separated protuberances anterior to the anal slit. The last or 12th segment is terminated by two conical projections bearing a bristle at their apices. Spiracles. There are nine pairs of spiracles, normally situated, one on the first thoracic segment and one each on segments 4-11. The last pair is placed apparently on the last segment, but Osten Sacken (14) has pointed out that this segment is in reality the 8th abdominal, the 9th being in C. pint and C. pini-inopis, (the latter being considered a synonym of C. resinicola) unusually 48 Annals Entomological Society of America [Vol. III, small and concealed under the 8th abdominal segment. The spiracles are horny, conical projections placed somewhat above the lateral line, especially on segment 1. The anal spiracles are far larger than the lateral ones, and are of most importance to the larva. They are truncated conical, with four stout spines. No young larvae were examined, but Eckel (5) says of the related resinicola: ‘‘The tracheal system undergoes no change with the larval growth except the odd development about the anal spiracles.” In the young resinicola larva each terminal spiracle is a long horny tube, whereas in full-grown specimens they resemble those of resinicoloides. The 3rd thoracic segment has no spiracles although the tracheal tube is present here and well developed. Eckel has noted this in the eastern species. Hairs and Bristles. These may be divided into two classes: (1) longer sparse hairs; (2) shorter numerous bristles. Those of the first class are, as far as could be ascertained, present on all but the supernumerary segment. On segments 4-10 are two larger hairs situated one below the other, the first close to the base of the spiracle. On segment 11, laterally and just below the spiracle is a large hair. On segment 12 are two apical hairs already referred to. These hairs, commonly near the spiracles, may assist in keeping the latter clear of the soft resin in which the larva hes. The bristles* are arranged in transverse rows forming long patches on the dorsal and ventral surface of the body and thin- ning out laterally. Dorsally these bands are on the anterior portion of segments 2, 3, 4, 5 and 6, and on the posterior part of segments 8,9,and 10. The band is heaviest on segment 4 where a blank space cuts out a broad mesal notch on its posterior bor- der. On segments 2 and 3 the band is broad and almost divided by a transverse space, and on segment 5 is further reduced to three patches. The patch on segment 6 is almost obsolete. These bristles point posteriorly while the simple bands on seg- ments 8, g and ro have them pointing anteriorly. Ventrally the bristle rows are not broken up but are simple, and are present on the anterior part of segments 2, 3, 4, 5, 6, 7, 8 and 9 where they point posteriorly and on the posterior portion of segments 4, 5, 6, 7, 8, 9, ro and 11, where they point anteriorly. These rows pointing forward thin out anteriorly, and on seg- * Illustrated in Ent. News, XX, Jan., 1909, Plate I. TQTO} Anatomy of Cecidomyta resinicoloides 49 ment 4 are quite small and inconspicuous, whereas the sets point- ing backward become thicker anteriorly. We see then that these rows of bristles are heaviest at either end of the larva and absent or imperfectly developed along the middle of the body, there being a gradual transition from the posterior—to the forward— pointing bristles. The bristles thin out laterally and are more numerous on the venter than on the dorsum. Although the larva is not an extensive traveller, it is never- theless quite active and moves about considerably within the limited space of the resinous mass in which it lives. It cannot survive very long when completely submerged in the gummy exudation, but lives with its anal spiracles at the surface of the mass or in a hollow therein, while its mouthparts are in close proximity to the abraded cambium. It must also turn itself around in its cocoon before pupation. It was observed that the posterior portion of the larva is usually more extended and the cephalic end often more contracted. When the larva desires to advance, its anterior segments are extended, thus exposing the strong rows of backward-pointing bristles, while the hinder end is somewhat drawn in, thus concealing its bristles, and, by con- tracting and expanding the anterior segments the desired move- ment can be obtained. It would not seem of so much importance that the larva move its whole body backwards, but that it extend its caudal spiracles to an opening for air is quite imperative. Thus it could firmly anchor its cephalic end to a desired spot, and by means of the forward-pointing bristle rows extend its spiracles to the surface. By sufficient contraction the rows of bristles at one end could be completely concealed and those of the other end fully exposed and brought into play. The pseudopods may also assist in the movements of the larva. INTERNAL.ANATOMY. Tracheal System. P\. V1, Figs. 4 and 5. This was very care- fully worked out, the tracheal arrangement being followed to the finest ramifications. As in other insects it consists of longi- tudinal trunks connected with the spiracles by lateral branches. Figs. 4 and 5 of plate VI shows this system from a dorsal and ventral aspect, of a larva submitted to some pressure under a coverglass. It consists of a pair of longitudinal dorsal trunks beginning at the spiracles of segment 1 and terminating in the caudal spiracles of segment 11. At G, Fig. 4, a stout branch is 50 Annals Entomological Society of America [Vol. III, sent ventrad, which divides almost immediately, the one branch proceeding to the head while the other, bending posteriorly, sup- plies the brain and sub-cesophageal ganglion. This posterior branch is F,, in Plate VI, Fig. 4. The two dorsal trunks are connected with the spiracles by short branches and are joined to each other in segments 4-9 and ti by cross tubes, that of segment 11 being quite stout while the rest are slender and send out two re-dividing branches. These cross-tubes usually lose their striated appearance at their middle, appearing composed of two tubes fused together by a pair of sponge-like knobs. In segment ro, there is no fusion, the tubes remaining separate. Fig. 6, E,,, Pl. VI, showsa condition observed in two larvae in which the left branch is quite swollen basally, which perhaps was the normal thickness of both tubes, through- out before atrophy occurred and the tubes separated. The fact that in C. resinicola there are complete cross tubes in segments 4-11 would appear to strengthen this view. Furthermore it will be noticed that in C. restnicoloides the cross tube in segment rr is quite stout and may substitute in a measure for the tubes in the above segment. Ramifying tracheae are apparently not so necessary in the last few segments. The transverse lateral tubes D,—D,,, Fig. 4, terminate in the spiracles except in segment 3 where there is no spiracle, in which case the tube disappears near the body surface. They are con- nected with one another either by the lateral longitudinal system through the short tube R,, R,, etc., as in segments 3-8, or they may run directly through the former, as in segments g and to. This arrangement, however, is frequently asymmetrical. The lateral longitudinal trunks arise from the dorsal ones in segment 2, Fig. 4, A,, Pl. VI, and terminate as free branches in segment 10, C,,, Fig. 4, and apparently correspond to the ventral tubes as illustrated by Eckel (5), but differ widely from the latter in that they do not join the dorsal trunk posteriorly. They send out many branches F,, F;, K,, K,, etc., but the arrangement of these is not always constant. The primary function of the more anterior of these branches is to aerate and support the ner- vous system, that of the more posterior ones to supply the Mal- pighian tubules. The ventral longitudinal system Fig. 5, Pl. VI, arises from the main dorsal one close to the anal spiracles in segment 11, and proceeding anteriorly commences ramifying in the middle of TQIO] Anatomy of Cecidomyia restnicoloides 51 segment 8, where it turns dorsally, closely following the Mal- pighian tubes to which short branches are sent. Circulatory System. This system owing to its transparency and delicacy was quite difficult to follow, and so a detailed description of 1t cannot be presented. The dorsal vessel is a thin-walled, transparent tube which arises in front of and below the brain, and proceeds obliquely upwards until it meets the body wall in the posterior part of segment 4 where it is secured by the suspensorium. Thence it runs along the dorsum extending at least to the posterior transverse tracheal tube. It is divided into chambers and the valvular orifices (ostia) on either side open and close inwardly. The dorsal vessel pulsates especially in its anterior portion where there are several orifices. On each side of the heart in the middle of segment 4 is a strip of loose pericardial cells which proceed posteriorly, closely following the vessel. Alimentary Canal. Pl. VII, Fig. 1. The alimentary canal consists of the long slender cesophagus D, the larger scarcely differentiated fore- and mid-stomachs (proventriculus and ven- triculus) G, the slender and much curved ileum or small intestine I, the colon K, and the rectum L. A pair of long, recurved salivary glands C. F. open into the mouth, and two Malpighian tubules H, likewise recurved, are fastened to the fore part of the ileum. The whole system 1s supported chiefly by the trachez. Lying above the cesophagus and salivary glands and extending well beyond the rods of the head skeleton is a large blind sac, S, Figs. 1 and 2, Pl. VI, which probably opens into the mouth. It contains in its middle a curious, dark purplish-brown object, A, Figs. 1 and 2, Pl. VI, which is lobed anteriorly, and posteriorly where it becomes semi-translucent and breaks up into small granules. This object is evidently the “‘point oculiforme’’ of Giard (8), who noted it also in a Cecidomyia, but what its function is he does not state. The large blind sac is apparently not muscular, though it 1s possible that it may serve as a food reser- voir. The “point oculiforme’’ suggests a strainer of some sort, especially if the sac were muscular. The cesophagus extends to the end of segment 3 as a very slender and delicate tube, and is chitinized within the rods of the head skeleton, C, Fig. 2, Pl. VII, and Fig..6, Pl. VII. The large, straight, muscular stomach extends from the 4th to the 8th segment inclusive, and its cellular coat is made up of 52 Annals Entomological Soctety of America [Vol. III, more or less regular longitudinal rows of large protruding cells. The whole canal, including the salivary glands and the Mal- pighian tubules is clearly visible through the larval body wall. In the posterior portion of segment 8 the slender ileum arises abruptly from the thick ventriculus as a curved and folded tube. Its cellular wall consists of several rows of large cells distinct in the anterior orange portion, less so in the whitish translucent middle part, and disappearing in the slenderer hinder end, which opens into the delicate grayish colon. The rectum is a rather colourless tube of considerable length terminating in the anus on segment 11. The extreme hind part of the canal probably functions but little, except perhaps when the larva is near pupation. It-was not seen to contain any waste matter, and if the larva was submitted to considerable pressure under a coverglass no excre- ment would be discharged. Glandular and Excretory Appendages of the Alimentary Canal. The salivary glands are of large size and extend along each side of the ventriculus to the middle of segment 6 where they recurve dorsally and proceed to the anterior end of segment 5, being fastened to the ventriculus at that point. The larger posterior portion of the glands are glassy white and of monili- forme aspect, with several rows of large glandular cells containing large nuclei. At about the middle of segment 3 the glands become narrower, lose their glassy, large-celled appearance, and twisting spirally downwards, then upwards, dilate abruptly into granular ampullae of a pale yellowish-white color. These are three in number, two larger posterior ones and a single incon- spicuous anterior swelling. From this latter point the glands taper gently, and at about the posterior border of the super- numerary segment give place to the tracheoid ducts. These are short slender tubes and join to form the common duct, B, Fig. 2, Pl. VI, a short distance within the chitinous rods of the head skeleton. Whether the secretion of the salivary glands of the larva produces an irritation in the plant tissue thereby causing a flow of resin, or whether “‘the twistings of perhaps a dozen spiny- skinned larvae smooth out a round cavity’’ (in the resin), ‘‘the irritation causing a constant supply of fresh resin,’ to use Eckel’s own words as regards C. resinicola, is a difficult thing to determine. Giard (9) thinks that the ‘‘zymase’”’ secreted by the salivary ns o Ta | Igo] Anatomy of Cecidomyta resinicoloides glands of the Cecidomyid larva determines the gall. No gall is formed in the case of the pitch-inhabiting midges, but the salivary secretion in this case might well stimulate and maintain the resinous exudation. Attached interiorly to the second ampulla of the salivary glands and stretching in a curve across the brain which it overlies very closely is a curious, loose ill-defined structure of granular appearance, F, Fig. VII, Pl. 2, and containing brown patches, especially towards the bases. This I believe is Weismann’s ‘cell chaplet,’’ which he defines thus: ‘“‘It consists of a string of large cells closely united which hangs like a garland, free in the body cavity. Its two ends are connected with the salivary glands,” etc. What its function is remains to be ascertained. The Malpighian tubules are of a deep orange color and arise from the anterior end of the ileum immediately where the latter is joined to the ventriculus by a delicate membrane. The arrangement of these two tubules with regard to the ileum and colon is not symmetrical since the left one follows rather closely those two divisions of the proctodzeum, while the right tubule is quite free. Fig. 5, Pl. VII shows the left tubule in its relation to the proctodeum. Arising from the ileum each tube proceeds dorsally but not quite in symmetry as the right tube bends anter- iorly and outwardly while the left first curves somewhat poster- iorly, P, Fig. 1, Pl. VII and then follows the common course. Each tube is closely appressed to the ventriculus to which it is slightly fastened and proceeds anteriorly past the middle of segment 7. Here however the left one by reason of its posterior bend does not reach quite so far forward as its mate, but terminates slightly farther posteriorly and ends somewhat beyond the middle of the colon.. Both tubes run along the ventral tracheal trunks in their terminal portions. Giard says: (10):—‘‘In all the larvae of the Cecidomyidz which I have studied, these tubes, two in number are united into an elegantly recurved handle, and open near the anus, the proctodeum being excessively short.’ In C. resinicoloides the tubes end freely, as we have seen, and the proctodzeum, if extended would probably at least be as long as the rest of the alimentary canal. It may be well said here that the alimentary canal (and its appendages) of C. restnicoloides much resembles that of Dzplosis buxt illustrated by Berlese (2). 54 Annals Entomological Society of America [Vol. III, Organs of Reproduction? At about the termination of the Malpighian tubules, and below the same is a pair of elongate, fusiform, transparent bodies terminating anteriorly and poster- iorly as slender tubes. These bodies were not observed in all the larvae examined. They are probably the developing repro- ductive organs. Nervous System. Plate VIII, Figs. 1 and 2. The nervous system 1s more specialized than in many other dipterous larvae, in that there is considerable cephalization of the ganglionic chain. This chain is composed of ten distinct ganglia, exclusive of the sub-cesophageal one, and extends from the middle of segment 2 to the middle of segment 4. It is supported chiefly by the con- verging branches of the lateral longitudinal tracheae, while the brain is kept in place and aerated largely by the stouter pair of tracheal branches, F,, which arise from the main dorsal system. The brain or supra-cesophageal ganglion consists of two lobes of larger size than the rest and broadly joined to the sub-cesopha- geal ganglion, the commissure not being apparent. The cesopha- gus passes between these two ganglia, as probably also the dorsal vessel, the salivary glands running on either side. At least one pair of nerve cords from the brain, and two or more from the sub- cesophageal ganglion proceed anteriorly, and each of the ganglia of the chain sends out several nerves, some of which run along the lengths of the supporting tracheae. The last ganglion of the chain is more elongate than the others. Very little could be made out of the sympathetic nervous system, and the paired ganglia, C, Figs. 1 and 2, Pl. VIII, arising -posteriorly from the lobes of the brain are all that I feel sure of. Musculature. ‘The muscles were not studied though it was observed that they were numerous and well developed, a condi- tion quite necessary in such a restless insect. Several muscles are fastened to the chitinous prongs of the head skeleton, and a number of large muscles radiate from the caudal cleft, being fastened to the body wall. Thus the larva is enabled to retract quickly its caudal end into the resinous mass if disturbed. Adtpose tissue. This 1s very well developed, so much so in- deed that it obscures in a great measure the different visceral systems. The fat body is of a dirty yellowish-white color, and consists of a loose network of lobes with long meshes, and is con- tinuous throughout the body, thickest below the middle, thence tapering to either extremity, extending anteriorly to the base of the TgIo] Anatomy of Cecidomyia resinicoloides 55 first pair of spiracles, and posteriorly to near the anus. It is circumscribed and largely supported by the tracheal system. Some time before pupation this fat tissue loses much of its com- pactness, becoming more or less watery in appearance and assumes a reddish color. Imaginal Discs of Wings and Legs. Plate VIII, Figs. 6 and 7. These are easily seen in large full-fed larvae where they appear as more or less circular discs attached to the body wall. The wing-buds, Pl. VIII, Fig. 7, A,A,, are larger than those of the legs, lie very slightly above the lateral line of the body, and are supplied with branches of the lateral and dorsal tracheae. Both wing- and balancer-discs arise from about the middle of their respective segments, ending posteriorly in a lobe and tapering anteriorly into the body wall. The three pairs of leg discs, Fig. 6 A,, A,, Aj, are situated on either side of the ventral line and are more distinctly circular in outline than the wing discs and more strongly and completely cut by an inner ring. They are supplied by small branches of the lateral tracheae. BIBLIOGRAPHY. 1. Aldrich, J. M. (1905). Catalogue of N. A. Diptera, pp. 151-152, 160. 2. Berlese, A. (1906). Sistema Ghiandolare, Ghiandole Confinate, (Gli Insetti, i Gosgavsy|e 3. Comstock, J. H. (1879). Diplosis resinicola, (Rept. Dep. Agr., 1879, p. 256). 4, Cannon, W. A. (1900). A Botanical Study of the Gall formed by Diplosis pini-radiatae, (Am. Nat., XXXIV, No. 406, pp. 801-810.) 5. Eckel, L. S. (1903). The Resin-Gnat Diplosis and Three of Its Parasites, (Ent. News, XVI, pp. 279-284). 6. Felt, E. P. (1903). C. resinicola, (N. Y. State Mus. Memoir 8, II, pp. 410-413.) 7. Felt, E. P. (1903). C. pini-rigidae, (N. Y. State Mus. Memoir 8, II, pp. 423-425). 8. Giard, A. (1893). Note sur l’organe appelé Spatula sternalis chez peptaines larves de Cecidomyes. (Bull. Ent. Soc. France, Bebe 22-1893). 9. Giard, A. (1893). Sur un genre nouveau et une espece nouvelle de Cecidomyide (Drisina glutinosa), eer soc. Ent. France, Dec. 27, 1893). Annals Entomological Society of America [Vol. III, Giard, A. (1893). Sur less tubes de Malpighi de les larves de Cecidomyes, (Bull. Ent. Soc. France, Feb. 22, 1893). Howard, L. O. (1898). The Hessian Fly in U.S., (Bull. 16, Dep. Agr., Div. Ent.) Kertesz, C. (1902). Catalogus Dipterorum, II, pp. 116-117. Mik, Jos. (1897). Einiges uber Gallmucken, (Wein. Ent. Zeit., XVI, pp. 290-292.) Osten Sacken (1862). On the Habits of the Cecidomyide, (Mon. Dip: N.-A., Pt. ©, pp: 179=198:) Osten Sacken (1871). Biological Notes on Diptera, (Trans. Am. Ent. Soc., III, pp. 345-346). D. resinicola, etc. Packard, A. S. Diplosis resinicola and Pini-rigidae,. (5th Rep. U.S. Ent. Comm., 1886-1890, pp. 797 et seq.) Sharp, D. (1901). Use of Spatula in Cecidomyide, (Camb. Nat. Hist lass et lil, p.460); Snow, W. A. (1900). The Destructive Diplosis of the Monterey Pine, (Ent. News, XI, pp. 489-494). Townsend, C. H. (1893). Notes on Some Cecidomyiide of the Vicnity “of Wash, (Di C..(Proc: “Ent. "soe! Wash: le ipp: 389-390). D. resinicola, etc. Ig10} Anatomy of Cecidomyra resinicoloides U1 ~ EXPLANATION OF PLATES. All figures greatly enlarged. Prats: Ni. Fic. 1. Dorsal view of head skeleton; S. blind sac; A, point oculiforme; O, cesophagus; B, salivary duct; N, nerve. Fic. 2. Lateral view of head skeleton; S, blind sac; A, point oculiforme; B, tracheoid duct; O, cesophagus; N. nerve. Fic. 3. Sternal spatula; A. A, A, dorsal view; B, lateral view. Fic. 4. Dorsal view of tracheal system of mature larva; A,—A,,, dorsal longitudinal trunk; B,,, ventral longitudinal trunk; C,—C,,, C,,, lateral longi- tudinal trunk; D,—D,,, lateral transverse tubes to spiracles; E,—E,,, E,,, cross tubes; F, F,, F,—F,, branches to nervous system; K,, K,, branches to nervous system; G, branch from dorsal trunk; H, branch to head; R,, Ry, R;, connecting branch between C and D; M, head; S, supernumerary segment. Fic. 5. Ventral view of tracheal system of mature larva, lettering same as in Fig. 4, B,, dorsal turn of ventral longitudinal trunk. Fic. 6. Aborted cross tubes of Seg. 10. Pirate VII. Fic. 1. Dorsal view of alimentary canal of mature larva; A, head; B, blind sac; C, ampullae of salivary glands; D, cesophagus; E, Weisman’s cell chaplet; F, large-celled portion of salivary glands; G, ventriculus; H. Malpighian tubule; I, ileum; J, ventral longitudinal trachea; K, colon; L, rectum; M, anus; P, pos- terior bend of left Malpighian tubule Fic. 2. Head skeleton and portion of salivary glands; A, antenna; B, head skeleton; C, cesophagus; D, common duct of salivary glands showing tracheal character; E, ampulla of salivary glands; F, Weisman’s cell chaplet. Fic. 3. Sagittal section through larva; A, adipose tissue; B, muscle; C, ali- mentary canal; O, cesophagus; I, ileum; P, Malpighian tubule; M, M,, salivary glands; H, heart; S, blind sac. Fic. 4. Cross trachea in segment 11 showing central spongy mass. Fic. 5. Lateral view of proctodeum; A, posterior end of ventriculus; B, ileum; C, colon; D, rectum; E, anus; F, Malpighian tubule; L, L,, slender portion of ileum; O, point of origin of Malpighian tubules. Fic. 6. Chitinized portion of anterior end of cesophagus. Fic. 7. Dorsal view of anal spiracles, 12—last segment. PratEe VIII Fic. 1. Dorsal view of nervous system: A, sub-cesophageal ganglion; B, brain; C, one of the paired ganglia of the sympathetic system; F, trachea feeding brain and sub-cesophageal ganglion. Fic. 2. Lateral view of nervous system; lettering as in figure 1. T. trachea. Fic. 3. Cross section through posterior portion of larva; A, adipose tissue; B, muscle; C, ventriculus; H, heart; M, Malphigian tubule; L, L,, slender portion of ileum corresponding to L, L.,, of Fig. 5, Plate VII. Fic. 4. Cross section through ventriculus showing epithelial cells. Fic. 5. Cross section through anterior portion of larva; A, adipose tissue; B, muscle; C. ventriculus; M. salivary glands; H, heart; F, pericardial fat cells. Fic. 6. Ventral view of anterior portion of larva (portions of segments 1, 2 and 3) showing imaginal leg discs, A,, A,, A,; B, dorsal longitudinal trunk of tracheal system; C, lateral longitudinal trunk; D, tracheal branch feeding brain, this is F,, Fig. 4, plate VI. Fic. 7. Dorsal view of portion of segments 2 and 3, showing imaginal -wing discs, A, and A,; B, dorsal longitudinal trunk: C, lateral longitudinal trunk. ANNAIS E. S. A. VoL. III, PLATE VI.. rears 20 : — = - S ip ll / = ,y A = } ; — | ‘fd / SS . 7 brs, {2 \ . Pp ea \ \ \ ‘ a = = \ — - CS EAN ; \ (Rena ir ' MRR aot ey ! \ i ORAS PPS STA Dh +s ' ——— . rx) + aeons te WA } : Gare a ie Wz aa WIM ite. rent, Se somes rer eee Ba es ea F. X. Williams. VoL. III, PLATE VII. Williams. XxX, F. Vou. III, PLATE VIII. ANNALS E. 5S. A. \f- Sbiracle zi yan F. X. Williams. SOME NEW FACTS ON THE BIONOMICS OF THE CALIFORNIA RODENT FLEAS. M. Bruin MitzMain, B. S., University of California. The thorough investigations which the United States Public Health and Marine-Hospital Service are at present undertaking in suppressing bubonic plague in California presented the writer with an opportunity to study the parasites associated with the transmission of the disease among rodents. In this contribution we have not taken into account the fac- tors involved in the epidemiology, but have restricted the scope of the paper to a discussion of the general habits and transfor- mations of the California rodent fleas. Many of the observations on the bionomics of the fleas are quite different from those re- corded by the investigators of the English Plague Commission in India. This is due in a measure to difference in climatic condi- tions, to which fleas as a rule are easily influenced; and the differ- ence in the species of flea. The English workers confined their experiments to one species, namely, the Loemopsylla cheopis Roth. We have considered the rodent fleas generally, giving most attention to the squirrel flea C. acutus Baker. FLEAS IN RELATION TO THEIR ENVIRONMENTS. We have made an effort to rear fleas in the laboratory by attempting to duplicate conditions found to exist in nature. The human flea, P. arritans Linn, was found to develop very satisfac- torily in a medium composed of floor sweepings taken from the cracks in the floor. The squirrel fleas and rat fleas were found to develop well in material taken from the nests of their respective hosts. We have experienced no difficulty in raising these para- sites by placing animals covered with fleas in cages with a bed- ding of sawdust; and the only provision necessary for complete development was the addition of fresh sawdust to prevent the accumulation of too much moisture. An experiment was made to determine in what media fleas away from the host would survive longest. For this purpose we tested various materials shown in the following table: 61 62 Annals Entomological Society of America _[Vol. III, LENGTH OF LIFE IN VARIOUS MEDIA. C. acutus. DrySand| Moist at Sawdust | Moistened Days Re- | with Sand Dry eine Moistened | Dee Sawdust moved from| Squirrel| from Pie ac with S; na Os with Host Drop- | Squirrel | * Nace Horse % ; Wheat pings Nest ae Serum Grains pean it 2 DM so se | i OUR 10 F. 4M. 6F. 4M.6F. 4M.6F. 7 2M. 5F. 4F. |Alldead| 1M.3F. 4F. IME or 9 LF; 2F. 26H ayo Mien: 10 2M. 5F./All dead All dead All dead 1M.4F. 14 ei Bey Ls 1M.3F. 16 LANES a 17 All dead DMs seh: 20 3 F. 25 a 26 All dead *M—Male; +F—Female. The controls 5 M. 5 F. were all dead on or before the seventh day. In the medium of moistened sawdust mixed with a few wheat grains it was found that the wheat sprouted in the sawdust and held sufficient moisture to provide a suitable condition for larvae as well as for adults. The mould which formed in a short time did not seem to affect the insect life. It is seen that the fleas did not fare well in the medium of dry sand. The sand was mixed with clay dust which would rise whenever the fleas hopped in the vial. As a consequence death resulted presumably by stoppage of the spiracles. It will be seen from a survey of this table that fleas taken from the natural host may be kept alive without food for a considerable time. The medium of moistened sawdust with a few grains of wheat seemed to answer the moisture requirements for flea life. LOCOMOTION. The only literature that has come to our notice on the jump- ing powers of fleas appears in the Journal of Hygiene, 1906, Vol. 6, p. 464. Here wehaveanote: ‘‘It had previously been found that a rat flea could not hop farther than five inches.” The species in question in these experiments was the L. cheopis, which is found to the extent of 99% on the rats in India. IgIo] California Rodent Fleas 63 It seemed desirable to learn the jumping abilities of the com- mon California species. An attempt was made to determine the distance upon a horizontal plane, as well as the height that the insect could jump. In the broad jump a few experiments were conducted with P. irritans, the most active of our California fleas. In this species we have found the jumping distance varies considerably with the nature of the container from which the insect is observed, e. g., a foothold of wood enables the flea to jump a greater distance than one of glass. Since the zrritans is found to predominate on floors of houses, a surface of wood was selected as the footing 1n our tests. In one experiment, five specimens were permitted to jump at will and the jumps of each were recorded. The mean average of ten jumps of each speci- men gave a distance of 7 3-10 inches. The longest jump recorded was 11 inches; this was made by a female. In another experi- ment, a female, which was starved for five days prior to the test, made four jumps of respectively 10 5-10, 11, 12, and 13 inches; averaging 11 5-8 inches. Thirteen inches was the longest jump recorded on a horizontal plane. A jump of 15 7-8 inches was made downward at an angle of thirty degrees. This specimen was then permitted to feed fifteen minutes on the arm of an attendant, after which it was carefully returned to the container, and its jumps for five minutes recorded. The longest jump after feeding was 12 inches. The jumping powers of squirrel and rat fleas were tested in this manner: Three hundred and seventy-five live squirrel fleas (C. acutus) and one hundred and fifty live rat fleas (C. fasciatus and L. cheopis) were placed in two distinct lots in fifteen open specimen vials in a water bath and left undisturbed for two days. At the end of the period, the vials were examined, the water bath and the surroundings being carefully inspected. No fleas could be detected in the water bath or in the vicinity; the fleas in the vials were counted, the original number being present. They were apparently as active as when removed from their hosts. The containers were cylindrical vials 3 3-8 inches in height and I 3-16 inches in diameter. The same test was tried in open shell vials of the next size smaller (3 1-8 x 1 inch) and the fleas were found jumping into the water bath. L. cheopis, the rat flea, was tried for its jumping power. It had been previously observed that members of this species were unable to leap out of open shell vials 3 3-8 inches in height. It 64 Annals Entomological Society of America [Vol. III, was thought desirable to permit the flea greater latitude making the trial fairer and more practical. Twenty fleas (eight males and twelve females) were placed in a rectangular porcelain dish, the dimensions of which were 11 7-8 inches by 7 5-8 inches by 2+ inches deep. A lght sprinkling of moist sand was placed in the bottom of the dish to provide a firm footing and the sides of the dish were extended to the height of 44 inches, enclosing the container entirely by strips of ‘‘tangle-foot”’ paper. At the end of twenty-four hours the dish with the fleas was examined, and it was: observed that several fleas had leaped high enough to become entangled on the adhesive paper. The position of these was noted, the distance from the bottom of the pan was meas- ured. The entangled insects were pried off, the sticky material ‘dissolved with alcohol and the insects examined microscopically. Five specimens were collected, two males and three females. ‘The locations on the adhesive paper relative to the base of the dish were as follows: 1 F. 24 inches 1 F. 3 1-8 inches 1 F. 2 9-16 inches t 1 M. 2 5-8 inches 1 M. 3 1-16 inches A census of the fleas remaining in the bottom of the dish gave 6 M. and g F. showing that none jumped over the surrounding paper, and that all jumping over 24 inches were embedded in the adhesive paper. The jumping powers of P. zrritans were further tested, twenty fleas of this species being placed in a rectangular museum jar, the inside dimensions of which were 2} x 54 x 7? inches depth. The glass lid of the jar was coated with ‘‘tangle-foot’’ and the sides of the jar within two inches of the top were likewise coated, with the same material. The following morning the jar was examined and two specimens were seen embedded in the “‘tangle- foot’? on the lid. The inside height of the jar being 7? inches from the base gives then this height as the perpendicular jump. Concerning the fleas ability to walk upwards on glass, we have noted that they cannot climb to any considerable distance. A ereat number of counts were made, the greatest distance observed being ? inches. The climb appears very laborious and in all cases the flea dropped to the bottom of the jar after a few efforts. We have observed the manner in which these insects can find their way about upon their natural hosts. They pilot their paths among the dense hairs, walking on the flat of their tarsi, seeming to shuffle along. | When on the animal they seldom hop about, TgIo] California Rodent Fleas 65 unless disturbed or unless the host snaps at them or scratches when unusually annoyed. They hop freely when jumping from the animal to the ground or vice versa, or from one host to another. When a flea is cornered, that is, when it experiences difficulty in passing an obstruction, it proceeds like a swimmer using the side stroke. The parasite drops to its side and locomotion ensues by a vigorous sweeping movement of the legs, almost entirely by the use of the hairs and spines, especially through the medium of the powerful spines of the tibia, when it ambles along on the flat side in a striking manner. The spines of the leg seem to be peculiarly adapted for this side motion: it is in this fashion that the nimble parasite manages to become so very elusive. We refer especially to the P. irritans. This may be tested by holding a live flea between the thumb and forefinger and unless you chance to be a flea trapper of long and painful experience the ingenious parasite will surely escape. CONSIDERATION OF COLOR ATTRACTION. We havea little experimental evidence on the question of attrac- tion of these insects towards color in animals. It is the prevailing opinion that white animals attract the greatest number of fleas. We have to report a few tests of color attraction. The material used in the first experiment was six guinea pigs showing extreme variations of color: three of these were pure white and three pronouncedly dark. These were placed in a large cage, the bedding of which was infested with squirrel fleas. After forty- eight hours the guinea pigs were removed simultaneously and a census of the fleas on the individual animals was made immediate- ly. Fleas and animals were anaesthetized at the same time, the white guinea pigs yielding respectively ten fleas, five fleas, and eleven fleas. The black guinea pigs yielded eleven, seven and four fleas; the last number came from a mixed black and white guinea pig. A summary of the fleas from the three white guinea pigs gave twenty-six fleas, or an average of about nine each; the two dark guinea gigs giving eighteen, averaging nine each, and the black and white giving four fleas, all of which were of the species, C. acutus, the squirrel flea. A second experiment was as follows: Four guinea pigs were placed in a large container, which was swarming with squir- rel fleas. After three days the animals were removed as before, and a census of the parasites was taken: one brown and black 66 Annals Entomological Society of America [Vol. III, euinea pig yielded eighty-nine fleas; another brown and black guinea pig yielded fifty-seven fleas; a pure white guinea pig yielded fifty-seven fleas; and the other white guinea pig yielded sixty fleas. These guinea pigs were kept together in the open vessel under the same conditions of light and temperature. One experiment was conducted by using guinea pigs as traps for fleas in an infested house. The results shown are relatively meager due to the previous use of adhesive paper for flea trapping. In this test six guinea pigs were permitted to wander for twenty- four hours through the basement of the house. The fleas col- lected from the guinea pigs were P. zrritans found on the hosts as follows: Color of animal Number of fleas Black guinea pig 1 White guinea pig 1 Black guinea pig 1 White guinea pig 1 Black guinea pig ) White guinea pig 0 It is obvious from these experiments that color does not exert the influence generally claimed for it. The white animals are no more attractive to the fleas we have used than are the dark colored ones. TRAPPING OF FLEAS WITH MEAT AS A BAIT. An idea prevails that fleas can be attracted and trapped on account of their predilection for the odor of fresh meat. This idea was put to test in an experimental way. On the 23rd day of August, 1909, during a season in the year in which fleas were extremely abundant, a vacant house, which was found to be flea infested, was chosen for the experiment. To give an idea of the abundance of the parasites, two attendants, who had occasion to enter the first floor of the dwelling, remained for the period of five minutes and emerged covered with fleas. Approximately two hundred fleas were taken from their clothing and persons. The materials used for the experiment were twelve sheets of “tangle-foot”’ fly paper, distributed in pairs among three rooms of the house. One sheet of each pair was supplied with a small fresh piece of cow’s liver. Care was taken to distribute the sheets in such a manner that the influence of light would be the same for each pair of sheets in the series. The sticky fly papers were left undisturbed for a period of three days, then collected and exam- ined. 1g10] California Rodent Fleas 67 A similar experiment was conducted at a later date, December sth, rgo9, when San Francisco dwellings were still flea infested. The sheets of fly paper were left in the basement of the house for twenty-four hours. The fleas in both experiments were removed by dissolving the “tangle-foot”’ in alcohol. Nine parasites which were all identified as Pulex irritans, were dis- tributed on the sheets as follows: Total number of fleas trapped. Experiment 1 | Experiment 2 Adhesive paper, Adhesive paper || Adhesive paper Adhesive paper Pair| - with meat | without meat with meat without meat iB 0 0 | 2 4 Pa, a 2 2 | 1 3 Si 2 1 6 4 4 | 0 3 47 49 5 | 6 12 | 1 0 6. | 1 1 | 4 1 Totals| 14 19 61 61 The table shows that according to our experience, though the data is limited, meat used as a bait under the conditions stated does not exert any special attraction for fleas. COPULATION AND EGG LAYING. The most prolific of the rodent fleas, Ceratophyllus acutus was set aside for studies in mating. The fleas were observed in nature in relation to the host, then microscopically. For the purpose of close observation, a live ground squirrel (Citellus beecheyi) was placed in a mouse jar, where it was kept for a few hours under surveillance. It was noticed that when the animal’s body was pressed against the glass that the pelage parted in furrows made by some active objects close to the skin. It was apparent that the squirrel was harboring a large number of active fleas, which occasionally came to view mounted on the hairs of the host. The attention of the observer was attracted by a number of fleas which appeared gigantic in size. These proved on close inspection to be paired. Approximately 1 to every 3 of the total number of fleas infesting the squirrel appeared thus in copulation. The female of the pair covered almost entirely the body of the 68 Annals Entomological Society of America [Vol. III, male, which was quite lost to view, only the hind legs projecting beneath the abdomen of the female. The pair in copulation was observed never to feed, moving about as one insect; and anoma- lous as it appears, the member on the ventral side, the male, con- tributed no effort in the locomotion during the nuptial wander- ings. His fore and middle legs doubled under the sternum, the hind pair extending stiffly as though paralyzed beneath the abdomen, where they were held securely by the tarsal spurs of the female interlocking with the tibial spines of the male. The locomotory function was accomplished by the female, the front and middle legs of which were used freely in walking; by releasing temporarily the entwining spines of the hind legs she was able to leap at will. In this manner, the female moved about on the body of the host as though unhampered by its anchored mate. Seven pairs of these fleas were collected from the live squirrel with the aid of a camel’s hair brush, placed in cell slides and covered with glass slips. They were observed under the low power microscope, then kept in separate vials, one to several hours until the function was completed. As observed microscopically, the male adheres to the female by the pseudo-joints of the terminal segment of the antennae. This attached appendage is extended and pendant, interlocking by its hairs the bristles on the ventral side of the second abdomi- nal segment of the female. The hind femora of the latter grasps the male by the head at a point anterior to the antennal -groove. The head of the male is held 1n position between the inner sides of the meta-femora of the female. The tarsal spurs of the female entangle the spines of the hind tibia of the male. The smaller of the pair thus suspended assumes the passive role moving at the volition of the female. The copulatory act is manifested in the male by the watch- spring like action of the spiral, which, coiling and uncoiling inter- mittently extends and withdraws the terminal apparatus. The male claspers are seen on either side of the vaginal groove of the female directly below the anal flap. The male makes a dis- tinct concavity in the caudal portion of the back; with the strong action of the muscle attachments the plate of the penis is made very elastic. ‘The spasmodic pulsations of the curled up abdomen are followed by a current of bubbles passing through the genitalia of the male upwards to the female. The spermatheca of the latter 1910] California Rodent Fleas 69 is faintly discernible during the act as it swells and ange sa The length of time involving coition in the fleas under observation varied a trifle less than two hours to four and a half hours. The movements preliminary to the act were not noted. The male makes the first effort to disconnect or disentangle the pair, In one instance, the male attempted to free itself by press- ing vigorously the mid and hind tarsi against the hind tibia of its mate. Simultaneously, the antennae were torn from the recumbent spines of the abdominal segments of the female. The abdomen was straightened, relaxing the clutch of the claspers; the spiral contracted releasing the intromittent organ, and a final kick of the hind legs gave the male its liberty. The female was seen on ten distinct occasions to approach the male, and the latter repelled the advances of its larger companion ; at each con- tact of the female, it would jump in the air, attempting to remain on the opposite side of the vial. Mammalian blood appears essential for fleas to partake of normal functions of copulation and oviposition. In our exper- ience insects kept constantly in jars and reared from cocoons never having been fed upon a host have not been observed to copulate or oviposit. In thirty specimens of P. irritans taken from a house which had been vacant for six weeks we found, after three days observation, that the fleas were perfectly healthy; and although females predominated no eggs were found at the end of this period. As a control, six females of this species collected from human hosts when kept in separate vials laid eggs normally, depositing from 5~12 in each instance. Four experiments with C. acutus, the squirrel flea, have given results similar to the foregoing, namely, that this species when reared from cocoons and kept eeived in jars at room tempera- ture has not been observed to mate or lay eggs. In twenty-five paired specimens of several species kept under observation it appears that the male does not long survive the act of mating. It dies even before the female has laid its first batch of eggs. When fertile females are kept under artificial conditions the eggs are laid in one laying in a period beginning two hours after copulation and extending to a maximum of thirty-six hours. When deprived of food the female has never been observed to oviposit after this length of time. We may note here that when used experimentally the female is invariably longer lived. Experiments to determine length of 70 Annals Entomological Society of America [Vol. III, life with human blood diet show that female fleas of all species outlive the male by several weeks. This is doubtless true also under natural conditions, where we find in collecting fleas from the host that the females predominate markedly. The eggs require optimum conditions of temperature and mois- ture for hatching. They have never been found on the host’ except in one instance. In this case a dog was used in the laboratory for supplying fresh fleas; this was done by placing the canine on sheets of paper which in a few hours were littered with a large number of flea eggs. These were laid loosely on the host by the fleas, the eggs falling to the paper, where they were collected. Flea eggs have never been found on man, and if present, would not hatch under normal conditions of the body temperature. We have found that keeping eggs in the incubator at blood heat is sufficient to prevent hatching. If the eggs were laid on the host we would certainly expect to find them on the squirrel, on which animal fleas are most abundant. We have taken from two squirrels, respectively, 225 C. acutus and 376 of the same species; but in no instance even where fleas are present in such large numbers have we been able to find eggs even after carefully combing the host. The C. acutus is by all means the best criterion in this matter, since we have found that it lays more eggs than any other of the rodent fleas. The eggs are laid singly in small clusters, and may be viscous as in P. awritans, C. fasciatus, L. cheopis, C. acutus; or dry as in C. canis and C. muscult. The former adhere to the medium in which they are laid, and the eggs of the last two species are laid loosely, so that they roll when shaken in the vial containing them. Eggs may be laid while the insect is still under the influence of an anaesthetic; when covered by a glass slip; and when exposed to strong sun light. It appears to be the first impulse for the female to lay its eggs when removed from the host and placed under artificial conditions. The great majority of the eggs obtained were laid on the first day, beginning almost immediately after the fleas were captured. It is a common observation that many females with their abdomens distended lay their eggs as soon as the vial is closed over them. The number of eggs laid at one laying by different species, varies from 3 to 18; the rat fleas averaging 6 and the squirrel fleas lay as many as 18. Tg10] California Rodent Fleas 7 THE PROCESS OF HATCHING. C. acutus. Six eggs laid while the females were kept under observation were examined from time to time until hatched seven to nine days later. One egg, was observed microscopically during the entire process; the other five eggs used as controls were examined occasionally. On the seventh day of incubation at room temperature the premonitory signs of hatching were discerned ina very faint rising and falling of the exochorion (outer shell layer) on one side of the shell. During the night, seven hours later, the movement grew more extensive, the pulsations becoming quite pronounced, caus- ing the egg to shift slightly from its position. On the morning of the eighth day a deep gash was observed in the side of the egg. The gash is made by the egg opener, a wedge-shaped, horny, claw- like structure on the dorsal side back of the head of the embryo. This is operated so as to hew through the resistant shell by a series of slits or gashes. The initial gash increases slowly in length, encircling the egg within an hour. In the course of a few hours, seven successive gashes are made, the location of these is quite constant; two-on each side Othe. egs and three at the center, the middle of the latter being the most distinct. The young worm produces these slits through the chorion (inner skin of shell) by turning with its dorsal side against the shell, striking with the egg pick first against the base of the egg, rising on its hind prolegs and striking rapidly as it climbs upwards. Prior to each stroke the egg pick is poised deliberately, the weight of the head accelerating the blow; this is augmented occasionally by a lateral shaking of the head. When the last slit encircles the shell, the embryo is at the most active stage, it effects a complete rotation in its shell at intervals of ten minutes. The gashes make the shell appear as though cut into ribbons. The rotary movement gradually subsides until a complete turning consumes twenty minutes, and almost imperceptibly there follows a lull. An inactive almost comatose condition prevails for a period of about seventeen hours. After the resting period, it appears that the egg pick is no longer functional; no new gashes are made, the embryo seeming contented to concentrate its energies against the middle gash. 72 Annals Entomological Soctety of America [Vol. III,. This is enlarged by a puffing and dilating of the head; through the semitransparent shell is seen a constant bubbling and a subse- quent dilatation of the cuticle. On the third day of the hatching process, a strong movement. of the embryo gives decided evidence of the progress of hatching. Immediately behind the egg pick, a triangular slt appears through which bubbles emerge to the surface of the exochorion; thus the amnion (the embryonic cuticle) makes its initial appear- ance through a crack in the exochorion of the shell. The amnion splits longitudinally adhering to the chorion and bulging out as the insect struggles. It is pushed out of the chorion as the abdominal segments are projecting. The emer- gence is furthered by the young maggot pressing the head against its tail causing the middle of its body to bulge through the central gap of the shell. The amnion becomes noticeably darker as it is exposed to the air; it is now a light brownish- vellow. The pressure of the body against the shell forms nearly a round hole through which the larva eventually emerges. The opening is enlarged by pressure of the head against the tail, raising the body like a hoop, causing an arch or a hump to appear with dorsal side outwards. The amnion sheds slowly on either side from the middle of the arched abdomen ventrally and with a movement of fluid beneath it cracks across the abdomen, peeling and wrinkling as the segments telescope. With a final vigorous bubbling and wrinkling the amnion sheds off, the moulted skin falling on either side, exposing the quite colorless cuticle of the maggot roughly wrinkled and bristling with slender hairs. These hairs which at. first appear transparent turn grayish when exposed to the air. When the tail of the larva has been torn loose from the shell, the head and thorax are still imprisoned within the egg, necessi- tating a maneuvering by waving its tail in the air, twisting and squirming while standing on its head. The abdomen doubles up ventrally and finally the young larva supported on its tail extricates the head by a violent shaking. When the head is torn from its fastenings, it is found that the amnion has been holding it within the shell. The adhering membrane is cast out when the shell is shaken off. The larva has now fully emerged, the colorless. cuticle has turned grayish and the slender threadlike bristles have: assumed an iridescent hue. TgI0] California Rodent Fleas a The larvae upon hatching busy themselves immediately in the quest for food. They experience little difficulty in locating it, for at the time of birth a supply of food is found upon the egg shell. Here they feed from the first on the tiny blood pellicles surrounding the egg shell; this the mother furnishes when the egg is laid. When the last of the egg pellicles of blood are con- sumed, the insatiable worms look about them for other sustaining morsels. The dejecta of the adult flea seems to provide the desired ingredients. The young larvae feed ravenously on the bloody deposits, apparently satisfied to pass the first few days on this unique diet. They can subsist entirely on the bloody fragments (flea faeces) when no other food is available, for a period of five to six days. THE REACTION TO LIGHT. The larva is positively heliotropic up to the stage of the initial moult. The more advanced sluggish larvae are repelled by the light. This is seen when examining the flea breeding cages a slight stirring of the nesting material attracts to the surface the tiny very active larvae. If the older larvae are desired, it is found necessary to nearly invert the container. Prior to the final moult when the larva is in readiness to pupate it can be seen almost invariably along the edges at the bottom of the box, where the greatest number of cocoons are brought to view. TROPIC INFLUENCES IN THE ADULT FLEAS. Rodent fleas are negatively phototaxic (repelled by light) to a very striking degree. The first impulse seems to be to seek protection from the light. This is seen in combing a squirrel or rat recently killed; the fleas will retreat constantly to the under- side, always in the direction away from the light. When shaken off, they return to the shadow of the host; in numerous instances even when the animal was dead for a period ranging from 24 to 50 hours the fleas when shaken off would seek the host and bury themselves under the hairs away from the light. When a number of live squirrel fleas and rat fleas were placed in an open test tube and held horizontally with the operator’s thumb covering the mouth of the vial and the bottom held against the window, the fleas crowded towards the open mouth in the direction of the thumb, bounding away from the window in an excited manner. When the tube was reversed with the open mouth towards the window, it was found unnecessary to v4 Annals Entomological Society of America [Vol. III, plug the mouth of the tube as the fleas did not attempt to jump out when given an opportunity to do so. Even when placed within a half inch of the open end of the tube with the head of the insect turned towards the light, the fleas reversed and jumped towards the closed end of the tube. This was repeated by tilting the mouth downward to offer an easier exit through the open mouth; but even this inducement did not influence the fleas, which invariably sought the closed end of the tube in the shadow. This was tried with squirrel and rat fleas as well as with human fleas and always with the same result. LONGEVITY OF THE CALIFORNIA FLEA. The great life of our native parasites seems to be anticipated at the very incubation. Eggs are laid at all times. We have observed oviposition during every month. The broods appear to be constant and the number irregular. The incubation stage of our Eastern forms taken from the observations of Pergande consumes 2—4 days ordinarily. Two days is given as the incuba- tion period for the cheopis in India. Our experience with this species is a duration varying from 9g to 13 days at room tempera- ture. When eggs of this species were subjected to identical con- ditions in which wild rats were caged in the laboratory basement (at a temperature of 20° to 25° C.) the length of the egg stage was 1g days. Larval life is cited by Eastern and English authorities as a minimum of 8 daysanda maximum of 24 days. The larval stage of our California fleas under laboratory conditions is never less than 28 days, often 30 days and sometimes longer. Pergande found that the pupal or cocoon stage of the Pulex irritans varied from five to seven days in the summer months at Washington. Other authors working with this species give twelve days during summer months. We have observed several instances of cocoon life during the warm days of September. Thirty days appear to be spent in this stage. The entire life cycle including adult life is given by several authors as four to six weeks. The British Indian Plague Com- mission gives as the time necessary for the completion of the cycle of development (in the case of L. cheopis) that 1s from the egg to the imago, as 21 to 22 days. The following is given as a type of the life cycle observed in C. acutus. This specimen was kept under observation from the Igo] California Rodent Fleas 75 moment the egg was laid by its parent. An hour after the egg was laid it was placed in a small vial with a little sawdust, sand, wheat grains and squirrel hair. It was allowed to develop in this environment until matured. Stage of development Date Age of flea Egg laid May 4th, 1909 : Hatched May 12th, 1909 8 days Moulted (second stage) May 18th, 1909 14 days Moulted (third stage) May 28th, 1909 24 days Cocoon June 9th, 1909 36 days Adult July 10, 1909 67 days Alive (never having been fed) Aug. 11, 1900 99 days The English workers in India ascertain the length of time which adult fleas live on rats as 41 days. The longest life of this species on an exclusive diet of human blood was observed to be 27 days. We have managed to keep this species (cheopis) alive for a period of 36 days by feeding on man. The length of life without feeding was also noted. This period varied in the experi- ments of the English Commission according to the environments in which the fleas were maintained. In the absence of liquid food supply, fleas could live in bran for six days, in gunny sacking a similar time, and in sand with moist cowdung for 13 days. From onr observations we have found the majority of fleas of all species to die in five days unless a moist medium was provided. As we have stated above rat and squirrel fleas may be kept alive for a considerable time when moisture is provided in some form. It is interesting to note that fleas which have never tasted animal food, having emerged from the cocoon and kept under the same conditions in similar material as fleas taken direct- ly from the host, will prove longer lived. A number of specimens of Ceratophyllus acutus removed from a ground squirrel and kept in moistened wheat grains and sawdust lived for 26 days. An equal number of fleas of the same species bred from cocoons in the laboratory were kept without a host in a similar medium. One male lived 38 days and a female lived for 65 days. In a series of experiments in which fleas taken from healthy rats and squirrels were fed daily on the arm of a man we attempted to determine the maximum longevity of these para- sites. The fleas were placed individually in open test tubes and at feeding time the tubes were inverted over the arm of one of the laboratory attendants. Fleas were thus applied daily for a period of from 5 to 15 minutes, but only the actual feeding time was recorded. The average was about five minutes. 76 Annals Entomological Society of America [Vol. III, A Loemopsylla cheopis was fed for 35 days, escaping on the 36th day. One C. acutus died after 58 days and another one at the end of 51 days. The common brown rat flea C. fasczatus, proved to be the most persistent feeder of themall. Unfortunate- ly one of them was permitted to escape after feeding on its induced host for 63 days. Another of the group suffered no ill effects from its enforced diet for 98 days; and the sole survivor had been nourished by its foster host since its removal from the rodent host for a period of five months. The experiment was discontinued, but the parasite survived a week longer in a starved condition. LONGEVITY RELATIVE TO SEX. We have noted the relative longevity of the sexes under the conditions of experiments in which fleas were fed on human blood alone. We shall take for consideration the two species common to rats, L. cheopis and C. fasciatus and the predominant squirrel flea, C. acutus. Four tests with C. fasciatus gave the following data: C. fasciatus. Six males of this species averaged 84 days, the maximum life being 17 days. Fifteen females gave an average of 32 4-5 days with a maximum of 125 days. The two unfed controls (male) of this series lived for 3 days, and the two control females lived for 5 days. L. cheopts. Two tests with a total of seven males of this species gave an average of 10 1-7 days and a maximum period of 15 days. Three females lived 28 1-3 days as an average and a maximum period of 49 days. The two unfed male controls lived 5 days and in four females the average was 54 days and the maximum 7 days. C. acutus. One test with this species furnishes the following data: Three males averaged 11 days, and gave a maximum of 11 days. Five females averaged 15 1-5 days. The longest life was 53 days. The unfed males averaged 3 days and showed a maximum of 4 days. The female control lived 5 days. TQIO| California Rodent Fleas a7 The greatest length of life of a male of any species is seen to be 17 days; and the term of life under these conditions for a female was 125 days, somewhat over 4 months. In these tests for longevity, it should be borne in mind that no attempt is made to arrive at the initial age of the fleas, but the time is reckoned from the day of removal from the host. An effort was made to determine the length of adult life of one species, C. acutus, by feeding the insect newly emerged from the cocoon. Ten specimens emerging within a few hours of each other were fed on the same day on human blood. One specimen, a female, lived for sixty-four days; at which time the experiment was discontinued. The unfed controls, as well as the specimens given a human blood diet, were kept in ordinary test tubes at room temperature. The activities of these fleas may have been influenced by changes in the temperature and the absence of moisture in the tubes. NOTES ON THE FEEDING PROCESS. There is a remarkable degree of variation in the feeding habits of the different species of fleas. We have not attempted except in a superficial way, to study the idiocyncracies of the rodent fleas in regard to the biting of their normal hosts. We have, however, quite thoroughly observed the manner of biting under experimental condition with man as a host. Without taking into account the attraction or repulsion which may be exerted towards man as a host, we shall consider the more striking features > the biting of the parasites. We find that the species do not all attack with equal avidity. Pulex irritans the ectoparasite of man, 1s insatiable in its blood craving. It differs in its relation to man in being more fastidious in its feeding than the rodent fleas. Although its bite is painful, it does not voluntarily feed in one spot for any great length of time. The Pulex irritans dif- fers from all other species (hundreds of specimens of which were tested on human hosts) in that it squirts blood per anum during the act of biting. The L. cheopis and C. fasciatus, the normal rat parasites, are found to bite man with equal readiness and will live about the same length of time when fed on human blood. Their biting is well defined and effective, but not nearly so painful as that of Pulex irritans, nor so prolonged as the common squirrel flea, Ceratophyllus acutus. A specimen of C. acutusswhen starved for 78 Annals Entomological Society of America [Vol. III, several days has been observed to feed on man uninterruptedly for a period of nearly one hour (59 minutes) at one insertion of its proboscis. The bite of the Ctenopsyllus muscul1, the blind flea of the mouse, 1s the feeblest we have had inflicted from any flea tested. The short weak piercing organs of this species makes a puncture, which is scarcely perceptible. In observing the length of time this flea bites, it is necessary to depend as a guide on the distention of the abdomen with blood rather than the prick of the mandibles. The muscult seems not able to adapt itself as an induced parasite of man. From a few experiments it appears to live not longer than five days on a human host. The Ceratopyyllus acutus which proves a very ready parasite of man makes its attack even without inducing experimentally. Our data compiled from reports and collections of squirrel hunt- ers shows that this species, as well as the other common squirrel flea, Hoplopsyllus anomalus, will bite man when exposed to their attacks. The unusually long rostrum in this flea is presumed to be the cause of its prolonged feeding at one insertion of the mouth parts. This principal is not unprecedented judging from a study of numerous parasites in the insect world. The following is the description of the method of feeding observed in C. acutus. This method is typical. The flea when permitted to walk freely on the arm selects in a few minutes a suitable hairy space where it ceases abruptly in its locomotion, takes a firm hold, with the tarsi, projects its proboscis and prepares to puncture the skin. A puncture is drilled by the pricking epipharynx, the saw- tooth mandibles supplementing the movement by lacerating the cavity formed. The two organs of the rostrum work alternately, the middle piece boring, while the two lateral elements execute asawing movement. The mandibles, owing to their basal attach- ments are, as is expressed by the Journal of Hygiene, Vol. 6, No. 4, p. 499, ‘‘capable of independent action, sliding up and down but maintaining their relative positions and preserving the lumen of the aspiratory channel.’ The labium doubles back, the V-shaped groove of this organ guiding the mandibles on either side. The action of the proboscis is executed with a forward move- ment of the head and a lateral and downward thrust of the entire body. As the mouth parts are sharply inserted, the T9to] California Rodent Fleas 79 abdomen rises simultaneously. The head and middle legs are elevated resembling oars. The fore legs are doubled under the thorax, the tibia and tarsi resting firmly on the epidermis serve as a support for the body during feeding. The maxillary palpi are retracted beneath the head and thorax. The labium con- tinues to bend, at first acting as a sheath for the sawing mandibles, and as these are more deeply inserted, it bends beneath the head with the elasticity of a bow, forcing the mandibles into the wound until the maxillae are embedded in the skin of the victim. When the proboscis is fully inserted, the abdomen ceases for a time its lateral swinging. The acute pain of biting is first felt when the mandibles have not quite penetrated and subsequently, during each distinct movement of the abdomen. The swinging of the abdomen gradually ceases as it becomes filled with blood. The sting of the biting becomes gradually duller and less sensitive as feeding progresses. The movements of the elevated abdomen grow noticeably feebler as the downward thrusts of the springy bow- like labium become less frequent. As the feeding process advances, one can discern through the translucent walls of the abdomen, a constant flow of blood, caudally from the pharynx, accompanied by a peristaltic move- ment. The end of the meal is signified in an abrupt manner. The flea shakes its entire body, gradually withdraws its proboscis by lowering the abdomen and legs, and violently twisting the head. When starved for several days, the feeding of the rat fleas is conducted in a rather vigorous manner. As soon as the proboscis is buried to full length, the abdomen is raised and there ensues a gradual lateral swaying motion, increasing the altitude of the raised end of the abdomen until it assumes the perpendicular. The flea is observed at this point to gain a better foothold by advancing the fore tarsi, and, then gradually doubling back the abdomen, it turns a back spring, with extreme agility, nearly touching with its dorsal side the skin of the hand upon which it is feeding. Meanwhile, the hungry parasite feeds ravenously. It is interesting to note the peculiar nervous action which the rodent fleas exhibit immediately when the feeding process is completed, or when disturbed during the biting. Even while the rostrum is inserted to the fullest, the parasite shakes its head spasmodically, in a twinkling the mouth is withdrawn and the flea hops away. 8o Annals Entomological Society of America [Vol. III, POSSIBLE VITAL CONSIDERATIONS INVOLVED IN FEEDING HABITS. We have previously noted that rodent fleas can live in a starved condition, away from the host, during a period of three to ten days, when kept in dry test tubes; as long as twenty-eight days when a suitable moist medium is furnished. The long periods of starvation appear not to affect the vitality of the parasite to such an extent that the ability to feed is impaired. We have recorded instances in which a specimen of the squirrel flea H. anomalus starved for sixteen days, and several specimens of C. acutus starved for twenty-seven days had sufficient energy’ to feed 10-15 minutes when an arm was placed in the breeding jar. These facts lead us with others to accredit in a measure the claims of older authorities on plague, who contended before the flea theory was recognized, that clothes and baggage of an infected community harbored the germs of plague, which re- main viable during long period of time. Modern writers have called our attention to the plausibility of infection in clothes and baggage due to the agency of fleas, which may be transported from infected communities. Bannerman (Journal of Hygiene, Vol. 6, 1906, pp. 189-191), writing on the possibility of the spread of infection by means of clothes, brings forth several instances in which conveyance of infection by clothes seemed the most likely means of introduction of plague in some villages in India. Plague infected rat fleas secreted in the clothes en route are held responsible for the transmission. Bannerman cites an instance of plague transmission through clothing where as much as ten days elapsed in the transportation of clothing removed from a plague infected victim and worn by a relative of the deceased, who in turn contracted the disease. It has not been our intention to enter into a discussion of the flea from the standpoint of plague epidemiology; it is desired merely to indicate the possibility that starvation of infected fleas may not eliminate the danger of transmission. ‘The presence of plague bacilli in rodent fleas as pointed out by the English Plague Commission (Journal of Hygiene, Vol. VIII, p. 261) does not appear to affect the rate of mortality in these insects. 1910] California Rodent Fleas 81 SUMMARY OF RESULTS. 1. The breeding of fleas under laboratory conditions can be carried on quite satisfactorily when there is furnished a medium simulating the nest of the host. It will be seen that fleas taken from the natural host can be kept alive without food for a con- siderable time, providing the proper moisture conditions are maintained. A medium of moist sawdust with a few grains of wheat prolonged the life of these insects to twenty-six days. Those without this medium under the same conditions, died on the seventh day. 2. Rat fleas namely, L. cheopis, C. fasciatus can not jump higher than 3 1-8 inches. P. writans, the human flea, was found to make a perpendicular jump of 7 3-4 inches and as far as 13 inches on a horizontal plane. 3. Experiments with black and white guinea pigs show that in relation to color attraction white animals are no more attractive to the fleas than are the dark colored ones. 4. Experiments with adhesive paper baited with meat, demonstrate that fleas can not be trapped by the odor of meat. 5. In the mating of fleas, the female takes the initiative and the male assumes the passive role. 6. Fleas reared from the cocoon kept without a host have never been observed to copulate or oviposit. 7. The eggs are never laid on the host. Oviposition takes place within thirty-six hours after the female is removed from the host. | 8. The process of hatching consumes a period of 3-4 days. The embryo breaks through the shell by means of a special egg pick, which is instrumental in the emergence. g. The larvae can live on the bloody egg pellicles and the dejecta of the parent for a period of 5-6 days. to. The larvae are positively heliotropic in the early stages and repelled by light in the later stages. t1. Adult fleas are negatively phototaxic. 12. The California rodent fleas have a greater life in all stages than fleas of the Eastern United States and India. The 82 Annals Entomological Society of America [Vol. III, average length of the stages of development, in C. acutus as a type, are as follows: Egg stage 8 days Larval stage 28 days Cocoon 31 days Adult without host 32 days and longer 13. Fleas which have never been fed from the time of emergence from the cocoon prove longer lived when starved than fleas removed from the host. 14. Rat fleas may be kept alive on a human host for a considerable time, one specimen being kept alive as long as five months. 15. The females of all species of fleas are considerably longer lived than the males. 16. Induced parasitism of rodent fleas on man seems to be influenced by the length of mouth parts in the different species. One specimen of C. acutus, the species with the longest rostrum, fed for a period of nearly one hour at one insertion of the mouth. 17. It is indicated that starvation of infected fleas, when these insects are transported in clothing, may not eliminate the danger of transmission of plague. NOTICE TO MEMBERS AND CONTRIBUTORS. “The Annals of the Entomological Society,of America will be published by the Society quarterly and will include the Proceed- - ings of the Annual meetings and such papers as may be selected by the Editorial Board. Papers may be submitted to any member of the Editorial Board and should be as nearly as possible in the form desired as final, preferably typewritten, and illustrations must be finished complete ready for reproduction. Plates must not exceed 5 x 7 inches unless intended to fold. In general, papers to be accepted must be. original, complete and previously unpublished and, ex- cept in connection with the proceedings, it will not be the policy to publish preliminary announcements or notes. Authors will . be allo: wed fifty reprints gratis and additional copies at cost to the Society. The annual subscription to members is $1.00 in addition to the annual dues and may be sent to the Secretary-Treasurer. Mr. J. Chester Bradley, Cornell University, Ithaca, N. Y. - Communications relating to the Annals, and all orders for separate copies or reprints should be addressed to the Managing again Editor or to ANNALS OF THE ENTOMOLOGICAL Society oF AMER- ICA, Biological Building, 0. S. U., Columbus, Ohio. CONTENTS OF THIS NUMBER. BRADLEY, J. C.—Minutes of the Boston Meshes BS ae ee SmrtH, J. B.—Abstract of Address, Entomology and Ento- POLO ISER4 iin Ohre Den ES Se iy oe tg seca whic opts 12 ‘Wesster, F. M.—A Predaceous and Supposedly Beneficial » Mite, Pediculoides,.Becomes Noxious to Man.......0. 2.15 BANKs, Namnay.—Mysmeleonidae from Australia... .. = See 8 WILLIAMS, FRANCIS X.—The Anatomy “OF the Larva OES Cecidomyia Resinicoloides Walliams. go ves eee 45 Mrrzmam, M. B. —Some New Pacts in eas Bionomics of the - TNs as Rodent Pleas... ie eet uhe aston vies tases fcbE rece 7 The regular annual subscription price for the ANNALS is, in ~~ the United States, Cuba, “Porto Rico, Hawaii and Mexico, $3.00; Canada, $3.50, other countries, $4.00. Checks, drafts or money” : ~~. orders should be drawn payable to. ANNALS ENTOMOLOGICAL ~Socrety oF AMERICA, and: addressed to eee Pee Rife 0.8. G:, Columbus Ohio, U; ‘S. A.