nurs ny / x ENTOMOLOGICAL SERIES BULLETIN NO. 14 REPORT OF WORK \ 4 OF THE EXPERIMENT STATION OR. hE Ws X HAWAIIAN SUGAR PLANTERS’ ASSOCIATION Rit PHILIPPINE WASP STUDIES Part 1.---Descriptions of New Species. By S. A. ROHWER Part 2. ---Descriptions of New Species and Jin Qo SSNIAN Is7/BHe History Studies. TS "320248 BY/ps WILLIAMS ) Y \ JUN 4 1942 A Na , | es \) rt ae ‘ANCELLED- JAN 1920 HONOLULU, HAWAII j DECEMBER, 1919 — Sl nad HAWAIIAN SUGAR PLANTERS’ ASSOCIATION ; OFFICERS FOR 1919 ESCH WODETLOUS Braet sets kta cpa cick ho tse cade otc abla Tete President TORN, “WATERHOUSE 8 fic crair ce decressepten te en igs Vice- pe on Ru D PANERA ts ie Se copnlinas) abel EK eus Wate aeta ta ceromes Secretary Tr Fo WOW AL DRO No actos nth oes a aera ane ait iad dene + vay TRUSTEES FOR 1919 E. F. BISHOP = E. D. TENNEY A. W. T. BOTTOMLEY J. W. WALDRON J. M. DOWSETT JOHN WATERHOUSE JOHN HIND E. H. WODEHOUSE W. O. SMITH EXPERIMENT STATION COMMITTEE J. W. WALDRON, Chairman A. GARTLEY T. H. PETRIE Cc. R. HEMENWAY G. P. WILCOX - J. N. 8S. WILLIAMS D. A. MEEK, Secretary EXPERIMENT STATION STAFF HH SPENCE eae to eats, ce aS Sea Director Rs (PE REGIS as Reale ee BS a cieee Consulting Entomologist OTTOs-H. SWERZH EA. tess en tena Entomologist gel HUN Svein AB aa aes Reg Bnd ee eae Ot eat ee see Entomologist ROS BORN: oussr SaaS fe Ree, a tae Assistant Entomologist rd Sr DIVER FOR AICE sn ores Sates ovsnsioce iow are Assistant Entomologist se WN cL Nea CAVES 5 Seong SR i ae ga cw Ao Assistant Entomologist oo EE rIVES Ex byl OUN.. ets uae sa hte re re dae Assistant Entomologist T H x E se Deets AQUINAS ota ccc ney Sika hea ea es ea ees ame Botany and Forestry, in charge OW “OA RIREIN LT Eikirsvcrn coc te ane eee eee Associate Pathologist Test CUI i ea oh ee) at Soa Assistant Pathologist is DOT ge a ad oe eee dats ae Assistant in Cane Diseases ODP ERG T Me oe tcee iano eo lense Supt. Forest Nurseries DoT O OF DB Hr ae ore. vas Oe ioe Asst. in Pineapple Investigations Dee AR IVLAN No tinsenc ena: esas one wea Asst. in Pineapple Investigations oy H P ne Cc H Cc E R AD E M BigS ciNO BRT Gg ccc tress state WiSwis co. wien eeeese, Sugar Technologist PWR ECA By ye Noo eh al meee ad Of eae aR Acting Sugar Technologist ARO DE a ii ah atk bie onl a Vici haw eho oeeen Assistant Chemist a Sohee & Maca 'h0 Sl od G'S BAL fi cr O41 Hrd SH a ae eee Assistant Chemist Ts. oe GENCE eet cik, BRE ied wy ne see Assistant Chemist CicBZ WARRINER 08 15 eRe aed Magn Rae ORO eS rT 37 METROCORUINDTECOTNUS) 4x0 Selon ooo enn Cherie eet Ee eRe 38 WU GUBDOBAGIG DOT Sica cys CRORE CERES ORDERS LCI ONCE Soto ee, Oe 8 REO 41 VUE UR OCCUR DUNGE AUG Ws tat nctn tas: o Boer eC oiae eee) OTe OC Lae 42 IMICUROCEPIVONULCOUOR Cusco nice sco erie uci nee Reo Poe eae 43 WHCUHOCO US CLICTUIUUS, 15°50) ohne a) opayasee oh ct lich tes ore ole eho) eerily epahe eae 45 PSAMMOCHARIDAE: IPSCMDADENIG. COCTULESCENS ac Paes 2 pues ee eee ee eee een oe 123 Chilorion ourulentus, var. ferrugiéneus..2. 22. 2bs pe eee oe ele 124 Chlorion wmbrosus, Var. PUWMAP CETUS. 22. -Ger estes obese eees 125 Chiorion MUSCUPCNIYR cel fo lce. cles iceie 6h oe tise reise pe EEE 127 ADVNOVIR- INULO) oe Foe. Pere Leb Ole oo se Ree OL eee 128 IGARRINAW) 32 acon. b civ.ciece Soe ee eile eine oe cee AlABL LOTTE VUZONENBIE: feo eS face ns2 ee ee cio ee Oe Eee 132 Notogonided willigmst: tx. cfeb 7 series c ciscic's ciple pies oe nee 134 Cratolosra pifgmawe. oS ao. bios. note jocie ob bee Dee eee 136 Notoegonidea. VicOnensis 2 eo sise's 26 oe 8 eee Hee Pe Oe ee 137 TATIE QUITI sob 1s Cb ea So ee eO SE Os HE en Ee 138 LACHYLLS. DANDENBIE 2 Ooo aisitd a Bee aie pc ain puis copa ee ee 139 Diacranorhina. (Piagetia) Wzonensis ....--.52-2 eer - es cee 140 Halouris Mandipilgnis =o ois bebe neo cee eros eee 141 TRY POXYLONIN AG pe oe eee oe soe piece eee 142 Lrypomylon ClONGOAWID 2 secu cos Baer reer Spee Gree 143 GBBECERIDAS .o25 33.722 co Sel ce - callers apie 03 6 eee ere 145 COTCLIIS: GNGUIOTIB: Po. bese ee oe Asics bo 2 coc Je eee eee 146 CERCETIB. BUNGE eo ne ee oe Hep dol oie cies aS Oe Le 148 VESPIDAE: HOUMENUNAD: sess oc: eet oe ce emis oo ee nee bone an nee 149 ERY GOWUUN OLTWI Oa. o vive Boers pe OES Ht bie 6 ba eee PE 150 Odynerus TUZONENBIB Do: 55 obec Lt oe eke ce eee 151 OGYRETUB VEBDOUBES 2 oo eB tc che ce tne noe Oe ee CREEL 151 EWRONES QUTVLED? 23552 cao he DEE OE Oe ee DO Ce 152 EWMENES™ FUNVUPCHIAB = do prt vice Feed oe pic pike hice Cee en Ee 156 ZLethus (Calligaster) cyanopterus 2.0 250000200500-00%5 02506 157 VESPINA® sce c2scutec sen cee ts ree eee ee eee 164 TCO) eins cig ola oie ohe eae rh Fhe Oe OO Ee 164 V C8p0 VUCTIOSE 3 oe bie oe eee ths Doge bite Jae Oe Eee 164 Vespa GAUBEH; Fo cinctgicsoss + tts bwente ee tice ee 165 Stenogasver (Tschnogaster) 722-465. shee eee ee eee 166 Btenogaster Gepreswigaste? .fo.36o cece cece et ee eee 167 Stenogastér Barigicius pcos F< pti eR Cee eo Pm oe aval SECROGOELEF BP ooo 'o ovis: 15 ee oe ee ee RIe Ee Oe ee Oe 172 Stenogaster micans, yar. WZONENSIS ........--0+00%%r-rere- 173 CONCLUDING BEMARBES | 3000 secu Seon s Pncbdteew te ee eee 177 LITERATURE Part i Descriptions of Philippine Wasps By S.A. ROHWER Forest Insects, Bureau of Entomology, Washington, D. C. Descriptions of these new species are offered at this time so the names may be used in an article by F. X. Williams, dealing with the habits of Philippine wasps. The descriptions were based largely on material collected by Dr. Williams, but occasionally specimens which had been accumulated by Professor C. F. Baker have been examined and made part of the type series. The types of these species are in the National Museum. SUPERFAMILY SPHECOIDEA. FAMILY SPHECIDAE. Subfamily Pemphredoninae. Polemistus luzonensis, new species. Female—Length, 5 mm. Anterior margin of the clypeus strongly triangularly produced, the surface smooth, shining ; inner orbital carina complete to level of anterior ocellus, the area be- tween it and the eyes foveolate; frontal carina curved and joining inner orbital carina opposite the top of scape (when extended dorsally) ; a strong median carina from anterior ocellus to front carina and also to lateral ones which diverge from the ocellus and set off a wedge-shaped area; frons and vertex shining with small, poorly defined punctures ; ocelli in a low triangle, the posto- cellar line distinctly greater than ocellocular line; eyes slightly converging below, the distance between them at the clypeus sub- equal with the length of the scape; posterior orbital carina strong; complete above middle of eye and with a branch on the vertex extending as far as the lateral ocelli; area between posterior orbital carina and eye foveolate; occipital carina strong, some- 6 what reflexed, the area in front of it foveolate; posterior orbits with small punctures and also with large irregular punctures be- low ; flagellum stout, short, somewhat longer than the scape, the first joint a trifle shorter than the fourth; anterior dorsal mar- gin of the pronotum carinate, dentate laterally and_ slightly emarginate medianly ; scutum with small, close distinct punctures ; notauli complete, feebly foveolate; lateral depressed lines nearly complete; two parallel impressed, feebly foveolate lines, which terminate caudally in two deep punctures, between the notauli; reflex edge of scutum above the tegulae foveolate; suture in front of the scutellum with one median ruga; scutellum with fine small punctures; propodeum with dorsal aspect shining and covered with large reticulations, the posterior aspect with a large concave area which is transversely wrinkled and has a shining, deep dorsadventrad depression, the lower part of the posterior aspect reticulate; lateral aspect of propodeum finely punctate and with irregular oblique wrinkles; mesepisternum punctured, the sutures feebly foveolate ; first intercubitus straight, the second curved outwardly; second recurrent almost interstitial; petiole short, depressed dorsally and with transverse rugae; abdomen shining, highly polished. Black ; basal joints of flagellum piceous ; four anterior tibiae and tarsi and posterior tarsi testaceous ; wings hyaline, iridescent; venation pale brown, stigma dark brown. Type-locality—Los Banos, Luzon, P. I. Described from one female collected in 1917 by F. = Williams. Type Cat, Noe221)16- US: Nat. Mas: Subfamily Larrinae. Dicranorhina luzonensis, new species. In color and general appearance the female resembles closely D. kohli (Brauns ) ee South Africa. Female —Length, 9.5 mm. Anterior margin of the clypeus broadly produced medianly, the projection broadly rounded later- ally and medianly with a narrow notch; clypeus with a median longitudinal ridge; head opaque, finely granular; median im- pressed line complete but not as deep at frontal crest; vertex finely punctured; supra-interorbital line slightly less than half the length of the hypo-interorbital line and a little shorter than the second and third antennal joints; third and fourth antennal joints subequal ; thorax opaque, finely granular; scutum with two impressed, slightly diverging median lines anteriorly ; propodeum subequal in length with the scutum, truncate posteriorly, the 7 dorsal aspect with a median longitudinal carina and transversely reticulato-aciculate ; posterior aspect with a deep, elongate median depression and with a few transverse wrinkles; sides granular with a few dosad-ventrad wrinkles; legs rather feebly spined; longer calcarium of the posterior tibiae three-fourths the length -of the hind basitarsis; abdomen normal; pygidium narrowly rounded apically, the surface shining and with separate distinct punctures; second, third and fifth abcissae of radius subequal, the fourth abcissa one-fourth longer than the first; recurrents nearly interstitial on cubitus. Black; mandibles, apical margin of clypeus, scape, tegulae, apices of anterior femora, anterior tibiae and tarsi rufo-ferruginous ; the appressed pile silvery ; wings hyaline, a dusky band across wing in front of stigma (darker above) ; venation pale brown, stigma black. Male——Length, 7.5 mm. Produced part of clypeus with two long (three times as long as basal width), parallel-sided teeth, the area between these teeth U-shaped; second abcissa of radius a little shorter than fifth and distinctly shorter than third; recur- rents well separated on cubitus ; otherwise as in female except the scape 1s piceous-black. Type-locality—Los Banos, Luzon, P.I. Described from one female collected by C. F. Baker and one male collected by F. X. Williams. Type.—Cat. No. 22104, U. S. Nat. Mus. Cratolarra pitamawa, new species. IT am not certain that the assigning of this species to Cameron’s genus is correct, but it agrees well with his characters and as it cannot be assigned to any of the other described genera I have thought best to place it here for the time being. Of the genera with bare pygidium it is probably closest to Larropsis, but is readily separated from that genus by the elliptical lateral ocelli and triangularly shaped pronotum. The shape of the pronotum will also separate it from Larra. Although much like Noto goni- dea it is at once distinguished by the bare pygidium. Female—Length, 9 mm. Anterior margin of the clypeus smooth, polished, practically truncate ; raised area of face smooth ; frons opaque, finely granulato-punctate; frontal furrow com- plete to anterior ocellus; vertex shining, with small separate punctures ; depression behind lateral ocelli U-shaped ; supra-inter- orbital line half as long as the hypo-interorbital line, but not as long as the second and third antennal joints; antennae slender, the third joint slightly longer than the fourth; scutum with close, 8 small, distinct punctures and a short impressed line above tegulae ; scutellum not impressed, more sparsely punctured than the scutum; dorsal aspect of the propodeum with an incomplete longitudinal carina, coarsely coriaceous medianly, laterally granu- lar and with few feeble transverse wrinkles; posterior aspect per- pendicular, granular and with feeble transverse wrinkles, the median longitudinal depression deep, complete, wide above; sides of propodeum obliquely striate; mesepisternum sculptured like the scutum; legs very feebly spined; longer calcarium of hind tibiae somewhat more than half as long as hind beat ; abcissae of radius in order of greatest leneth (1, +), 3, 5, 2; abdomen shin- ing, very finely sculptured ; pyg cidium fully three times as long as basal width, truncate apically. Black, with dense silvery pile, first three tergites with apical silvery bands ; wings smoky hyaline ; venation dark brown. Type-locality—Los Banos, Luzon, P.I. Described from one female collected July, 1916, Be F. X. Williams. Type—Cat. No. 22112, U. S. Nat. Mus. Tachytes banoensis, new species. Female—Length, 12 mm. Anterior margin of the clypeus de- pressed, with three nearly equal lateral teeth and a U-shaped median notch; the surface of the clypeus convex, and with close punctures ; frons with separate, poorly defined punctures; frontal line complete and extending above the anterior ocellus to level of lateral ocelli; vertex with punctures well separated, depression behind lateral ocelli horseshoe-shaped and with a deeply im- pressed line extending to supraorbital line; supra-interorbital line distinctly less than half as long as hypo-interorbital line, but subequal in length with antennal joints two and three; antennae rather stout, the third joint somewhat shorter than the fourth; thorax punctato-granular; scutellum not impressed; propodeum distinctly shorter than the scutum, not sharply truncate, dorsal aspect coarsely granular and with a narrow median impressed line which on the posterior face becomes wider; abdomen subshining ; pygidium rounded apically covered with short brown bristles ; legs moderately spined, longer calearium of hind tibiae as long as hind basitarsis; second and third abcissae of radius subequal. Black, with dense appressed pubescence, that of thorax faintiv golden; abdcmen with four silvery bands; thorax with rather dense, erect, yellowish-white hair; base of mandibles and the tegulae ferruginous; wings yellowish-hyaline; venation fer- ruginous. 2 Type-locality—Los Banos, Luzon, P.I. Described from one female collected by F. X. W illiams in 1917. Type.—Cat. No. 22108, U.S. Nat. Mus. Notogonidea luzonensis, new species. Closely allied to N. crawfordi (Rohwer) but is more slender and the pubescence of the pygidium is silvery. Female—Length, 11 mm. Anterior margin of the clypeus bare, shining, with a narrow median notch, without lateral teeth frons and vertex opaque, punctato-granular; frontal impressed line complete; supra-interorbital line less than half as long as the hypo-interorbial line and distinctly less than the second and third antennal joints; antennae stout, the third joint distinctly longer than the fourth; thorax opaque with close fine punctures ; mesopleural furrows crenulate ; a short impressed line in anterior middle of scutum; dorsal aspect of propodeum granular and in addition with transverse wrinkles, posterior aspect truncate, eranular, with strong wrinkles laterally, median impression spear- shaped, deeper dorsally , lateral aspect granular ain with oblique weak wrinkles; legs rather feebly spined ; longer calcarium of hind tibiae three-fourths as long as hind basitarsis; third abcissa of radius subequal with fifth, but distinctly shorter than the fourth; abdomen opaque; pygidium one-third longer than basal width. Black, with abundant appressed, silvery pile; silvery bands on all the tergites; posterior femora red; wings yellowish- hyaline ; venation pale brown. The paratype has the hind femora black. Type-locality—Los Bafios, Luzon, P. I. Described from two females collected July, 1916, by F. X. Williams. Type—Cat. No. 22107, U. S: Nat. Mus. Notogonidea williamsi, new species. Close to N. manilae (Ashmead) but the eyes are somewhat closer together at the vertex, the lower part of the prepectus is not wrinkled and the clypeus is without a median notch. Female—Length, 7mm. Anterior margin of the clypeus broadly truncate, the lateral angles sub- angulate ; head opaque, finely. closely punctured ; median "depression Gi facial basin broad ; a deep channel in front of anterior ocellus ; the transverse furrow behind lateral ocelli angulate caudally in the middle; distance between the eyes at vertex slightly less than half the distance between them at the antennae; scape opaque, with a distinct 10 lateral keel, as long as the first two joints of flagellum; pedi- cellum half as long as first flagellar joint; first two joints of flagellum subequal; scutum subopaque with distinct, close pune tures; scutellum more shining, with slightly smaller and move widely separated punctures; at the base of the scutellum are two circular pits; mesepisternum shining, with well separated distinct punctures ; prepectus without sculpture; prepectal furrow crenu- late; dorsal aspect of propodeum coarsely reticulate ; the posterior aspect vertical, transversely wrinkled and with a narrow longi- tudinal channel; sides of the propodeum irregularly rugulose ; tergites shining; pygidium, closely punctured, acute apically, not twice as long as basal width; legs feebly spined; hind basitarsis one-third longer than the longer calcarium ; third abcissa of radius slightly shorter than the second which is distinctly shorter than the first; recurrents meeting on the cubitus; second abcissa of cubitus half as long as third. Black; apices of mandibles and the palpi piceous; face, sides of propodeum and apical margins of tergites with sparse silvery pile; wings yellowish hyaline, vena- tion dark brown. Type-locality—Los Banos, Luzon, P.I. Described from two females, one (type) collected in 1917, and the other June, 1416, by F. X. Williams, for whom the species is named. Tyvpe—Cat. No. 22105, U. S. Nat. Mus. Larra lugonensis, new species. Female.—Length, 12 mm. Anterior margin of the clypeus regularly rounded, without teeth, smooth; basal part of clypeus and lower inner orbits with small, rather close punctures; facial basin deeply depressed, walls rounded, shining, and with a median furrow; vertex shining with small, scattered punctures ; anterior ocellus in a broad U-shaped depression which is parted in the middle by a distinct furrow from the ocellus; behind the lateral ocelli is a transverse furrow which in the middle extends caudad as a U; distance between the eyes at the vertex half as great as at the clypeus; scape and pedicellum shining, the apical part of the scape as long as flagellar joints one plus two; pedicellum but little shorter than the first flagellar joint; flagellum sericeous, the first three joints subequal; mesoscutum opaque with distinct punctures which are separated by about twice their diameter ; scutellum shining, with well separated distinct punctures ; mese- pisternum opaque, sculptured like the scutum; prepectal furrow crenulate; sides of the propodeum shining, with well-separated setigerous punctures, basally becoming smaller and almost disap- 11 pearing; dorsal aspect of propodeum transversely punctato- striate, more distinctly so basally, with a distinct median longi- tudinal raised line; the posterior aspect almost vertical, trans- versely punctato-striate and with a deep rather broad median channel ; tergites shining ; pygidium polished, with a few scattered punctures, rounded apically, about one-third longer than basal width; femora, especially the hind, robust; legs rather strongly spined; hind basitarsis about one-fifth shorter than the longer calearium of hind tibiae; hind calcaria pectinate; first three abcissae of radius subequal; third abcissa of cubitus distinctly shorter than the second or the second abcissa of radius. Black; basai knob of scape, mandibies, palpi and spines rufo-piceous ; hind femora red; fore wings subhyaline, venation dark brown; hind wings hyaline, venation pale brown; face, sides of pro- podeum and apical margins of tergites with appressed silvery pile. Type-locality—Los Banos, Luzon, P. 1. Described from two females one (type) collected by C. F. Baker, the other collected by F. X. Williams. Type.—Cat. No. 22106, U. S. Nat. Mus. Two females from Baguio, Benguet, Luzon, collected by C. F. Baker are only 9 mm. long and have the third antennal joint shehtly longer than the fourth. Otherwise, I see no differences and for the time being consider them the same species. SUPERFAMILY MUTILLOIDEA. FAMILY PSAMMOCHARIDAE. Subfamily Ceropalinae. Genus Xanthampulex Schulz. This genus belongs to the Psammocharidae, as has already been noted by Turner, and would be placed in the subfamily Cero- palinae as defined by Banks. Xanthampulex luzonensis, new species. Although allied to trifur (Schulz) and pernix (Bingham) this new species differs in a number of ways from the descriptions of these species. Female —Length, 8.5 mm. Anterior margin of the labrum narrowly arcuately emarginate; anterior margin of the clypeus truncate; inner orbits slightly raised; prominence between the 12 antennae truncate below, with the lateral angles rounded, the median depression line-like ; frontal impressed line complete, es- pecially deep where it breaks through the raised area below the anterior ocellus; ocelli in nearly an equilateral triangle, the posto- cellar line about half the length of the ocellocular line ; interocel- lar area convex; a narrow furrow on posterior orbits next the eye; antennae as long as head and thorax, the third joint dis- tinctly longer than the fourth; anterior margin of the pronotum truncate, the lateral angles prominent but not tuberculate ; notauli not foveolate; metanotum pyramidal; metapostnotum rather nar- row, transversely striate; propodeum sloping; abcissae of radius in order of greatest length (3, 4), 2, 1; nervulus interstitial ; cubitella well beyond nervellus; abdomen smooth, compressed apically; sheath truncate apically. Rufoferruginous; most of clypeus, face, protuberance between antennae, complete orbits and posterior line on pronotum white; front below lateral ocelli (except orbits) black; scape, pedicellum and first flagellar joint beneath reddish, rest of antennae black; wings clear hyaline, venation pale brown. Male—Length, 7.5 mm. Very much like the female, differ- ing as follows: pedicellum and third joint black; black of head extending over occiput; four hind tarsi black; inner orbits not raised; transverse swelling below anterior ocellus almost obso- lete; pronotum not so sharply perpendicular anteriorly ; posto- cellar line less than half the length of the ocellocular line. Type-locality—Los Banos, Luzon, P.I. Described from one - female collected by F. X. Williams. Also one paratype male from Mt. Makiling, Luzon, P.1., collected by C. F. Baker. Pyvpe—Cat. No. 22103, U. S. Nat. Mus. Subfamily Pepsiinae. Pseudagenia okawa, new species. From the descriptions this new species seems to be allied to tincta, but the legs are entirely black. Female.—Length, 9.5 mm., slender. Anterior margin of the clypeus gently rounded; eyes reniform, the greatest distance be- tween them but little less than their length; frons punctato-granu- lar; frontal line faint but complete; interocellar area convex, shining, sparsely punctured ; postocellar line subequal with the ocellocular line; third antennal joint distinctly longer than the fourth; dorsal aspect of the pronotum about half as long as the scutellum, the posterior margin arcuately emarginate; scutum 13 punctato-granular; scutellum punctured, more sparsely so me- dianly; metanotum closely punctured; metapostnotum narrow, with one transverse carina; propodeum sloping, granular and in addition with irregular transverse wrinkles, the posterior lateral area with large scattered punctures in addition, median long1- tudinal groove complete; radial abcissae in order of the greatest length, 3, 2, 4, 1; nervulus postfurcal by half its length; cubitella and nervellus interstitial; abdomen subopaque, pygidial area not shining. Black, with appressed silvery pile; wings clear hyaline; venation dark brown. Type-locality—Los Banos, Luzon, P.I. Described from one female collected June and July, 1917, by F. X. Williams. Type—Cat. No. 22101, U. S. Nat. Mus. Pseudagenia nyemutawa, new species. Of the described species this is nearest to P. blanda (Guerin), but is at once distinguished by the metallic hind femora. Female —Length, 10 mm. Anterior margin of the clypeus truncate; greatest distance between the eyes about one-fourth less than their length; frons finely granular; frontal impressed line obsolete above the middle, distinct below; interocellar area strongly convex, shining, sparsely punctured; postocellar line slightly shorter than the ocellocular line; an impressed line be- hind each lateral ocellus; third antennal joint distinctly longer than fourth; dorsal aspect or pronotum about two-thirds as long as the scutellum, the posterior margin broadly, angularly emargi- nate; thorax granular; metapostnotum more than half as long as the metanotum, shining and with a few transverse rugae; propodeum rather short, rounded, without a median impressed line, covered with irregular transverse wrinkles; second and third abcissae of the radium subequal, much longer than the fourth which is nearly three times as long as the first; nervulus postfurcal by more than half its length; cubitella distinctly beyond nervellus ; abdomen shining, pygidial area highly polished. Bright metallic blue with a faint purplish tinge; antennae, palpi and tibiae and tarsi black; pubescence white; wings clear hyaline, venation dark brown. Type-locality—Los Bafios, Luzon, P. I. Described from one female collected in 1917 by F. X. Williams. Type—Cat. No. 22100, U. S. Nat. Mus. 14 Ageniella banoensis, new species. Apparently allied to erigone Bingham, but is not as coarsely sculptured. Female.—Length, 10.5 mm. Anterior margin of the clypeus smooth, obtusely triangular ; inner margins of the eyes curved, the greatest distance between them about one-third less than their length ; head finely granular; front without a complete impressed line; ocelli in somewhat less than an equilateral triangle, the postocellar line subequal with ocellocular line; third antennal joint somewhat longer than the fourth; dorsal aspect of pronotum shorter than the scutellum, the posterior margin arcuately emargi- nate ; thorax opaque, the scutum closely eranular ; metapostnotum transversely striate and medianly impressed; propodeum slop- ing, granular, with irregular transverse wrinkles all over, and with a median impressed, longitudinal line; abcissae of radius in order of greatest length are 3, 2, 4, 1; nervulus postfurcal by nore than its length; abdomen subopaque. Black, with rather dense gray pile; wings hyaline, the apical margin narrowly dusky and with a broad dusky band just beyond stigma; venation dark brown. Type-locality—Los Banos, Luzon, P.I. Described from two females (one type) collected in July and August, 1917, by F. X. Williams. Type.—Cat. No. 22099, U.S. Nat. Mus. Ageniella williamsi, new species. Similar to 4. unifasciata (Ashmead) but the clypeus is dif- ferent, the pronotum is longer and the brownish spot in the wings is larger. Female.—Length, 6.5 mm. Anterior margin of the clypeus truncate and with a rather broad median tooth, the apex of which is rounded; eyes reniform, the greatest distance between them about one-fourth shorter than their length; ocelli nearly in an equilateral triangle, postocellar line one- fourth shorter than the ocellocular line; third antennal joint much longer than the fourth; face finely granular; dorsal aspect of the pronotum sub- equal in length with the scutellum, the posterior margin very gently arcuately emarginate ; mesoscutum with close dense punc- tures; punctures of the scutellum smaller and more separated; metascutum granular; metapostnotum shining, transversely striate; propodeum granular and with irregular transverse wrinkles on the posterior two-thirds; first and third ‘abcissae of 15 radius subequal, the second abcissa a little longer than the fourth ; abdomen shining; apical tergite obtusely pointed. Black; with short white hair which is especially dense and appressed on pos- terior lateral sides of propodeum and on tergites; wings hyaline, the radial, second and third cubital and apical spot in second dis- coidal cells brown; venation pale brown. Type-locality—Los Banos, Luzon, P.I. Described from two females (one type) collected by F. X. Williams for whom the species is named. Tyvpe——Cat. No. 22098, U. S. Nat. Mus. SUPERFAMILY VESPOIDEA. FAMILY VESPIDAE. Subfamily Stenogasterinae. Stenogaster varipictus, new species. Structurally like S. nitidipennis Saussure (as determined from Philippine Islands) but is easily distinguished by the yellow lines on the scutum and the V-shaped black mark on the pro- podeum (in nitidipennis the base is black and the apex pale). From the description it seems also to be allied with faviplagiatus (Cameron) but does not agree with that description in all ways. Female —Length, 13 mm. Head smooth; ocelli in slightly less than an equilateral triangle, the postocellar line less than the ocellocular line; antennae thickened to apex, the third joint as long as fourth and fifth, apical joint pointed, distinctly longer than preceding; thorax smooth; scutum with two impressed, median, longitudinal lines anteriorly; propodeum smooth, with a faint median, longitudinal groove which is more distinct pos- teriorly; second tergite with a short petiole, then widening to apical fifth, then parallel sided. Black or blackish with yellow marks as follows: face and clypeus, except a median dorsal spot, inner and posterior orbits, dots between antennae and laterad of anterior ocellus, pronotum anteriorly, posteriorly and a line later- ally, lateral borders and two median lines on scutum, spots on scutellum, line on metanotum, propodeum except a \-shaped dorsal spot and a lateral line, irregular spot on mesepisternum dorsally and a cuneate line below and also the posterior margin, metapleurae dorsally, base and irregular lateral spots on second tergite, narrow lines on following tergites; legs yellowish, inter- mediate and hind tarsi and spots on hind coxae, femora and tibiae 16 black or blackish; wings hyaline, venation pale brown; hody clothed with scattering erect whitish hair. Paratypes show that the extent of the yellow markings is vari- able. Type-locality—Los Banos, Luzon, P. I. Described from three females, two collected by C. F. Baker and one by F. X. Williams. Other localities —Two females from Mt. Makiling, Luzon, col- lected by C. F. Baker and one female from Butuan, Mindanao, collected by C. FP. Baker. Type—Cat. No. 22049, U. S. Nat. Mus. Stenogaster depressigaster, new species. Apparently allied to S. nigrifrons (Smith), but it can easily be distinguished from that species by the smooth scutellum and silvery (not golden) pubescence on the clypeus. Male—Length, 16 mm. Clypeus as long as broad; frons reticulato-punctuate ; vertex smooth; ocelli in an acute triangle, the diameter of the anterior one greater than the postocellar line; antennae stout, short, third joint nearly as long as the two fol- lowing; pronotum. practically smooth; notauli well defined an- teriorly ; scutellum and metanotum smooth; propodeum faintly transversely aciculate, and with a distinct median longitudinal furrow ; abdomen smooth, second segment with a short petiole, then gradually widening to apex, most of its dorsal surface. de- pressed, the depressed area with dense hair; apical sternite strongly concave, narrowing to a truncate apex. Black; base of antennae rufoferruginous; thorax somewhat brownish with the following yellow marks; anterior and posterior margin of prono- tum, scutellum, metanotum, four spots on propodeum and a spot below each pair of wings; legs brownish, the anterior femora and tibiae and the intermediate femora beneath yellow; wings sub- hyaline, venation dark brown. The paratype from Mt. Makiling has the dorsal pair of snots on the propodeum reduced in size and in addition has a spot at about the middle of the mesepisternum. Type-locality—Los Banos, Luzon, P.I. Described from two males collected July and August, 1917, by F. X. Williams, and one male from Mt. Makiling, Luzon, collected by C. F. Baker. Type—Cat. No. 22048, U. S: Nat. Mus. Stenogaster micans var. luzonensis, new variety. Very like the typical form but in both sexes the yellow spot 17 above the antenna, the one in the emargination of the eye and the spot on the occiput is wanting. The yellow on the propodeum is divided into four spots the lower of which is somewhat T-shaped. In the male the clypeus 1s all yellow. Type-locahty—Mt. Makiling, Luzon, P. I. Described from two females collected by C. F. Baker and one male from Los Banos collected June and July, 1917, by F. X. Williams. Type—Cat No. 21928, U. S. Nat. Mus. Female is type. Subfamily Eumeninae. Odynerus lusonensis, new species. Female—Length to end of second tergite, 7.5 mm. Robust. Sides of the clypeus fully two-thirds as long as the basal width, anterior margin deeply angulately emarginate, the lobes narrow- ly triangular, acute apically; emargination of eyes broad, the front with large, distinct punctures (larger and more distinct centrally); a distinct impressed line between bases of antennae; interocellar area raised but hardly tuberculate; postocellar line slightly shorter than ocellocular line; vertex and posterior orbits punctured like the front; antennae stout thickening apically, the third joint slightly longer than the fourth; anterior margin of the pronotum truncate, tuberculate laterally; prothorax, scutum and scutellum with close, distinct punctures; scutellum not impressed; metanotum truncate posteriorly, crested and deeply emarginate so as to be bi-tuberculate; propodeum not angulate, with a well defined carina above a depressed basin ; first tergite short, apical width greater than length, punctured like the second; second tergite slightly wider than long, shining, sparsely punctured, apex not at all reflexed; legs without spines. Black ; clypeus except median spot, mandibles, scape beneath and basally, a trapezoidal frontal spot, emargination of eyes and above, elongate spots on posterior orbits, anterior margin of prothorax, tegulae, large circular spot below tegulae, scutellum, metanotum, sides of posterior face of propodeum, apical margin of first tergite (broader laterally), circular spots near base of second tergite, apex of second tergite and sternite broadly, nar- row lateral posterior margin of third tergite and sternite and a large spot on apical tergite yellow ; legs yellow, coxae, trochanters and posterior femora and anterior femora above black; apical four joints of four hind tarsi and flagellum beneath apically piceous; practically without hair or pile; wings brownish, darker along costa and in radial cell; venation very dark brown. 18 Male.—Length to apex of second tergite, 6.5 mm. LDiffers from the above description of female in having the interocellar area less strongly raised; metanotum mostly black; yellow of first tergite not as broad laterally, the spots on the second tergite larger and not circular in outline ; apical margin of second ter- gite more coarsely sculptured but not reflexed; emargination of clypeus deeper and W-shaped basally. Type-locality—Manila, Luzon, P. I. Described from three (one-type) females and three (one allotype) males collected by Robert Brown. Also one female from Los Banos, Luzon, P. I., collected by F. X. Williams. Type.—Cat. No. 22119, U. S. Nat. Mus. Part Te Descriptions of New Species and Life History Studies By Francis X. WILLIAMS Hawatian Sugar Planters’ Association Experiment Station. PRERACE: The great wealth of insect life in the tropics has long been a matter of comment among naturalists. And since insects are so abundant there, it follows that, living in such varied environment, many curious habits and structures are developed among them. The ants, bees and wasps form no exception to this statement and we find that these vary more both in themselves and in their handiwork than do those of temperate Europe and America. As a student of wasp-life, first in the United States and later in the Philippine Islands, the superiority of the tropical over the temperate regions as a field less ex xplored and offering greater opportunity for biological work, strongly impresses my memory. In glancing over the literature which relates to the habits of ants, bees and wasps however, it is seen that with some notable ex- ceptions the works leave much to be desired. This is due chiefly to the fact that the field entomologist is often handicapped for time and so is unable to make a prolonged study of the fauna of one locality. This applies especially to the tropics. I con- sider myself fortunate, therefore, when in the Philippines, from June, 1916, to September, 1917, I was able to make a resident study of many wasp activities and thus observe them from day to day or week to week. The results obtained indicate, as usual, hardlly more than a skirmish in the vast field that here lies spread before the entomologist who wishes to take up studies along these lines. The opportunity for this study presented itself when I was en- gaged by the Experiment Station of the Hawatian Sugar Plant- ers’ Association, in entomological work at the Philippine College of Agriculture, Los Banos, Luzon Island. The college is ideally situated for the study of insect life. It is about 40 miles by rail- road south of Manila, and lies immediately at the base of Mt. Makiling. This well forested peak is an ancient volcano which 20 rises to a height of nearly 4000 feet above sea level. Several thermal springs flow from its sides, and hence we have the name Los Banos (the baths) for one of the towns at the foot of the mountain. Thus the College offers at once a study of the lowland Fig. 1. College of Agriculture, Los Banos. Mt. Makiling in the background. Photo by L. B. Uichaneo. and generally more widespread fauna of the cultivated and semi- cultivated areas as well as that of the forest itself. The com- parative isolation of this mountain probably accounts, in a large measure, for the very rich and rather distinct fauna and flora it supports. Entomologically, thanks chiefly to the efforts of Prof: C. F. Baker, Dean of the College-of- Agricultures sand*his collector, Julian, it is probably the best known region in the Philippines. The Philippine Islands form a part of the Indo- Malayan region, which includes Java, Sumatra, Borneo, India, portions of China, etc. Los Banos lies approximately in latitude 14° 10’ N. and longitude 121° 10’ E. and has a distinctly tropical climate. Its seasons may be considered three: the warm and moist period, from about June to September; the cool, to January; and the dry period, to May. The rainfall is fairly heavy, and violent storms (baguios) sometimes occur. During the single cool sea- 21 son which I experienced, the mercury seldom fell a degree or two below 70° F., and in the warm season rose to over “90° F At no time, therefore, can the weather, at low levels, be termed cold—though the natives themselves will not agree with you on this point—but while flowers and fruits of one sort or another may be found throughout the year, there is a relative scarcity of insect life during the less warm and dryer months, just as occurs in temperate regions in a much more pronounced degree, and 1f we investigate, it will be found that many insects suspend their activities for a time and hibernate or aestivate as elsewhere. For the study of insect life the tropics offer decided advan- tages over the temperate regions; in the former the life-cycles of insects are generally more rapid and in many cases uninterrupted throughout the year. Insects are more numerous, many are of greater size, and the climate, while not always most salubrious, seldom interrupts entomological enterprise. Cloudy weather, or even at times a light rain in the tropics does not spell a cessation of insect activities. I have derived much information from the works of other entomologists, and it will be found that related wasps of temper- ate and tropical regions have much the same habits. Those who would gain a better idea of the habits of wasps will do well to consult “Fabre’ s Souvenirs Entomologiques, a large part of which has been translated into English; Peckhams’ Social and Solitary Wasps; Latter’s Bees and Wasps; Hartman’s Observations on Some Solitary Wasps of Texas; Ashmead’s Habits of Aculeate Hymenoptera ; Dutt, on Life Histories of Indian Insects; Sharp, in the Cambridge Natural History of Insects, V, Pt. 2; Maindron, Notes pour servir a Vhistoire des Hymenopterés de l’Archipel Indien et de la Nouvelle Guinée; Howes, in Tropical Wild Life in British Guiana; Isely’s Biology of Some Kansas Eumenidae ; Bequaert’s Revision of the Vespidae of the Belgian Congo; and Rau & Rau, Wasp Studies Afield. I owe special thanks to Mr. F. Lutz of the American Museum of Natural pes New York, for copying literature inacces- sible to me; to Mr. S. A. Rohwer of the United States National Museum, for identifying a number of the wasps considered here; and he has kindly consented to publish his paper as a part of this Bulletin; to Mr. Nathan Banks of Cambridge, Mass., for spider determinations; to Mr. J. A. G. Rehn of Philadelphia, for naming Orthoptera; to Mr. J. C. Bridwell, for information and criticism ; to Mr. W. R. R. Potter, illustrator for the Experiment Station here, for valuable hints on drawings, many of the latter being his own work; to Mr. O. H. Swezey for proof-reading, etc.; and to Prof. C. F. Baker, Dean of the College of Agriculture, P. I. The types of the species described by me are all in the collection of the H. S. P. A. Experiment Station, Honolulu. tifa INTRODUCTION. The Hymenoptera may be roughly divided into a stinging or aculeate group and a non- stinging or non-aculeate group. The first comprises ants, bees and wasps, and here the sting (modi- fied egg-laying apparatus or ovipositor) issues from the tip or apex of the abdomen, while the trochanters of the legs are gen- erally simple; the second division includes leaf wasps, chalcis wasps, ichneumon wasps, etc., these having the ovipositor issuing (having its external origin) from before the end of the body, and the trochanters more often two-jointed. Ants practically always live in communities commonly made up of males, functional females or queens, and more or less sterile females or workers. All these phases have one or two node-like joints on the base of the abdomen. Bees as a rule are more hairy than wasps, though many of the parasitic groups are comparatively naked. At least, some of their hairs are branched. They feed their young with flower products, while most wasps provide theirs with animal food, usually insects and_ spiders, entire, in chunks, or well masticated. Social wasps, the “Yellow Jacket” (Vespa), for example, live in communities. As in the ants and social bees, the most spe- cialized kinds have three forms of individuals to a nest. These work for the good of the community. With the solitary wasps there are only males and functional females. Each of the latter provides solely for her own young, and even when living in dense colonies, as in certain sand-nesting forms (Bembex) and mud- daubers (Sceliphron), exhibit no cooperation. Nests oF WASPS. Since in their immature stages wasps are helpless creatures, it devolves upon the mother to provide them with food and shelter. This brings us to the subject of nests. Properly speaking, wasps of the most primitive type make no nests, but sting and parasitize their prey where they find them, and as these usually live in places well concealed from other enemies, the wasp’s brood 1s fairly secure. Thus here the victim’s abode substitutes in a way for the wasp’s inability to make a nest. The Bethylidae, usually small black wasps, often classed with the parasitic Hymenoptera, but more properly associated with the Aculeata, illustrate this point. Many prey on lepidopterous larvae, w hich they attack either in their retreat—as borings in decayed wood, or in rolled- up leaves—or in their cocoons (Sclerodermus, Sierola and Holepyris). The Hawaiian Sclerodermus tmmigrans Brid- 23 well has been found by Bridwell (1918) parasitizing in its cocoon the larva of Car yoborus gonagra F abricius, one of the so-called bean-weevils. The genus Epyris, comprising some of the larger Bethylidae, prey on the larvae of tenebrionid beetles, which they often sting when underground. But where the victim is subdued on the surface of the ground this wasp proves equal to the emer- gency and buries it. So, too, with the Scoliidae, an extensive and useful family of thick-set wasps that prey on lamellicorn beetle grubs. The latter usually feed in decayed wood, compost heaps or underground at roots, and are there sought and para- sitized by the wasps and left without further ado. But there are a few records of Scoliids also burying these grubs. (Forbes, 1908, p. 159.) The prey of some wasps of more advanced type live in well-defined burrows which are utilized as nest holes by these wasps. For example, Methoca, a genus placed sometimes in the Mutillidae, but more often with the Thynnidae, parasitizes the larvae of tiger-beetles, and if these, in their struggle for life, crawl out of their retreat, they are afterwards pulled in and buried. The Mutillidae or ‘velvet ants” mostly enter the nests of bees and other wasps and ees their young. Some of the Psammocharidae, also known as Pompilidae or “spider wasps,” use the burrows of their victims as a cell for their young. The Ampulicidae or cockroach hunters drag their subdued prey into some hole or other shelter. But a ‘large proportion of the ground wasps, sometimes also called “Fossores” or “digger wasps, are more advanced in development and thus fitted to dig burrows or construct other definite nests, of which there exist a great variety. For instance, Tachysphex and Priononyx, sphecid wasps, both hunters of Orthoptera (grasshoppers, roaches, etc.), quickly excavate and store their short burrows, which terminate in a single cell or enlargement. Liris, Cerceris and Tachytes are more extensive diggers and may require a week or more in which to dig and provision their tunnels of sev- eral cells.t. Many wasps are architects, building’ nests above ground ; fewer combine their architecture with burrows. Thus we have the mud nests of many Sphecidae, Pompilidae and Eument- dae, and then there are many Odynerus and some Masarids which dig holes, divide off the cells with a partition of clay, and with the same material build a delicate tube over the entrance to the bur- row. Trypoxrylon and some Pseudagenia, spider-wasps, fre- quently divide off cells in pre-existing hollows (bamboos, reeds, y The order of business in wasp life—mainly nesting activities—is not always the same. We have some species, as Priononyx, Dolichurus and some Psammo- charidae, which first secure their provisions and then dig a hole, while others, probably the majority, make their nests and then search for victims. 24 mud-dauber nests, etc.). Some mud nests are rainproof, others are not, and so the latter are built in sheltered places; hence we find the nests of mud-daubers (Sceliphronini) often placed under eaves of houses, against ceilings and in other safe positions. Some mud architects cover their cells with tree gum and lichens (Pseu- dagenia sp.), or with debris of various sort (Podium). Macro- meris uses a clay-like material of composite ners Some of the more specialized solitary wasps (Zethus, Zethusculus) use leaves or moss for nest building, and the very elongate shade-loving Stenogaster, material much like, though more frail than used by the paper-making hornets. It seems, then, that the higher wasps very commonly make use of superior building materials—superior in that less of it is required in the construction of a cell than in mud-daubing wasps, for example. Space may be economized, but time does not appear to be. REE GG: The cell made or the victim secured, the egg is the next se- quence in wasp life. The queen hornet (Vespa) may in her lifetime of some months lay several thousand eggs; this is all out of proportion to the egg-laying powers of solitary wasps. Zethus cyanopterus, a highly “specialized solitary wasp, probably lays fewer than a dozen eggs. Tachytes, a locust wasp less highly de- veloped, may produce as many as fifty, and some of the Scoliidae probably somewhat exceed this number. The wasp eggs that I have studied ranged from less than a millimeter in length (Methoca, Epyris) to seven millimeters (Chlorion, Macromeris). They are whitish or yellowish in color, two or more times longer than thick, usually slightly arched and commonly a little thicker at the head than at the posterior extremity. They may be placed near the food or on its surface.* The Psammocharidae glue their eggs near the base of the spider’s abdomen; most Sphecidae that attack grasshoppers, crickets or locusts affix theirs under or partly alongside the thorax, while Wethoca and Scolia sp. place their egg on the underside of the abdomen of the beetle grub. The head end of the egg is often against some tender spot of the victim’s anatomy. In those cases where the prey is paralyzed so as to be incapable of motion, the wasp’s egg may be only lightly glued to it (Methoca sp., Sceliphron, Scolia sp., in Scolia ‘the : egg stands on its head as it were) ; but where the prey largely recovers from the wasp’s sting, it is more securely glued (Tiphia, Methoca sp.) or placed in a well-protected position on y (Larra, Noto- * A great many of the non-aculeates insert their eggs in the food, whether it be plant or insect tissue. 25 gonidea). Most Eumeninae, chiefly potter wasps, suspend their egg by a thread to the roof of the cell, which is provisioned with more or less active caterpillars. Some of the more special- ized wasps of this group (Synagris, Zethus) lay their egg freely in the bottom of the cell. Stenogaster and the social wasps build cells in a more or less inverted position, and there the egg is glued in the bottom of the cell itself. Wasp eggs may hatch in less than two days (Dolichurus, Methoca sp., Notogonidea) to four, five or more days (7iphia sp.). Cool weather greatly retards the devel- opment of the egg, so that two weeks or more may be required to hatch it. THE Larva. The next stage in wasp development is that of the larva. The larva is a more or less whitish grub, devoid of legs, but possess- ing adequate jaws. In many cases so gradually does the young erub break through the very thin and closely -adhering egg- shell that it is hardly possible to determine just when the egg hatches; in these cases it seems that the grub punctures the eny elope under the mouth and feeds at that point and simply outgrows its en- velope. Or else the larva may crawl out of the shell (Methoca sp. Epyris). The eggs of Eumeninae are reasonably firm, so that after hatching, the shell may collapse but little, and since the eggs of these wasps are not attached to their food, it follows that, in order to feed of themselves, the young grubs must be active from the very first, and this is true of those that I have seen. In the case of the Sphecidae, Psammocharidae, and others, the newly-hatched grub is usually a very helpless creature. \When it breaks through the egg-shell it finds food practically in contact with its mouth and so may feed at once. But if at this stage it be removed to another portion of its victim’s anatomy it very frequently perishes. Later on, however, after several moults, which are often difficult to follow (except in the Eumeninae), it acquires stouter mandibles, and, becoming more able-bodied, largely abandons its first method of feeding by suction, for a chewing one; thus it reaches out to the different victims or portions thereof, to devour the tougher parts—head, legs, etc. The soft, footless wasp grubs, one appearing much like another, should, if sufficiently studied, present characters that would place them in natural groups. There is here a field for the taxonomist. The feeding period of the larva is usually brief, perhaps three to fourteen days, according to species, food and climatic con- ditions. bo Or THE Cocoon. Directly after it has become full-fed, the grub forms a cocoon.+ As a rule the species that nest underground have the tougher cocoons ; they may be composed of from one (Dolichurus, Epyr is ) to many separable layers of silk (Methoca). Other cocoons may be formed of particles of sand, wood-dust, etc., agglutinized by means of the grub’s saliva into a tough shell (Bembex, Larra). The mud-daubers, already protected in a thick cell of mud, make a tough but very thin cocoon; so also with the mud-using Psammocharidae. It is noteworthy, however, that the parasitic Psammocharids, Ceropales, spin stout cocoons, even when occupy- ing the mud cells of Pseudagenia. Might this indicate that they were formerly burrowing wasps? The cocoons of Pseudagenia are thinner than those of their ground-nesting relatives that have to resist the moisture. Some Eumeninae make a thin but rather tough cocoon ; others, however (some Zethus), merely construct a thin leather-like partition which extends from near the top to a short distance down the sides of their cells. Several species at least of Stenogaster make no cocoons, but coat their cell with a shiny substance. Some wasps, of the family Pemphredonidae, are said to spin no cocoons. (Green, 1902.) The inner wall of the cocoon is smooth and varnished. Cocoons usually taper more at the posterior extremity, wherein the larva voids for the first and last time the contents of the intestine. There is some ad- vantage in this, as the cell is kept comparatively clean during the feeding stage, while later the walls of the cocoon readily absorb the moisture of the faeces which gives that end its characteristic color. Before spinning the body of the cocoon the larva usually makes a sort of scaffolding of silken strands to support it. The cocoon is commonly simple, but in some cases one end or even the side may be provided with a collar, knob or chamber (Me- thoca, Dolichurus, Epyris) The larva now encased within its cocoon has a resting period therein; this may be a question of a few days or of many months, for it is thus that solitary wasps usually pass the winter or the dry season.* The resting larva is usually more yellowish than yA few make no cocoons. * A rare exception, at least in temperate regions, is cited by Fabre in the case of a caterpillar wasp, Ammophila hirsuta Kirby, which hibernates as a full-grown wasp and in consequence begins nesting operations at a much earlier date than do related species. Davidson (1899), in speaking of Sphex elegans of Southern Cali- fornia, says that the majority of these wasps hatch out in July and August and pass the winter concealed in crevices, etc., and that probably not more than fifteen or twenty per cent remain in the larval stage until the following May. Bees very often hibernate as freshly-hatched adults. : when feeding, deeply segmented and possessed of more or less activity. iia Rupa. The next change comes with the final moult of the larva, the result being the pupa. Before shedding its skin the larva grad- ually acquires a slightly different form; the thorax becomes dif- ferentiated in assuming a smoother, cone-like shape, and pupal appendages may be discerned dimly outlined beneath the skin, which, like a tight-fitting garment, bursts at the back of the thorax and gradually slips down, exposing the forming pupa. This is especially delicate in the Sphecidae, Psammocharidae, ete., but less so in the potter wasps or Eumeninae. The appendages— legs, wings, antennae, etc.—are free, i. e., not fused flatly into the body, as in the moths and butterflies. The pupa is shaped much like the adult w asp, though it may be considerably bent or doubled up upon itself (Eumenes, Zethus, Stenogaster) and adorned with spines or tubercles. This stage is of comparatively brief duration—two weeks, more or less. The pupa, at first whitish, darkens with age into the color of the adult wasp. In due time its thin envelope ruptures, and through the insect’s motions 1s gradually worked back to the end of the body, until the wasp is entirely free from it. The wings grow almost visibly as they leave the pupal sheath, so also with the other appen- dages. It remains within the cocoon and cell until sufficiently hardened and strong, when, with the aid of its jaws and some- times a little mouth liquid, it soon cuts its way out and emerges an active and independent wasp. As in bees, it seems that the males are first to emerge and they soon busy themselves with find- ing the females. Thus male Scoliidae, Odyneri and others may be observed awaiting in their breeding grounds the appearance of the females. The life cycle of wasps for the summer brood or broods may be estimated as about from twenty-five to forty days. The adult wasps, even in the solitary species, may live for sev eral months. They do not always go about nest-building immedaitely, but first enjoy sunshine and flowers before commencing their arduous task. THE Foop oF ApuLT WaASsps. Here flower products predominate. In the Philippines, Premmna odorata Blanco, a wide-spreading verbenaceous tree with flat in- florescences, often beyond reach of the net attracted very diverse insects—among the wasps were Tachytes banoensis, Chlorion um- 28 brosus and aurulenta, Eumenes curvata and fulvipennis, Zethus cyanopterus, Rygchium atrum and Scolia, near procer, and Liacos analis. The small border-plant, Alternanthera versicolor Regel, (Amarantaceae) attracted Scolia manilae Ashm. and others. Compositae such as Solidago and many Umbelliferae yield good wasp catches in temperate regions. The sweet-potato vine /po- moea batatas (Convolvulaceae) has the petioles of its leaves provided with nectar glands which furnish food for flies, bees, wasps and other insects. In Kansas, I have seen the stems of the sunflower, Helianthus, that had been pierced by a_ small lepidepterous larva attract many Hymenoptera by the exuding sap. This was also the case with willow trees which exuded a liquid where punctured by the borings of beetles. Bridwell has observed Hymenoptera in California pierce with their mandibles the base of the glandular spines on the stems of sunflowers (H. annuus) and draw nourishment therefrom. Veg- etation attacked by leaf-hoppers, plant lice, mealy bugs, lantern flies, etc., frequently becomes plentifully sprinkled with honey dew. Such foliage later becomes blackened by a fungus growth and is the favorite rendezvous for numerous insects—flies, beetles, ants, wasps, etc., and an entomologist may reap a harvest there. When a wasp overcomes a victim destined to serve as food for her young, she may first help herself, feeding at its mouth juices and perhaps also at the sting wounds. The bee- catching wasps (Philanthus) appreciate the honey which they force out of their prey’s crop, and the little bethylid wasp (Holepyris hawatiensis) gorges itself with the juices which she extracts from her caterpillar prey by piercing with her mandibles, the base of one of its thoracic legs. The Pseudageniae or leg-amputators no doubt derive some nourishment from the leg stumps of their spider victims. Roubaud (1916) has observed that the larvae of the social wasps Jcaria, Belonogaster and Polistes exude upon stimulation by the male or female wasp, a salivary fluid which the wasp laps up. THE Foop-oF Wasp LARVAE. Wasps as a rule provide food of animal nature for their grubs; each group is usually addicted to a certain kind of prey—this may consist of spiders, beetles, grasshoppers, cicadas roaches, bugs, bees, flies, moths, skipper butterflies, caterpillars, etc. Sometimes the food is served in the living (paralyzed) and sometimes in the dead state. The insect or spider captured may often be much larger than its captor and may put up a stiff Zo battle before it is overcome. A wasp hunting and subduing her prey is an interesting spectacle though frequently tedious to observe. It is interesting to note that some of the ponerine ants as Leptogenys and Lobopelta, the lowest subfamily of the Formi- cidae, and presumably derived from solitary wasps, have a method of hunting their prey—in this case small land crustaceans much like that employed by some of the more active terrestrial Fossores. In the Philippines I have seen ants of the genus Lobopelta, very active and nervous insects, searching beneath leaves, small lumps of soil, etc., in damp, shady places. Every now and then a crustacean would be flushed, to leap away to safety; eventually a less active or a cornered one would be pounced upon by the ant, which quickly paralyzed it by one or more stings on the underside of the body. The victim was then borne away nestward under the ant’s body. A single victim may prove sufficient food for a wasp grub; with other species a number may be required.* Rarely dead and dry food is furnished the grub (Microbembex). Sometimes the young wasp is fed from day 3 day by its parents, being given either entire insects (Bembex) or a chewed-up portion of cater- pillars (Synagris). A few wasps feed their young a honey-like substance (Masaridae; and perhaps Stenogaster). Many social wasps give their young a wide range of food. SLEEP OF WASPS: Not much is known about this interesting subject. In the United States, Rau and Rau (1916) have made a number of observations on the sleep of Hymenoptera. They found that some of the Sceliphronini or mud-daubers assemble in sheltered places for the night. Many other Sphecidae, Eumeninae, and some male Scoliidae pass the night on weeds. This applies more to the male sex. They may assume odd positions on these weeds which they grip with their jaws and legs or with the jaws only. Other wasps, mainly Fossores, occupy their nest-burrows or other holes for the night. The more social or specialized wasps sleep in or about their nests—thus we find some of the psammocharid mason wasps keeping watch in the shelter wherein their cells are situated and resting on them. Higher Eumeninae may occupy one of the cells of their nests (Zethus, Synagris, Zethusculus ), the Stenogaster or slender-bodied forest wasps most certainly sleep or rest on or in their nests. In the tropics however, where *T know of no case in the Aculeata except in many Bethylidae where more than one egg is laid on a victim. 30 the hours of darkness are often not particularly cool, wasps that do not hide in the ground may not really sleep a great deal. On moonlight nights i have heard the large Carpenter bees buzzing at Solanaceous flowers; the few wasps “that I have inspected dur- ing the night, no matter how cautiously approached, were always wide awake and alert... But where chilly benumbing nights pre- vail, wasps directly exposed to the atmosphere are quite dead to the world—until the sun arouses them. This 1s common in tem- perate regions. Bequaert (1918), p. 23, speaks of a wasp, Provespa, as being “adapted to nocturnal habits and restricted to the Oriental re- gion.” He also makes a note of two species of crepuscular bees taken in S. W. United States. ENEMIES. Both young and adult wasps have numerous enemies. Perhaps the most severe check to many species are ants. These obnoxious creatures, belonging to the genera Pheidole, Pheidolegiton, So- lenopsis and Monomorium, for example, often rob the wasp of her prey, occupy her nest, and even penetrate cocoons buried in the ground. It is very noticeable that while a few wasps actually provision their burrows with ants (Aphilanthops, Fertonius), they have in general a decided aversion and fear of them. Para- sitic wasps, beetles and flies are other serious enemies, and a cer- tain large tropical Vespa or hornet plunders many a nest of solitary and social wasps. Certain Aculeate wasps (Ceropales, Mutilla, some Stizini, etc.), prey upon the young grubs or their food. Economic STATUS. Taken as a whole wasps are beneficial insects and destroy an incalculable number of harmful insects, as well as some beneficial ones; among those addicted to the latter habit are species of Philanthus and Palarus which prey on the hive bee and others ; Mutillidae or “Velvet Ants” destroy the young of other w asps, as do the brilliant cuckoo wasps and a host of other kinds. Asa rule, people who are not afraid of spiders regard them as destroy- ers of obnoxious flies, millers, ete. Perhaps spiders, even up to the giants of their kind, have more wasp enemies than do insects. A whole family, the Psammocharidae or Pompilidae, bring up their young upon spiders. Many genera in other families prey upon them and the mud-daubers (Sceliphronini) literally stuff their cells with such provender. Thus one wasp may destroy up- Sil wards of a hundred spiders. And yet the latter are usually abun- dant enough. Their beneficial nature may sometimes be ques- tioned, inasmuch as they are often cannibals, and do not discrimi- nate between beneficial and injurious insects as food. A very large number of wasps prey upon caterpillars, locusts, flies and beetles ; others attack bugs. One genus of wasp ( Larra) makes the mole-cricket, Gryllotalpa, its common prey; and the observing person living in districts where grasshoppers abound cannot fail to note that these are preyed upon by certain wasps. Hacker (1918) records the big Australian eumenid Monerebia (Abispa) ephippium Fabr. ee on the larva of Clania igno- bilis Wlk., one of the bagworms (Psychidae), a family of moths containing some noxious species. The big wasps of the genus Monedula prey largely upon Taba- nidae @Elonce. flies ). Of M. surinamensis Bape. Belt (1874), p. 239, says: “Horse-flies (Tabanus) were too numerous, an drops of blood trickled down our mules’ faces where they had feasted. In some parts large, banded black and yellow wasps (Monedula SUYINAMENSIS asia ) came flying round us and had a threatening look as they hovered before our faces, but they were old acquaint- ances of mine in Brazil, and I knew that they were only searching about for the horse-flies with which they store their nests. * * *” The Scoliidae are eminently beneficial—as is known to econo- mic entomologists. They destroy the larvae of the cocoanut beetles (X ylotr upes and Oryctes), and smaller species are known to prey upon the larvae of the so-called May or June beetles. The marked success of Scolia manilae Ashm., a rather small Philippine wasp introduced by this Experiment Station, into the cane fields of Hawaii, in checking the grub of the Anomala beetle, indicates that the Aculeate wasps have a place in the list of bene- ficial insect importations. In concluding, I may say that the nests and the burrows of solitary wasps and of some social kinds as well may be robbed with almost invariable impunity. Solitary wasps are never ag- gressive to the degree exhibited by their social brethren, and while they would sting or endeavor to do so were they incau- tiously held between the fingers, their weapon is directed mainly aoaimsptieir prey... hey do not appear to sting in the quarrels among themselves. But we know from reliable sources—such as experience—that hornets and many other social wasps behave otherwise ; they are bold in their strong communities and protect * According to Froggatt (1907), the large, metallic-blue thynnid (Diamma bicolor) of SE. Australia will, in the female sex, which is equipped with a powerful sting, turn on her back and fight furiously when molested. Gs 32 their molested paper nests with impressionable vigor. This home- defending instinct is also well developed in social bees and many ants. DESCRIPTIONS: OF NEW SPECIES, INCL@DING-ONE SPECIES: BROMTORMOSs: THYNNIDAE. Methoca. The six species of this genus taken on the Island of Luzon, P. I., as well as the single specimen from Formosa, while differ- ing from each other, at least in the female sex, by well-marked characters—as body proportions and sculpture—all agree in be- longing to quite another section than those of ‘temperate North America, Europe, and even as far as studied—of South A frica.* The following are the more marked differences between these two groups: THE PHILIPPINES AND FORMOSA. FROM o Philippines. Compound eyes proportion- ately larger and closer together ; ocelli in a more elevated, some- times an equilateral triangle; head shorter or less gibbous behind eyes; proportions of thorax more slender; no en- closed area at apex of propo- deum (except in fimbricornis). @ Philippines and Formosa. Compound eyes proportion- ately much larger; scutum of thorax reduced to a flattened or depressed area. FroM TEMPERATE NorTH AMERICA, EUROPE AND SOUTH AFRICA, ¢ Temperate North America and Europe. Compound eyes proportion- ately smaller and more widely separated ; ocelli in a rather low triangle; head longer or more gibbous behind eyes; thorax stouter ; enclosed area at top of propodeum = 3 United States species. 2 Temperate North Amer- _tca, Europe and South Africa. Compound eyes proportion- ately smaller and more widely separated; scutum convex and better developed. i to S18) KEY TO SEPARATE THE SPECIES OF PHILIPPINE AND FORMOSAN METHOCA., Females. Dise of pronotum finely but distinctly striate, frontal process Obsolete s color blacks: ., Mess Ghai os 1 eet eee striatella Disc of pronotum not striate or with only a very few median striae; frontal process strong (except in the impunctate OULOSCLAUD are Sen Alera ee RO ear OLIN ee hake cs hire Z Head, thorax and abdomen glossy, almost wholly impunctate ; clypeus simple, i. e., no dull thin apical portion; jet black SOLS CUTS As as ok eam cern re Cam seeles ign Menara Seite Ve formosana Head at least, somewhat punctate or striate, clypeus with a thin dull apical portion (do not mistake the hair-fringed 2 LAD UIMMGLOT PES )iivine swans ete, soit act merotae a on 3 Thorax reddish ; propodeum widest posteriorly and there sub- IC DUIKCE SS Se She ise RE SMe or aeeiC nhs im lem OTN ies i punctata Thorax black ; propodeum not subtruncate posteriorly.... 4 No striae on vertex or upper part of frons; form very slen- CUS Trae a See aio rio. cose are. 8 hohe eel ae eae fimbricornis Distinct striae on frons and vertex, form moderately slender Sts OASIS LarCRON PACA NADER SEE OR ie ark RO Oech eee monticola Males. Antennae beneath with two ranks of curved hair; posterior face of propodeum truncate and bounded by a carina.... RRNA sls x! Ss ale Mates relay teen ct ad ke fimbricornis Not havineatheaboyercharacters ses. 2 ie acacia tei ee 2 The sternite before the apical spine, deeply and narrowly cleft; clypeus without a pale and thin apical portion... 3 The sternite before the apical spine, at most shallowly emar- ginate; clypeus with a thin pale apical portion....... 4 Propodeum with large well-marked areolate spaces through- out the disc; a row of vertical carinae on propleura...... SPR casa dace one's! Sn tar ota hss aes sail ohigeal a Bxslee aye eneaet striatella Propodeum comparatively, rather finely rugulose; no row of Mewmed| LOhOplelinal Carinae ycmy vars tac eas debilis Gular suture marked by transverse foveae extending from three-quarters to the entire distance between the occipital ridge and the labium; puncturation near the suture deep and general; well-defined punctures on disc of at least the anterior two-thirds of pronotum; wings clear... punctata Wit SAIMEMSCAtEl Er 7a, es ches shew yesh clientes oats aoe fuscipennis 34 Gular suture with only about two foveae, and these not dis- tinct and occupying only a small length of distance; punc- turation here more sparse and shallower; pronotum, with puncturation distinct only on the anterior half or less; WINGS AMT UMA oy esas oletegentions ooerns oree cee monticola The males of Section 2 are difficult to separate, and even the genitalia do not appear to differ here. Additional material and more study will probably improve the status of the species of this genus in the Philippines. ~~ SS a SS —- A Fig. 2. Fig. 3. Fig. 4. Fig. 5 Many more species remain to be discovered, and if the giants among the Cicindelidae, as Manticorna of Africa and Amblyc hila of North America, and the remarkably-formed arboreal species of Indo-Malaya and other parts of the tropics, some of whose larvae dwell in holes in twigs instead of in the soil, have their respective Methoca or allied enemies, then we must look for some wonderful species among these wasps. Fig. 2. Methoca monticola; aedeagus; stipe and sagittae. From above. Fig. 3. Methoca punctata; sagittae; the apical portion is curled laterally and probably unnaturally here. From above. Fig. 4. Methoca striatella; aedeagus; 3-piece stipe and median portion (sagittae). From above. Fig. 5. Methoca debilis; aedeagus; stipe and sagittae. From above. Methoca striatella, n. sp. (Figs. 4, 6 and 7.) Female, type: Length 6.15 mm. Shining black; mandibles, antennae except the 4 or 5 terminal articles, the dorsal portion of the neck, legs except fore coxae, reddish, apical segment and a half of abdomen brownish yellow. Quite slender. Head ) l - ~ ¢ nearly twice as wide as prothorax, but not large; mandibles bi- dentate, with a few coarse confluent punctures except at apical third. Clypeus simple, subtruncate and very little emarginate, somewhat raised on median line, margin with scattered punc- tures. A low frontal tubercle, but no bituberculate process above base of antennae; front, except antennal foveae, as far as posterior ocelli with scattered punctures and very fine, interrupted longi- Fig. 6. Methoca striatella, 9, X* 13. tudinal striae; a furrow extending from the anterior ocellus towards clypeus; vertex polished, with sparse punctures, ocelli distinct forming nearly an equilateral triangle; eyes rather large, diverging from inner lower portion tow ards mandibles ; antennae rather stout, articles 3 and 4 subequal and all more or less punc- tate. Prothorax above with cervical part finely reticulate and separated posteriorly by a transverse short longitudinally carinate furrow. Anterior half of pronotum finely and obliquely striate, posterior half transversely striate and with a few punctures, striae becoming obsolete along pleurae; mesonotum with the an- 30 terior part (scutum) slender and coarsely rugose, then follows a small median pit, and then the large oblong and convex scutel- lum, which is well margined and sparsely punctate and a little aciculate transversely ; pleura of anterior portion of mesothorax with a few large and curved longitudinal carinae; posterior to this they are striate and punctate. Metathorax inconspicuous, propodeum nodose, polished, with sparse large punctures and Fig. 7. Methoca striatella, @, X< 7.5. some aciculations, laterally and posteriorly mostly smooth, stigma oval; legs slender; claws with a basal tooth. Abdomen dorsally impunctate, a few scattered punctures ventrally. Generally clothed with rather sparse, erect white hair. Male, allotype: Length to extremity of anal spine, 9mm. Black; apical half of mandibles, ridge under first article of antennae brownish; tarsi largely brow nish, with rest of legs almost black, tegulae in part ye ellowish ; wings hyaline; head much wider than thorax; mandibles bidentate, with coarse confluent punctures except at tip; clypeus truncate, rather shallowly emarginate at truncation, with rather sparse punctures and a medain pointed tubercle; front closely punctate except at furrow between an- tennae, and becoming scattered punctate towards vertex, which is sparsely punctate; ocelli large and forming almost an equilat- 37, eral triangle; compound eyes large; a horn-like process at the base of each antenna above; antennae except the two basal arti- cles opaque; 3 shorter than 4, 9-13 arched. Pronotum polished, subangulate posteriorly, with a very few punctures, and becoming transversely rugose at anterior upturned collar; ‘pleurae with a V-shaped area with large dorso-ventral striae, more distinct for anterior half. Scutum of mesothorax with anterior median por- tion convex, posterior part flattened, disc with a few coarse con- fluent punctures and coarse transverse striae; scutellum sparsely punctate with the anterior transverse sulcus carinate, mesopleurae polished, with rather sparse shallow punctures and with short longitudinal carinae on anterior and antero-dorsal edge; metano- tum with a semicircular median excavation. Wings normal : vena- tion yellowish brown. Propodeum everyw here coarsely reticu- late, with the cells generally transverse, the largest of which begins at the slope to apex; the basal one is longitudinal and median. Abdomen dorsally smooth and polished, with fine punc- tures for hair; ventrally 1 and 2 have sparse large punctures; others with their apical portion (beyond transverse row of hairs) rather closely punctate ; last sternite mesad with some fine reticu- lations, its apical margin deeply cleft; ventral portion of anal spine well margined and with a few punctures except in the middle line. Insect set with rather sparse white hairs. Genitalia (Fig. 4): Outer lobe of stipes broad, basal third polished, middle lobe bilobed, inner lobe simple. Type @. 119 and numerous ¢ paratypes; Los Banos, 1916- 1917. Some variation in size in both sexes, and the females may have more reddish brown than in type. Methoca formosana, n. sp. (Fig. 8.) Female, type: Length 4.5 mm. Black; apical portion of mandibles and most of first two antennal joints reddish brown; legs with terminal joints in part and coxae apically, yellowish brown. Quite slender, polished and_ practically impunctate. Head a little less than twice as wide as thorax, well extended behind eyes, and subtriangular as viewed from above. Mandibles comparatively short, bidentate; clypeus rather dull, subtruncate, raised mesially, frons without bituberculation near base of an- tennae, but with a median pit; eyes unusually large and proxi- mate; vertex rather pointed (rounded subconical), anterior ocel- lus rather distant from posterior ones ; antennae subclavate, basal articles long, 3 and 4 subequal; neck of pronotum with some fine transverse striae, scutum (smaller anterior portion of meso- thorax) with two incomplete longitudinal furrows on each side 38 of median line, a semicircular bordered impression on posterior border of scutellum; propodeum widest behind middle; legs moderately stout, abdomen normal. Adorned with sparse erect white hair. Type 1¢, Taihoku, Formosa, January, 1916; collected by F. Muir. Fig. 8. Methoca formosana, 9, Doe o< alee Methoca fimbricornis, n. sp. (Figs. 9 and 10.) Male, type: Length 7.5 mm. Shining black except reddish tip of mandibles, yellowish brown palpi and apical portion of legs. Moderately stout. Head much wider than thorax and subquadrate ; mandibles bidentate, with coarse punctures, malar space longer than usual and punctate, palpi long, clypeus rugulose, its anterior portion gently rounded out and there impunctate; a sharp median clypeal compressed process ending before anterior margin; clypeo-frontal fossa deep; frons polished, with large, deep and well-spaced punctures, no tubercles above antennae, a median frontal depression ; punctures smaller and closer together on ocellar area ; ocelli distinct, arranged in a triangle; vertex with sparse smaller punctures. Eyes nearly parallel inwardly ; anten- 39 nae rather short, first article without carina beneath, third article hardly one-half length of fourth, articles 3-10 or 11 beneath with two rows of long incurved hairs, becoming shorter towards apex of antennae, articles cylindrical; gular region transversely striate. Prothorax subcylindrical, subtruncate in front, collar with large scattered punctures and transversely strigose, pos- terior portion with large transversely confluent punctures ante- Pig. 9. Methoca fimbricornis, 6, X 9.3. riorly and sparsely punctate posteriorly. Mesonotum very coarsely and closely punctate, mesopleurae coarsely punctate and with indications of longitudinal rugae. Scutellum long and rather narrow, very coarsely rugose-punctate and with the trans- verse, anterior depressed area with 3-4 strong carinae, meso- sternum transversely rugose; forewings with two submarginal cells, the basal and the transverse median nervure quite inter- stitial; legs rather slender. Propodeum coarsely areolate. Ab- domen rather depressed, tergites with some fine subapical punc- tures, sternite 1 irregularly rugose-areolate basally and with large, well-separated apical portion, remaining sternites with fine scat- tered punctures, apical sternite with posterior margin nearly straight, a little depressed mesially there ; curved spine with some scattered punctures on ventral base. 40 Female allotype: Length, 7 mm. Polished black, mandibles and legs except tarsi, dark red, basal third or more of antennae brown- ish, the remainder almost black; tarsi, and apex of abdomen in part yellowish brown. Quite slender, especially the thorax. Head about twice as wide as prothorax and narrowing behind the eyes; mandibles bidentate, only sparsely punctate; clypeus long, its membranous apical portion dull and truncate, the remainder Sty Fig. 10. Methoca fimbricornis, 9, * 11. smooth and with a strong median pointed tubercle; frons with a bituberculate process above antennae, and a shallow depression, a few vertical striae on raised portion on each side, the remainder of frons, together with the vertex, with rather large and very scattered punctures; ocelli arranged in an equilateral triangle ; eyes converging somewhat towards vertex; antennae with the basal articles of flagellum glabrous, these rather long and swollen apically, article 2 more than half as long as 3, which is a little longer than 4; apical joints finely punctate. Thorax and abdo- men almost impunctate; the cervical region of pronotum with 41 transverse striae followed posteriorly by transverse ones curved mesially into longitudinal ones. Scutum of mesothorax nearly flat, with one median carina and a pair of lateral ones; legs rather slender. Very sparsely clothed with pale hairs, with a few dark ones on head and dorsum of thorax and abdomen. Type 14, paratype 10’, both College of Agriculture, Los Banos ; 2 allotype, low Makiling forest, 1917. The male is very distinct from others of the genus; the female I place here with some doubt, though she is also well removed from the other species. Fig. 11. Methocas debilis, 6, X 9. Methoca debilis, n. sp. (Figs. 5 and 11.) Male, type: Length, 7.25 mm. Related to striatella. Shining black. Differs from striatella in its smaller average size, finer sculpture, slighter build, ete. The propodeum is much _ less coarsely sculptured, being rugose and showing well-defined areolae only on the middle basal portion; the propleurae have no vertical carinae, the first sternite is separately, punctured almost to its base, while in striatella it becomes rugose on the basal third. The apical sternites are about as in striatella. The genitalia are 42 much as in striatella, but the outer lobe is not so broad as in the latter, and the sagittae or median part somewhat different. This species varies in size from 6 to 9 mm. Type 1¢, 1916; numerous paratypes; all from College of Agri- culture, Los Banos, 1916-1917. De ita Fig. 12. Methoca punctata, 9, X 11. Methoca punctata, n. sp. (Figs. 3 and 12.) Female, type: Length, 7 mm. Head black, thorax red, abdo- men black with reddish apex, apex of mandibles dark brown, last 5 or 6 articles of antennae with brown beneath, coxae reddish, tarsi brownish black. Fairly stout for the genus. Manduibles bidentate and with large confluent punctures except apically ; cly- peus long, subtruncate and consisting of a thinner, more opaque and impunctate apical portion and a thicker, more polished basal piece armed with a small median tubercle, disc impunctate except for the irregular marginal hair punctures ; front with a bitubercu- late process above antennae, the front except antennal fossae, 43 with large scattered punctures and longitudinal striae as far as posterior ocelli, furrow extending down from anterior ocellus, almost obsolete; vertex with large punctures and finer striae; ocelli in a nearly equilateral triangle ; eyes large; antennae punc- tate, rather stout, articles 3 and “4 subequal. Prothorax above with cervical part, first with transverse and then with longitudinal striae, pronotum with large scattered punctures except on sides; scutum with three carinae; scutellum large, almost impunctate, mesopleurae sparsely punctured ; propodeum stout sparsely punc- tate, wider and subtruncate posteriorly, stigma oval ; eles of legs with a basal tooth. Abdomen dorsally impunctate; a few ventral abdominal punctures. Generally clothed with panes sparse, erect white hair. Male, allotype: Length to extrimity of spine, 10.5 mm. Black, colored as in M. striatella, except that the tegulae are all black, wings hyaline. Coarsely punctate. Clypeus truncate with the transverse apical portion thin and yellowish, the thick emargi- nate basal portion punctate and with a low median tubercle ; ocelli forming nearly an equilateral triangle. A tubercle above the base of each antenna; frons coarsely and closely punc- tate ; vertex not so closely punctate, sparsely so posteriorly. Dor- sum of thorax punctate, hardly so on posterior part of pronotum, scutum rather sparsely punctate, with a few transverse and lateral rugae, scutellum rather finely punctate; propodeum rather coarsely rugose or rugose punctate, somewhat areolate mesially above; wings normal. Abdomen above with fine hair punctures, beneath with alternate antero-posterior bands of coarser and finer and closer punctures; sternite before spine gently emarginate ; ventral side of spine with coarse punctures. Genitalia: Outer lobe of stipes narrower than in striatella, middle lobe bilobed, inner lobe blunt ; median portion with a thin flat process or flagel- lum; the more or less membranous portion evidently varies in form according as to whether it is more or less distended or buckled. Rather sparsely clothed with white hairs. Type 12; numerous @ and 2 ¢ paratypes, all Los Bafos, Luzon, 1916-1917. Methoca monticola, n. sp. (Figs. 2 and 13.) Female, type: Length, 6.5 mm. Shining black; mandibles and basal part of antennae dark red, legs very dark reddish, almost black, trochanters and tarsi not so dark, apex of abdomen yellowish brown. Rather slender ; head rounded behind the eyes ; mandibles with some large confiuent punctures ; clypeus long, the 44 anterior portion dull and subtruncate, the stouter basal piece shin- ing and impunctate and with the median tubercle prominent ; frons with the bituberculate process present, coarse vertical striae mixed with puncturations, but the antennal fossae rather smooth ; vertex with the striae finer and the punctures sparser and large; ocelli in a nearly equilateral triangle, no striae back of a line behind ocelli; antennae with articles of basal half of flagellum rather EE a Re S Fig. 13. Methoca monticola, 9, X 13. fusiform, article 3 a very little longer than 4. Prothorax polished, with a few scattered punctures ; dorsum of cervical part anteriorly with some transverse striae and posteriorly with some short longitudinal ones. The small scutum has 3-4 sublongitudi- nal carinae; scutellum, rest of thorax and propodeum polished, with only a few scattered punctures. Abdomen smooth and pol- ished and almost impunctate. Sparsely clothed with pale erect hairs and fewer darker ones, especially on dorsum of head and thorax. 45 Male allotype: Length to extremity of spine, 9 mm. Black; apical part of mandibles reddish, carina under first antennal joint yellowish brown, tarsi more or less brown, wings infuscate. Sculp- tured and punctate very much as in M. punctata, it being very close to the latter and the slightly larger fuscipennis. Antennae with article 3 shorter than 4, the terminal articles well rounded out beneath. It may be separated from punctata chiefly by its darker wings. I am unable to find differences in the genitalia of the three species. The foveae in the gular suture of monticola are nearly obsolete, whereas they are well developed in the other two species. The pronotum is practically impunctate for all but its anterior third or so, and the head about the gular region is not strongly punctate. Type 19 ; paratypes 12 and 1’; all from Baguio, 5000 feet, Luzon, end of June, 1917. Methoca fuscipennis, n. sp. Male, type: Length, 11 mm. Black, wings infuscate. Much like the two preceding species, though more coarsely punctate. From punctata it is separated chiefly by its smoke-tinged wings, and it differs from monticola by the presence of w ell- marked transverse foveae, strong puncturation about gular area and by its pronotum, which is punctate for its first two-thirds or so. Type 1¢, Makiling forest, Los Banos, March, 1917; paratypes 24, College of Agriculture, June and July, 1917. PSAMMOCHARIDAE. Pseudagenia caerulescens, n. sp. (Fig. 14.) Female, type: Length, 8.75 mm. Head rather dull greenish blue; thorax metallic blue, with a little purplish lustre ; abdomen blue with a greenish lustre; antennae rather light brown; tarsi of forelegs, more or less, apical portion of mandibles, posterior margin of terminal abdominal segments, and pygidium brown. Head wider than thorax, ae oe. finely and closely punc- tured clypeus with disc evenly convex, its marginal profile a low obtuse wedge with the basal portion steeper ‘and the whole somewhat rounded ; clypeo-frontal suture with somewhat more than its middle third subtruncate into the frons; antennae with article 3 longer than 4; only a slight swelling between the ocell1. Pronotum with posterior margin broadly angulate ; pro- and meso- notum exceedingly finely and close punctured; metanotum with a few transverse striae on sides; postnotum about one-third as long as metanotum, sulcate in the middle and with one or more trans- verse carinae ; propodeum gently rounded, with a shallow, rather 46 indistinct median furrow and with well-spaced, broken and rather irregular transverse carinulae, spiracle opening widest uppermost ; legs hardly spinose, larger spur of hind tarsi about half as long ae : 2 =e ; Fig. 14. Bio. 15: Fig. 16. Pseudagenia caerules- Pseudagenia blanda, @. Pseudagenia macromeroides, GCS mee View of the two ventral : View of the two ventral apical segments. View of the two ventral apical segments. apical segments. Fig. 18. Bios lie Pseudagenia makilingi, Pseudagenia nyemitawa, 2. On View of the two ventral View of the two ventral apical segments. apical segments. as basal joint of tibiae ; wings clear, second submarginal cell about twice as long as wide; secondaries with the transverse median vein ending well before the origin of the cubital. Abdomen smooth; pygidium polished, impunctate, somewhat concave in profile. Rather sparsely white pubescent with some erect hairs on head, thorax and propodeum. 47 Male, allotype: Length, 6 mm. Slender. Colored about as in the female, but the antennae are blackish, except for the two basal articles, which are brownish, the first being quite pale beneath ; the clypeus is rather narrowly margined with white, as also the face along the lower half or more of the inner eye margin; the abdomen is a dark blue-black with a portion of the sides of the first two segments brownish in certain lights, and the apical mar- gins of the segments whitish. The clypeus is only slightly convex; it 1s basally well rounded into the face and the apical margin is broadly truncate; antennae not long, nor very slender, articles 3 and 4 subequal. The sculpture and hair are about as in the female, but the propodeum is flatter, with a wide, well- marked median furrow and the transverse rugulae reduced and substituted by reticulations. Venation as in the female. Sixth visible ventral segment with a low subtruncate median lobe; spine of lateral lobes proximate to median lobe; seventh ventral seg- ment sharply wedge-shaped, sides a little concave below apex; the carina apically linear, becoming broader and flat posteriorly. Type 1 @, building a clay nest at Los Banos, July, 1917. Para- tiyipemle oe. Allotype is a son of type, Los Bafios, 1917. The paratypes have on the w hole darker antennae. Resembles makilingi in color, but is smaller, and has a shorter postnotum. Pseudagenia macromeroides, n. sp. (Fig. 16.) Female, type: Length, 11.5 mm. Rather dull black with sil- very grey pubescence; apex of trochanters partly, and posterior edge of abdominal segments narrowly, yellowish brown, antennae and legs with a tinge of brownish, wings hyaline. Head short, wider “than thorax, very finely and closely punctured ; clypeus convex, its marginal profile well rounded out, slightly w edge- shaped, median third of clypeo-frontal suture truncate into frons and there very shallowly emarginate; a delicate shallow fur- row from anterior ocellus to betw een antennae ; article 3 of anten- nae considerably longer than 4, which is subequal to 5; ocelli in a rather low triangle. Pronotum with well-rounded shoulders, the posterior margin very broadly angulate, almost arcuate, and to- gether with the meso- and metanotum exceedingly finely punc- tured, the last with transverse striae at sides; postnotum about one-half as long as metanotum, medially sulcate for its posterior portion and with about 4 transverse carinae; propodeum gently rounded, emarginate at base and with hardly a trace of a furrow except on posterior face, and with fine and close transverse rugu- lae, almost granulate ; spiracle opening narrowly oblong, pidece at 4 48 about middle; legs hardly spinose, larger spur of hind tibiae about half as long as the first joint of hind tarsi; hind wings with the transverse median vein ending just beyond the origin of the cubitus. Abdomen smooth; pygidium polished mesad, with a few punctures towards apex and slightly convex in longitudinal profile. Some erect hair on head, thorax and abdomen. Male, allotype: Length, 8 mm. Slender, clothed and colored about as in female. Clypeus convex, a little more than twice as wide as long, marginal profile broadly truncate ; ocelli in a rather low triangle ; antennae moderately slender ; article 3 a little longer than 4; postonum about as in the female ; propodeum flatter, finely rugulose-granulate with a fairly distinct basal furrow ; hind wings with the transverse median nervure and origin of cubital intersti- tial. Abdomen subpetiolate, brownish at apex. Sixth ventral seg- ment only gently but widely rounded out mesially ; seventh visible ventral segment broad, finely reticulate and nearly naked, wedge- shaped apically, a sharp median carina apically, broadening and becoming rounded posteriorly. Type 1 fresh 2°, Los Bafios, July, 1917; paratype 19 ; allotype 1 reared ¢, Los Bafios, May, 1917. Builds mud cells similar to the cells of Macromeris. Pseudagenia makilingt, n. sp. (Fig. 18.) Female, type: Length, 11.5 mm. Metallic blue with a trace of purple on head and thorax; antennae and distal portion of legs brownish black; mandibles blackish, posterior margin of abdom- inal segments very narrowly pale brownish; wings ‘clear. Head a good de eal wider than thorax, which when view ed from above is rather long and of uniform width. Head and thorax with very close and exceedingly fine punctures; clypeus convex, its mar- ginal profile a blunted wedge with a wide but shallow emargina- tion on each side of the point; clypeo-frontal suture truncate into the face for about a third of its width; article 3 of antennae nearly one-third longer than 4; a well-marked frontal impressed line from anterior ocellus to between antennae. Pronotum with pos- terior margin broadly angulate; metanotum with some _ lateral striae ; postnotum about two-thirds as long as metanotum, broadly and shallowly sulcate and with about five transverse carinae and a delicate median longitudinal carina; propodeum well-rounded, hardly sulcate, with well separated, rather irregular transverse striae on disc, spiracle openings broadest uppermost; legs long, hardly spinose, hind wings with transverse median vein ending well before origin of cubital. Abdomen smooth; pygidium pol- ished, nearly flat in longitudinal profile and with some well- 49 marked scattered punctures on distal half or more. Fine silvery pubescence and some rather short erect hairs on body. Male, allotype: Length, 10 mm. Much slenderer and duller than female, the blue hardly metallic; clypeus, frons to above base of antennae, basal joints of antennae beneath carneous white, articles 6 to about 9 yellowish beneath; legs with a portion of coxae be- neath pale } yellowish or brownish, femora light and dark brownish with tinge of purplish, rest of legs more or less dull brownish, side of first abdominal segment and apical margin of this and fol- lowing segments pale. Antennae as in the female, quite long and slender, article 3 longer than 4; postnotum two-thirds or more as long as metanotum, “shallow ly suleate and with about six trans- verse carinae ; propodeum shallowly suleate, with fine, rather ir- regular transverse striae or rugulae. Clothed with silvery pube- scence and some erect hair. Sixth ventral segment with a fine, rather thin and flat median tooth, as long as it is wide at base, its sides rather concave, sides of segment drawn out into a somewhat obliquely pointed tooth or process. Seventh sternite shaped like the end of a boat, longer than wide, and with an elevated shallowly depressed, margined slender we edge tapering to a point. Type 12, Lower Makiling forest, Los Bafios, March, 1917. Paratypes 1¢@, Los Banos, July, 1917; 1 bred (allotype) and 1 other %, same locality, 1917. SPHECIDAE. Larrinae. Hyloliris, new genus. (Figs. 19, 20 and 70.) (Forest Liris.) Related to Liris and Dicranorhina. Head wider than thorax and proportionately large; face much as in Liris, with a swelling along inner eye margin, a transverse swelling on forehead; the two posterior ocelli reduced, trans- verse and proximate ; clypeus very wide, with the median portion well produced and bilobed; eyes strongly converging towards vertex; mandibles not excised beneath; maxillary palpi with six and the labial with four articles. Pronotum well beneath the level of mesonotum and fitting into the latter at an obtuse angle; mesonotum quite wide, propodeum truncate at apex. Legs rather long and stout; wings with venation as in Liris except that the second submarginal cell is narrower along the marginal, and the second discoidal is wider along the cubitus than in Liris. Abdomen short, subsessile. 50 2 A tooth just below apex and a large one just before its middle length ; pygidium well margined, tapering and constricted near the narrow tip; disc naked w ith an irregular median carina. g mandibles very long and slender; a small tooth at apical third and a larger one near base; hind femora beneath with an emar- gination near base and a long basal curved spine, as in Piagetia; pygidial area not defined. Type of genus Hyloliris mandibularis, n. sp. Hyloliris mandibularis, n. sp. Female, type: Length, 12.5 mm. Dull black; mandibles ex- cept apex, and tegulae ‘light brown; basal joint of antennae, pos- terior edge of fifth tergite and pygidium apically, brownish; fore- Fig. 19. oa ‘ a Hyloliris mandi- a ea bularis, @. Mandibles. Fig. 20. Hyloliris mandibu- laris, @. Hind femur. wings dusky along basal margin of costa, and all the marginal cell and beyond on either side and below, ee ak mandibles long, crossing; clypeus with the central third o - less well produced, cleft mesad and obliquely emarginate at ore angles; antennae with article 3 longer than 4; interocular space at vertex less than one-half as wide as the width of the clypeus. Head and thorax very finely granulate ; postscutellum lightly sulcate mesially ; pro- podeum with disc not margined by a distinct carina, and indis- tinctly transverse striate, coarser and subrugulose at base, and with a shallow median subapical depression; sides more or less obliquely striate, the posterior face obliquely truncate, with a deep sulcation. Legs rather long and moderately spinose; larger 3! spur of hind tibia about three-fourths as long as first joint of hind tarsus; tarsal claws long, slender and simple; wings ample, mar- ginal cell truncate and appendiculate. Abdomen stout, a little depressed and abscess dull and very finely reticulate; apical sericeous bands on 1, 2 and sides of 3; pygidial area well mar- gined, narrow, tapering rather bluntly, sides constricted, apex with several very small and inconspicuous short finger-like processes ; disc rugulose, nearly twice as long as wide at base, with a rather irregular median carina. Golden pubescence on clypeus frons, genae, part of occiput and slightly at porterior angles of scutum. Otherwise nearly devoid of hair. Male, allotype: length, 11.3 mm. Colored as in the 9 ; clypeus about as in @ except that the lateral emarginations of the pro- duced portion are more pronounced, forming an acute tooth later- ally; mandibles very long and curved; antennae with article 3 longer than 4. Sculpture as in @ with following exceptions: propodeum with the posterior face more gently sloping, the disc distinctly margined except posteriorly ; narrowed at base and nearly flat ; rather finely graulate except along borders, where it is somewhat rugose striate; shallowly sulcate; a smooth median place on posterior border ; the pleurae and most of posterior face finely granulate, with indications of striae at the angles. Hind femora well excavate beneath near base and with a curved inner tooth. Last dorsal segment finely reticulate, broad and subtrun- cate. Type, the described @ ; paratypes,8 9 9 and6¢. All taken or reared August, 1917, lower portion of Makiling forest, near College of Agr iculture. The specimens tend to become greasy, and vary in size from Sai toclZ-o5( 9"). Im the: the mandibles and armature of hind femora vary somewhat in proportions. VESPIDAE. Eumeninae. Odynerus vespoides, n. sp. (Fig. 81.) Female, type: Length (h.th.+-t.1--2), 10 mm. Thickset. Rather dull black, with yellow as follows: a lunate area along base and basal sides of clypeus and extending as a detached spot on each side of apex, stripe on outer side of mandibles, along pos- terior margin of eyes, on the middle of frons and one along lower inner eye margin to and filling the ocular emargination, along underside of first article of antenna, a band across the pronotum, the anterior and posterior third of tegulae, two mesopleural spots, DZ the upper round, two spots on anterior part of scutellum, one on the middle of mesonotum and a large one on each side of pro- podeum, a small spot under hind wing base and a larger one on basal pleurae of propodeum, a spot on outer edge of coxae 2 and 3 and on anterior face of 1, on femora beneath except basally, and quite apical on hind femora, most of tibiae above, along basal portion of first joint of hind tarsi; abdomen with two bands, one pre-median, the second subapical on tergites 1 and 2, the two on tergite 1 united laterally, 3-5 with a subapical band, all subapical bands slightly thickened and notched mesially; sternite 2 with a large spot on each side partly split posteriorly, 3-5 with a sinuate sub: ipical band, not meeting mesially ; two dull orange stripes on disc of scutum. Wings lightly infuscate, more deeply so along costa. Head hardly as wide as thorax; clypeus convex, about as wide as high, with a moderate emargination apically and which is about one-half as wide as greatest ‘clypeal breadth, closely and coarsely punctate, mandibles “about as long as interocular diam- eter at vertex, with four low rounded teeth, exclusive of apex ; labial palpi 4-jointed, fourth joint minute, third about one-half as long as first, second intermediate ; maxillary palpi 6-jointed, three apical joints comparatively small; frons closely and coarsely punctate, vertex somewhat less so; a sharp carina between base of antennae ; ocelli deep-set, in a low triangle ; a large median depres- sion on vertex with a thick stand of erect golden brown hair, protruding well above surface of head; temples gibbous behind eyes and somewhat more than one-half as long as latter ; antennae stout, articles cylindrical. Thorax almost uniformly, very coarsely separately and deeply punctured, the largest punctures on propo- deum, merging on the pleurae into coarse irregular longitudinal striae ; propodeum short, truncate rather shallowly concave pos- teriorly and there nearly smooth and with a sharp median carina. Venation of ordinary type, second submarginal cell very narrow along marginal; legs normal, tibiae with single apical spur, claws with one tooth before apex. Abdomen stout, rounded, with rather large shallow hair punctures, finely aciculate; tergite 1 hardly half as long as 2, steep anteriorly where there is a low gently rounded pre-median carina; a very shallow median sulca- tion on 1, punctures along posterior border of segment deeper and closer; tergite 2 slightly narrowed at base; its posterior edge produced into a short thin margin; tergites 3 and 4 with a much wider strip along posterior margin, the strip flaring somewhat so that it extends outwards at a different angle; these extensions appear as a flap at their pleural origin. Anal tergite subconic with a delicate median groove terminating before apex; sulcature on sternite 2 with about 12 complete costae; no posterior depression, 53 but this segment has large, well- separated punctures, that of the others much finer. W hile there is some dark pubescence on thorax, the whole upper portion of the insect is covered with rather coarse, suberect to erect bristly black pile arising from the large punctures on the head and thorax, and the smaller shallow ones on abdomen; this pile is rather conspicuous along the base of the abdominal strips and is paler on underside of body. Described from 1 @ from the lower portion of the Makiling forest, Los Banos, taken in the summer of 1917. This stout, much banded insect might perhaps form a subgenus of Odynerus; it bears some resemblance to the subgenus Rhyg- chium. The rounded transverse ridge on the first tergite and the much developed hind margins of tergites 3 and 4, mark this wasp as a peculiar one. LIFE HISTORY STUDIES. SCOLIIDAE. This family comprises a large number of digging wasps of rather depressed and compact form that prey almost wholly on the larvae of lamellicorn beetles.; They have a very general dis- tribution and in the tropics particularly, are represented by many species of formidable size. They are not highly specialized wasps and form no nests; yet their olfactory—or more probably, their auditory powers—must be highly dev eloped since female Scoliidae are able to locate their prey well buried in the soil, decayed wood, etc. Each species is more or less restricted to a certain species of lamellicorn grub, so that we may regard the diversity and abund- ance of the latter in a given region as an index to the scolid fauna of that place. The adults are often found at flowers; the female wasp how- ever usually spends the afternoon hours seeking her prey, and both sexes of many species pass the night underground. In cer- tain cases the males, as in the American Elis, sleep congregated on weeds, Sweet Clover (Melilotus alba) being a special favorite. The beetle grub may be stung to permanent paralysis, as in Scolia sp., or it may be only temporarily immobilized in the case of Tiphia. Scolia fastens her egg delicately so that the head end is cemented against the venter ‘of. the helpless grub, whereas Tiphia, as if mindful of the eventual activity of her now quieted victim, affixes her egg much more securely—but not always so + According to Burkill (1917), Scolia erratica Sm. in the Straits Settlements, ete., attacks the grubs of the Red Coconut Weevil (Rhyncophorus ferrugineus) as well as those of a Rhinoceros Beetle (Oryctes). 54 firmly that it may not sometimes be rubbed off by the wandering host—gluing it for its length transversely on a certain portion (varying with the species) of the venter, more rarely on the dor- sum.* The Scolia grub then, secure on its motionless victim, transforms in its birth place, but the young Tiphia may be treated to a more or less perilous ride underground, surviving which, it spins up where its victim perishes. In order to escape from the cocoon, the wasps of the genus Scolia, at least in part, neatly cut off the top of the cocoon; Tiphia on the other hand, gnaws a somewhat irregular hole near or at the head end of its prison. The Scolidae of the Philippines are represented by numerous species, some of them giants in size. Many are forest insects, some appear partial to “bamboo groves, while others prefer the open lowlands. The family is of decided economic importance, as the nature of the literature on these wasps clearly shows. The data which follow on this family are largely derived from notes and observations by Messrs. Osborn, Timberlake, Swezey and Muir. The identification of the 7iphia and the keys to their species is the work of Mr. P. H. Timberlake. The work of rear- ing and shipping the scoliid and other parasites was carried on under the direction of Mr. F. Muir, who was also responsible for the discovery as a desirable parasite, and the successful establish- ment of Scolia manilae in the Hawaiian Islands. Scolia manilae Ashmead. Length, 8-12 mm.; black and yellow. Scolia manilae (Figs. 21 and 22) was described by Ashmead in 1904, from two specimens collected at Manila by Father W. A. Stanton. It is a rather plain, “untropical’ looking wasp. Its black color is variegated with some yellow spots and bands on the abdomen and with a rather thin covering of stiff white hair. The male is more slender than the female and has longer antennae. It is perhaps the smallest and one of the most abundant and widely distributed of the Philippine Scolia. It seems to be con- fined to the lowlands and therefore often found in cultivated areas where there is food in plenty both for itself and its young. I have taken it on Luzon I., at Manila and Los Bafios, and to the south on Cebu and Negros Is. It probably inhabits all islands of any size in the archipelago. *“In many cases, at least, the beetle grubs stung by Tiphia, while quite active, yet exhibit, when handled, a comparative muscular weakness in the thoracic region. 5 tn While our breeding records at Los Banos show that Scolia manilae occurs as an adult throughout the year, it is, as with most other insects, more abundant during the warmer rainy months—from June to October. At this time the males, though hardly more numerous than the females—which they somewhat antedate—are, because of their habits, much more in evidence. Fig. 21. Scolia manilae, 9, X 5 (after Swezey). Large numbers may often be seen flying back and forth over a bare space of ground evidently in anticipation of emerging females. They are on the wing both earlier and later in the “day than the latter, and both sexes are low fliers. The females are not often seen except at flowers, and. these they patronize only during the warmer hours of the morning. In the city of Cebu, Cebu I., the wasp was abundant on a decum- bent weed in the public square, and near Dumaguete, Negros I., on the blue- ae verbenaceous weed, Stachytarphe ta jamat- censis (L). At Los Bafios it was pretty w ell restricted to the in- ee blossoms of “Cucharitas,” Alternanthera versicolor Regel, one of the Amarantaceae. This little border plant was common in the garden plots at the College of Agriculture. Under favorable conditions both sexes could be taken in some num- bers at these flowers, the males somewhat more abundantly. Scolia is rather wary. A female on alighting on a flower will often, as if suspecting danger, remain alert and motionless for some moments before commencing to feed. At other times when busy feeding, she has the habit, when lightly touched on the back, of elevating the posterior pair of legs as if to ward off the offending object. In cool weather the wasps fall to ground when disturbed and ‘ ‘play possum.” 56 After the morning meal the female wasp sets about finding her prey—lamellicorn beetle grubs of the subfamily Rutelides. While very few or no grubs were found in the Philippines para- sitized in the field by Scolia, 1t is most probable that the abundant Adoretus and Anomala are her common victims. The former particularly was used for breeding this wasp. The grubs which Scolia prefers are probably in the last stage of their growth, when they are 20 mm. or‘more in length. But it is evident from the diminutive size of some of these wasps, especially of the male sex, that small grubs must sometimes be utilized. Fig. 22. Scolia manilae, , X 5 (after Swezey). fo) e The grubs live from one to several inches in the ground and feed at the roots of plants, particularly of grasses. In spite of the frequent hardness of the soil the Scolia wasp by some sense, probably situated in the antennae, succeeds in locating her prey and in digging down to it. ~The latter -resents “the iutrusion of this unwelcome visitor, but to no avail, for it is soon put into a comatose state, from which it never awakens, by one or more stings in the nerve center of the throat or breast. It now rests quite limply on its back, but may exhibit signs of life in slight movements of the mouth parts. The wasp presses out a sort of cell about it and resting longitudinally on the grub, attaches her pearly white egg so that it stands upright on its head end, on about the seventh ventral segment of the abdomen, (Fig. 23). The incubation period is from two to about three days. The wasp grub feeds with its mouth applied always to the same spot on its host, (Fig. 24+). It moults several times, and in the latter part of its life, breaks through the grub’s skin and, inserting head and thorax into its victim, soon reduces it to a mere shell. From five to about seven days after hatching it J/ Fig. 23. Egg of S. manilae on Anomala grub, X 5 (after Swezey). spins a light brown, cylindrical oval cocoon (Fig. 25), in the space formerly occupied by the grub. The cocoon is about 12 mm. long and more broadly rounded at the head than at the caudal extremity. It 1s frequently spun against the remains of the larva and further supported in its cell by some loose periphe- ral strands of silk. The enclosed grub now has a resting stage of from a few days to some weeks, depending chiefly on the pe and humidity. It finally sheds its skin for the last (probably about the fourth) time and thus becomes a pupa, delicate, whitish and with appressed or folded appendages. Dur- ing this stage, which is always brief, it is capable of wriggling its abdomen a little. The cocoon stage lasts three weeks or more—- rarely less—according to season. The wasp casts off the thin pupal envelope but waits until it gains sufficient strength and hardness before neatly cutting off a lid from the head end of the cocoon to make its escape. The life-cycle of Scolia manilae in the Philippines varies from about thirty days to over two months—being longest during the 58 winter season. Hence there are several broods a vear, a condi- tion which adds greatly to its efficiency in reducing the Anomala erubs in the canefields of Hawaii. Fig. 24. Young S. manilae larva on Anomala grub, 5 (after Swezey). The wasp was bred in large num- bers at Los Banos for shipment to Honolulu, Hawaii. Approximately 100 females, each kept separately in a jelly tumbler, were used for breeding. Each tumbler was half filled with soil, into which was placed daily an Adoretus grub, and a sprig of “‘Cucharitas’” moistened with water and honey furnished food for the Scolia, All apparatus were kept on ant-proof shelves and tables. The Adoretus grubs in the tumblers were examined daily, uusally in the morning, and it was found that on an average about 50% were parasitized, though some which had been stung had no Scolia egg upon them. Each parasitized grub was placed on its back in a clay cell, which was then plugged up and placed in soil in a tightly- closing’ varnish can. The ‘latter packed in a box or basket were then ready for shipment. The mortality of the wasps in the tumblers was not great. The larger number lived several weeks in these unnatural conditions and we have a record of one female living for 79 days. Sometimes both grub and wasp would succumb to fungous diseases. While some of these wasps would lay quite satisfactorily others laid very few eggs. The one that lived for 79 days in captivity laid 52 eggs, the largest number recorded. Granting an abundance of grubs and no accidents in the natural outdoor life, Scolia manilae may well lay over 50 eggs during her existence. experiments have shown that an unfertilized female may lay eggs that will develop into adult wasps, but thus far only a few males have been reared. On the Island of Oahu, Hawaii, where this insect has been 9 on established, it increased very rapidly in numbers. Owing to the less tropical climate of the Hawaiian Is. as compared with its native home the Philippines, the life-cycle seems to average slight- ly longer in the former place. Thus records show that the egg hatches in from three to five days and that the feeding period for the Scolia larva is from six to ten days, and that the whole life-cycle occupies from 40 to 55 days. One captive female lived for nearly four months, but only laid nine eggs during that time. Five bred females on which Mr. Osborn kept records at the insectary at Waipio (near Honolulu), averaged 23 eggs apiece, and the average ee of life was 49 days. Thirty-two was the highest number of e gs by one female in this lot. Fig. 25. Cocoon of S. manilae on Anomala grub, * 5 (after Swezey). Scolia manilae is now far more abundant in the cane fields near Honolulu than we ever found it at Los Banos, Philippines. Indeed, in view of the rapidly disappearing Anomala orientalis beetle from the cane fields here and the scattered distribution of the Rose beetle (Adoretus tenuimaculatus Waterhouse ), cue ele ders how the wasp manages to remain so plentiful. Mr. H. Swezey has found it in the immature stages on Anomala ae in the field, and we have found it in one or two cases on Adoretus grubs in a cane field. On Oahu, we have caught for distribution many thousands of females. They were taken with a net, forceps or fingers,—their sting is not very startling—as they were feed- ing at the flowers of Stachytarpheta, Bidens or Euphorbia growing along canefield roads. Scolia hunting is profitable only on fay orable mornings, as during the afternoon both sexes— though the males w ith probably some exceptions—bury them- selves and remain underground until the following morning. 60 KEY TO THE SPECIES OF LOS BANOS TIPHIA. By P. H. TIMBERLAKE. Males. e> Head andethorax -notrre tcullathe dei ter. cree myst et eles peteone et sete Head and thorax with a fine but distinct surface reticulation. Punctures generally sparser in area in front of ocelli; carina on anterior margin of pronotum well elevated; submedian carinae on propodeum straight, generally converging towards apex (in one reared specimen perfectly parallel), median carina reaching two-thirds or quite to apex; 2nd and 8rd joints of antennae nearly or quite equal, the 4th hardly or not over a third longer than the 3rd. On Autoserica grubs. Tiphia ashmeadi Cwtd. 2. Joints 2 and 3 of antennae nearly or quite equal in length, the 4th nearly twice as long as the 3rd; anterior margin of pronotum behind the transverse carina without longitudinal sulci. Head and thorax rather sparsely punctured (except on frons just above antenna); truncated part of propodeum finely rugosely sculptured throughout; median carina on propodeum complete. On Anomala grubs. Tiphia segregata Cwfad. Joint 2 of antennae distinctly shorter than 3rd, the 4th generally hardly over a third longer than the 5rd; anterior margin of prono- tum, with distinct longitudinal sulci behind the transverse carina. Head and thorax closely punctured; truncated part of propodeum generally smoother and more polished above; median carina on propodeum often incomplete, sometimes not reaching more than one-third of distance to apex. On Adoretus grubs. Tiphia lucida Cwfd. Females. il velead and: ithorax snot reticulatede ee ac. seers imei ees nee ees 2 Head and thorax with a very fine surface reticulation. Punctures on head and thorax close set and coarse; inner spur on hind tibia reaching about to two-thirds of the 1st tarsal joint; legs except all coxae, tarsi and front femora reddish, the middle and hind femora more or less suffused with blackish, especially on their outer surface. On Adoretus grubs. T. lucida Cwfd. 2. Punetures on vertex rather sparse, especially just in front of ocelli; sarinae on propodeum with a row of punctures alongside each, sub- lateral carinae very indistinct or practically absent; inner surface of hind femora highly polshed and with a few minute scattered fine punctures, which become much larger and rather close-set along the lower margin; inner surface of hind tibiae very finely and closely pubescent-punctate on upper half, and much more thinly so on lower half, the two parts separated by a bare, shining longitudinal, narrow 61 area; upper half of propleura highly polished and smooth, the lower half rather coarsely obliquely lineolate. On Anomala grubs. T. segregata Cwfd. Punctures on vertex more numerous than in segregata, but considerably thinner than in /weida; carinae on propodeum sharply defined, with- out punctures alongside, the sublateral carinae distinet and reaching a little beyond the spiracles, and enclosing a distinetly sericeous area; truncated part of propodeum with a rather distinct median carina; hind femora on inner surface very highly polished and with the pune- tures slightly more prominent than in segregata; hind tibiae as in segregata except that the lower half of their inner surface is almost entirely bare, and polished without punctures; propleura finely obliquely lineolate on upper half, more coarsely so below. On Autoserica grubs. T. ashmeadi Cwfd. Fig. 26. Tiphia lucida, 9, X Tiphia lucida Ashmead. Length 8-10 mm. ; black. This insect (Fig. 26), originally described from Manila, is the commonest Tiphia around Los Banos. It was abundant from June to September at the College of Agriculture, and occurred chiefly on the patches of sweet-potato vine (/pomoea batatas (L) ). It appears to be a lowland form. It parasitizes the grub of Adoretus, which it temporarily immobilizes and lays an egg transversely on the ventral side of the body, to one side of the middle, in the incision separating the last thoracic from the first 62 abdominal segment. The egg is rather securely glued to the grub, parasitized specimens of which were eke brought in from the field, and also grubs which showed, in the brown “blotch behind the posterior coxae, that they had rubbed off the parasite’s egg. Nothing very definite is known concerning the duration of the early stages of this w asp. At most, it probably has not more than two or three generations a year, for while the summer life- cycle in three instances was found to be 44, 51 and 60 days, re- spectively, it is very much longer for the rest of the year. We have a record of T. lucida remaining in the cocoon stage for over twelve months. The 7iphia larva adheres rather firmly across the base of the thorax of its prey, and when it sheds its skin, this shrivels up into a pad beneath it. The last stage of growth is rapid, so that the Adoretus grub soon perishes. ‘The wasp lays fairly well in captivity, as the following breeding record at Los Banos for 1916 shows: Average No. co} enna Mee ee Roe ee No. Tiphia | Average No. Adoretus grubs | Average No. used days each kept oiven | eggs laid 5 | 62 | 11.27 21.4 | 13.8 The best layers of this lot, which also lived the longest and were given the largest number of grubs, were as follows | Tiphia | No. days lived | No. grubs given | No. eggs laid | | INO 48 Omri crate 32 wit 47 INGOs pil nome peso 3 97 58 IN OS 2 rarest ece as 23 59 47 INO REC Solas a eiciees cea 28 Ct 55 INGORE DOs verte ee 27 81 53 NON OA eerie ste sest 27 72 47 The parasitized grubs were, on account of their activity, placed in soil-filled jars or tins and not in clay cells. This apples for all the 7iphia. This wasp passes the winter months in the cocoon stage, and we have found that at least in the cocoons shipped to the Ha- waiian Islands this stage may be very much protracted. Thus, of cocoons sent from Los Banos during March, 1916, the last to 63 hatch was on December 9, 1916, or in about nine and a half months. Other Tiphia lucida showed a cocoon stage of eleven, twelve and thirteen and a half months. The emergences from a shipment were very irregular and often extended over several months. Pig. 27. Mutilla sp., which parasitizes the cocoons of Tiphia, X 14. The cocoons of this Tiphia were the ones most frequently dug up in the fields by Filipino grub-diggers. Their parasitism did not appear great, one or two yielding a rhipiphorid beetle, but numbers show ed small perforations indicz ating that ants hz id de- stroyed the contents. The country about Los Banos abounded in “velvet ants” or Mutillidae—many of small size. As these insects are known to parasitize the cocoons of various solitary bees and wasps, I was able to rear a small, two-spotted species (Fig. 27) from a bred cocoon of Tiphia lucida. On August 30, 1916, I buried seven cocoons, containing the quiescent pale yellowish 7iphia larva, in soil in a tumbler and introduced a small Mutilla with two white spots on her abdomen. (Fig. 27.) On September 3, I cut open these cocoons and ceana that two of the quiescent 7iphia larvae had been parasitized by Mutilla. In the one case the Tiphia larva Or 64 had a Mutilla egg transversely arranged on its dorsum between the first and second thoracic segments. The egg was semi- transparent whitish, about five times as long as thick, some- what curved, rather broadly rounded at one end and conical at the other. Under a hand lens its surface showed numerous pointed granulations. The second parasitized Tiphia had an active Mutilla larva a day or two old, on the underside of the body. Like the egg, its skin was also finely roughened. It re- quired about four days to become full-fed, consuming all its prey. It failed to spin a perfect cocoon, but remained as a Fig. 28. Tiphia lucida, Aedeagus; 2, side view of median portion; 3, side view of lateral portion. quiescent, dull whitish larva for about eight days before changing into a pupa, which a few days later hatched into a female wasp. I succeeded in getting several other Mutilla ovipositions. One cocoon contained two Mutilla eggs. The cocoon’s envelope is not visibly perforated by the ovipositor of the parasite. Tiphia segregata Crawford. Length 7-10 mm.; black This insect (Fig. 29) parasitizes the grubs of several species of Anomala and lays a comparatively small egg to one side near the ventral tip of the abdomen between segments 7 and &. Be- sides occurring on the lowlands about Los “Bafios, segregata was also found at Baguio, in central Luzon, at an altitude of nearly 65 5000 feet. There, Tiphia and other Hymenoptera begin to dis- appear in July, with the advent of the rains, but in the lowlands it has a more prolonged season, where it appears early in the year. The life-cycle is rather long, and the broods probably not more than ae OQ, Of three per annum. Mr. O. H. Swezey describes the stinging and oviposition as eae : Fig. 29. Tiphia segregata, 9, X 7. “September 3, 1916, while attending to breeding jars, a female Tiphia was noticed to be giving attention to an Anomala grub that was on the surface of the soil trying to burrow in. They were both removed to a vial for closer observations. The Tiphia seemed to be most concerned with the posterior part of the grub, continually feeling it all over with her antennae. This apparently annoyed the erub, for it would squirm and bend backward and try to attack the Tiphia with legs and mandibles. It would seem as though she was endangered, but she paid little attention to it, just moving out of reach, although it seemed at times as though she would suffer injury. Sometimes she would crawl around on the grub. After about twenty minutes to half an hour the 7iphia gave indications of trying to sting the grub. She finally, while perched transversely on the middle of the back of the grub, which was lying on its left side, extended her abdomen down under and forward between the legs nearly to the head and stung the grub, 66 paralyzing it. She may have stung it more than once. The actual stinging could not be seen, but the grub quickly became quieted, while the 7iphia abdomen could be seen in motion as if trying to find the proper place to sting—or perhaps stinging more than once. As soon as she was satisfied, she crawled around on the grub a bit, lengthwise of it, apparently to satisfy herself as to which was the posterior end and the correct place to put her egg, for she was continually vibrating her antennae, touching the surface of the grub continually. Finally after a few moments she seemed satisfied with her exploration, and embraced the posterior seg- ment transversely, being curved around the dorsal side so that Fig. 30. Tiphia segregata, Aedeagus; 2 and 3, side views of lateral portion; 4, of median portion. the apex of her abdomen came in contact with the right side of the grub just below the lateral fold and in the groove between the last segment and the one preceding it. At this point she rubbed around on the surface of the grub with the apex of her abdomen for about a minute, probably spreading on some substance to stick the egg in position. Finally the egg was extruded somewhat sud- denly. It adhered in the spot prepared, and the 7iphia immedi- ately left the grub. This was at 3:00 p. m., and the grub had recovered from the paralysis in fifteen ae and commenced burrowing into the soil of the tin box in which it was placed.” While this wasp laid well in captivity, only a small portion 67 of the young reached the cocoon stage, and this was especially true of Baguio examples. ves following is a breeding record for T. segregata at Ho- nolulu: Wasp No. days kept No. eggs laid INO nye tlie iions cravareiieyie cc c.tes 60 40) IN 033 \Gib cic olor NECN Coe 60 28 INOS Ae rayecete anise os 45 Gal INOS 2 Sree ce epee sie 41 64 IN (Oba Oboe Waverserstoelivete 34 90 TNI@S ats sie orcya © ORO CECE 27 27 These and other 7Tiphia often frequent honey-dewed bushes, but I also took a number of 7. segregata by searching the tops of Cosmos plants growing in a Japanese garden near Baguio. The wasps had evidently retired there that afternoon to pass tne night. Tiphia ashmeadi Crawford. Length 8-10 mm.; black. T. ashmeadi is a fairly common insect at Los Banos. Adults were taken from the middle of May to the following February. Like the others of the genus, it passes the dryer season in the cocoon, though beginning to hibernate at a rather late date. Females were often seen on the surface of the soil in well-shaded places, where they were sometimes the center of attraction for several males and the scene of quite a tussle for her possession. The wasp parasitizes the grub of a lamellicorn beetle of the genus Autoserica or an ally, and which is readily identified by its large and very mobile palpi. This grub is more rarely found and more active than either Anomala or Adoretus. Philippine Anomala were not parasitized by this Tiphia, although the Jap- anese Anomala orientalis in Hawaii were to some extent. The egg is laid on the thorax, between the second and third pair of legs. What appears to be the same 7iphia also oviposits on segment 4 or 5 of the abdomen. The life-cycle is quite the briefest of the Scoliidae studied, and in the Philippines the wasp was the easiest species to handle. The cycle for the winter months at Honolulu, ranged from about eighteen to twenty-eight days, of which about three days were passed in the egg, and about thirteen 68 for the larva outside its cocoon. At Honolulu and Waipio Sub- station on Oahu, T. H., some of the wasps lived for a consider- able period; one male in a life of six weeks mated with twenty- one females, and the latter sometimes lived for more than a month. The following table shows some of the best layers: Fig. 31. Tiphia ashmeadi, Aedeagus; la, side view of same; 2, side view of lateral piece; 3, side view of median portion. | Approximate Wasp | No. days kept | No, Anomala | No. eggs laid | | grubs given Bigeye at | 32 | 81 | 43 Dia eer kage al teeeees 26 66 9 Hee eA are | 16 | 45 | 29 CN ted ee ec | 24 | 66 | 33 Only a few of the 7. ashmeadi eggs laid on Anomala grubs in Honolulu developed as far as the cocoon stage. Out of 291 Ovipositions but twenty-six cocoons resulted. 69 The cocoon stage—at least, for those wasps shipped to Hono- lulu—was often very great. A certain number shipped during the latter part of 1916 hatched very shortly—or even sometimes en route from the Philippines—and up to the month of January, 1917, but other cocoons of this fall shipment did not produce wasps until from June to September of the following year. Thus in a few cases nearly a year was consumed in the cocoon stage. Cocoons shipped early in 1917 all issued the same year, though covering a period of several months. While a series of Tiphia from Baguio seems identical with IT. Ashmeadi at Los Banos, the former parasitized mainly Ano- mala, more rarely Adoretus. The egg is deposited on the under- side of the thorax or immediately posterior to it. It is possible that here is a distinct variety. THYNNIDAE. Methoca. The remarkable little wasps of this genus, which numbers per- haps upwards of thirty species, are found in all parts of the world except Australia. The best known is Methoca ichneumonides Latr. of Europe, and it was upon this insect that Adlerz (’03, ’05) made his biological studies and first discovered that the genus preys upon the larva of the Tiger-beetle (Cicindela). By a layman, the female Methoca would undoubtedly be mis- taken for an ant. She is entirely without wings, and her body is slender and polished; nevertheless, she is easily distinguished from any ant, for her abdomen is not notched above at the base nor are the antennae elbowed. Furthermore, her gait is different and her long curved sting may project beyond the tip of the abdomen. The male, on the other hand, is not at all ant-like. Of greater size than his mate, possessed of stout thorax and strong wings, is it any wonder that he was first named Tengyra! He, indeed, much resembles some of the scoliid wasps. The grub which the slender wasp must conquer is a formidable creature. Many times larger than her aggressor, a flesh-eater armed with sickle-like jaws, it would seem proof against the attack of such a puny antagonist. The top of its head and first segment are strongly armored, fit snugly together, and being held in a horizontal position at the surface of its trap-burrow whence it snaps at unwary insects, effectively block the aperture. But the wasp is more than a match for the beetle larva. She herself 1s exceedingly nimble and hard-shelled, and her very slenderness is an advantage. Moreover, she is cool and cautious and knows when and where to use her long sting—as we shall presently see. 70 Methoca striatella Williams. Female, length 5 mm.; black. I found the female Methoca (Fig. 6) at the College of Agricul- ture from August to November. She was not plentiful enough to be studied in her natural home, and so, like other observers, I had recourse to glass receptacles. Here I met with considerable success. A jelly tumbler two-thirds filled with well-packed soil got rs VB CG gla 50 2S aoe SYR} 3 Oe ey Le eter BE: Fig. 32. Methoca striatella awaiting her opportu- nity to sting the tiger-beetle larva, XK 10/3. into which and next the glass several holes, each to accommo- date a Cicindela larva, had been made, furnished the scene of operations. Provided I had sufficient patience, I encountered no difficulty in obtaining beetle larvae. These dwell in more or less well-defined colonies, each insect occupying a neat, nearly or quite vertical burrow in the soil. The tunnels have no debris nor heap of soil about their mouth, which when stoppered by the ant- mated but ground-colored lid, the head and first portion of the thorax of the larva, form a deadly trap for the unwary insect which walks within reach of the curved mandibles. Being pro- vided with a pair of forward-pointing hooks arising from a hump on the upper side of the abdomen, the larva by anchoring 7A itself firmly in its burrow with this apparatus is able to with- stand the strong pulls of an extra large captive. At your heavy tread the larva vanishes down its tunnel and may not reappear for many minutes. Digging up these insects is laborious; in- serting a straw in the burrow for the irate proprietor to grasp and be suddenly hauled out is feasible only in certain cases; lying very quietly in wait beside the hole, with an obliquely-held trowel just denting the soil an inch or two before the aperture, and pushing in this trowel with a sudden speed, thus cutting off the insect’s retreat when it is at the top of the burrow, was found to bring the best results. One beetle larva was placed tail-first into each hole in the soil in the tum- bler, a sprig of a weed stuck in the earth and daubed with honey, and fi- nally a larger and a smaller female wasp were liberated in the glass and the lid put on. Soon quieting down after liberation into the tumbler, the Fig. 33. Egg of M. actions of the methocas were various; STEN on 5th ven- they sipped up honey, rested or walked fae eae ee about their prison, or finally they be- Veg) came interested in their natural prey, the tiger-beetle larvae. These very often stopped up the burrows for a time and worked their way down deeper. During their long captivity (27 days) the wasps, naturally very shy creatures, soon became accustomed to the small quarters, they seldom stampeded unless unduly disturbed, and per- mitted the removal of the metal lid of the tumbler for their very close inspection. Since the predatory operations of these two little creatures were not without variation, I will recount several of the affrays between wasp and beetle larva. October 5, 3 p. m.: The bigger Methoca soon became inter- ested in a large larva which rested in its burrow at a distance of three or four millimeters below the surface of the ground, (Fig. 32). The wasp crouching, though with her abdomen well clear of the ground, pushed her rapidly-vibrating antennae over the hole. This caused the larva to slightly raise its head and snap weakly, at which Methoca backed away a little but immediately returned to repeat the performance. The larva snapped a few more times. Since the conditions were unnatural, it is probable that the wary tiger-beetle larva was not normally excited. The antennae which waved and wriggled sideways over its bur- Ie row certainly seemed very alluring; nevertheless, the fishing was poor and the wasp, becoming bolder at her lukewarm reception, poked her head over the burrow—as one would crouch at the brink of a precipice—bent her antennae downwards once or twice and, after the larva had snapped weakly a few times, made two or three feints at darting down towards her prey as if to goad it into action. The last feint aroused the larva a little, and Methoca, seeming determined to end matters, faced her adversary and, raising herself up somewhat, brought the abdo- men forward so as to be able to sting readily, quickly backed / Fig. 36. Cocoon of M. Fig. 34. Full-grown ; : larva of WM. stria- striatella, X 4/3. tella. eb: Fig. 35. Female pupa of M. striatella, x 7. down into the burrow, descending along the unarmored back or between the back and side of the Cicindela. There were some very quick movements from both parties, but almost 1mmedi- ately the little wasp rushed out headfirst, her prey retiring deeper in the tunnel. The procedure to this point occupied about fifteen minutes. I thought that the wasp’s attack on her large opponent had been unsuccessful. Not so, however; the long curved sting had done its work and the cicindelid was soon in a nearly coma- tose condition. Methoca, though somewhat ruffled, was un- harmed; she brushed herself but did not return to the scene of operations for some minutes. When she finally went to the burrow she approached her helpless victim with great caution and speedily backed out at its movements; eventually becoming bolder, she passed in and out of the burrow, inspected her prize, and gave it one or two strong pulls with her jaws. It seemed that she was not prepared to lay an egg, and so I left her late in the afternoon to find an egg on the larva early the next morning. October 6, 2 p. m.: The larger wasp was placed in another prepared tumbler, and, finding a medium-sized larva therein, promptly busied herself in subduing it. As before, she teased 73 or urged it, but becoming impatient at its poor response, she backed down into the burrow at an opportune moment and, always facing her prey, stung it instantly and rushed out. The act of entering the hole appeared deliberate if quick, and the stinging was done with a sort of jumbling speed. October 7, 8 a. m.: A fresh beetle larva was placed in a shallow hole in the tumbler which contained the larger Methoca. The latter vibrated her antennae before it, but the larva was uncomfortable in its short and well-illuminated tunnel and started to crawl out. The wasp, as if standing on her hind legs, bent her abdomen forward so as to have her sting in position, much as do braconid parasites, etc., in stinging caterpillars, and stung the cicindelid during a vigorous scuffle of less than a second while on the surface of the ground. I replaced the helpless larva in the tunnel, but Methoca, her possessive instinct at its height, would not be driven away from her prey, which I gently essayed to do with a pair of forceps. By the way of further experiment | grasped the jaws of the larva with the forceps and, pulling up, caused it to move; at this, Methoca again curled her abdomen as if to sting, but she evidently felt sure of her previous work and soon walked about her booty. October 9, p. m.: A tiger-beetle larva had plugged up its burrow for one and one-half inches and rested at the bottom of the tumbler. None the less, the smaller wasp located this plugged tunnel and, after a little preliminary caution, commenced digging out a passage through and in it. But being such a slender insect, with legs hardly fitted for digging, she accomplished little, pene- trating a half inch in about twenty minutes. She soon aban- doned the work. October 14: The larger wasp slid down into a burrow and alongside the cicindelid larva, which promptly seized her amid- ships in her jaws, but was immediately and fatally stung for her pains. ‘The released wasp was apparently unhurt. October 23, p. m.: The larger wasp became engrossed in a recently inserted fee and leaned inquiringly over its burrow. She grew impatient at the inactivity of the occupant, and at a movement from the latter deliberately lowered herself down, tail-first. At her entry the larva retreated further down, but Methoca appeared to be doing her best to place herself in a posi- tion to be seized. I was not able to clearly witness the ensuing struggle, but the doomed larva squirmed about violently and took many minutes to succumb. October 24, a. m.: The larger wasp attacked a larva, which for a brief instant held her in its sickle-like jaws. The released 74 wasp climbed up the burrow and then descended again tailfirst, stung the larva deliberately, but made haste to retreat. Although I watched all these operations very closely, it was difficult to see clearly the final, very rapid stinging action of Methoca; her prey was presumably stung about the unprotected throat, as Adlerz (1905), Bouwman (1909) and the Champions (1915); have ees ed in another species. The tiger-beetle larva is severely stung so as to remain in a death-like torpor; indeed, it sometimes perishes prematurely and the wasp grub with it. After the larva has become quiet and the wasp has cleaned and rested herself, the latter descends into the burrow and devotes further attention to her prey, giving it an occasional pull as it to straighten it out, and crawling over the body and pinching it with her mandibles, particularly about the venter. The egg may not be deposited for some time, and I was never able to see the operation in this species. The egg (Fig. 33) is pearly whitish, slender and slightly curved, rather pointed at the posterior end and more rounded at the other. It is about one millimeter in length. In specimens under observation, eggs were deposited, one to each Cicindela larva, from between ventral abdominal segments one and two to as far back as segment six, where one was placed latero-dorsally. As a rule, however, it is to be found on the underside of segment three or four of the abdomen, in the middle or to one side thereof, its long axis more or less paralici with that of the larva, and the broader head end pointing ante- riorly. After egg-laying, the wasp fills up the burrow with soil. As will be seen later, a majority of the solitary wasps that nest in the ground drag their prey into some hole well removed from the scene of combat, but the nature of her victim makes this added labor unnecessary for our wasp. Here is a ready-made tunnel of considerable length, the like of which the slender Methoca could not begin to excavate—why not bury the tiger- beetle larva in its self-made burrow? The prey is here of ex- ceptional size and the wasp has work enough burying it. The Scoliidae, a family of thick-set powerful wasps, readi ly pene- trate, the soil or decayed tree trunk to reach and paraly ze the sluggish chafer beetle grubs, but as a rule they find no need for excavating a special grave. This habit, then, does not appear to be the result of inferior talent, but one governed by the nature of the prey. The egg hatches in from forty to fifty-six hours. Once or twice the larva was seen to have crawled out of the shell, but on other occasions no egg-shell could be seen. The larva in its earlier life is comparatively large-headed, well segmented and rather stout and fusiform. Both ends of the body are rather Ze) glassy translucent, but the median or gut portion is pale orange with scattered white granules. During this first period it is an external feeder, its mouth being pressed against the body of its host.. Thus at this stage I was successful in transferring a Methoca grub from one larva to another. During about the last two days of its feeding life it develops a greater appetite and, inserting its head and the now curved portion of the thorax into its victim, rapidly sucks up the body juices. The wasp larva is now small-headed and slender, and the mid-gut is dark wine or brown color. It is full grown and ready to spin its cocoon in about five or six days, when it measures from ten to twelve milli- meters in length. (Fig. 34.) Even larvae which spin big cocoons do not always find it necessary to completely devour their large prey; in two cases observed the posterior third remained un- sucked, the cocoon later abutting on this mass. Methoca usually eat out at least the anterior portion of their prey, and when full-fed may sometimes be seen with the head within that of the hollowed Cicindela, waving vigorously. The following are notes on the growth of a single Methoca larva: October 26, October 27, October 28, oO athyca im. latched: pe imc Length, 1esO) nim, peem.2~ Wbenethy .3.50) mm: NT bo BS BDO DO io) October 29, ms. length, 7.10 mms @ctober 30; 2p. m.: Length, 10:00) mm: October 31, 7 a. m.: Beginning to spin. The cocoon (Fig. 36) varies in length from eleven to seventeen millimeters, and is from about three to four and a half millimeters in diameter. Abnormally small ones occur. As can be seen from the figure, the cocoon tapers to more or less of a point posteriorly, while anteriorly it ends in a sort of calyx or collar. In spinning the cocoon a rather flimsy outer env elope with a wide flaring mouth is made, and within this the more compact inner cocoon, with a constriction to mark the base of the collar, is spun, this collar being united along its edge with the thin, flaring mouth. Within the calyx or cup may often be found the chitinized por- tions of the Cicindela larva. The cocoon is of a light brown color. Larvae placed in a narrow clay trough spun perfect co- coons, which, as they consist of many well-separated layers of silk, require considerable time for their completion. The wasp bites its way out through the base of the calyx. I have seen only the female pupa, (Fig. 35). It is whitish, not quite as slender as the adult, but of about the same length. It Oo’ 7 bears some tubercles on the head and back of the thorax, and the abdomen is also armed, especially by lateral segmental processes. The sting is very conspicuous, and bent back along apical third of the abdomen, and from its position and extruded condition recalls the pupae of the non-aculeate, parasitic Hymenoptera, such as the Ichneumonidae and Braconidae. The cocoon stage may occupy from ten to twenty or more days for the summer cycle, or it may extend over a period of months for the drier season. Thus the development of Methoca from egg to adult may occupy less than a month. The following data show some Methoca life-cycles Egg laid, September .29, p. m.; hatched, October 1, p. m.; spinning, October 6, a. m.; adult, October 29; @. ras 30 days. Egg laid, October 1 or 2; spinning, October 9, a. m.; adult, October 29> 9.2 'Cycle 23 days: Egg laid, October 5, p. m.; hatched, October 8, a. m.; spinning, October 14; adult, November 11; ¢. . Cycle 37 days. Egg laid, October 5, p. m.; hatched, October 8, p. m.; spinning, October 13; adult, November 3; (7. | Cycle 29 days. Egg laid, October 4, p. m.; adult, November 9; ¢@. Cycle 36 days. As a rule the male and female insects were not taken in the same localities; the former I never saw on the ground, but took these rather handsome wasps flying or running jerkily on the leaves of low weeds or bushes that either harbored honey-dew- producing bugs or were in the shelter of such trees or shrubs. The wings of the male wasp rest flat on the body and reflect a purplish “black iridescence. The female, which seems the rarer of the two, was at times similarly attracted to honey-dewed bushes, but was taken more often in rather bare, shaded paths and about leaf trash. Methoca punctata Williams. Female ; length 6 mm.; red and black. My notes on this somewhat larger and stouter species (Fig. 12), while fragmentary, are sufficient to show that it differs to some extent in its biology from the wasp just considered. I did not devote sufficient care to the rearing of this insect and so cannot go much beyond the egg stage. As w ith M. striatella, wasps were 1m- prisoned in jelly tumblers with the same species of Cicindela. Yet this insect, which is larger and more powerful than the pre- Wh, ceding species, stings her prey much less severely than the latter, so that finally the ‘beetle larva attacked may for a time regain most of its normal activity. The following are some of my observations on the onslaughts of Methoca: @ctober 24°10) a: ms: I placed a female M. punctata in a tumbler supplied with soil and tiger-beetle larvae. On looking into the tumbler a moment later I saw one of the larvae holding the wasp amidships in her jaws. She was held thus for over a second, during which period she w as endeavoring to reach and sting the throat of her captor, and, accomplishing this, was re- leased and came to the surface of the ground, apparently rather dazed or crippled. She crawled about flatly on the soil and brushed herself. But placing her near the scene of the recent struggle, she reconnoitered and, entering the burrow again, ap- parently with care, stung the Cicindela, now nearly quiet, in or near the mouth, crawled over its body, Aas it here and there, and, pulling it upwards, worked upon the ventral part of the thorax. She remained in the burrow for about thirty min- utes, but the egg was deposited some time later. At 3:50 p. m. I noticed that this wasp had stung another cicindelid larva, the latter being one-third out of the burrow and resting upon its back. Methoca soon entered the burrow, and seizing her prey by the posterior end, pulled it within, where it was subjected to a vigorous kneading process, the wasp pinching it especially about fe thorax. Finally she placed herself venter to venter and head to head on the thorax of her prey and remained thus for a few seconds, presumably ovipositing. Then she climbed out of the hole and brushed herself. Like the first larva, this one was left with the head at the top of the burrow. October 25, 10:50 a. m.: Methoca placed with tiger-beetle larva which was some way down its tunnel. Down the latter the wasp crept carefully, tailfirst and facing her big opponent. The larva soon snapped, and thereupon the wasp grasped in her man- dibles a small root that projected into the burrow above her prey. She evidently thought this rootlet a part of the larva’s head, for holding it firmly fie stung at it repeatedly. But a moment later I saw the wasp _a little further in the hole and firmly held the cicindelid’s jaws, the larva moving violently and emitting a brownish mouth fluid. The larva stung, soon quieted down, was duly malaxated, oviposited on and the burrow filled up above it. October 26, 11 a. m.: Wasp soon located her snapping prey, and proceeding carefully down its burrow, grasped in her man- dibles a palpus or mandible of the cicindelid, and holding fast, bent her body beneath the head of her prey, endeavored to sting 78 it in the neck. She was not immediately successful, as the larva struggled fiercely and, emitting a blackish brown fluid from her mouth, soon messed up her enemy. But in the course of a very few seconds the larva grew quiet and Methoca, releasing her hold, got out of the burrow, flattened and scraped her venter on the soil, and rubbed her abdomen with her hind legs in an effort to free herself of the juice so liberally smeared on her body by her prey. October 26, 3:30 p. m.: The wasp walked down another bur- row in a bold but alert manner, and while the cicindelid was snapping viciously she grasped the side of its head or thoracic shield in her mandibles and, bending her body under the larva’s head, endeavored, free of and to one side of the big mandibles, to sting it in the neighborhood of the throat. The larva did not immediately succumb, but struggling violently, crawled out of its burrow and wriggled about until its tenaciously clinging and doubled-up foe stung it. The wasp, however, perhaps owing to the unwelcome coating of juice on her body, released her hold before the grub had become quiet. It is to be noted that although this wasp handles her victims more easily than does her slenderer darker relative, she does not reduce them to permanent immo- bility like the latter. One Cicindela which had been stung, but unsuccessfully oviposited on by M. punctata lived for five weeks thereafter. The egg is not pearly white as in the first species, but is lightly suffused with carneous or flesh color. It is, moreover, secured longitudinally on the underside of the third thoracic segment, the head end of the egg being anterior. Its incubation period is longer than that of striatella; two eggs under observation each re- quired about 72 hours to hatch, the larvae crawling out of the egg-shells. None was reared through the larval stage, but sev- eral cocoons were dug out of the soil by Filipinos. The cocoons resemble those of the other species in being similarly provided with a cephalic collar. The winter season is passed in this stage. Methoca ichneumonides Latreille of Europe completely or al- most completely paralyzes her prey. The egg is laid transversely behind one of the posterior coxae, and from eighteen to twenty- eight days elapse from oviposition to the spinning of the cocoon. Methoca stygia Say of Eastern United States and Canada usually only partially paraly zes her prey, so that it again becomes rather active. The egg is laid as in M. ichneumonides. Methoca striatella Williams of the Philippines completely para- lyzes her prey. The egg is placed more or less longitudinally on the underside of one of the abdominal segments. From seven 79 to nine days elapse from oviposition to the spinning of the cocoon. Methoca punctata Williams of the Philippines only partially paralyzes her prey, so that it again becomes rather active. The egg is laid longitudinally on the underside of the third thoracic segment. Incubation is longer than in M. striatella. PSAMMOCHARIDAE (POMPILIDAB). This is an extensive family consisting of small to very large wasps noted for their long legs and consequent great agility. The neck is short, and the thorax being usually higher than the head, gives these insects a rather hunchback appearance. The abdomen is without a noticeable pedicel or stem. A number of the species are grey-black or jet black, many have orange wings, while in others they may be iridescent violet black. They are probably the most active runners of all wasps, and inasmuch as the great majority of species prey on spiders, there is much need for swilft- ness. Bingham (1900-1901) has found Pompilus bracatus Bing- ham and Salius verticalis Smith which store their nests with cockroaches. Another species preys on crickets. The genera Macromeris, Paragenia, Pseudagenia and others are mason wasps, having advanced beyond the digging stage still adhered to by perhaps the majority of the family. ey build cells out of clay or other earth-like material; they may construct these in sheltered or unsheltered places above ground, more rarely in burrows. Cell-groups may be the work of several females, which live in apparently semi-social harmony for months at a time. The method of building cells is here dis- tinctly different from that employed by the other mud-dauber wasps, as Sceliphron, Pison, Trypoxylon, etc., for where these latter spread on the mud with the mandibles, sometimes aided by the forelegs, pompilids use the end of the body for this purpose, bending the abdomen under the thorax in an ‘ungainly manner.* In these w asps the dorsal part of the last segment of the body (pygidium), which forms the mud-manipulating apparatus, is more or less devoid of hair. The genera Pompilus, Pepsis, Salius, Aporus, etc., are emi- nently diggers. Of these wasps Latter (1913) says, p. 7: “The enlarged and closely approximated coxae are of great value to the + Froggatt in ‘‘Australian Insects,’’ p. 106, credits Salius (Priocnemis) bicolor with sometimes storing her burrow with Cicadas. In this he is evidently mistaken, for the insect which he figures (p. 105) for Salius is not a pompilid but probably Exeirus lateritius, a large wasp related to the Nyssonidae. * Howes (1917) noted one of these mud- daubing spider wasps in British Guiana thus using the end of her body as a trowel. The insect is wrongly identified there as a sphecid wasp. 6 80 insects when excavating their burrows. These nest-tunnels are often driven to a depth of several inches, and all the soil to be removed is brought to the surface by means of the enlarged and rather flattened coxae which act like hoe-heads beneath the body and drag the soil along the floor of the tunnel as the animal backs up to daylight from the dark recesses of her gallery: the close apposition of the middle pairs of coxae ensures that very little soil slips between to be left behind.” A few pompilids, however, are satisfied with making no nests of any kind. Hart- man (1905) speaks of a brilliant blue Texas pompilid which de- posited her spider prey on the margin of her own web, laid an egg upon it and left it to fate. Sharp (1901) speaks of Emery, who states that some pompilids merely sting and parasitize but do not bury their prey. A curious relation exists between Pompilus pectinipes V. de L. and P. rufipes L. and argyrolepis Costa. Ferton (1905) studied the habits of these insects in Bonifacio. He found that P. rufipes and argyrolepis, as is usual in the genus Pompilus, store their burrows with spiders which they themsely es capture, but the related P. pectinipes has parasitic habits, for it evidently smells out the closed burrows of the foregoing species, digs them up, removes the wasp egg from the spider’s abdomen and, laying her own in its place, refills the burrow. Ferton thinks that the para- sitic habit of pectinipes is a comparatively recent acquisition, for the insect, unlike the pompilid Ceropales, is still armed with a good sting and legs fitted for digging. The genus Ceropales is parasitic on the genera Pompilus, Pseudagenia and probably others, and is more or less specialized or modified for her work. According to Adlerz (1903), Cero- pales watches the uneasy Pompilus dragging a spider homewards, and at an opportune moment runs up and lays her egg in one of the cleft-shaped stigmata or breathing pores of the spider’s abdo- men. Subsequently Pompilus oviposits on this spider and buries it, but her own young is soon overcome by the Ceropales grub, which, hatching earlier from its place of concealment, consumes all. I noticed a similar parasitism in the Philippines with Nan- thampulex (Ceropales) vs. Pseudagenia and Pompilus, but did not notice the oviposition of Nanthampuler. Spiders of many kinds are overcome by the Pompilidae. Pepsis formosa, the big “Tarantula Hawk” of the West and Southwest United States, has often come into literature because of its duels with the great spider, and both spider and wasp, mounted on cards, are included among the souvenirs sold to tourists who visit * An account of a battle between Pepsis and the spider in Texas is given by S. B. Buckley (1861). (070) — the habitat of these creatures.* No less interesting combats take place in the tropical forests, and Bingham (1900-1901) tells of -a fight to a finish in the forests of Tennasserim, India, between a very large wasp (Salius sycophanta) and a powerful spider of the genus Galeodes. Salius is about two inches long and is fully as large as our famous “Tarantula Hawk,” while the spider Galeodes is a rather long-bodied, very active and repulsive crea- ture nearly two inches long, and belonging to the family Solpu- gidae. The account reads as follows: “*T once saw and timed a fight between the huge fossorial wasp Salius sycophanta Grebodo, and a very large species of spider (Galeodes) which is common in the forests of Tenasserim, lving in holes at the roots of trees and clumps of bamboo. I give a copy of the note I made at the time :— Camp ATARAN, 16th October, 1891.—Found the nest of a large hairy spider, a Galeodes, behind my tent among the roots of a clump of bamboos. The animal was outside and I tried to catch it, repulsive looking as it was, with a pair of forceps, but he, or she rather, escaped and ran down a hole. On digging it up I found the hole ran obliquely into the ground for about two feet and ended in a slightly enlarged chamber. I had dug gently and earefully to avoid injuring the spider and thus came on her crouching inside. I tried to pick her up with the forceps, but with a sudden leap she sprang on to my sleeve and began running up my arm. This was too much for me, and I gave a violent jerk to my arm, sending spider and for- ceps flying. The spider fell close to the side of my tent on a rather bare space and then, to my astonishment, scuttled for its life. I followed and eaught sight of the cause of its alarm. A beautiful large Salius with black body and tawny red wings, which I recognized as S. sycophanta, was flying in an excited way round and round. Looking for the spider, after a hunt, I spied her crouching between a tuft of grass and a bundle of wooden tent pegs that had been thrown down. She had tucked in her legs and made herself as small as she could, and I was struck by the resemblance of her reddish-brown furry body to the color of the ground, Quite a long time the Salius kept quartering the ground, and it seemed to me quite evi- dent she was hunting by sight, for as long as the spider remained motion- less, although the wasp passed two or three times within an inch or so, she never seemed to find it out. At last, why I don’t know, the spider made a rush trying to get to the shelter of a biggish clump of thatching grass about a yard away. Like a flash the Salius swooped down on her and then commenced a rough-and-tumble fight on the ground, both combatants roll- ing over and over so fast that I was quite unable to see whether the wasp managed to sting the spider or the spider managed to bite the wasp. How- ever, the spider in a few seconds shook herself free and again tried to reach shelter, but in vain—the Salius with a loud buzz flew right over and, alighting on the ground beyond, faced round and intercepted her. Then commenced a curious scene: the spider stood on the defensive, turning and facing the wasp, as the latter with quivering wings and abdomen paced round and round, evidently watching an opportunity to close again, and the spider quite on its guard, standing up and keeping a vigilant eye on her enemy. This went on, I should think, for fully ten minutes until, perhaps, tired out and taking advantage of the wasp’s stopping still for a second, the Galeodes made a second rush only to find herself grappled by her enemy. 82 This time the hand-to-hand or rather sting-to-jaws fight did not last so long; when the spider shook herself clear of her antagonist she was ev1- dently injured. Her rush for shelter was a crippled run, and it was clear she was nearly done for. The wasp, apparently as fresh as ever, leisurely flew after her, caught her up and this time I distinctly saw the sting go in onee, twice, three times, the poor spider making but feeble resistance and, unable to bite her enemy, lay still at length writhing a little. Then for a minute or so Salius danced her dance of triumph, parading round her prey in a quick jerky sort of walk, flirting her wings and quivering her antennae. Finally, she approached the comatose, if not dead, spider and deliberately bit off the long strong legs. Then, half flying with it and anon dragging it along the ground, she conveyed her shikar to her nest, which was exea- vated at the foot of a large padouk tree (Pterocarpus indicus) about fifty yards from my tent. Here she disappeared, and I had little difficulty in finding the entrance to her burrow. Placing my butterfly net over its mouth, I waited. In about five minutes out she came and I bottled her. Then I dug up her nest and found the bodies of no less than five Galeodes, all deprived of their legs and all with a single egg attached to the fur on the underside of their stomachs. So far as I could make out, all the spiders were quite dead, except the last caught, which still moved feebly when touched. As I said above, I timed the fight I have deseribed. From the time I saw the wasp looking for the spider to the time the latter lay moribund, having its legs sawed off, was exactly thirty-five minutes. C. T. BINGHAM, Conservator of Forests.’’ Mandalay, 19th October, 1899. Ferton (1905) states that Pompilus crassitarsus Costa and Planiceps helveticus Town. probably prey on trap-door spiders in Europe; a Pepsis in Brazil catches such spiders (Poulton, 1917), and Davidson (1905) has found the same relation between Para- pompilus planatus Fox and the trap-door spider Cteniza cali- fornica in Southern California. Pompilids do not always attack spiders in the open, where there is plenty of elbow-room, though probably they more com- monly do. Sometimes the wasp will grapple with the spider, at the same time evading its fangs, and during the rough- and- tumble fight that ensues, ‘cripple it with a sting; others are more careful and wound their powerful antagonist without closing in.* Many of the species, particularly the architect Pompilidae, have the curious habit of snipping off the legs of their paralyzed vic- tims close to the body, the work being done with speed and dexterity, and little or no blood will ooze from the wounds. Not all the legs may be cut off, and occasionally none. Perhaps this snipping ‘off is done so as to economize cell- space, but we find that the wasps addicted to this practice carry their victims quite * Belt (1874, Chapter VIII) speaks of a wasp, evidently a pompilid, as follows: “‘In Australia, I often witnessed a wasp combating with a large flat spider that is found on the bark of trees. It would fall to the ground and lie on its back, so as to be able to gr apple with its opponent; but the wasp was always the victor in the encounters I saw. 83 handily beneath them, while those that are not leg cutters almost always adopt the clumsy method of dragging the spider as they run backwards. No Pompilidae, so far as I know, stores more than one spider in a cell, each victim being of sufficient size to feed the wasp grub to maturity. The sting inflicted by these spider wasps is said to be burning and painful, and in this they differ from many of the other soli- tary wasps. : Macromeris violacea Lepeletier. Length of body, female, about one inch (25 mm.) ; body black ; wings proportionate- ly large, iridescent violet black ; legs long. Macromeris is alike a deni- zen of the dark Makiling for- est and of the more cultivated areas about the base of the mountain. Conspicuous for her size and dusky coloration, she is occasionally seen mak- ing short flights in the woods, banana groves or about ; palms, and searching the hol- f lows of trees, trunks or other places likely to harbor HHeter- opoda venatoria Linn., a large and decidedly active spider, a well-known creature of the tropics and a frequent resi- dent of the houses there. This Fig. 37. Portion of a tree hollow spider forms the food of the showing a Macromeris violacea Vacromeris grub. At other guarding her cells, X 1/2. : Sere : : times the wasp is occupied in building her cells, selecting the material from tree trunks or the COV eel runways of termites or white ants. She 1s not a common insect, and if you are able to follow one to her nest, placed well within the hollow of a tree or in a partly open bamboo stem, you will likely find one or more alert and aggressive fe- males perched in the semi-gloom, about a group of large oblique cells, (Hig: 37). Perhaps a male will be present also. He 84 is a stouter-legged creature that makes a show at being fero- cious—a mere bluff on his part, retreating as he does before your advances. Madam Macromeris, on the other hand, has much at stake and is far more in earnest, especially if she be the foundress of the little colony, when she is often bold and fearless. She will elevate her wings in an angry buzz and frequently refuses to be routed. This wasp is particularly warlike when she has just brought in a spider; indeed, I was once compelled to imprison an enraged Macromeris, an unusual necessity with solitary wasps, in order to examine her nest. Like her smaller and more amiable cousin, Paragenia argenti- frons, next treated, these black ogresses may live in small groups or communities in fair harmony for months at a time. But | have found their nesting sites usually less accessible to vision than those of Paragenia, and necessitating an awkward pose on the part of the observer, preclude’ as close a-istudy; The cell material seems to be very largely composed of the earth-like substance employed by white ants in making their long cover- ways along tree trunks. Bark is also used and doubtless water 1s added to the material. Maindron (1878), in describing the cells of M. splendida, which he found at Ternate and Gilolo ( Moluc- cas), mentions as cell materials, vegetable debris, the bark of trees, and a sort of gummy cement which is insoluble in water. Dutt (1912) in his “Life Histories of Indian Insects,” speaks of clay, sand, chewed-up vegetable matter, gummy substances, etc., as used by M. violacea. We are accustomed to see the spider-wasps or Pompildae running along the ground in search of their victims, but Macro- meris seems to keep off mother earth as much as possible, and this is quite a natural procedure if we consider the more or less arboreal habits of the big spider which she seeks. Likewise, some of the Pseudagenia, smaller but related architects, prefer a tree “life: On November 15, 1916, I first located the nest of Macromerts. It consisted ultimately of fifteen neat cells, a compact lot ar- ranged in more or less vertical rows in the darkness of a tree (Cordia sp.) hollow, at an elevation of two or three feet from the ground. The cells, their aperture upwards, were obliquely placed. From one to three wasps were present, astraddle the cells. On December 22 I detached the cells. Fourteen of these were completed, while one at the lower end of the inner row was just commenced. A cell recently vacated had the aperture bitten out by the emerged wasp, neatly enlarged and rimmed again for further use; a second cell contained a large spider with a wasp egg on it, a third cell contained a spider and a young wasp larva, a fourth cell had a nearly replete larva, while a fifth was vacant. The nine remaining cells were closed and 85 contained cocoons. As we shall see later, one is often able to tell whether Macromeris is utilizing a cell for the first or for the second time or more. I kept these cells in a receptacle, and all the stored cells save one whose contents I had considerably dis- turbed came through successfully. In my brief and limited ex- perience with both Macromeris and Paragenia I have found no parasitism in the nests, whereas the cells of the several Pseuda- genia which I have studied produced a large percentage of ichneu- monid parasites (Cryptinae). Dutt (1912) had similar experi- ences with Indian Pseudagenia. This may be partly explained by the habits of Macromeris and Paragenia, which, being semi- social, are thus better able to guard their nests, while Pseuda- genia, quite solitary as a rule, cannot protect their cells and hunt for spiders at the same time, and, furthermore, they desert their completed nests to form others elsewhere. But one might, in consequence, expect to find the semi-social genera the more nu- merous in individuals, though this does not seem to be the case. The Macromeris whose nest I had destroyed stayed in the tree hollow for some time; they appeared much concerned as they walked about the nesting site, tapping here and there with the antennae.* On November 28, I discovered that a hollow in a large branch of this same Cordia tree, an isolated specimen near the edge of a banana plantation, harbored another Macromeris nest. One of these wasps dragging a spider on the roadside was a conspicuous object. She hailed from the banana plantation across the road, which, of course, she was unable to bridge through the medium of trees, heavily burdened as she was. There was no doubt, how- ever, that she was very anxious to climb up anything that of- fered, in order that she might make a sailing flight towards home, for, utterly unmindful of me, she immediately crawled up my bamboo net handle which I presented to her, then along my hand and arm, and had I not diverted her course, again by inter- posing the bamboo stick, she certainly would have scaled my person to its highest point. From the end of this stick she para- chuted heavily to the tree. Here I noted that she carried the spider close beneth her body, grasping it with her mandibles by the underside and posterior end of the abdomen, near or at the paired papillae-like processes there. All but three of the spider’s walking legs had been bitten off close to the body. Macromeris, with raised wings, traveling as hastily as possible, climbed up and sailed from one limb to another. I soon saw that she did not belong to the household of which I first spoke, for there she * Hollow occupied again by August, 1917. S6 could not be persuaded to enter, but after some angry buzzing she finally disappeared seven or eight feet up, inside a hollow branch. This nest, rather inaccessibly placed, contained three or four wasps at one time. I ascertained that three were females. It had a long existence, as it was not abandoned until the latter part of July. Apparently the faithful wasps await the hatching of their progeny. Within a week of the time of abandonment of the nest above referred to, I located an old female Macromeris which had just begun to build a cell in the partly exposed node of a bamboo stump, said stump also containing a thriving colony of Paragenia spider-wasps, which occupied the node below Macromeris and is referred to extensivély in the next article. This was on the morning of July 28, and I have good reasons to believe that the rather dilapidated Macromeris had been the proprietress of some of the recently-abandoned cells in the Cordia tree, perhaps a third of a mile away. The bamboo stump was excellently situated for obser\ ation and the old and sagacious architect almost immune to fear. By the afternoon of July 29, she had made little prog- ress, the cell being less than one-third complete. But since the interior of the bamboo was smooth, Macromeris could only wath difficulty secure a sufficiently firm foothold to plaster away; ac- cordingly, I scratched up and dented the wall about her ceil, and this assisted her v rastly. Like Paragenia and Pseudagenia, our black giantess chews up and revolves her building material in her mouth, and smears it on the cell with the dorsal tip of her abdo- men. By the afternoon of August 1 the cell was complete. It was placed with its long axis vertical, and was furnished with a short nipple at the lower end. It was a well-made affair of a rather springy nature. Its base was the bamboo wall. At 3 p.m., August 2, Macromeris had nearly closed up her cell, within which could be seen a portly spider. Whereas we are accustomed to note a cat’s change of nature after she has acquired the responsibility of a family, so also is this change evident in our big black wasp. Heretofore fearless and impa- tient, she is doubly so now that she has captured her spider, stored it and placed a precious egg thereon. I would not hesi- tate to call her vicious as she buzzed angrily at my curiosity. By 5:00 p. m. the cell was capped, the wasp putting on the finishing touches. Unlike Paragenia, she fetched her material—after the spider was stored—from without the nest. During one of these trips to the field a Diacamma, a ponerid, one of the largest Phil- ippine ants, came upon the cell and began biting into its. still moist cover, but the ant was angrily nipped and put to flight by the irate Macromeris returning with cell material. The wasp, 87 disburdening, examined the intruder’s damage, patching up the spot, biting dry bits from the cell for this purpose. The cell cover was smoother than the rest of the exterior, and its fovt- rest or nipple had evidently been used up in sealing. The old mother had about completed her life’s work, her worn body Fig. 38. MM. violacea with her spider prey, Heteropoda vena- toria, whose legs she has bitten off, and which she is carrying up a tree, X 2. probably contained no more eggs, and she devoted her few re- maining days to guarding the cell. Perched upon the object of her solicitude, her tarsi would every now and then twitch ner- vously. It was seldom that she took a brief trip to the field, and I observed her upon the cell during the night. Finally, on the morning of August 15, she was found dead, but hardly stif- fened, in the bamboo partition or internode below her cell, which now enclosed a fine large cocoon. 88 I was able to study another Macromeris nest over a period of some weeks. On February 18, 1917, two nests were located higher up on the slopes of Makiling ; one was hopelessly hidden far within a hollow tree, but the “second was quite accessible, placed as it was in the stump of a deformed old tree, open above. Under the overhang within this hollowed top and concealed from view mainly by a few vines, was an angry wasp and her Wee celled nest. Two of these cells were closed. At 11:15 a. 1 March 2, there were four cells, while the wasp, with a load, ws cement in her mouth, appeared to be contemplating a fifth; this was practically completed two days later at 5:30 p.m. Next day there were two wasps about the cells, and though they did not really quarrel, their attitude towards one another was very war- like. At 1:40 p. m., March /th, I heard a patter and buzz above me, and looking upwards, Macromeris came into view on the trunk over her dwelling. She was heavily laden with a big spider, with which she had leaped, evidently, from the higher branch of a neighboring tree to her present position. She carried the spider c losely pressed beneath her, underside upwards ( Fig. 38), grasping it by the two papillae-like processes near the hind end of the body. The victim was quite large, apparently full of eggs, and had been stung into insensibility. Much heavier than its captor, it was depriv ed of its four pairs of walking legs Macromeris lost no time in stowing her prize in a cell, first insert- ing therein the spider’s abdomen so that the creature would face out. Then the wasp set to work examining her prey; at this juncture I broke off the cell, but let it lie, its base now open, in the near bottom of the tree hollow. Here, a few “small ants promptly invaded the cell; this was soon discovered by Macro- meris, who became greatly enraged thereat. Buzzing in loud anger, her wings widely spread, ‘she grasped the spider by the cephalo- thorax ‘and jerked it violently out of the loosened cell. This freed it of the ants and she dragged the carcass up the hollow, while I cleaned out the cell and plugged up the opened base with cotton and wedged it among the upper cells. Macromeris soon stored the spider once more, but she objected much to the protruding cotton and made repeated pulls and digs at it, so that the cell became loosened again and all went tumbling down into the hollow. But the wasp clung angrily to the cell and could not be driven away, and threatened to climb up my forceps. I finally took the cell outside the hollow; at this she was greatly perplexed and hunted for it even on the ground about the base of the treex In order to replace theicell: winch I did more firmly, I was obliged to imprison the warlike insect. Upon being re- leased shortly afterwards the wasp went directly to her cell. The 89 next afternoon, March 9th, at 2:05, the cell had been sealed up. I watched this nest for two weeks thereafter, but there was little work added. She probably considered her cell-group sufficiently large and would be content to see her brood hatch and perhaps use over again the cells thus made vacant. I have found Macromeris nesting well in the forest at an alti- tude of perhaps 1000 feet. Here. one was seen gathering cell material from a standing tree trunk 100 feet from her nesting site. She, however, was quite shy. Her single capped cell was well ensconced in a natural hollow formed by the root and but- tresses of a large tree. Sometimes a male is to be found in the nests. He is tolerated by the fe- males, often perched for long periods on a cell, as he parades about with ele- vated wings. -One of these males I found badly crippled, two of his legs on one side being lacking. The wasp may search in quite con- fined quarters for the large spider ; this shows that she does not fear it much. Occasionally one finds this spider in Fig. 39. M. violacea cell close proximity to a wasp’s nest and viewed from beneath, Jacking much of its great activity, sug- een Mere gcouve that Macromeris had stung it ee arall Size: but not sufficiently. She may tolerate small gecko lizards in her nest hollow, though I have seen her rout, by a quick dash, a greenish tree lizard. Evidently other wasps than Macromeris prey upon the big Heteropoda spider. A Salius sp., an insect fully as large as Macromeris, but with black and orange wings, and much like Pepsis formosa, or the “Tarantula Hawk” of the West and South- west United States, was once seen hunting for a spider which she had flushed in the neighborhood of a Macromeris nest. At intervals Salius would fly about the Cordia tree to alight and pur- sue usually a definite set of trails, and as one of these led very close to the interior of the Macromeris hollow, one of its denizens would frustrate the strange wasp’s venture into the forbidden ground by rushing fiercely out and compelling flight. The cells of Macromeris are fairly uniform, being on the aver- age 33 millimeters long by 20 wide and 20 in height. Their form is shown in Figs. 39 and 41. Externally the walls are rough, with the cap smoother, and the interior still smoother. The egg is deposited on the underside of the spider’s abdomen 90 (Fig. 39) near its base. Two were examined—one measured 6.5 Mitinelees long by 1.4 in diameter; the other 7. by 1.5 mm. They were rather dull creamy white, slightly thickened at the head end, as viewed from above almost straight, laterally slightly curved so as to conform with the curvature of the spider’s body. Incubation in one case was a little over three days. The wasp grub feeds four or five days fixed almost immovably on the spider’s abdomen, and during this time seems to have cast off its skin once. Thereafter becoming more active, it uses its jaws more for chewing than for pricking, and rapidly consumes the harder portions of the victim’s anatomy. It finally makes use of the more anterior part of the underside of its body as a table, eating into a dark Fig. 40. Mature larva of uninviting mess thereon—the remnants of M. violacea with rem- the spider plus debris, ete.y Finally hav- nant of spider on its . 5 lap. About natural size, 4g consumed all, the grub measures, when straightened out, from 27 to 34 millime- ters, or over an inch long, with the greatest width about 9 mm. It is shining white, smooth and well segmented, with the lateral fold strong. The fat bodies are small, intersegmental and dor- sal. The anal segments are yellowish. The head is moderately large with the usual brownish oblique facial lines. About seven cays after hatching it is full fed (Fig. 40), and soon commences spinning a cocoon. ‘This is a matter of two days or more. When complete the cocoon 1s about 28 mm. long and 1 mm. wide, broader at the head end and tapering posteriorly, (Fig. 41). It is golden brown and rather shiny, and it lies be the longer axis of the cell, its darkened, more pointed tail end touching the lower or pos- terior end of the cell; while the body of the structure is well hung by supporting threads. The broader, rounded head end is very nearly in contact with the lower part of the cell cap. When held up to the light the resting larva or pupa is easily visible through the thin walls of the cocoon. The larva may wriggle quite audibly in this enclosure. We must bear in mind that the cell rests directly upon the tree hollow, which usually forms a large part of the floor over which, and prior to spinning its cocoon, the larva of Macromeris weaves a sheet of silk, thus making its cell more impervious and secure. In examining the cells of one of these insects it will y It is interesting to note that this method of feeding is not confined to solitary wasps. Wheeler (1900), p. 295, says regarding the ponerine ant Odontomachus clarus in Texas: ‘‘These larvae are placed on their broad backs and their heads and necks are folded over onto the concave ventral surface, which serves as a table or trough on which the foode is placed by the workers.’’ 91 be found that some lack these silken sheets; these are the newer cells, still unoccupied or containing an egg or a larva. Others containing an egg or larva already have this sheet. We find, then, that the wasp economizes in using a cell over again, as do social wasps. Fig. 42. Ventral view of pupa of MW. violacea, X »/”) 9/¢ o/s. Fig. 41. M. viola- cea cell in verti- cal section show- ing cocoon with- in. About. nat- ural size. The pupa (Fig. 42) is, in the female, about 23 mm. long, large- headed and long-limbed. At first it is whitish, but soon darkens with age. The 1 upper portion of the body bears several processes and spines. The abdomen has four forked puplecetes on each side, and a fifth unbranched one. The tarsi are swollen at their tips, and the tibial spines prominent. On January 29, a. m., I saw female Macromeris emerge from her cocoon; she took about thirty minutes to bite and force her way out of it. A copious brown juice which she exuded from her mouth softened the silk at the broader end of the cocoon and thus aided her efforts. Finally an irregular hole was formed and, struggling actively, she parted the last resisting strands. Macromeris came out with wings fully developed though some- what bent down at their tips. She was quite sprightly. A rough hole is bitten out of the cell cap by the emerging wasp. The life-cycle of the single Macromeris that I followed through was: December-February: Egg stage, 3 days; larval stage, 6 feed- 92 ing days and 9 days in cocoon; pupal stage, 24 days. Total, 42 days. M. violacea as observed by Dutt in India was: Egg, 2 days; larva (including resting stage), 10-11 days; pupal stage, 14 days. Total, 26-27 days There are many details in the life history of this handsome wasp that remain to be worked out. Paragenia argentifrons Smith. Length of body, 17 mm. ; grey-black, wings transparent, legs long. The life of this rather widespread Oriental insect is a most interesting one, for, like its large relative Macromeris, it 1s fond of company and, moreover, is a wasp whose domestic activities are often easy to observe. Its favorite nesting place is the hollow of a bamboo, whether it be a split stump, fence-post or beam, whose aperture does not permit too much daylight to enter and isso inclined as to exclude the ram. It seems to, prefereto work in the shaded bamboo groves and indeed is not averse to carrying on its building operations during a light rain or, if needs be, when the stin is on. the point of setting. Although more sociable than Macromeris, it is not so aggressive, while in the work about the domicile (Fig. 43) it often appears like a case of “too many cooks spoil the broth,” since the wasps do not always mind each other’s affairs, but, wishing to contribute, interfere. Nevertheless, the little group of Paragenia wasps presents a peaceful aspect ; in my many observations I have never seen what must be the aggravating conduct of one wasp resented by another. It is but fair to state, however, that some of these insects may bear permanent evidence of rough treatment, in the form of broken antennae or dislocated legs. There seems to be a sort of password system ina nest—often when a female wasp enters the hollow she is challenged by a female within, which makes a short dash at the new arrival and, giving her a brief facial in- spection at very close quarters, engages her no further. Every- thing proceeds quietly. More timid males frequently hang about the entrance, while a confident one feels quite at home within. Several nests, all in split or cut bamboo, were located, but a large one, first observed on July 10, 1917, proved very easy of access and furnished the bulk of my notes on Paragenia. It was situated in a node of a bamboo stump in a grove growing by the side of a stream. This stump, some three feet tall and three and a half inches in diameter and composed of three or four nodes, had been cleft longitudinally so that a slab about one- FORE Fig. 43. Node of bam- boo stump with one Se LZ7ZTILy side removed to show cells of Paragenia ar- gentifrons within. At A, is shown a cell from Fa —— which a wasp has re- OPT A a ae aw ZEIISIST TILIA cently emerged; at Bb, is a cell with the top portion removed to show the delegged spi- der, Heteropoda ge- mella, with a wasp’s egg upon its abdomen; at C, is a cell contain- ing a Paragenia c¢o- coon; at D, is a wasp which has just stored a spider; while at E, a mars LE beat a on ara are two cells ready for storing. Length of cell- mass, 6 inches. 94 third the circumference of the bamboo was separated by a large crack from the main body for about two-thirds the length of the stump. The elastic slab was on the overhanging side and could be further drawn away, thus more fully exposing the contents of two and a half nodes. The topmost of these was soon des- tined to contain a single cell of an old Macromeris wasp, the second sheltered the mud cells of Paragenia, while a portion of the node near the base appeared to furnish a nest for Diacamma, one of the large ponerine ants. On July 10, 1917, this Paragenia nest was composed of eight mud cells, all sealed and vertically arranged, one touching the other, in one side of the node; on August ‘16 there were twenty- two sealed cells, one unsealed, and the rudiment of a twenty- fourth. The nest was examined almost daily up to September 12, or for more than two months. It is probable that on July 10 this nest was already two or three weeks old. The greatest number of wasps present at one time in or about it was eight. It is probable that in most cases but one wasp begins the con- struction of the nest. The cells (Fig. 43) are longer and broader than high, and their point of attachment forms the basal side. They are composed entirely of mud collected in some moist spot. This mud, gathered by the jaws of the wasp, is turned over and over in her mouth-parts until it assumes the right consistency, when, lowering her head and at the same time bending the abdo- men forward beneath her until its dorsal tip pierces the ball of mud, now almost touching the bamboo, she applies part of the mud to the wood, where it is spread on as desired with dorsal end of the abdomen. This mason work is repeated until the mouthful of mud is expended, when the architect flies away again to renew the supply. The walls of the cells rise evenly, and since the builder apples and plasters on the mud from the inside, the structure is always smoother within, and the wasp often assumes a very cramped position as she works in the nearly completed cell. The length of time occupied in building such a cell is variable, as the work is frequently done intermittently, but often extends beyond a day. When the cell is ready for the reception of a spider it has a large well-rounded orifice. I did not see any of these insects capture their prey (/Hetero- poda gemella Linn.). Paragenia, though alert, has by reason of its very long legs a draggling or drooping sort of gait as she searches about the trash at the base of bamboo clumps and other likely places for suitable spiders. These are more or less de- legged in the field; perhaps all but the short anterior pair of legs (pedipalps) are severed from the body, or one or more of the true legs may remain attached. The wasp grasps her prey 95 as does Macromeris— that is, by the ventral end of the abdomen —and carries it beneath her body. Thus laden, she backs into the cell partly bringing in the spider with her, and then as she emerges from the cell she tucks in her burden at the same time, using the end of her body very deftly for this. Thus the spider’s underside faces the mouth of the cell and lies with the head end upwards. It may be many hours before the egg is laid and the cell closed; in fact, it may remain open overnight. The wasp sticks her head into the cell, attending to the spider in some manner—she often does further tucking in with the extremity of her abdomen, or she may have company, one or more compan- ions, who act towards this spider much like herself. This fuss- ing may be partly due to unpreparedness for oviposition. Finally, after a careful look at her prey, she reverses her position and, sticking her abdomen within the cell, remains thus for some seconds, evidently laying an egg. This is placed near the base of the underside of the abdomen. Then she sets to work promptly to close the cell with a cap of mud. But we must note that now she no longer gathers mud from some moist spot, for, as if con- scious of the danger of parasites, she bites off pieces lower down on the cell itself or at its juncture with another and uses this ma- terial, moistened in her jaws, for capping. Only occasionally does she fly out for a quick sip of water wherewith to soften the dried mud. Owing to the diminishing size of the cell’s aperture she is eventually forced to plaster from the outside, and that 1s why a portion of the cap is somewhat smoother than the sides. The length of Paragenia’s life-cycle was not determined, save that, barring accidents, the mother lives to see at least a good part of her progeny issue from the cocoons and cells. The wasp grub hatching from the egg, proceeds to consume the para- lyzed spider; in a few days it is full grown, having devoured practically the whole of her prey, and spins a thin yellowish- brown cocoon, tapering at one end and rounded at the other. As in Pseudagenia and Macromeris, the cocoon, about 19 mm. long, is secured to the inside of the cell by its tapering end, and also supported elsewhere by strands of silk. The pupa 1s an interesting object. It much resembles that of Macromeris, but in addition to the lateral Y-shaped processes of the abdomen, bears a forked thorn and two successive knobs on the back of the thorax and three pairs of thorns on the top of the head. There are also thorns near the base of the antennae. The cocoon and pupa lie in the long axis of the cell. The wasp in escaping from the cell bites a small hole through the cap. If this cell happens to be needed by one of the mothers of the colony, the debris is removed from it and the emergence hole, made by its “J 96 former inhabitant, neatly enlarged and rounded off so that it has the appearance of an unused cell. The bits of mud bitten off in this work are not cast aside, but plastered on the cell’s walls, where they will come in handy for recapping the cell after another spider is stored in it. It is probable that the wasps use any cell in the nest they find vacant and do not concern themselves about its builder. The operations as related above are frequently rendered more complex or irregular through the erratic behavior of one or more of the female wasps. Thus it often happens that when a spider has been tucked away in a cell, not one, but two or three wasps give it their attention, so that we can only guess which is the lawful owner of the prize. The wasps will crowd about the cell; one will poke in her head and become interested in the spider, then she will withdraw and a second Paragenia now steps forward and investigates likewise. Perhaps they feed on the spider’s mouth or leg juices. On one occasion a wasp hauled the spider out of the ‘cell and let it fall to the bottom of the node, much to the consternation of the others. Again, late in the afternoon a Paragenia brought in a spider from the field, but there was no cell ready to store it in. Nothing daunted, however, the wasp, after wandering about a little, dropped her burden on the internode and set to work to make a cell, building it up from a mere rudiment—two short strips of mud_ laid down many hours previously. She worked rapidly, nevertheless darkness put an end to her industry. 1 watched her plastering away until 6:10 p. m., and after supper visited the nest again at 7:50 p. m. to find the cell incomplete, the spider unstored, and three female Paragenia about the cell-mass. Thus it 1s evident that the females, at least in part, stand by their nest during the night. At 7 o’clock the next morning the spider still lay in last night’s position, but by 8:45 a. m. had been stored in an old cell whose newly-made emergence hole had been enlarged. I pulled the spider out of this cell, and the alert wasp, after cutting off the remaining leg, once more stored it within. Sometimes a couple of wasps will simultaneously build cells side by side—close quar- ters indeed for such long-legged sprawling creatures—and though they may necessarily interfere with each other’s work, there is no quarreling. Again, capped cells may for no obvious reason be opened and sealed up again. One such cell contained a Paragenia cocoon, another a spider five days old. In another case one wasp was plastering up a cell which contained a spider while a second wasp was plastering as well as biting open this same cell. But when the egg (here perhaps not the first one) was deposited, the cell was closed without interference. OF, So that one is inclined to marvel at both the stupidity as well as the common sense of these creatures. Pseudagenia blanda (Guérin Méneville), Length 13 mm.; metallic blue; hind femora red. This very pretty insect is strictly a forest dweller; here she fre- quents the muddy paths at lower elevations, seeking with much fastidiousness the mud to build her stout cells. She is a shy wasp, as she walks about nervously here and there inspecting the mud. Finally arriving at a suitable spot, she bites out small pieces which she works up in her mouth parts to a more homo- geneous and fluid consistency. When a sufficiently large ball has been accumulated she suddenly straightens out, raises herself up and takes wing. Though her cells may be only fifteen or twenty feet away, it is very difficult to fol- low her flight. During March and Apri, 1917, I located a few nests; the cells which compose them are cylindrical and rather thick-walled, ranged one alongside another. ._. There may be as many as four cells to a Fig. 44. Ventral view i ae ; : ae ooo Aree re pee, nest — (Pig, at) and in each is stored a dagenia blanda.. Re- delegged spider on whose abdomen is dueed. placed the Pseudagenia egg; the latter is curved and measures 2.60x0.65 mm. I found one nest in a fissure in a stout vine and several at the base of large trees, and as they were roughly plastered over with mud, they blended well with their surroundings. I once saw this or an allied species hunting her prey. She ran swiftly, and with the abdomen curved forward beneath her, along tree trunks, inspecting her path with her antennae. This peculiar hunting posture might lead one to suppose that the par- ticular spider which forms her prey is likely to pounce upon her and so Pseudagenia has need of her sting well to the front. Pseudagenia makilingi is likewise a forest insect; she lacks the purple effulgence of blanda and has no red on the legs. She fashions a two-celled mud nest, which she hides in a curled-up leaf, etc. Pseudagenia nyemitawa Rohwer. Length 12 mm.; metallic blue. This very agile and handsome wasp, though closely related to the forest-dwelling P. blanda, builds a very different sort of 98 nest. She is occasionally seen within the forest, but I have found her more at home along the wooded banks of a stream at the College of Agriculture. The two or three-celled nest (Figs. 45,47) is placed in rather exposed situations. It is pasted on tree trunks or twigs, but as the structure seems proof against wind and weather, being varnished over with a tree gum, it remains in position long after the brood or parasite has emerged. The energetic architect is rather shy and a swift work- er. Several females were ob- served gathering, at the base of tree trunks,-the main building material of the nest, earth - like substance that forms the coverways for the termites or white ants. The wasps first sip up water from some convenient hollow or edge of stream, and the ball of earth subsequently gath- ered is worked around in her mouth until it assumes the right consistency, when it is plastered on the building site with the dorsal tip of the abdomen, (Fig. 45) The first cel lis secuhed to the bark; ete., along its side, the Fig. 45. Pseudagenia nyemitawa putting OPC end upwards; it is the finishing touches to a two-cell nest stored as soon as possible which is fastened to a tree trunk. The with a spider, which, how- wasp is revolving a semi-hquid ball of gum in her mouth, and is smearing a ©V€T; may not be captured bit from this gum on the top of the soon. I happened to see a nest with the dorsal tip of her abdo- spider stung by the wasp, its masa ans legs snipped off in a minute or less, then it was grabbed by the anterior part of the body, dorsum up, and rapidly dragged beneath the wasp, over boulders and thence to a small tree, the couple disappearing in the crown of the latter. The egg is soon laid on the base of the spider’s abdomen beneath, the cell plugged up and another added along- side. the first. Before the cell group of two or. three cells is completed, a partial coating of varnish may be put on. The cells finished, are more closely united with mud and the ineri- table ‘‘footrest,’ a small ledge or projection to the upper and 99 with the mud she turns her attention to a tree gum, which she works over in her mouthparts and spreads on precisely as she did the mud. Then, without intermission, she brings in another final coating or, rather, patching material—a pale grey lichen— which she works up in her mouth into a sort of viscid paste, also applied like the mud and varnish, but the nest is blotched rather than completely covered: with it. We now have as a completed nest, an object which might well pass for an ex- crescence or other irregularity in the bark to which it is fastened. outer side of the last cell, is added. When Pseudagenia is through Fig.. 46. A delegged spider on whose ab- domen a half-grown larva of P. nyemi Fig. 47. Three-cell nest of tawa is feeding. En- P. nyemitawa, showing larged. doors by which wasp has left cell. Natural size. When on thin twigs, however, it is somewhat disproportionate to be regarded in the light of concealed coloration, particularly when it is plastered on to a thin yellow bamboo twig. The wasp’s egg is over 2 mm. long; the grub (Fig. 46) seems to be of the usual Pseudagenia type and spins a rather thin and pallid cocoon, the head end of which points upwards. In break- ing its way out of the cell the young wasp attacks the upper end, which it bites in circular fashion, so that when the resulting disc is forced out, it usually. stands ajar like a door, the gummy cell covering serving as a hinge. (Fig. 47.) In examining the cells of this species I came across one con- taining a darker brown thicker cocoon, which finally disclosed a reddish Xanthampulex, a parasitic pompilid, which doubtless laid her eggs on the Pseudagenia’s spider in the field. The nest 1s also parasitized by an ichneumonid, several of which I reared. 100 Pseudagenia nye mitawa, unlike the larger Macromeris and Paragenia, lives in solitude and deserts her exposed nests when they are completed; thus parasites have a good chance of carrying on their nefarious trade Pseudagenia caerulescens Williams. Length 9 mm.; metallic greenish blue. This very pretty spider wasp was observed nest-building but once. On August 11, 1917, I saw her erecting her little mud nest within the silken retreat of a jumping spider, the web being in a vertical crack in a bamboo stump. Psewdagenia was in the act of bringing in rather large helpings of mud with which she capped the first cell. This completed, she commenced another cell, building it with the base against and below the first. On the morning of August 12 the cell was finished. and at 9:55 a. m. she had already stored a spider and was ovipositing on it; in doing this she rested quietly for a minute or so, her abdo- men “partly inserted in the cell. Immediately thereafter she took flight, and returning with a supply of mud in her mandibles, commenced sealing up the cell, using the tip of her abdomen as a trowel. But at this point I captured the architect, well knowing that 1f I waited a little longer she would be gone for good. The two cells, which were thickest in the middle, were 10-11 mm. long by 6.5 mm. in diameter; exteriorly they were rough and not artistic. Each contained a species of jumping spider with a long conical abdomen and large grasping legs or chelae. Six legs had been cut off. The egg, which was pearly white, was a little arched in con- forming with the curve of the spider’s abdomen, and a very little less rounded and more pointed at the tail end. It measured about 2 mm. long by 0.6 mm. in diameter, and was transversely fast- ened by the head end, on the side and near the base of the spider’s abdomen. On August 17th, one of the wasp larvae commenced spinning, and on the thirty-first a male emerged from the first cell. In making his exit therefrom he bit a small hole through the clay near the end of the cell. Pseudagenia sp. Length about 8 mm. Unfortunately I secured no adult females of this species and saw but a single one engaged in nest building, and this one I noted was of a steel gray color with clear wings and red hind 101 femora. In habits she resembles P. caerulescens, for she also uti- lizes the more or less cylindrical silk nests of attid spiders as a shelter for her mud cells, (Fig. 48). These webs were quite com- mon on ornamental shrubs, es- pecially Croton (Cordaeium), a couple of the tough leaves or a rolled-up one concealing the nest. The spiders’ nests are from about 11% to 2% inches long, and the wasp places from two to four cells, usually arranged in a string, within this tube. Pseudagenia uses the spider’s doorway in bring- ing in her prey and balls of mud. She does not appear to molest what spider egg-co- coons there may be present. In one case a web sheltered also a one-individual Poly- rhachis ant nest, but this small jug-like affair, the work of the queen, was separated by a fine web from the body of the tube. = 5 Fig. 48. a, Croton leaf with edges The mud cells (Fig. 48, d trimmed to expose silken retreat of and b) are short and blunt; jumping-spider (Kugasmia sp.). the cocoon within is thin and Towards the petiole the retreat is tan color and 8 mm. long, and occupied by the nest formed by a the pupa well armed with ab- queen ant of Polyrhachis sp. Three | ae , RE i ee a Pseudagenia cells are shown within, @OMmal and other processes. The opening to the nest is towards Notwithstanding the fact that the top of the figure. b, Cells the Pseudagenia cells are well showing cocoon of Pseudagenia; hidden in the spider’s web. a e, Cell with cocoon of an ichneu- S28 6 iat monid parasite; d, Cells of Psey- black and white ichneumonid dagenia. b and @ X 4/3; other Wasp gains access to them and figures reduced. parasitizes them heavily, (Fig. AS, c). Hence the reason 71 reared none of the architects. Pseudagenia sp. Lower Makiling Forest, August 17. While observing /Hylo- liris mandibularis nesting in a large decaying tree trunk, I noticed a small spider-wasp enter a hole in the soft wood. Some saw- 102 dust just outside the aperture indicated that Pseudagenia had en- larged the burrow in this old honeycombed log. Upon exca- vating I found that the short tunnel contained flirée cells, each supplied with an immobilized spider. One of these had a wasp egg on it and another a third-grown larva. The spiders appeared to be ground forms; one was accidentally destroyed, while of the other two, the first had only the last pair of legs snipped off, while the second had all four pairs severed. Pseudagenia macromeroides \Villiams. Length 11 mm.; grey-black. This ordinary-looking little wasp was not found to be common. From what few observations I have made on this species, I con- clude that the larger cell groups indicate a semi-social habit ; 25 or more amassed cells must exceed the labors of one w asp. The cells, which are made entirely of mud, are placed in groups in sheltered places, as within a bamboo node which has been partly hacked open. They are much smaller and more crudely made than those of the big Macromeris violacea, but re- semble them in general form, (Fig. 49). In a rather far-gone 16-cell nest | found an old crippled female and a male; other nests (cell groups) were deserted or occupied by ants. One cell contained a typical Pseudagenia pupa. A Pseudagenia, which I never cap- tured for identification, was common . nesting in the bamboo wall-supports of Fig. 49. Cells of Pseuda- nipa ROTOR Several times have I seen ae mac eS About natural size. this wasp, in my room, carrying her de- legged spider beneath her and enter with it through the cut made for. slats, into the node itself. Evidently this insect makes use of the mud from the cells of the house Sceliphron (S. deformis) to fashion her own cells. Water for moistening the clay was handily situated on my washstand, where some had been spilled on the oil-cloth. A good-sized white and black cryptid wasp hung around the architect’s doorway, but I drove her away before any damage was done. The household wasps find but little shelter in houses except from the inclemencies of the weather. Ichneumonid and * Nipa fruticans Wurmb. is a native palm in tidewater lands, and whose leaves furnish ‘‘shingles’’ for bamboo frame houses and huts. 103 cuckoo wasps, bombyliid flies, ants and others follow her within and decimate her brood as they do outdoors. Pseudagenia okawa Rohwer. Length 9 mm.; gray-black. Fig. 50. Bamboo twig split open to show the three cells of Pseudagenia okawa. Natural size. * Nipa fructicans Wurmb. is a native palm common in leaves furnish ‘‘shingles’’ Unless we make use of a good magnifying glass in our field study of wasps, many spe- cies are not to be separated from others. This is particularly applicable to certain wasps that prey on Orthoptera (grasshop- pers, crickets, etc.) and to the large psam- mocharid genus Pseudagenia. Here, how- ever, their nesting habits often help us; the shape, number and arrangement of the Pseu- dagenia cells aid in differentiating the spe- cies. So, too, with the twig-nesting wasp P. okawa. She looks much like some others, but nests differently. I found her partition- ing off the hollows of slender twigs into from two to four cells and storing them with de- legged spiders, (Fig. 50). I found but three nests—all A twigs on the ground in the shade of a large mango tree. The inner cell plugs are of mud, and the outer one, which shuts off her nest from the outside world—moisture and ants—is in addition, smeared with a gummy substance. I reared four Pseudagenia okawa in June, 1917. Ageniella unifasciata (Ashm.) and Ageniella williamsi Rohwer Length about 7 mm.; fuscous spot on fore wing. These two little species, save for the pres- ence in A. unifasciata of a distinct median tooth on the margin of the clypeus, are very much alike. Both are abundant lowland in- sects. A. unifasciata may have as many as nine cells to a nest. The cells are com- monly arranged in rows on walls, tree trunks, etc., and provisioned with a delegged spider. The cells may be also built on twigs. tidewater lands, and for bamboo-frame houses and huts. 104 A. williamsi builds a several-cell nest on small plant stems (Fig. 50a). under a clod of earth,-ete) " As im the) first species mtmese are about 11-13 mm. long and more or less cylindrical. Both species pursue their spid- ers quite fearlessly and bite some or all the legs off before dragging them, beneath their body, to the nest. I once came across a small Ageniella | excitedly chasing a ground spider; she finally stung it, bit off some of the legs and proceeded to carry it away, wher I caught a glimpse of a larger spider dashing toward it, and then our wasp appeared burdenless a ; : foot or so away. Thus the larger Hig. 0a. Wour-cell nest of Age. spiders do mot fear he «smaller niella williamsi. About natural - : siie. wasps, and rob them of their prey. So also will a larger Ageniella of the same species take over the pursuit and possible capture of a flushed spider, driving the rightful huntress away, who looks on while the chase is resumed. The cells of these wasps are much infested by an ichneumonid. Ageniella banoensis Rohwer. Length 13 mm.; grey-black, a fuscous spot on fore wing. Except for her double size, A. banoensis is much the counter- part of the two preceding species. Not so common, she has rather different nesting habits than either, for although a true architect, she builds her mud cells in a hole in the ground, thus perhaps doubly insuring her brood. On capturing a spider: she has the most exasperating way of lugging it about and of making lengthy stops, thus taxing one’s time and patience. Once, when under the shelter of a_ big mango tree, I noticed one of these wasps searching about a lim- ited space of ground; it was evident that she had already flushed a spider and was now trying to find the fugitive. After circling over the ground for some time she aroused a rather long-legged but small-bodied spider, which she overhauled in an open place. The spider was courageous, however, and showing fight, tried to embrace her more agile foe and to leap upon her back. This ageniella frustrated, and in a skirmish crippled the spider with a sting; then she pounced upon its back and completely quieted it 105 by a sting in the neighborhood of the mouth. Then, following a brief delay, the wasp bit off the four pairs of legs at apparently one strong snip for each leg, the member being cut off at the next to last joint, leaving the coxal stub only. This operation. notwithstanding the sev eel interruptions by wandering ants, occupied but a minute and a half. Thereafter she grabbed her prey by the anterior and dorsal portion of the body; “she carried it beneath her, halting every now and then to fuss with her burden. This procedure was kept up for half an hour, when I captured the pair. Another of these wasps was carrying a spider with three uncut legs. This burden was borne for about 100 feet before both disappeared in a shallow hole in a bank. This burrow, later on blocked up, contained three rather short but ample cells, in enc of which was a more or less delegged spider with a wasp eg secured on the underside of its abdomen, The eggs were ease white, somewhat curved, and 2.35 x 0.58 mm. in dimensions. The second nest I located was a little way inside a rat hole in a bank. A short tunnel leading to a single cell contained a de- legged spider. I interrupted the proprietress at work, and I here noted that she operates like the rest of her cell-making tribe. On this occasion she was gathering mouthfuls of moist soil about a foot away from her nest to seal up the entrance to her burrow, using the end of the body as a plasterer. My observations on this insect were from March to August, 1917. Pseudagenia aegina Smith. Length 9 mm.; thorax red, two bands on forewing. The nest of this pretty species is shown in Fig. 51. It was found well in the forest and was suspended from a bank by a small rootlet. It is made of rather dark soil or mud arranged so as to give the nest a granulated texture. It is one-celled and yielded a male wasp. The insect is not uncommon about the edge of the forest. Batozonus bioculatus (Bingham) Length 15 mm.; black; antennae, legs and wings largely orange. During the rainy season this conspicuous orange-winged insect might often be seen searching for the corpulent web- spinning spider Poltys sp., V with which she provisions her nest-hole. She is a handsome pompilid, and, apart from her gaily-colored w ings, 106 has the antennae and one or two basal abdominal spots of the same orange color. On two occasions I saw this wasp capture her prey, and on the first of these the performance on the part of the huntress was somewhat remarkable. During the latter part of August, 1916, I noticed one of these wasps hunting excitedly in a small garden plot at the College of Agriculture. The area had a few small bushes about its center and was bordered by a small ornamental plant, “Cucharitas” (Alternanthera versicolor Regel). To this Cuchar- itas border the wasp presently flew and, searching about, soon located a double spider thread, the two strands being quite close together and _per- haps uniting further on, leading from this low border to a_ bush about four feet away and secured thereto, at a height of a tittle less than three feet from the ground, by several guy-threads. Batozonus was | at first able to walk along this double Fig. 51. Cell of Pseudagenia aegina suspended from a rope, but farther along ran and flew jy ootlet. Natural size. rapidly over the single part, selected the correct guy-rope and thus located her prey in the bush. The spider did not question the wasp’s motives, but dropped to the ground, crawled up the bush, and repeated this performance in an effort to escape. The poor arachnid, however, fat and stiff- bodied, was not of those swift species which frequently evade their would-be wasp captors, and so in a short time she was over- hauled and effectively stung. Then the victorious wasp appar- ently fed for a while on the mouth-juices of her prey, but sud- denly noticing a stream of small dark ants passing nearby, low- ered her antennae in consternation and seemingly upon the nearest ants, and then grabbing her very large prize, she dragged it away to place it three-quarters of an inch or so upon a tuft of grass. The spider was eventually borne away and interred. On August 9, 1917, while on a hillside, I brushed a large- bodied spider off my person and then noticed that one of these yellow-winged wasps was searching the ground about my feet. She soon located this spider, which being rather clumsy hardly served to intimidate her, and so after a struggle she grasped and climbed upon the dorsum, and with her head towards the caudal end of the spider, stung it in the region of the chelae or jaws and immediately stopped its activity. "Then seizing the spider with her mandibles by the base of one of the forelegs, she backed 107 away with it and placed it upon a little seedling plant two and a half inches in height. She circled about it so as to impress the locality in her memory and flew away in search of a suitable place to make a burrow. Soon coming back, she perched the spider upon another weed, commenced one hole, but abandoning this work, examined the spider once more, and going to another spot twenty to thirty feet away from her prey, she began dig- ging for good, at9 a.m. She bit out much of the soil and used her forefeet also in excavating. At 9:22 a. m. the burrow was about complete. It was nearly vertical and only about 1% inches deep. Batozonus now returned to the spider, grasped it by one of the hind legs and swiftly backed away with it towards the burrow. On her way thereto she encountered a cast-off skin of a spider; this she charged furiously for some seconds, and after this ridiculous performance resumed her journey. Halting about eleven feet from her burrow, she pulled her prey six inches up on a fallen twig, placing it in a fork; this act was care- fully done, for the wasp saw to it, by quick pulls at the spider, that the latter was in a secure position. She resumed her digging, but visited the crotch several times to see that the spider was unmolested, and at 9:35 a. m. dragged the spider down and to within two inches of the burrow. Soon she seized it by the base of one of its legs and backed down into the hole and out of sight with it, the passage being a tight fit. Batozonus remained below for a few seconds only for the act of laying her egg, and could presently be seen, her head almost on a level with the sur- face of the ground; she was engaged in filling up the burrow, tamping down the soil vigorously with the end of her abdomen. This tamping was done with such a rapid motion that the wasp was seen to fairly vibrate in a longitudinal plane. “She bit off the top edges of the burrow and sometimes brushed in a little soil with her forefeet, then tamped again. Tamping was long- sustained and constituted by far the principal filling operation, being very effective in making a firm blending core of soil. When the burrow was but little filled, the tamping insect was in an approximately vertical position. The tamping process was fin- ished at 10:09, when Batozonus commenced backing over to the nest-site, carrying in her jaws lumps of soil and other material wherewith to conceal the position of the burrow. I caught this wasp and dug up the spider, which lay on its side, practically immovable at the bottom of the burrow. The wasp’s egg was placed longitudinally on the abdomen, tar to one side of the mid-ventral line, near the base. It was pearly white and polished, a very little thickest below its middle length and more rounded posteriorly. Moderately stout, it was gently curved 108 to conform with the profile of the spider’s abdomen. Length about 3 mm.; thickness .8 mm. UT mown ti a a as py Fig. 76. Cerceris angularis, 9, X 3. My attention was first drawn to this nesting site on seeing a large wasp flying heavily laden with a brilliant blue- black and red long- horn beetle which she clasped beneath her. Subsequently I saw other beetles, different species of Lamiinae, in transport to this bank. Setting to work with the trowel revealed at a horizontal depth of 6-12 inches several fresh Cerceris burrows and a quantity of old and deserted cells, containing cocoon and beetle debris. The galleries appeared to be largely stuffed with soil, but I located numerous cells quite close together; this would indicate that each female had a many-celled tunnel. These cells were of 147 ample proportions, a good inch long, oval and horizontally formed and with smooth solid walls. Many held beetles and wasp larvae or eggs, while others contained cocoons. The provender was more or less flexible and well preserved, but in all cases stung to complete immobility. All told, I dug out 14 species of beetles from these cells, two large species of Anthribidae and the re- mainder long-horn beetles of the subfamily Lamuinae; the largest of these measured 21 mm. long or about ‘the length of the wasp herself. Another beetle measured 19 mm. The two Anthri- bidae, with their long antennae, much recall a long-horn beetle and one might suggest that the similarity deceived Cerceris into catching beetles outside her accustomed group. Perhaps my ob- servations on the next species will dispel this idea. ; The egg of Cerceris angularis is proportionately large, for it measures 6 x 1.08 mm. and is thus the equal of those of some of Io Te Fig. 78. mien 79: Beetle showing egg Full- grown larva Cocoon of C. angu- of C. angularis, X of C. angularis. laris. Natural 4/3. Natural size. size. the large Chlorion. It is shining white, slightly curved and a little broadest at the head end. It is fixed along the underside of the beetle (Fig. 77) in such wise that the head end rests upon one of the fore coxae. Cerceris grubs seem difficult to rear ; their natural life in the damp cells is a condition hard to imitate. Thus I did not get much data on their growth. Younger individuals are whitish, and those about to spin cocoons lemon yellow. A full-fed indi- vidual is 28 mm. or over an inch long, well segmented and pro- portionately slender. There is no need here even of a moderate- sized head : the small and comparatively long head with sharp and slender mandibles and elongate underlip fit it well for forcing apart the segments of the corselet, etc., of its heavily-armored food supply, and the grub’s slender thorax, capable of much extension, follows the head into the beetle’s body to get at the meat. The abdomen terminates in an unevenly-lobed and narrow cylinder (see Fig. 78), apparently a character of Cerceris larvae 148 and perhaps useful in forming a sort of lever or brace when the larva needs the force to push and bite into the joints of its prey. The mother wasp does not feed her young from day to day, but stuffs a cell with beetles, lays her egg on one of the latter and closes the chamber. Consequently, when spinning time arrives some days later, the big grub finds itself surrounded by a debris of beetle fragments (for it devours only the tender interior), wing cases, antennae, legs, etc.; these, with an addition of grains of earth, compose a loose outer covering for her cocoon. It gathers up the fragments one by one and loosel y fastens them together with silk. In this framework the real, thin but opaque rich brown cocoon is spun. The head end of the cocoon (Fig. 79) is quite the widest and the posterior extremity somewhat knobby. The smaller cocoons measure 24 x 6 mm. and the larger 28x9 mm. In emerging, the wasp makes a couple of clean scissor-like cuts which free it at the broader end of its cocoon. My observations are from July to September, 1917. Sse, Ss > Fig. 80. Cerceris spiniger, 9, X 3.4. Cerceris spiniger Rohwer. Length 12 mm.; black, marked with yellow; wings fuscous on costa near apex; head remarkably broad. This species (Fig. 80), which I observed from June to August, 1917, lives in the woods in small colonies, digging neat and steeply-inclined tunnels to a depth of about 16 inches in the rich soil. The soil which is brought up from below by the burrowing insect contrasts in color with that of the surface and is in part formed into a sort of short horizontal vestibule or half-tube lead- ing along the surface to the tunnel. One nest had two cells with long passageways, one cell being provisioned with nine small 149 beetles and contained also a quarter-grown wasp larva, while the other held the mother wasp and 17 beetles. The beetles were of eight species—Anthribidae, 1 species; Curculionidae, 1, and Chrysomelidae, 6, the last comprising the bulk of the store. The larva is of the same general type as that of the preceding species. Later on I dug up another nest and secured five dull brown baggy cocoons of weak texture. A pupa secured from one of these measured 11.5 mm.; it had a broad head bearing in the middle of the vertex a porrect, forked finger-like process; the thorax and appendages were stout, the legs being rather spinous, while the comparatively slender abdomen was provided with sub- dorsal and lateral processes. VESPIDAE. Eumeninae. Here belongs a very large assemblage of wasps related to the social species, and while the Eumeninae have but two phases (male and female), it is also true that sometimes, as in the lower Vespinae, the third caste or differentiated egg-laying female or queen cannot be distinguished. The Eumeninae vary considerably in form—there are the thick- set Odynerus and the slender species of Ewmenes. Some are over an inch in length. Most of the species whose nesting habits are known use clay for making cells, and while a large number employ this material merely to partition off cells in some suitable cavity, as the hollow in a twig, a deserted mud-dauber nest, etc., many build their own nests of mud. Among the latter are the jug-making Ewmenes or “Potter Wasps.” Fabre, Dutt, Hart- man, Cretin and Howes give good accounts of the nesting habits of wasps of this genus. Some of the Odynerus build mud cells, others use pre-existing hollows, while many dig their own neat burrows, over the entrance to which they may or may not con- struct a clay tube. Isely (1913) gives excellent accounts of the life-history of some of the Kansan Odyneri. A few tropical wasps of the Zethus group make nests of pieces of leaves. Caterpillars practically always are stored in the cells. In a few cases sawfly and chrysomelid beetle larvae are used. The egg is usually suspended from the ceilmio sof the) cell bynva filament. On one occasion I saw a very small species of Philip- pine Eumeninae hunting caterpillars on the leaves of a shrub. A wily caterpillar hanging from her thread dropped part-way to the ground. The wasp appeared conscious of this trick and began an extended search for the caterpillar, circling again and again around the latter’s jumping-off place, but she could not locate her prey. Some of the more highly specialized Eumeninae have habits in common with their social brethren, i. e., feeding their young from day to day and abiding by their nests during the night. This applies to Zethus (Calligaster ), some Zethusculus and. Synagris. Odynerus tropicalis H. de Saussure of Africa feeds her young from day to day (Bequaert, 1918). This latter habit occurs also in other solitary wasps of entirely different groups ( Bembecidae and Sphecidae), but I do not regard it here as a development towards social life. Rhygchium atrum Saussure. Length 20 mm.; thick-set; black; antennae and most of wings, orange. This is a tame household insect and in a sense therefore rather a nuisance. But inasmuch as it seems to prey wholly on the larvae of certain moths of the leaf-roller type (Pyr alidina), this fact should counterbalance any annoyance it may occasion. While some species of this genus fashion neat cells of clay, the handiwork of atrum is rather crude and consists solely 1 partitioning off a pre-existing cavity with plugs of mud. The mother Rhygchium was a frequent visitor to the chemistry lab- oratory and insectary at the College of Agriculture, where she sometimes attracted unfriendly notice by plugging up Bunsen burners and other apparatus with mud. In dwellings she makes a specialty of nesting in the bamboo furniture, and as this is often shifted about, the poor insect is then at a loss where to stuff the heavy caterpillar she is carrying and which she may perforce exhibit to the unappreciative household while assembled at the noon meal. One of these insects made her nest in a reclining bamboo chair; she scarcely objected to my presence in this chair, but that afternoon brought in seven caterpillars. Other nests were found in shallow rung holes, the wasp coiling up her slender prey within them. The empty cells of the leaf-bit nests of Zethus (Calligaster) are not infrequently appropriated by Rhygchium, w stores them and seals them up with mud. In another case, a large empty mud nest of Eumenes fulvipennis, lying on a shelf in the insec- tary, was utilized by a couple of Rhygchium; the empty cells were close together, and so there was quite a skirmish whenever the two mothers met. Then they faced one another in hostile attitude and made snaps at their mandibles. 151 The wasp may often be seen hunting for caterpillars, and her actions are then much like Odynerus, to which genus she is closely allied and sometimes even included therein. The provender, an active green larva, nests in seeming security in a leaf, a portion of which it rolls up and fastens THES with silk. I observed the wasp examining the leaves of a weed (Amarantus sp.). With alert antennae she would alight on a curled leaf, bite first into one end and then the other, and wheel about swiftly in fear that the uneasy occupant would escape. She did not sever the silken guys, but bit holes through the leaf and at an opportune moment hauled out, stung and carried away the unfortunate larva. Several caterpillars are placed in one cell and a plug of mud plastered over the aperture. The egg is suspended by a filament. The wasp is an indefatigable worker, starting the day early and sometimes laboring until so late an hour that in the approaching dusk she has difficulty in locating her cell. Stilbum, no doubt, and other Chrysididae are often household insects and probably one of the worst parasitic enemies of Rhygchium. Odynerus luzonensis Rohwer. Length 8 mm; black, with yellow marks I found a single burrow of this insect; 1t extended obliquely into the side of a small termite mound. * put on, and smaller portions of leaf are applied as an interior finish, and these being vigorously chewed up form a compara- tively smooth surface. The cell (Fig. 93) has now an interior depth of about 30-32 millimeters, with a mouth of 8 mm. in width and 12.5 mm. below the middle. This, the first cell, probably requires the longest time of all to construct. About five days of variable and rather cool January weather were required by a Zethus in making the fourth cell of her nest. | Hardly is the cell completed when an egg is deposited therein. Unlike Eumenes, Odynerus, Rhyg- chium and most other Eumenidae whose eggs are suspended from the ceiling of their cells by a filament, but as in Synagris, that of Zethus is laid unattached to the bottom of the cell, so that if the latter be un- Fig. 93. The first cell of a Z. duly inclined the egg will tumble cyanopterus nest, X 3/4. out, The ego (Fig, 91) as creamy 161 white and fairly polished. It is very little curved, the rounded ends are practically similar, and it measures about 4.70 mm. in length by 1.20 mm. in thickness. The incubation period was not determined beyond the fact that, for a wasp, it was not rapid, From oviposition up to the time when the larva is full-grown, the mother exercises great care in protecting her young. During the day, if the weather be promising, she occupies herself in part with improving the nest, strengthening the cell, extending its mouth, forming the protective roof, or putting on an extra layer or two of leaf-bits, and in part with keeping guard within the ample cell, her stout head blocking the entrance, her antennae coming to attention at your movement, and, if hard pressed, retreating well within the cell and biting any offending object. She is, however, a really timid insect and when frightened away keeps at a distance until the danger seems past. She always stays in the cell at night and never appears to be asleep, as I have found her ever awake and watchful far into the night. The egg-shell is quite tough, as is usual with the Eumeninae, so that it retains much the same shape as the egg. The young larva is more or less curved into an arc, whitish, cylindrical, and has a large head. It sheds its skin very shortly after hatching, and is active and tough from the first, wriggling its body and working its needle-like mandibles hungrily. The mother feeds it from time to time with freshly-killed moth caterpillars, appar- ently Pyralidina. One such caterpillar I saw captured. The wasp flew about the crown of a small tree ( Litsea sp.) and exam- ined the leaves for her prey. Finally she found a leaf rolled up lengthwise, perched herself upon its distal end and, cutting through it with her stout jaws, inserted her head into the open- ing thus enlarged and hauled out a small caterpillar. This she did not appear to sting, but chewed off the head and most if not all of the thorax of the unfortunate captive. It is this remnant of caterpillar—all good and tender meat—that Zethus carries away beneath her and delivers to her offspring. At the close of the day the wasp enters the cell occupied by her young and stands watch. This she does also quite frequently during the day, especially in dull weather. She goes into the cell headfirst and, turning about in the enlarged bottom, advances to guard the entrance. But soon the growing larva attains such proportions that it occupies most of the enlarged portion of the cell, and assuming a curled position, refuses to be dumped out 162 through human agency, and while there is still ample room in the cell for both mother and young, the former can no longer turn about therein, though she may attempt it, and so holding her wings close to the body, she backs in. The mature larva is quite stout, with the head rather small. When it is full-fed the mother closes up the mouth of the cell with leaf-bits, and while the inner side of this plug is rather crude, the outer face, which is slightly concave, is quite neat. The occupant can hardly be said to spin silk, though it elaborates a thin leath- ery partition (Fig. 94) some 6 mm., Fig. 94. Vertical section more or less, below the leaf partition through 4. cyanopterus and extends it slightly «down the cell cell containing a pupa and Ba Ope : showing the two dises Walls. The remainder of the walls is closing the c2ll, X 3/4. thinly coated with a larval juice of some sort. The pupa (Fig. 94) as it lies in its bent attitude is about 22 mim. long. Ina rather advanced stage it is creamy yellow. When first freed from the pupal envelope, the adult is quite feeble in its humped attitude, and the wings not yet longitudinally folded, are bent apically. Some time must elapse before it ac- quires sufficient strength to bite its way to freedom through the two doors of the cell. Unless there is an emergence from another cell before the last is sealed up, the wasp may set about making another cell. She is therefore at first compelled to forego the comforts of a bed- chamber and spends the night out of doors, or partly so, using the commenced cell as a shelter. Otherwise a cell from which an adult has issued is cleared out for another brood. But prob- ably this process is not often repeated. The Zethus nest in the hibiscus bush I observed almost daily for a period of 127 days (December 24, 1916-April 22, 1917), and when I first found this nest, its proprietress had commenced to build the third of what proved to be a 4-celled nest, and so on December 24 it must have been at least two weeks old. The male wasp (Fig. 88) may linger about the nest for sev- eral Gays, and he was noted resting on or near it at night. But the progeny does not appear to stay around the old homestead. What enemies does the mother Zethus strive to guard 163 against? No doubt the large solitary hornet (Vespa deusta), a very destructive creature, may sometimes try to gain entrance into the open cell. Ants frequently swarm on bushes and must be considered a menace. The ubiquitous eumenid wasp Ihyg- chium appropriates the cells of old nests and would not be so harmless if given free access to an occupied nest. But per- haps the most successful foe of our wasp is a large chalcid wasp, Leucospis, a cosmopolitan genus of mostly black and yellow species which have the peculiarity of longitudinally folded wings. This insidious creature has been found by Fabre to parasitize the cells of Chalicodoma and Anthidium, both solitary bees. No doubt it preys upon many other species of solitary bees and wasps. Apparently Leucospis does not enter the cells of Zethus, but hangs around for days and at an opportune time thrusts her long ovipositor to the base, through the leafy layers, and may thus reach the young wasp within. Sometimes the watchful proprietress perceives the pest and routs it by a short dash; on other occasions, however, the parasite works in safety, hidden behind the bulk of the nest, the poor mother wasp stand- ing guard at the entrance to the cell. One finds many incomplete nests, usually full of ants. The rutelid beetle Adoretus sometimes uses deserted nests during the day as a retreat, crawling in among the leaf layers, and a small moth caterpillar spins up and feeds on the leaf-bits. This gifted wasp is to be found throughout the year, and while not abundant was more frequently taken during the rainy sea- son. It seems to have a special fondness for the umbellate flower heads of Premna odorata, a small wide-spreading tree of the Verbena family. The wasps of ne genus Synagris, though not very closely re- lated to Zethus, Hee some habits in common with tt. According to Roubaud (1910), who studied this genus in Africa, the ee specialized kinds of Synagris carry on their nesting activities much as in Odynerus and Eumenes, storing the cells with cater- pillars and then closing them. But more highly-developed forms, although they also make mud cells, feed hei young from time to time and guard the nest just as in Zethus. S. cornuta goes further still, in that she feeds her young, not with a somewhat chewed-up caterpillar, but with one worked into a mass. The cells, although of mud, have the narrowed and bent neck and enlarged bottom of the leaf-bit cells of Zethus. Zethus seems to differ in structure from some of the Amer- ican forms and appears restricted to the Old World tropics. Some of the American species of Zethus build mud nests. 164 Zethusculus hamatus Zay. has been studied by Howes (1917) in British Guiana; it makes delicate cells of ribbon moss peeled into little hoops. These cells resemble the shell of certain snails. ihe egg of this Zethusculus, though attached to the roof of the cell, is apparently sessile. The food of the larva are caterpillars. Zethusculus lobulatus De Saussure has been studied by Ducke (1914) in Para, who illustrates a compound nest of chewed-up pieces of leaves and built by several females. Each female here attends to her own cells. Other species nest in holes in wood. Vespinae, The Vespinae or social wasps, known in temperate regions by only a few genera, are represented in the tropics by many. W hile some species can scarcely be induced to sting, there are es merous others which defend their nests very promptly, and i is a divided question whether all the species are docile ee unmolested. Perhaps the best represented Philippine genus and the one commonest in species, is /caria. These are usually rather small wasps, resembling somewhat an undersized Polistes. They build in many situations ; some make an uncovered pedicelled nest as in Polistes and Polybia, while others make a paper cover for their cells; such nests, often flat or oval in shape, are placed between palm leaflets, or other leaves, against boulders, tree trunks, etc. The wasps are stingers of good or moderate ability, and at your first venture usually stand at attention with alert wings and visage, awaiting definite hostilities. Though there is scant temperate weather in this part of the tropics, the /caria communities do not appear to be perennial; numerous deserted nests are to be found, though many of these are doubtless ruined by the big Vespa deusta. It seems, however, that at least some species of /caria have a sort of swarming habit, as one may meet with a good wasp colony engaged in building a nest, where a day or two previous none was to be seen. Icaria is a very interesting genus which, owing to the abun- dance of its species and their often comparative mildness or dis- position, should be thoroughly studied. Of the Philippine Vespa I met with two species, both very large wasps. lespa luctuwosa Saussure, the smaller of the two, is a grey- black insect with nearly transparent wings and measures in the worker about 20 mm. long. It builds large paper nests on trees or bushes. | found the insect common in a banana grove where it fed at the elongate flowers of this plant. A couple of these w asps made repeated clumsy and unsuccessful attempts to pounce on 165 blue-bottle flies which were feeding at a bunch of the decayed fruit. A very familiar sight to the casual observer is the big yellow- winged Vespa deusta, Lepeletier (Fig. 95) as it flies about the more rural dwellings, trees, banks, etc., searching for something it may destroy. The queen is nearly 32 mm. long, black, with orange wings; the worker is about 25 mm. long. This wasp is a slow flyer, and though com- mon, only one or two at a time are to be seen. Both castes are often met with. She is a most destructive creature and a per- fect bane to some of the Steno- gaster and Icaria, whose nests she destroys with impunity. Being of great size and power- ful build, she probably meets with little opposition among the majority of other wasps. A note of January 14, 1917, reads: ‘A.M. “I “saw. ene. of these wasps alight on a leaf of a Buri palm (Corypha elata Roxb.) said leaf being curled up into a cone within which a good-sized paper nest of a small pale yel- lowish Jcaria was secured for its ventral length along and to eachside’ of a rib: of the leat. The nest was only a few inches long, and the more or less hort- zontal comb was covered witha low paper roof. The entrance Fig. 95. Queen of Vespa deusta, X 2. to the interior of the nest faced - outwardly. Without hesitation the burly wasp squeezed through the door, the owners standing aside and looking on. Vespa stayed inside for some minutes; in the meantime ten or fifteen /caria arrived from the field, but these as well patiently awaited at the door for the big predator to finish her pillage. I became impatient, and when | jarred the leaf the wasp finally backed out, carrying in her mandibles a white object, probably one or more /caria larvae, and flew away in a good straight line. Further shaking showed that the nest contained a number of /caria. The next day, as I passed by the nest, I noticed that there was a hole in the roof, probably the ") 166 work of the robber, made in entering or leaving the frail struc- ture. Some time later the break was repaired. I soon discov- ered that if an /caria or a Stenogaster nest were desired it was better to collect it immediately rather than to wait a few days at the risk of having it mutilated by espa. Another note of July 7 reads: “3 p.m. A Vespa hunting among some Buri palms chanced upon a nearly completed nest of the big mud-dauber (Sceliphron intrudens). The nest was al- ready “thick- walled with mud and needed only the more or less artistic ribbing for its completion. Our wasp alighted upon it, but finding no open cells, set to work biting at the softer recent daubings of the proprietress. But the mud beneath was stone- like and her efforts were futile. In the meantime Sceliphron returned with a ball of mud, but Vespa, with a dash from the nest, promptly drove her away. The safe-breaker, stout- jawed as she was, chewed vainly for five minutes at the cement- like material before she flew off in disgust.” I soon came to regard these wasps as a nuisance and destroyed them when possible. Stenogaster (Ischnogaster). This is a genus of exceedingly slender forest wasps typ- ical. or Oriental and Australian regions. Although usually placed with the Vespinae or social wasps, they have some char- acters of the Eumeninae and some of neither subfamily. Those that I have observed do not fold their wings longitudinally in repose. Some are solitary wasps and others are social, and al- though the latter condition may here be somewhat rudimentary, the wasps themselves are highly specialized and architects of surpassing ability. I have never found the social Stenogaster in any but quite small and harmless communities, and have not studied these sufficiently to determine if they live as a true society. Among the solitary species are S. eximius (Bingham) of Ceylon, S. micans (Saussure) var. luzonensis Rohwer of the Philippines, and a small unidentified one from the same Islands ; of the social ones are S. nigrifrons (Smith) of Burma, S. melleyi (Saussure) of Java, and S. depressigaster Rohwer and S. vari- pictus Rohwer of the Philippines. It seems doubtful if they have three phases The nests are very diverse and delicate and fashioned from decaying wood, grains of earth, etc. In some cases the cells are plastered along one of their sides to a stone or tree trunk; more commonly, however, they are suspended as a nest, from a fine 167 rootlet, fern, or from a species of fine hair-like fungus which grows on dead leaves, stems, etc. Some of the nests are ex- tremely slender, with more or less vermiform passages and cylin- drical cells; others, as in S. melleyi and S. micans var. luzonensis, have six-sided cells. The egg is glued to the bottom of the cell and the larva fed from time to time with a soft paste whose com- position I did not ascertain, but suspect it to be a plant product. At least some of the larvae do not form cocoons, and in one case the cells are not closed, though far within a tube; the pupa is so fashioned here as to exceed the cell opening in dimensions. The wasps are eminently lovers of the shady forest, in whose uncertain light they frequently escape notice; some have the habit of patronizing certain leaves or fine hanging rootlets, upon which they alight again and again. The flight is easy, in grace- ful jerks, and quite noiseless. The Stenogaster are among the most remarkable and _ least known of wasps, and their peculiar form, more or less inter- mediate position, their love for the forest’s shade and, lastly, their beautiful and varied nests, will always make me regret that I did not become acquainted with them sooner. Stenogaster depressigaster Rohwer. Length 15 mm.; very dark brown, with paler markings; pedicel very slender. Although the shady swimming pool, deeply ensconced between high forested banks, was an object of my frequent patronage, yet a full year passed before I became aware that its immediate region abounded in nests of solitary wasps, several species of which did not appear to venture outside these dense woods. A radius of 100 feet would include the four species of Stenogaster I took in the Philippines. The fine Cerceris angularis here mined the vertical bank in an old established colony; the red and black Pseudagenia aegina hung her granular cylinder from some root, and Eumenes, Sceliphron, Trypoxylon, etc., favored the spot. A very peculiar, fragile and exceedingly slender nest suspended cmder a bank by one or two aut black horsehair- like threads— ay of a roar- ing oe cate came to my notice. cies to this frail piece of workmanship were two or three very elongate and grace- ful little wasps, which at my too near approach took wing and, poising in air a few feet away, eyed me, no doubt with nervous apprehension. When _ the danger seemed past, they cautiously and in characteristic jerky flight approached their domicile and 168 once again settled on it. A brief examination of the external features of a nest (Fig. 96) convinced me that it was made up of cells placed in irregular tiers, each consisting of four or five cells arranged in more or less spiral fashion, the lowest and terminal open passageways or cells now being the only objects of the wasp’s solicitude. But when I saw two of these slender insects each enter a terminal narrow passage and disappear far within, I wondered at the nest’s structure. Dissection revealed an entirely different household arrangement. The serpentine passage* ex- tended the length of the nest and gave forth vertical pockets or cells used as brood chambers by the wasps. Fig. 97 shows the arrangement of a nest of medium size, and although there is more or less irregularity in cell disposition, the vertical entrances to the cells are almost invariably, if not always, at Fig. 96. approximately right angles to the one Nest of above and below, and thus alternate ee saatids openings (passages at the terminal por- ee tion of the inner cell wall) are in the About x same vertical plane. This probably comes 3/4 about from the spiral arrangement of the passageway. In commencing a nest, one or several wasps may do the work; naturally, cells are first made, and these harbor an egg before their com- pletion, and when sufficiently lengthened turn into a passageway. A brief search revealed several nests; one was sus- pended from a slender fern, while another proved the largest of the lot, an affair of over 20 cells and har- boring at least six wasps, several of which were fe- males. This is the nest illustrated by Fig. 96, and it measures 170 mm. in length by 7.5 mm. in diameter, or about the dimensions of an ordinary lead pencil. The nest suspended from the fern was a beautiful structure, showing in bands, the differently-shaded nest material, probably earth and decayed wood. I thought to secure this also, but desirous of having the unsuspecting inhabitants add to its length, delayed the robbery. I was anticipated, however, by another Stenogaster enemy, no doubt the large pillaging Vespa deusta, who wrecked this home four times in eight “TI suspect that there may be sometimes two passages in the stouter nests. 169 days, tearing apart the cells, punching holes into others, and mak- ing great hav oc with the contents. Nevertheless, the little build- ers were very persevering and at least three times repaired their ruined domicile. Some “of my other study similar fate. suffered a As in other social wasps, the egg (Fig. 98) is glued to the bot- tom of the cell. It is about 1.5 mm. long and very firmly secured along its curved middle to the cell, and upon its incurved outer surface the wasp deposits a sticky, rather transparent ball of jelly-like food. The young larva, hatching from the stout egg, the shell of which retains much of its shape, curls and partly imbeds itself in thts food, which it proceeds to devour, (Fig. 98, lower). Gaining in size, it curls itself at the bottom of the cell; its body contour, together with a few of its fine hairs, now keep 1 from tumbling out (Fig. 99), while the underside of ae body (the side towards the cell base), more tightly curled than the outer, forms a sort of funnel into which the wasp now places }the viscid. lump of jelly. When pupation time arrives, the grub merely smears a thin film on the walls of its abode and pupates head downwards in this inverted cup, whose open end, being a little eccen- tric and slightly narrower than the main bore of the cell, prevents the curious, humped, triangularly - bent pupa from falling out. The pupa (Fig. 100) is so bent upon itself that the mandibles touch the end of the body ; and it is the head and upper Fig. 97. Nest of S. depres- sigaster show- ing arrange- ment of cells and passage- ways. The numbers indi- eate the eells, and the tops of the pass- age-ways are shown in two planes by se- ries of paral- lel lines. The passage - way between cells Ti andealtoears in a_ plane parallel to the line of vi- sion; the wall above, as well as the round- ed top —por- tion of the passage,which is extended below, is shown to the left in broad- side view ar thorax side of the triangle which forms a base a little wider than the cell’s mouth. Thus it lies in an upside-down position. The exuvium, in some cases at least, remains in the cell. Measured from its longest side, i. e., from the surface of the mesonotum to the angle formed by the pedicel and the rest of the abdomen, the pupa is 9 mm. long. There is a pair of prong-like affairs on the mesonotum and a ‘low median spine on the metanotum; the head 170 bears a median, rounded tubercle just before the ocelli, and the abdomen and legs are not armed. When the adult hatches it remains head down in the pupal position for some hours. [Evidently it fears no effusion of blood to the head and eventually crawls down the tube to freedom. I found more of these nests higher up in the forest, hanging under the shelter of palms, leaves, overhanging boulders, under partly-fallen trees, and in the archways formed by the prop-lke roots of huge forest trees; they were practically always in posi- tions sheltered from rains. But in one or two instances, so close were these frail non-waterproof nests to a waterfall, that Fig. 98. Fig. 99. Fig. 100. Fig. 98. Above, vertical section through base of cell of S. depressigaster, showing egg with food-mass placed upon it. Below, ecll base showing young larva encircling food-mass. Enlarged. Fig. 99. Half-grown larva of S. depressigaster. Enlarged. Fig. 100. Pupa of S. depressigaster, as it lies in the cell, X 5/3. the spray frequently struck them and forced the proprietors to edge around to the lee side of their dwellings, to which they habitu uly cling and where no doubt they rest at night; and after a heavy downpour the swollen stream so augmented the cataract, that the nests, soaked with water, broke up and were abandoned. But once did I find one of these nests near habitations; it was suspended under the apex of the roof of a rustic sort of open seedling house, itself in a well-wooded country. Like the smaller solitary species whose nest is illustrated in Fig. 103, these insects will often suddenly buzz their wings when on the nest and thus cause the featherweight structure to sway as in a light breeze. The nests are sombre earth color, variegated with “lighter or darker bands, and are not at all conspicuous. The wasps are often seen in small companies hanging to some slender rootiet. extended observations on this species will doubtless reveal very interesting habits; the wasps are very homelike and will repair their nests time and again—even after Vespa deusta has destroyed or deformed every cell it contains—and bravely start life’s gamble anew. 171 Stenogaster varipictus Rohwer. Length 14mm.; brown, with pale yellow markings. This is not a rare forest insect—that is, if it be rightly looked for—since it is by no means conspicuous and usually nests in rather inaccessible places. Ina small way, it seems to be a social wasp; one to several insects attend to a cell group. It may be, however, that each female has her own lot of cells in this cell group. We usually find several nests in one place, such as on Fig. 102. Nest of S. var- Fig. 101. Nest of S. varipictus pictus, with cells sealed on the bark of a tree, X 1. by mother wasp, X 1. the smooth and somewhat overhanging side of the trunk of a large forest tree, on the overhang of boulders, especially when these are more or less surrounded by water, or they may even be fixed along the underside of green leaves, and one small nest so located had its sheltering leaf constantly struck by water drip- ping slowly from the bank above, much to the annoyance of tue proprietress. The nest (Figs. 101 and 102) consists of one or more rows of fragile cells, their mouths pointing downwards, and the object to which they are attached serving as part of their sides. The cell material appears to consist of fine grains of sandy mud mixed perhaps with particles of decayed wood. I did not witness nest- building, but the structures show that the cells may be built from either side of the row, of which there may be as many as four; the lowest being first made and the succeeding ones built partly on the bottom ends of the preceding. There may be thirty or more cells to a nest and one consisting of 16 cells had two wasp attendants. The egg is glued to the bottom of the cell; it may be placed in a partially completed one or in one which has already hatched a wasp. The larva is broad, arcuate and yellowish white. When it is full-fed the mother seals up the cell. It does not appear to 172 spin a cocoon, but impregnates the cell walls with a silvery sort of material. These nests, if they escape the searching eye of the big yellow-winged hornet, are occupied at least for some months; the colonies, at any rate. are maintained in the same localities for years. On May 3, 1917, | found one female wasp resting on a 3-cell nest; on visiting this nest again on August 17 it had been increased to seven cells, more or less complete, and at least two of which were being used again. The nest contained four Stenogaster larvae and one egg. Stenogaster ; smallest species, undetermined. Length about 12 mm. As may be seen from Fig. 103, this little wasp constructs a beautiful nest; in the single case ob- served it was ultimately a 7-celled affair secured to a slender rootless which hung out of a low bank in the forest. Although the shy PS tress was frequently seen, and she resembled < good deal a small S. depressigaster, she aad unidentified through my failure to capture her, for she deserted the nest before its completion. The-~cells, jas can: be’ seen, each open’ directly with the exterior and are arranged in a somewhat zigzag fashion. When on July 16, 1917, I first saw the nest it consisted of the two discs and one open cell, containing a wasp egg or young larva; on August 23, when I saw the wasp for the last time, there were six separate cells, the two basal ones being sealed up and a seventh cell com- menced. The two discs on the stem above the nest remind one a good deal of the metal plates fastened to the mooring lines of vessels and serv- ing as rat guards. Their function in the case of the nest may well be an imperfect protection from ants, or perhaps they may serve as um- brellas, though neither they nor the cells are Fig. 103. Nest — strictly rainproof. The wasp builds rather stead- of a speciesof ily and has several young or eggs to attend to solitary Ste- ae atthe z hs pene TD simultaneously. The cells each re equire a day ol 3/9. more to be completed. The nest, which is rather dark grey-brown, appears to be made of decayed 173 wood and is stouter and stronger than that of its social relative S. depressigaster. At first each cell is unembellished cylindrical, but eventually several longitudinal ridges or ribs are added, being made confluent with those of the cell above it. The egg is laid in the bottom of the cell, and it is probable that the grub is fed much as in the preceding species. The life-cycle is not very brief, if we base it on the two follow- ing observations: July 17 the first cell contained an egg or young larva; August 2 or 3 this cell was sealed by the mother; July 19 the second cell contained a larva, but was not closed until August 16 or 17. However, as these were the only cells closed, and as both produced an ichneumonid parasite September 4 and 8, the growth of the Stenogaster may have been retarded, or perhaps the wasp had been feeding the parasite itself for some time! The Stenogaster was a decidedly timid insect and difficult to perceive in the semi-gloom of her environs. She was away from her nest much of the time, and when at home was sometimes seen examining her cells, but more frequently resting along the middle of the nest, buzzing her wings once in a w hile and thus causing the lightly-hung structure to sway as in a gentle breeze. Stenogaster micans (Saussure) var. luzonensis Rohwer. Length 21 mm.; black, with yellow markings, wings pale smoky. The males of this rather large and graceful wasp, having much time at their disposal, are not infrequently seen in the lower Makiling forest, where they disport themselves at certain spots along the paths and glades. It is otherwise with the female, (Fig. 104). EGusehold duties seem always to keep her occupied and make her habitat a less conspicuous one. Although my observations on this insect are very incomplete, I can safely say that it is solitary in habits. The frail and exqui- site nest (Fig. 105) is a pear-shaped structure suspended at its globular end to a fern, horsehair fungus, etc., sheltered by an overhanging and often inaccessible part of a bank, or even under an imperfect cover of dead leaves or other forest trash, supported by twigs, vines, etc. The nest is made of moist and well-decayed wood chewed into a pulp and formed into a delicate paper which is not rainproof. The architect, then, is occasionally found about the decayed trunks of trees, patronizing her specially selected log for many days at a time. She is a cautious insect and, flying from her partially-built domicile to near the trunk, ap- proaches the latter with care and in light pro- gressive jerks, and aleri to any danger, alights on a chosen spot. Here material is bitten out, and when a_ sufficient load of it accumulates in her jaws she flies off and away in her peculiar graceful flight, which, though not rapid, and following each time the same path, 1s, owing to the wasp’s slenderness and the conditions of light in the forest, quite difficult to follow. Fur- thermore, she works slowly—the trips are at : ; Fig. 104. Stenogaster micans, var. luzon- intervals of approxi- ensis, 2, X 3. mately fifteen minutes, and so both time and patience are taxed. On July 20, 1917, I found one of these wasps building her nest, which was fastened to the delicate stem of a little fern growing well out of reach, under an overhang- ing ledge of the creek bank. This locality was within the forest and in the constant thunder of a small cascade. The nest was secured by a small central pedicel and _ con- sisted of one tier or~ horizontal comb of about 12 rather shallow cells. Desirous of identifying the ar- chitect, I caught and then released her; but from now on, however, the nest was deserted. On August 9, I visited this unfinished nest and was surprised to find the cells consider- Fig. 105. Completed nest of §, 2bly darkened again as well as micans, var. luzonensis, X 1. lengthened. Evidently another fe- on 17 male seeking a nesting site had chanced upon and welcomed this unfinished structure and was building it up. The two figures will serve to illustrate the growth of such a nest. The material used makes a more delicate nest than those of Vespa or Polistes. The basal portion of the nest forms the bottom of all the cells as well as the two outer sides of all peripheral cells, all being pretty hexagonal. Where two outer sides of the same cell meet, a sharp ridge or carina is formed, which is continued to near the tip of the nest, the walls of the main nest being simply extended down from these cells and drawn into a neck of filagree work and ending finally into an aperture with one of its sides formed into a spear-shaped extension. I found two incomplete and deserted nests, both in the shelter of banks; another in the course of construction was well up the side of a ravine, where it was suspended from a thread of horsehair fungus and imperfectly protected from the weather in being under a mix- ture of fallen leav es, twigs, etc. a doubtful shelter at best. I located this nest first by follow- ing the wasp’s flight and then by stumbling upon her domicile in the underbrush. This was on September 3, but I had previous- ly noted this (?) wasp (identi- fied by the direction and path of luz onensis, with only the basal her flight ) in company with One portion made. Viewed from be- OF two other females (which neath, X 1. flew in other directions) on July 26, gathering nest material from the trunk of a fallen tree which lay in the bed of a ravine. Here in a small forest opening were quite a number of males of this species. The nest up the slope was only a few inches above ground and about two-thirds built. But it was well hidden, as must needs be to escape the piracy of the large Vespa deusta. On September 6, after some heavy rains, I visited this nest again, to find it in a more unfinished state than on September 3, it having probably suffered as a result of the precipitation. On September 10, the nest was about three-fourths done, and on the 16th, it was completed, at least exteriorly. The male wasp has the habit of alighting again and again on the same leaf, carefully approaching this rest by a succession of jerky, pausing advances. He appears to be extremely nervous Fig. 106. Nest of S. micans var. « 12 176 and is seldom quiet for any length of time. Often two males will meet in mid-air to indulge in peculi ar vertical gyrations—a sort of loop-the-loop. Occasionally one meets with a specimen with large anthers glued to the back of the head. The males usually occur in small groups. Though doubtless more partial to bright forest weather, the wasp works and plays during dull days and even in light rain. A British Museum guide catalogue figures two Stenogaster (Ischnogaster) nests from Java, arid ihe: larger one exactly re- sembles that of S. micans. After describing J. eximius of Ceylon, Bingham (1890) quotes Mr. FE. E. Green, who sent him a few of the nests, which resemble those of S. micans var., as follows: “Each nest seems to be the property of one pair only. It is a low- country insect. * * * My friend (Mr. John Pole, a very accurate observer) assures me he has repeatedly seen this same species, and no other, associated with these nests. * * * Mr. Pole writes of the wasps: ‘Their habit seems to be to remain in the opening, using the lacework at the bottom as spyholes.’ J. nigrifrons of Burma constructs “a social many-celled nest, tier above tier,’ as the allied /. melleyi is stated to do by de Saussure. 177 CONCLUDING REMARKS. Approximately 182 species of aculeate wasps (including a few Bethylidae) were collected in the Philippines, and of these 52 are treated biologically in this paper. The list is as follows: Species. Retiylidae se see acta 5 Milymnidae: oreo st oo 6 All belong to Methoca. Mibttialinidaes sms aes ac 8 Scolticdae fracas ai. chs 15 Psammocharidae (Pom- pultclaeypeta sare ete erie = 54 Of these, 20 are Pseudagenia or related mud-workers. ZAG UCIG AC Aas aes 5: 3 SIOUNSCG HIG AEE n onereeie eertie 15 IDA SORTS. Re Octo aan 19 Crabromidae fea. if oa: 7 dinvpoxylominae ss... -). 8 Trypoxylon 6, Pison 2. Rempuredonidae 2.2.4. 1 Vine sid aes austen 5 Bemibecidae yan ask nee. 4 INWSSOMIGAe "So ocne cae 1 (GDS OVS te CTs sain Be ea ae 1 Genceridaenmims sen ata 5 tiimMeminae aes aoticn or 14 NWieSpindes aso at 11 MNO tale ney accent 182 The Psammocharidae furnished 15 species for biological data, the Sphegidae 8, and the Eumeninae 6. Some of the wasps, notably the highly-specialized genus Steno- gaster, are partial to the shady forests; others, as some Psammo- charidae and Larrinae, seem to prefer the hotter and more open lowlands, while others still, as various Eumeninae, are at home in either environment. A number of wasps hunt or build in and about the bamboo and palm houses, especially when these are in the vicinity of a woodland. These dwellings are very open, and their timbers often pierced by the borings of small beetles, the numerous bam- boo uprights regularly perforated for the support of other framework, and the thatching of Nipa palm leaves offer inviting places for nest-building. Thus in the house in which I lived for over a year, and which was situated on the edge of the Makiling forest, at least 20 species of wasps nested or sought their prey 178 within its walls. Three species of mud-daubers (Sceliphonin1) were the most conspicuous builders, and two species of these wasps, and the two, inch-long Eumenes dwelt in the forest as well. The various wasp enemies, Ichneumonidae, Bombyliidae and Chrysididae, were also present indoors. While there appears to be a tendency towards social life in several widely-separated groups of wasps, the vespoid forms are the only ones through which it has been brought to a conclusion and become an accomplished fact—in the Vespinae. We find true social life in the Hymenoptera characterized by several elements, chief of which are: the division of labor, marked in the higher forms by three phases, viz: egg-laying fe- males or queens, sterile females or workers, and males—then we have the more or less elaborate communal nest, which may afford shelter for both adult and young wasps—and finally, the habit of feeding the wasp-grub from time to time until it is full grown. A strongly-developed instinct to defend the nest is also present in the majority of social wasps. Now, there are a number of solitary wasps which possess one or more of these traits. Many of the Bembecidae, sand wasps which often live in colonies, feed their young from day to day; others of this family lay the egg before storing any food, and some Microbembex provide their young with food in several in- stalments and close up the cell before the grub is full-fed. The Peckhams (1898) have noted that the American larrid, Lyroda subita, feeds her young from day to day, while Ammobia mutica of the Philippines may continue to store a cell with grasshoppers ‘until her larva has reached a considerable size. It does not appear to the writer, however, that the cases just mentioned indicate progress towards social life. With some of these wasps, as Ammobia mutica, and possibly Microbembe.x, it suggests a prey or food difficult to secure at all times—hence the more or less protracted period required to fully store the cell. The erratic Bembex do not show an industry commensurate with their activity, and they probably feed their young with as many flies as they are able or feel inclined to catch in a day. Some Psammocharidae, as Macromeris, Paragenia, and prob- ably a few Pseudagenia, of the eastern tropics, manifest habits which more nearly approach those of the Vespinae. The Psam- mocharidae are lower down in the scale of development than the Vespinae, and have arrived at this relative specialization as an independent branch, which, if further developed, would termi- nate perhaps in a group of social wasps parallel, but not closely related to our present communal forms. While these interesting 1907. 1907. OMe 1903: NOW. 1918. 1835. 1905. Ileal Iie 183 FERNALD, H. The Digger Wasps of North America and the West Indies belonging to the Subfamily Chlorioninae. Proe. U. S. Nat. Mus. XXXI, pp. 291-423, pl. VI-X (p. 364, note on biology of C. elegans). FrErTON, C. Sur les Moeurs der Dolichurus haemorrhous Costa. Actes de la Soe. Linn, de Bordeaux, XLVII. FERTON, C. Pompilus pectinipes V. de lL. Formation d’une espece parasite. Ann. de la Société Entomologique de France, LXXTV, 1905, pp. 73-80. Forses, H. O. A. Naturalist’s Wanderings in the Eastern Arehi- pelago. London. (On pp. 72-73, Zethus cyanopterus, 1 fig.). Forses, 8. A. 24th Illinois Report, pp. 157-160, pl. X. (Tiphia vulgaris). Froceatt, W. W. Nests and Habits of Australian Vespidae and Larridae. Proc. Linn. Soc. N. S. W., IX, ser. 2, pp. 27-34. Froecart, W. W. Australian Insects. Sydney. Froccatrr, W. W. ‘‘ Policemen Flies’’; Fossoriai Wasps that Catch Flies. Agric. Gaz. N. S. W., pp. 667-669; 2 figs. GREEN, E. E. On the Nesting Habits of Trypoxylon intrudens and Stigmus niger. Spolia Zeylaniea I, pt. ILL. GRosSBECK, J. A. Habits of Cercéris fumipennis Say. Journ. INS nite SOC eXOXe piso: Hacker, H. Entomological Contributions (Notes on collecting Hy- menoptera). Mem. Queensland Mus. VI, pp. 106-111, pl. XXXI- XXXII. Haumpay, H. H. Notes on the Bethyli and on Dryinus pedestris. Ent. Mag. II, pp. 219-220. HARTMAN, C. Observations on the Habits of Some Solitary Wasps of Texas. Bull. Univ. Texas No. 65, Scientific Ser. No. 6, pp. 1-72, pl. I-IV: HarTMAN, C. The Habits of Humenes belfragei Cress. Jour. Ani- mal Behavior III, pp. 353-360, figs. 1-7. Howes, P. G. Tropical Wild Life in British Guiana (Beebe) ; Entomological Part III, pp. 371-450, pl. I and II, figs. 126-140. Huspsparp, H. G. Some Insects Which Brave the Dangers of the Pitcher Plant. Proe. Ent. Soc. Wash. III, p. 314-315. [Lsodontia (Sphex) philadelphica St. Farg.]. HUNGERFORD, H. B. and WinitaMs, F. X. Biological Notes on Some Kansas Hymenoptera. Ent. News. XXIII, pp. 241-260, pl. XIV- xaVale IseLy, D. The Biology of Some Kansas Eumenidae. Kansas Uniy. Science Bull. VIII, pp. 233-309, pl. XX XIV-XXXVII. JONES, F. M. Pitcher Plant Insects. Ent. News, pp. 14-17, pl. WIT and 1V. (Lsodontia philadelphicus Lep.). JORGENSEN, P. Los Crisididos y los Himipteros Aculeatos de Men- doza. Annales Musea Nacional de Buenos Aires 3, XV, pp. 267- 338. (Pompilid Brethesia catches Mygale spiders). KUNKEL D’HERCULAIS, J. Les Insectes. Paris. Latrer, O. H. Bees and Wasps. Cambridge University Press. The Cambridge Manuals of Science and Literature, p. 132, 22 figs. LEPELETIER DE SAINT-FARGEAU, AMEDEE. Histoire Naturelle des Insects. Hymenopteres. Tome 2°, Paris. 1878. 1882. 1893. 1893. 1909. UOUT. 1840. 1895. 1898. 1900. 1905. 1914. Mek LOM UGHE TONS: 184 Matnpron, M. Notes pour servir a |’histoire de Hyménoptéres de l’Archipel Indien et de la Nouvelle-Guinée. Observations sur quelque Sphegiens (g. Pelopoeus) de 1’Archipel Indien, Métamor- phoses, Descriptions d’espéces. Ann. Soe. Ent. Fr. (5) VIII, pp. 385-398, pl. IX. Mainpron, M. Notes pour servir a l|’histoire des Hyménoptéres de l’Archipel Indien et de la Nouvelle-Guineé, II, Observations sur quelque Sphégiens. Ann. Soc. Ent. Fr. (5) IX, pp. 173-182, pls Vi. MAtnpron, M. Histoire des Guépes solitaires (Enméniens) de 1’Archiped Indien et de la Nouvelle-Guineé. Ann, Soc. Ent. Fr. (6) II, pp. 69-76, 169-188 and 267-286, pls. III-V. MARCHAL, P. Remarques sur les Bembex. Ann. Soc. Ent. Fr., pp. 93-98. MarcuHat, P. Observations biologique sur les Crabronides. Ann. Soc. Ent. Fr., pp. 331-338. Planche 8. MAXWELL-LEFROY, H. Indian Insect Life. Agricultural Research Institute, Pusa. Merrit, E. D. A Flora of Manila. Bureau of Science Pub. No. 5, pp. 5-490. Manila. Muir, F. The Introduction of Scolia manilae Ashm. into the Ha- wallan Islands. Ann. Ent. Soc. Am. X, pp. 207-210. PackarD, A. S. Notes on the Transformations of the Higher Hy- menoptera I. Jour. N. Y. Ent. Soc. LV, pp. 155-166; 5 figs. ParKeER, J. B. Notes on the Nesting Habits of Bembex nubilipennis. Ohio Naturalist X, pp. 163-165. PARKER, J. B. Notes on the Nesting Habits of Some Solitary Wasps. Proc. Ent. Soc. Wash. XVII, pp. 70-77, pl. XI. Parker, J. B. A Revision of the Bembicine Wasps of America North of Mexico. Proc. U. S. Nat. Mus., Vol. 52, pp. 1-155, LOIT; 230) figs. PASSERINI, C. Osservazioni sulle larve, ninfe, ed abitudini della Seoha flavifrons, Pisa 1840, et Continuazione delle osservazioni nell anno 1841 sulle larve di Seolia flavifrons. Firenze, 1841. PECKHAM, G. W. and E. G. Notes on the Habits of Trypoxylon rubrocinctum and Trypoxylon albopilosum. Psyche, VII, pp. 303- 306, PrcKHAM, G. W and E. G. Instincts and Habits of the Solitary Wasps. Wisconsin Geol. & Nat. Hist. Survey. Bul. 2, Sci. Ser. 1, pp. 1-245; 3 figs. and 14 pl. PrcKkHAM, G. W. and E. G. Additional Observations on the In- stincts and Habits of the Solitary Wasps. Bull. Wis. Nat. Hist. Soc. I. New ser. No. 2, pp. 85-93. PECKHAM, G. W. and E. G. Wasps Social and Solitary, pp. 1-305; figs. PENAU, J. Social Evolution in Wasps. Scientific American, pp. 117-118; 14 figs. PERKINS, R. L. C. Fauna Hawaiiensis, Vol. I; Introductory Essay on the Fauna, Review of the Hymenoptera, pp. LX XIII-CXI. Picarp, F. Sur les Moeurs et le genre de proie de 1’ Ampulex fasciatus Jurine. Bull. Soc. Ent. France, pp. 113-116; 2 figs. PouuLtTon, EK. B. Predaceous Reduviid Bugs and Fossors, with their Prey, from the 8. Paulo District of Southeast Brazil. Trans. Ent, Soe. London, pp. XXXIV-XLI. Rau, P. and N. Some Observations on Mud Wasps. Ent. News XXIV, pp. 392-404; 4 figs. 1909. 1909. 1896. 1901. 1915. NOM. 1905. 1893. Oe OW 1914. 1896. 1868. 1897. 1900. 185 Rau, P. and N. The Biology of the Mud-daubing Wasps as _ re- vealed by the contents of their Nests. Journ. Anim. Behavior, Vole VAS pps ci-O3)) 0) ple Rau, P. anp N. The Sleep of Insects; An Ecological Study. Ann. Ent. Soc. Am. IX, pp. 227-274. Rav, P. and N. Wasp Studies Afield, pp. V-372; 69 figs. Prince- ton University Press. ReAuMurR, M. dE. Memoires pour servir a |’histoire des insectes Wale tie RaRee V. The Unadorned Tiphia, or White Grub Parasite— Tiphia inornata Say. 6th Missouri Report, pp. 123-126; fig. 34. Riney, C. V. The Larger Digger-Wasp (Sphecius spheciosus). In- sect Life IV, pp. 248-252, figs. 33-38. Rouwer, 8S. A. Some New Hymenopterous Insects from the Philip- pine Islands. Proc. U. S. Nat. Mus. XX XVII, pp. 657-660. RousBaupb, E. The Natural History of the Solitary Wasps of the Genus Synagris (translation from French from Paris. Ann. Soe. Ent., July 1910, Vol. 79, Pt. I, pp. 1-21). Annual Rept. Smith- sonian Inst., pp. 507-525, pl. I-IV. ScuHoiz, E. J. R. Die Lebensgewohnheiten schlesischer Grabwespen I, Zeit. f. Wiss. Insectenbiol V, pp. 21-26. ScHouz, E. J. R. Die Lebensgewohnheiten schlesischer Grabwespen II, Zeit. f. Wiss. Insectenbiol V, pp. 179-182; 2 figs. ScuHwarz, E. A. Sleeping Trees of Hymenoptera. Proc. Ent. Soe. Wash. IV, pp. 24-26. SHarp, D. Camb. Nat. Hist. Ins. VI, pt. 2. London. SmirH, H. E. The Grasshopper Outbreak in New Mexico during the Summer of 1913 (Priononyx atrata and Stizus wnicinctus). Bully 2935 U0. Ss Dept. Aerie: “Profess) paper: SmMyTHE, E. G. The White Grubs Injuring Sugar Cane in Porto Rico. I. Life Cycles of the May Beetles or Melolonthids. Jour. Dept. Agric. Porto Rico, 1 and 2, pp. 47-92; 9 pl. (Elis, Camp- someris and Tiphia). Snoperass, R. E. The ‘‘Coulee Cricket.’’ Journ. N. Y. Ent. Soe. XIII, June. (Palmodes morio Kohl, p. 81). SoutHwick, E. B. The Parsnip Web Worm (Depressaria hera- cliana De G.). Insect Life V, pp. 106-109. (Humenes fraterna preys on the caterpillars). Swezey, O. H. Scolia manilae, a successfully introduced parasite for the Anomala grub. Hawaiian Planters’ Record XVII, pp. 50- 55; 5 figs. TurRNER, R. E. On a Collection of Sphecoidea sent by the Agricul- tural Research Institute at Pusa, Bihar. Mem. Dept. Agric. in India V, No. 4, pp. 173-205. (Dolichurus gilberti Turn. of India preying on roaches). VASSILIEV, I. V. Mem. Bureau Ent. Scient. Committee, Central Bd. Land Admin. & Agric., St. Petersburg, VII, No. 2, second en- larged edit., 36 pp., 20 figs., 2 col. pl. (Short Notes on Anisoplia- austriaca Herbst. and Methods of Fighting It. (Scolia enemies). Von IHeERING, H. Zur Biologie der Sozialen Wespen Brasiliens. Zool. Anzeiger, XIX, pp. 233-309, pl. XXXIV-XXXVII. WausH, B. D. and RitEy, C. V. Wasps and Their Habits. Amer. Entomologist I, pp. 122-148 and 162-164; figs. 16. Weep, C. M. Life Histories of American Insects. New York. WHEELER, W. M. A Study of Some Texan Poneridae. Biol. Bull. Now 2) pps J3sie 10) figs: 1914. 1908. 186 WHEELER, W. M. A Solitary Wasp (Aphilanthops frigidus F. Smith) That Provisions Its Nest with Queen Ants. Journ. Anim. Behavior, Boston, III, pp. 374-387. WHEELER, W. M. A Study of Some Ant Larvae, with a Considera- tion of the Origin and Meaning of the Social Habit Among Insects. Proc. Amer. Phil. Soc. LVII, pp. 293-343; 12 figs. WHEELER, W. M. The Parasitic Aculeata; A Study in Evolution. Proc. Am. Phil. Soe., LVIII, No. 1, pp. 1-40. WicKHAM, H. F. Habits of a Wasp (Agenia architecta Say). Ent. News IX, p. 47. Wiuuams, F. X. Monograph of the Larridae of Kansas. Kansas Univ. Science Bull. VIII, pp. 119-213, pl. XXII-XXX. WiuuiaMs, F. X. Notes on the Habits of Some Solitary Wasps that Occur in Kansas. Kans. Univ. Science Bull. VIII, pp. 223- 230), pl: XOX. : WILLIAMS, F. X. Notes on the Life-History of Methoca Stygia Say. Psyche, XXIII, pp. 121-125, pl. XII. WILLIAMS, F. X. Epyris Extraneus Bridwell (Bethylidae), a Fos- sorial Wasp that Preys on the Larva of the Tenebrionid Beetle, Gonocephalum seriatum Boisduval. Proc. Hawaiian Ent. Soe. for 1918, pp 55-63; 8 figs. WILLIAMS, F. X. Some Observations on the Leafhopper Wasp, Neso- mimesa Hawaiiensis Perkins, at Pahala, Hawaii, Feb. 11-April 25, 1918. Proce. Hawaiian Ent. Soc. for 1918, pp. 63-68; 3 figs. Wotcort, G. N. Notes on the Life-History and Eeology of Tiphia inornata Say. Jour. Econ. Ent., Vol. VII, pp. 382-389. Guide to the Exhibited Series of Insects, in British Museum, Lon- don. (pp. 248, 249 show nest of Ischnogaster sp.) ~~ + SMITHSONIAN INSTITUTION LIBRARIES WNL Re spe { fe Work Poa’ 3 Ex per ri timen Fits 4 the pe awe fa ' be feel So Shea a . | 9 eval Jas} Odveies