'832 
 H9 
 
 the Museum of Comparative Zoology 
 .T HARVARD > i . ; , .: 
 
 VOL. L. No. -'. 
 
 
 HYMEXOi'TEKA FIJO.M FL(>UI>>ANT. CnL 
 
 BY T 1). A. C<x 
 
 CAMBRIDGE, MASS., U. S. A. : 
 
 PRINTED FOR THE MUSEUM. 
 
 JCNE, 1906.
 
 Bulletin of the Museum of Comparative Zoology 
 AT HARVARD COLLEGE. 
 
 VOL. L. No. 2. 
 
 FOSSIL HYMENOPTERA FROM FLORISSANT, COLORADO. 
 
 BY T. D. A. COCKERELL. 
 
 CAMBRIDGE, MASS., U.S.A.: 
 
 PRINTED FOR THE MUSEUM. 
 
 JUNE, 1906.
 
 fa?
 
 No. 2. Fossil Hymenoptera from Florissant, Colorado. 
 By T. D. A. COCKERELL 
 
 THE Tertiary shales of Florissant, Colorado, have been made famous 
 through the writings of Lesquereux and Scudder, wherein are described 
 hundreds of species of plants and insects preserved in fine volcanic ash 
 and sand. The vast multitudes of individuals and species, and the won- 
 derful state of their preservation, render the locality perhaps the richest 
 of its kind in the world, and afford us as good an opportunity as could 
 be looked for to reconstruct the fauna and flora of a remote age. Just 
 what age this is, is a matter in dispute ; but for various reasons, which 
 I give in a paper to be issued in the University of Colorado Studies, I 
 think it is almost surely Miocene. 
 
 Unfortunately, Mr. Scudder has not been able to finish the investi- 
 gation of the materials he secured at Florissant. In his work on Tertiary 
 Insects (1890) he indicated briefly the great wealth of undescribed species. 
 Since then he has published some miscellaneous species (Bull. 93, U. S. 
 Geol. Surv., 1892), the Khynchophorous Coleoptera (Monog. U. S. Geol. 
 Surv., 1893, 21), the Adephagous and Clavicorn Coleoptera (Monog. 
 U. S. Geol. Surv., 1900, 40), and the Tipulidae (Proc. Amer. Philos. 
 Soc., 1894, 32). The great work accomplished by Mr. Scudder can in 
 some measure be understood by one who has learned the difficulties of 
 this kind of investigation ; the eye-strain involved in determining 
 minute and often nearly obliterated features, and the wide knowledge 
 and good judgment necessary in order to classify specimens which only 
 exhibit part of the characters commonly used as diagnostic. It is not 
 to be expected that another such master of palaeoentomology will appear 
 to take up the work ; but the valuable materials must not be neglected, 
 and we may hope that with the aid of several workers they will all 
 be made known. 
 
 The present contribution deals with the bees and wasps, and one 
 species of Stephanidae, kindly entrusted to me by the Museum of 
 Comparative Zoology. In addition to the species described, I have 
 examined more imperfect specimens of perhaps as many others ; but it 
 has seemed best to publish only those which could be classified with 
 TOL. L. No. 2 3
 
 34 BULLETIN: MUSEUM OF COMPARATIVE ZOOLOGY. 
 
 some certainty, and adequately diagnosed. If the imperfect specimens 
 just mentioned had been all the Hymenoptera found at Florissant, it 
 would have seemed worth while to give them more careful scrutiny, and 
 to describe a number as well as possible. No doubt, by very careful and 
 prolonged comparisons, such portions of the venation as could be deter- 
 mined would be found in many cases to reveal probable or practically 
 certain affinities; but the work would be arduous in the extreme, and 
 would test one's skill to the utmost. As it is, the numerous well-pre- 
 served specimens give us an excellent idea of the fauna, and the deter- 
 mination of the poorer materials may be at least postponed without any 
 serious injury to science. 
 
 In numerous cases, owing to the wings being folded, or one beneath 
 another, the venation looks at first sight abnormal, and will appear to 
 disagree with the descriptions offered. The future student of these in- 
 sects should therefore not conclude too hastily that the descriptions are 
 inaccurate. 
 
 In general terms, it may be said that the Florissant Hymenoptera do 
 not differ greatly from their modern representatives. While some of the 
 extinct genera are apparently more primitive than the dominant genera 
 of the same groups to-day, they are scarcely more so than certain genera 
 which still exist in the modern fauna. Thus, among the Scoliids, we 
 naturally assume that those forms with regular venation, like that of 
 many other wasps, are more primitive at least in respect to this char- 
 acter than those with broken or irregular cells. The two fossil genera 
 of this group are therefore less specialized in venation than the common 
 species of to-day, but they are in the same general stage of development 
 as the rare American genus Engycystis, and the Australian Austratiphia. 
 Thus, if it were possible to restore the Florissant Hymenoptera to their 
 original state, and send them to some entomologist as coming from an out 
 of the way region, he would see in them nothing transcending the pos- 
 sibilities of the modern world. 
 
 It must further be said, that the types represented do not suggest 
 tropical or subtropical conditions ; they accord well with the vegetation 
 in indicating a cliinnte like that of the austral zones of the temperate re- 
 gion. The bees are principally of genera found flying in Colorado to-day, 
 and there is no indication of the types especially characteristic of Mexico. 
 Both among the bees and the wasps, the element which we regard as of 
 neotropical origin is conspicuously absent. It is only just to remark, with 
 regard to the bees especially, that the generic identity assumed from the 
 parts preserved might in some cases be belied, could we examine the
 
 COCKERELL: FOSSIL HYMENOPTERA FROM FLORISSANT. 35 
 
 mouth-parts, etc. The evolution of the bees has gone on principally in 
 the development of the mouth-structures, the venation remaining nearly 
 as in the fossorial wasps, or at any rate not undergoing any radical 
 changes. Hence it may be that if we could see the tongue, palpi, etc., of 
 the Florissant species of Halictus, Andrena, Authidium, etc., we should be 
 compelled to remove them from those genera ; but the agreement of the 
 wings and general appearance is such that I feel as confident of the gen- 
 eric determinations as is possible under the circumstances. 
 
 The families represented are exactly those dominant to-day in North 
 America, and the absence of certain groups must no doubt be regarded 
 as accidental. 
 
 One would infer from the evidence afforded by the Florissant Hymen- 
 optera, that the genera of this group are more persistent in time than the 
 genei-a of Mammalia, but less so than those of flowering plants, espe- 
 cially trees. The same conclusions might be reached independently by 
 a study of geographical distribution, at least so far as they relate to mam- 
 mals and Hymenoptera. No doubt the genera of Hymenoptera are 
 more widespread than some other groups of organisms which may 
 possess greater antiquity, owing to the ready locomotion of the 
 former. 
 
 Unfortunately, we have no series of mammals known to be of the same 
 age as the Florissant shales. The White River beds, which Matthew 
 (1899) calls Oligocene, have produced in Colorado some 63 species of 
 mammals, all referred to extinct genera except a few pertaining to Didel- 
 phys and Sciurus. These animals, very differently from the Florissant 
 Hymenoptera, if produced alive would excite the greatest amazement. 
 Species of Titanotheriidae, Elotheriidae, Hyaenodontidae, Rhinocerotidae, 
 Camelidae, Oreodontidae, etc., would cause bewilderment to a zoologist to- 
 day. Even those pertaining to families still inhabiting the earth would 
 for the most part look quite strange to us, being of extinct genera. 
 
 The Loup Fork beds, referred to the Upper Miocene, have produced 
 in Colorado about 28 species of mammals, but even these are nearly 
 all of extinct genera, though only two, possibly three, of the families are 
 extinct. We note the arrival of the Elephautidae, and the great abun- 
 dance and variety of Equidae. As the Florissant shales are certainly not 
 late) than the Loup Fork, but doubtless earlier, the opinion that the 
 families and genera of aculeate Hymenoptera are much more conservative 
 than those of Mammalia seems justified. The same facts lead us to 
 believe that the differences noted by Scudder between the insects of the 
 Green River series and Florissant surely indicate a considerable difference
 
 36 BULLETIN: MUSEUM OF COMPARATIVE ZOOLOGY. 
 
 in time ; and since the Florissant beds must for a variety of reasons be 
 held to be the later of the two, the probability that they are Miocene 
 is augmented. 
 
 APOIDEA. 
 
 TABLE OF SPECIES. 
 
 Three submarginal cells 1 
 
 Two submarginal cells 5 
 
 1. Basal nervure strongly curved; marginal cell ending in a point on costa; 
 
 insect small, about 6| mm. long, anterior wing somewhat over 4 mm. 
 
 Halictus florissantellus. 
 Still smaller ; length about 4i mm., intense black 
 
 Halictus scudderiellus. 
 
 Basal nervure not, or not strongly, curved; larger, anterior wing over 5 mm. 
 long 2 
 
 2. Second s. m. receiving first r. n. before the middle ; anterior wing about 
 
 8 mm. long Calyptapis florissantensis. 
 
 Second s. m. receiving first r. n. beyond the middle, or at apex ; anterior wing 
 less than 7 mm. long 3 
 
 3. Point of marginal cell a short distance from costa ; second r. n. bent near 
 
 upper end ; first s. m. shorter than second or third on cubital nervure 
 
 Lithandrena saxorum. 
 Point of marginal cell a short distance from costa ; second r. n. not bent near 
 
 upper end ; size small Ceratina disrupta. 
 
 Point of marginal cell on costa ; second r. n. not bent near upper end . . 4 
 
 4. Abdomen normal ; second s. m. just three times as broad below as above 
 
 Andrena sepulta. 
 Abdomen clavate ; second s. m. narrower, not nearly three times as broad 
 
 below as above Andrena (?) clavula. 
 
 6. Stigma small ; insect broad and robust 6 
 
 Stigma fairly or quite large ; insect smaller, or less robust 8 
 
 6. Abdomen subglobose, without visible markings ; second r. n. passing well 
 
 (about 120 M ) beyond apex of second s. m. ; breadth of marginal cell about 
 
 680 /j. . Dianthidium tertiaritim. 
 
 Abdomen longer, banded 7 
 
 7. Wings strongly infuscated ; marginal cell about 720 p broad 
 
 Anthidium scudderi. 
 Wings clear ; marginal cell about 570 n broad . . . . Anthidium exhumatum. 
 
 8. T. m. with the lower end most apicad, so that it forms an angle with b. n. ; 
 
 eyes prominent Libellulapis antiquorum. 
 
 T. m. with the lower end most basad so that it is in line with lower end of b. n. 9 
 
 9. Small, length slightly over 6 mm. ; abdomen dark brown ; width of marginal 
 
 cell 255 ft, Heriades haUctinus. 
 
 Larger, length 8 mm. or over ; width of marginal cell 300 n 10 
 
 10. Abdomen light reddish brown ; head smaller Heriades laminarum. 
 
 Abdomen banded ; head larger Heriades bowditchi.
 
 COCKERELL: FOSSIL HYMENOPTERA FROM FLORISSANT. 37 
 
 The following abbreviations are used : b. n. = basal nervure ; s. m. = submarginal 
 cell ; r. n. = recurrent nervure ; t. m. = transverso-medial nervure ; t. c. = trans- 
 verso-cubital nervure. In the wing, breadth always means in the direction of 
 the short axis in the case of the marginal cell and stigma. 
 
 CERATINIDAE. 
 Ceratina disrupta, sp. nor. 
 
 Black ; probable length about 8 mm., or perhaps less ; anterior wing about or 
 not quite 6 mm., dusky, especially in marginal cell and beyond ; head separated 
 a short distance from body in type and seen in side view, about 2175 n long and 
 1050 from back to front, with the broadly rounded cheeks and general appear- 
 ance of Ceratina ; stigma well-developed, its width about 150 (this and all follow- 
 ing measurements in ji), its margin bordering marginal cell about 300; marginal 
 cell 1350 long, 370 wide, ending in a point a little away from costa; first s. m. 800 
 long, its length on cubital nervure 525 ; second s. m. much narrowed above its 
 length on marginal 200, on cubital nervure 600 ; third s. m. 300 long on marginal, 
 nearly 600 on cubital ; lower section of b. n. (bordering first discoidal) gently, not 
 strongly, curved, about 676 long ; length of first discoidal 1500 ; first r. n. entering 
 second s. m. beyond middle ; second r. n. entering third s. m. 160 from its end, the 
 upper end of the nervure not bent. 
 
 Type. No. 2001, Mus. Comp. Zool. Florissant, Col. (No. 9355, S. H. 
 Scudder Coll.). 
 
 MEGACHILIDAB. 
 Anthidium FABRICITJS. 
 
 The species of Anthidium differ among themselves in the details of the 
 venation, as shown in the following table : 
 
 First r. n. meeting first t. c., or passing a very short distance beyond it. 
 
 A. oblongatum Latr. (Europe). 
 A. bernardinum Ckll. (So. Calif.). 
 A. illustre Cress. (Western U. S.). 
 First r. n. passing a fair or long distance beyond first t. c 1 
 
 1. B. n. meeting t. m. ; t. m. of hind wing only moderately oblique 
 
 A. stelvides Spin. (Chile). 
 
 B. n. passing basad of t. m. (sometimes very slightly) ; t. m. of hind wing 
 very oblique 2 
 
 2. Basal angle of first s. m. about as acute as angle formed by basal and sub- 
 
 costal nervures A. laterale Latr. (Palaearctic). 
 
 Basal angle of first s. m. obviously more obtuse than angle formed by basal 
 
 and subcostal nervures . . . . A. emarginatum Say (Colorado, etc.). 
 
 A. conspicuum Cress. (Colorado, etc.). 
 
 A. scudderi, sp. nov. (Florissant). 
 
 A. exhumatum, sp. nov. (Florissant).
 
 38 BULLETIN: MUSEUM OF COMPARATIVE ZOOLOGY. 
 
 It thus appears that so far as the venation shows, the Florissant species are 
 nearest to some of those flying in Colorado at the present day. In the follow- 
 ing descriptions, those characters are italicized which may especially be relied 
 upon for the separation of the fossil species. 
 
 Anthidium scudderi, sp. nov. 
 
 Robust, probably about 15 mm. long (the end of the abdomen is lacking) ; 
 width of thorax about 6 mm. (probably increased by flattening), of head about 4 ; 
 general appearance normal ; head and thorax black, with faintly indicated light 
 markings ; apparently the clypeus was light, and a large patch on vertex, and a 
 pair of longitudinal subdorsal stripes on anterior part of thorax (mesothorax), but 
 these markings, vaguely indicated by reddish color, may not truly represent the 
 tegumentary colors ; mandibles apparently short and heavy ; antennae and legs not 
 visible ; abdomen broad, very pale reddish, with the hind margins of the segments 
 infuscated, the darkening strongest on the actual margin, and gradually fading 
 anteriorly, the dark band occupying about a third of the visible part of the segment 
 (much less on the first) ; on the second to fourth segments are rather poorly indi- 
 cated dark marks in the subbasal region in the middle line, and on each extreme 
 lateral margin, apparently indicating a subbasal band very broadly interrupted in 
 the subdorsal region ; it is perhaps probable that the abdomen was in life yellow 
 marked with ferruginous ; the apparent pattern is not quite like that of any 
 modern species before me, but it is not difficult to see how it might become modified 
 into some of the patterns seen in modern Rocky Mountain species. Quite a close 
 general resemblance is shown by the abdominal pattern of A. bernardinum, but in 
 that species the lateral subbasal dark spots are much nearer the middle line. 
 
 Anterior wings about 8 mm. long ; wings strongly infuscated, except in the basal 
 region, conspicuously hairy. Venation in general quite normal ; marginal cell broad, 
 its width about 720 n; b. n. going only just basad of t. n. ; first r. n. joining 
 second s. m. a long distance (quite 420 ^u) from base ; discoidal nervure oblique 
 and curved, so that the second discoidal cell is conspicuously longer on lower than on 
 upper side, the upper outer corner being very obtusely rounded. Hind wing with 
 t. m. long and very oblique. 
 
 Type. No. 2002, Mus. Comp. Zobl. Florissant, Col. (No. 11,381, S. H. 
 Scudder Coll.). Dedicated to Mr. Scudder. 
 
 Anthidium exhumatum, sp. nov. 
 
 <f Robust; length 13$ mm. ; width of head a little over 4, of thorax about 5, of 
 abdomen about 6| mm., these measurements (particularly the last) no doubt 
 increased by crushing ; head and thorax black without any apparent markings ; 
 mesothorax coarsely roughened ; ocelli large, not approaching eyes, distance be- 
 tween middle and lateral ocelli a little less than the diameter of one ; abdomen 
 with broad pale reddish bands, the hind margins of the segments not darkened, nor any 
 spots visible ; apex broadly rounded, no processes or teeth being visible, but a large 
 quadrate area is occupied by the genitalia, the exact structure of which cannot be 
 made out ; hind tarsi apparently broad and flattened ; hind tibiae with a rather 
 abundant hairy scopa. Wings colorless ; nervures pale; marginal cell not so broad
 
 COCKERELL: FOSSIL HYMENOPTERA FROM FLORISSANT. 39 
 
 as in A. scudderi (its breadth about 570 /u) ; stigma shorter and smaller, more like that 
 of Megachile ; second discoidal cell with the upper and lower sides about equal, the 
 upper apical corner hardly depressed ; b. n. almost meeting t. m. ; first r. n. passing 
 some distance beyond first t. c. ; second perhaps passing slightly beyond apex of 
 second s. m. ; second (morphologically third) t. c. strongly bent near the middle; 
 t. m. of hind wing hard to see but oblique. 
 
 Type. No. 2003, Mus. Comp. Zool. Florissant, Col. (No. 13,709, S. H. 
 Scudder Coll.) and reverse of the same specimen (No. 11,388, S. H. Scudder 
 Coll.). 
 
 This is not quite so typical an Anthidiuni as A. scudderi, but I think it can- 
 not be referred to any other genus. These bees are referred to Anthidium and 
 not to Megachile, not only because of the color of the abdomen, but also on 
 account of the characters of the venation. The following comparison shows 
 the difference in venation between A . scudderi and Megachile calogaster : 
 
 A. SCUDDERI. M. CALOGASTER. 
 
 B. n. goes a little basad of t. m. B. n. falls a little short of t. m. 
 
 First r. n. joins second s. m. at a First r. n. joins second s. m. at a 
 
 distance from its base almost as great distance from its base not greater than 
 
 as half length of r. n. one quarter length of r. n. 
 
 Stigma larger and more pointed. Stigma small and truncate. 
 
 Marginal cell conspicuously broader Marginal cell conspicuously narrower 
 
 than greatest breadth of first s. m. than greatest breadth of first s. m. 
 
 First discoidal much broader. First discoidal long and narrow. 
 
 Second r. n. not well seen, but en- Second r. n. enters second s. m. 
 
 ters second s. m. at or very near tip. about as far from apex as first r. n. does 
 
 T. m. of hind wing very oblique. from base. 
 
 T. m. of hind wing not or hardly 
 oblique. 
 
 Specimen No. 8444 is an Anthidium exactly agreeing with A. exhumatum in 
 the width of the marginal cell and in the shape of the second discoidal, but 
 having most of the venation obliterated. It differs by the strongly banded 
 abdomen (suggestive of the living A. occidentale), which, however, is not 
 spotted. It appears to be a female, and I little doubt that it represents that 
 sex of A. exhumatum. 
 
 Dianthidium tertiarium, sp. nov. 
 
 <f Body black or dark brown, without visible markings ; head lacking ; length, 
 exclusive of head, about 8 mm. ; width of thorax about 4J, of abdomen about 4-J 
 mm.; anterior wing about 9 mm. lon<r, slightly dusky, with the nervures dark ; 
 abdomen subglobose, terminating in two rather small and obscure tubercles, which 
 are about as far distant as the breadth of the basal joint of hind tarsus at apex; 
 posterior claws with a strong inner tooth. Venation of anterior wings normal ; 
 first r. n. ending about 300 /u from base of second s. m. ; second r. n. passing about
 
 40 BULLETIN: MUSEUM OF COMPARATIVE ZOOLOGY. 
 
 120 n beyond apex of second s. m. ; marginal cell about 630 /* broad ; first s. m. 
 narrower than in Anthidium exhumatum ; t. m. obliterated. 
 
 The general shape and the structure of apex of abdomen seem to indicate a 
 species of the subgenus Anthidiellum, allied to the modern D. gilense Ckll. The 
 venation agrees well with D. gilense, except that the first r. n. enters the second 
 B. m. at a greater distance from the base and the marginal cell is more evenly 
 rounded at apex. 
 
 Type. No. 2004, Mus. Comp. Zool. Florissant, Col. (No. 806, S. H. Scud- 
 der Coll.). 
 
 Heriades laminarum, sp. nov. 
 
 Length 8 mm., robust, head and thorax black, abdomen very light reddish brown, 
 doubtless red in life ; width of abdomen 3 mm. ; length of anterior wing about 4 
 mm., venation pale reddish-brown ; stigma fairly large, vein separating first s. m. 
 from marginal cell , not quite as long as that separating stigma from marginal 
 cell; marginal cell narrow and long, its width 300 yu, its apex rounded; first dis- 
 coidal about 225 /x shorter than marginal ; first s. m., on cubital nervure, about 075 
 /i long, r. n. about but its total length is about 810 ; second s. m. about 645 long, 
 receiving first r. n. about 100 /* from base, and second r. n. hardly 30 from apex ; 
 b. n. bent, falling a little short of t. m. ; t. m. in a line with lower part of b. n. 
 (which shows that the bee is not a Panurgid) ; third discoidal shorter than in 
 H. truncorum. T. m. of hind wing not at all oblique. 
 
 Type. No. 2005, Mus. Comp., Zool. Florissant, Col. (No. 3062, S. H. 
 Scudder Coll.). This looks like Proteriades semirubra (Heriades semirubra 
 Ckll.), but it appears to be allied to the ordinary species of Heriades. 
 
 Heriades halictinus, sp. nov. 
 
 ? Length slightly over 6 mm. ; anterior wing about 4 mm. ; stout-bodied, head 
 and thorax black, abdomen dark brown ; eyes narrow. Stigma large ; marginal 
 cell about 975 /t long and 255 wide, end rounded ; length of first s. m. on cubital 
 nervure about 555; length of second s. m. about 610, it is not greatly narrowed 
 above, its outer margin presents a gentle double curve ; lower edge of first s. m. 
 straight; upper apical corner of second discoidal rounded; lower section of b. n. 
 curved as in Halictus, meeting t. m. ; first r. n. joining second s. m. at a distance 
 from its base equal to about half length of first t. c. ; second r . n. joining cell at 
 its extreme apex ; length of lower (curved) part of b. n. about 300 M- 
 
 Type. No. 2006, Mus. Comp. Zool. Florissant, Col. (No. 10,564, S. H- 
 Scudder Coll.). 
 
 At first sight one would take this for a small Halictus, but the t. m. and 
 various other characters indicate its true affinity. Compared with Halictus 
 similis, the principal differences in venation are as follows : 
 
 (1) Straight section of b. n. (bounding first s. m.) almost as long as curved 
 
 section (only about a quarter as long in H. similis). 
 
 (2) Only two submarginal cells. 
 
 (3) Second r. n. joining second (morphologically third) s. m. at extreme 
 
 apex (far from apex of third in H. similis).
 
 COCKERELL: FOSSIL HYMENOPTEKA FROM FLORISSANT. 41 
 
 (4) Second (morphologically third) t. c. strong (third weak in H. similis). 
 
 (5) B. n. meets t. m., which is curved in an opposite direction to b. n., 
 
 its lower end oblique and more basad than the upper (less basad in 
 H. similis'). 
 
 The H. similis used for comparison is the form obtained by Mr. Lovell in 
 Maine. 
 
 From Heriades laminarum, the present species is easily known by its darker 
 abdomen and smaller si/e, as shown especially in the wing-measurements. 
 
 Heriades bowditchi, sp. nov. 
 
 No. 13,761 is larger (length, with the head thrust forward, 10 mm.) ; the 
 thorax was evidently very coarsely punctate, the punctures contiguous ; the 
 following measurements are in /*; curved portion of b. n., 300; straight por- 
 tion a little longer ; width of marginal ceU, 300 ; its length, about 1350. The 
 t. m. is as in the other species. Head large, slightly wider than thorax ; 
 abdomen light-colored, with the apex brown (doubtless black in life), and two 
 broad entire brown bands on the apical half. The anterior wing (not perfectly 
 preserved) must have been a trifle over 5 mm. long. This differs from H. 
 laminarum by the very decidedly larger head, and the banded abdomen. The 
 apex of the marginal cell, seen from one direction, seems to be very obliquely 
 truncate, but this may be illusory. The stigma is pale, but it is certainly 
 much longer, and more slender than in H. laminarum. 
 
 A second example (No. 13,436, S. H. Scudder Coll.) confirms the validity of this 
 species. The specimen is clearly a ?. 
 
 The abdomen has broad entire reddish-brown banda on the first four segments, 
 that on the first being faint ; the marginal cell is pointed at tip, not obliquely trun- 
 cate ; its length, measured in this specimen, is the same as in the type. The t. m. 
 curves inwards below, as in the type. The legs are hairy. 
 
 Type. No. 2007, Mus. Comp. Zool. Florissant, Col. (No. 13,761, S. H. 
 Scudder Coll.). Named after Mr. F. C. Bowditch, Mr. Scudder's companion 
 at Florissant. 
 
 ANTHOPHORIDAE. 
 Calyptapis, gen. nov. 
 
 Stigma small but rather broad, about like that of Melissodes, the part within the 
 marginal cell smaller than that without; marginal cell large and broad, the tip 
 away from costa, obtusely rounded, not at all appendiculate ; three submarginal 
 cells, the third very long, and considerably the longest, narrowed a little more than half 
 to marginal ; the third t. c.with a very distinct double curve, but not abruptly bent, 
 the cell (third s. m.) slightly appendiculate at its apical point, which is not far from 
 the point of junction of the second r. n., the latter joining at the end of the straight 
 lower margin, at the beginning of the upward curve, about as in Melissodes atripes ; 
 second s. m. pentagonal, the lower inner corner produced to considerably less than a 
 right angle ; the cell is rather large, broader below than high, narrowing above,
 
 42 BULLETIN: MUSEUM OF COMPARATIVE ZOOLOGY. 
 
 from the obliquity of the first t. c. ; it receives first r. n. before the middle, at a point 
 almost beneath the upper insertion of the first t. c. ; first s. m. longer and larger 
 than second, but not very greatly so, its lower margin gently curved, giving it a 
 considerable breadtli ; b. n. straight, except hear the basad end, where it bends 
 downwards, and is attached a short distance basad of the t. m.; t. rn. not oblique; 
 second r. n. gently curved outwards, its junction with the third s. m. forming an angle 
 greater than a right angle ; second discoidal cell longer below than above, but not 
 very greatly ; first discoidal not so long as marginal, but not greatly shorter. The 
 structural characters of the body cannot be ascertained. 
 
 Calyptapis florissantensis, sp. nov. " 
 
 Black ; anterior wing 8 mm. long, venation distinct, brown. The following 
 measurements are in /x : width of marginal cell, 630 ; length of t. m., 300; width 
 of second discoidal cell at apex, 825 ; from insertion of second r. n. to appendix at 
 end of third s. m., 135 ; distance between insertion of first r. n. and base of second 
 s. m., 375 ; length of b. n. about 1875. 
 
 Type. 2008, Mus. Com. Zobl. Florissant, Col. (No. 4933, S. H. Scudder 
 Coll.). 
 
 So far as the venation goes, this genus is not far from certain species of the 
 modern Melissodes. If it were a living insect, differing from Melissodes only 
 in the manner indicated, it might be held to typify only a subgeneric group ; 
 but under the circumstances, and with a probability that the mouth-parts, etc., 
 if preserved, would afford additional characters, it seems best to treat it as a 
 distinct genus. It is probably too much to hope that fossil Anthophoridae will 
 ever be found, showing adequately the palpi and other minute characters so 
 useful in segregating modern genera. 
 
 ANDRBNIDAE. 
 Libellulapis, gen. nov. 
 
 ? Eyes apparently very prominent, the anterior part of face produced ; first 
 s. m. not so long ; second discoidal narrower at end ; second r. n. curved or bent out- 
 wards (straight in Parandrena) ; size small, abdomen conspicuously banded. The 
 head, as preserved, has a singular resemblance to that of a dragonfly. 
 
 Libellulapis antiquorum, sp. nov. 
 
 ? Length about 6 mm. ; anterior wing about 5 mm. ; width of thorax about 2|, of 
 abdomen about 2 mm. Head and thorax black ; eyes prominent ; flagellum stout ; 
 abdomen colorless, with a large brown patch on each side of middle of third seg- 
 ment ; segments 4 and 5 each with a very broad entire brown band ; 6 with a 
 fainter band ; middle and hind femora stout ; venation brown ; stigma large, but 
 rather slender, with a large part in marginal cell, width of stigma about 195 /x (all 
 the following measurements are in p) ; marginal cell long and narrow, the tip on 
 costa, width of cell about 300 ; b. n. practically straight, except a slight bend at 
 proximal end, meeting t. m., which is oblique, at least 45 out of the straight line
 
 COCKERELL: FOSSIL HYMENOPTERA FROM FLORISSANT. 43 
 
 with b. n. ; first s. m. on cubital nervure about 630 long, its lower edge straight 
 (which distinguishes it from Halictoides) ; second the same length, but only 360 long 
 on marginal, the second t. c. with a double curve ; first r. n. entering second s. m. 
 about 150 from base, second about 60 from apex ; breadth of second discoidal at 
 base 195, at apex about 390. Compared with Halictoides maurus it differs by the 
 second s. m. being much broader above, and receiving the second r. n. nearer its 
 end, by the larger and narrower second discoidal, and the lower edge of first s. m. 
 practically straight. Compared with Hesperapis rhodoceratus, the insertion of the re- 
 current nervures is different, and the second r. n. in particular is quite different in 
 its direction, etc. ; the straight lower edge of first s. m. agrees. Compared with 
 Parandrena andrenoides, the stigma is smaller, and the second discoidal is not so 
 broad apically. It does not agree with Diandrena or Biareolina. 
 
 Type. No. 2009, Mus. Comp. Zool. Florissant, Col. (No. 9061, S. H. 
 Scudder Coll.). 
 
 A second example (No. 8560, S. H. Scudder Coll.) shows that the legs are 
 dark and hairy ; the mandibles bidentate, the inner tooth rounded and small ; 
 flagellum about 195 n broad ; abdominal bands not well preserved, but a dark 
 patch at apex. This example shows the same curiously prominent eyes as the 
 type, hence it does not seem likely that the feature can be due in some acci- 
 dent of crushing. The eyes stand out on each side of the head to an extent of at 
 least 300 n, forming an angle with the anterior part of the face, which appears 
 quadrate, twice as broad as long. The eyes of Parandrena are prominent, espe- 
 cially in the male, but they do not look like those of Libellulapis. 
 
 Halictus florissantellus, sp. nov. 
 
 ? Length about 6 mm. ; stout-bodied ; head, thorax, abdomen, and legs black ; 
 width of thorax 2 mm., of abdomen slightly more ; length of anterior wing some- 
 what over 4 mm., stigma and nervures dark ; middle tibia very much broader than 
 basal joint of its tarsus (breadth of tibia 263 n, of basal joint of tarsus 120) ; b. n. 
 strongly curved, normal for Halictus, curved part 465 /tlong, straight (upper) part 
 about 150 ; t. m. a little oblique, a little basad of b. n., but not separated from it by 
 an interval, its lower end more apicad, as is normal for Halictus ; width of second 
 discoidal at base 225 ; stigma large, about 165 ju broad ; marginal cell about 270 
 broad, ending in a point on costa ; wings quite hairy in costal region anterior to 
 stigma ; first r. n. joining cubital nervure 690 /x from base ; submarginal cells not 
 traceable. 
 
 Type. No. 2010, Mus. Comp. Zool. Florissant, Col. (No. 921, S. H. Scudder 
 Coll.). 
 
 Although only part of the venation is preserved, this, and the general appear- 
 ance of the insect, agree with Halictus, and the generic reference seems safe. 
 
 Halictus scudderiellus, sp. nov. 
 
 ? Length about 4J mm., anterior wing about 2f ; intense black, including legs 
 except tarsi, which are pale reddish ; antennae stout, breadth of fiagellum about
 
 44 BULLETIN: MUSEUM OF COMPARATIVE ZOOLOGY. 
 
 150 ft ; legs normal, breadth of hind tibiae about 255 ft, of the basal joint of their 
 tarsi about 150 ; dorso-ventral diameter of abdomen about 1350 ft, length of head 
 about the same ; costa somewhat arched ; stigma large and black ; marginal 
 cell about 900 ft. long and 196 wide, ending in a point on costa; first section 
 of b. n. about 150 ft long ; second section curved, fully 375 ft long ; second s. m. 
 on marginal a little over 150 ft long, on cubital nervure slightly over 300, the second 
 t. c. curved outwards. 
 
 Type. No. 2011, Mus. Comp. Zool. Florissant, Col. (No. 1966, S. H. 
 Scudder Coll.). 
 
 Lithandrena, gen. nov. 
 
 A geuus of Andreninae, allied to Andrena. It differs from Andrena and 
 Nomia in the second r. n., which is strongly bent in its upper part, straight 
 but oblique below ; from Andrena it differs in the proportions of the subrnar- 
 ginal cells (see the dimensions given below) ; and from Nomia by the tip of 
 the marginal cell, which is pointed, and a little away from costa. The general 
 appearance is that of an Andrena, but it cannot be referred to that or any other 
 genus known to me. In Cresson's table it seems to run to Ceratina, but it is 
 not allied to that genus. 
 
 Lithandrena saxorum, sp. nov. 
 
 $ Length 8f mm., anterior wing about 5 ; diameter of thorax 3, head the same, 
 of abdomen 3 mm. ; head and thorax black ; abdomen light, with a broad entire 
 dark band on each segment ; legs hairy ; flagellum rather stout, diameter 195 ft. 
 In the following account of the anterior wing the measurements are in ft: stigma 
 well-developed, diameter 175, length of the part within marginal cell 360 ; marginal 
 cell long and pointed, length 1620, breadth 405, apex pointed, away from costa, but 
 distance from apex to opposite point on costa scarcely 65 ; three submarginal cells ; 
 total length of first s. m. 900, but its length on cubital nervure only 570, the first 
 t. c. being remarkably oblique, and having its lower part curved ; length of second 
 s. m. on cubital nervure 615, but it is greatly narrowed above, its length on mar- 
 ginal being only 150; first r. n. joining second s. m. 90 from end; length of third 
 s. m. about 676, but it is greatly narrowed to marginal, its length above being 
 about 270 ; second r. n. with a strong bend at the end of the upper two fifths, 
 the lower three fifths straight ; length of first discoidal cell 1650 ; b. n. meeting 
 t. m. ; lower section of b. n. slightly curved (but not more so than in some forms of 
 Andrena), and more than twice as long as upper one (length of lower section 630, 
 of upper 300) ; t. m. oblique, its lower end more apicad (as in Andrenines, Pan- 
 urgids, etc. ) 
 
 Type. No. 2012, Mus. Com. Zobl. Florissant, Col. (No. 8219, S. w H. 
 Scudder Coll.). 
 
 Andrena sepulta, sp. nov. 
 
 $. Length 9 mm.; width of thorax 2.J mm., of head and abdomen about the 
 same ; abdomen of normal shape ; flagellum stout ; there is an appearance as if 
 the eyes nearly met on the vertex, but I think this is illusory, resulting from the
 
 COCKERELL: FOSSIL HYMENOPTERA FROM FLORISSANT. 45 
 
 way the head is crushed ; head and thorax black ; abdomen nearly colorless, with 
 broad suffused reddish-brown bands on apical margins of the first three segments, 
 the apex also dark; abdomen hairy all over; legs light reddish-brown, hind tibia 
 about If mm. long, tarsi hairy ; wings very hairy. The following wing-measure- 
 ments are all in /i : stigma large, pointed apically, width about 300, part within 
 marginal cell about 450 long ; marginal cell long and pointed, the apex on costa, 
 length of cell 1605, breadth 375 ; three submarginals, first and third long, second 
 short, much narrowed above, almost triangular, third much narrowed above ; first 
 s. m. on cubital nervure 810, its total length 1125 ; second a. m. on cubital n., 
 405, on marginal, 135 ; third s. m. on cubital n., 720, on marginal, 330 ; bend of 
 third t. c. about 210 from cubital n. ; first r. n. enters second s. m. at extreme apex, 
 second enters third s. m. about 90 from apex ; second r. n. leaves cubital n. at a 
 right angle, but gently curves inwards, being nowhere at all bent ; upper section of 
 b. n. 330 ; lower section 630; lower section gently curved, but not at all as in the 
 Halictines ; b. n. falling about 60 short of t. m., which is oblique, its lower end 
 more apicad. 
 
 Type. No. 2013, Mus. Comp. Zobl. Florissant, Col. (No. 14,288, S. H. 
 Scudder Coll.). The venation is not exactly like that of any modern, species 
 with which I have compared, it, but the differences are unimportant. 
 
 Andrena ( ? ) clavula, sp. nov. 
 
 9. Length 8. mm.; width of thorax 3, of head 2; length of anterior wing 6f 
 mm. ; eyes ordinary ; flagellum stout, subclavate, rather short, about 300 ^ broad 
 near end ; head and thorax black, femora dark ; hind tibiae and tarsi apparently 
 pale, but middle tibiae dark ; wings somewhat dusky ; abdomen subclavate, dark 
 reddish-brown, with three rather narrow pale bands, occupying hind margins of 
 segments 2 to 4 and the extreme bases of the adjacent segments. 
 
 Venation (front wings) as in A. sepulta, except that second s. m. is narrower and 
 more parallel-sided. Measurements in /* : width of stigma 240; length of marginal 
 cell about 1455, its width 360 ; lower section of b. n., 630 ; second s. m. on cubital 
 n., 360, on marginal, 155. 
 
 Type. No. 2014, Mus. Comp. Zobl. Florissant, Col. (No. 6963, S. H. 
 Scudder Coll.). The shape of the abdomen is like that of a ? Ceratina, or 
 possibly certain Halictines, but the venation does not agree with these. As 
 the venation is exactly the same (speaking generically) as that of A. 
 it seems that the insect should be considered congeneric. 
 
 SPHECOIDEA. 
 
 CRABRONIDAE. 
 Tracheliodes mortuellus, sp. nov. 
 
 Black; length 7 mm. or somewhat more; abdomen petiolate; wings short; 
 metathorax coarsely striate or ridged ; upper posterior part of pleura finely striat- 
 ulate ; ocelli large, in a fairly high but not nearly equilateral triangle ; mandibles 
 stout, bent inwards apically (i. e. the outer edge becoming very convex), with the
 
 46 BULLETIN: MUSEUM OF COMPARATIVE ZOOLOGY. 
 
 cutting edge sinuate, but not distinctly bidentate ; venation nearly as Kohl figures 
 for Tracheliodes megerlei (Brachymerus megerlei Dahlb.), but having the stigma 
 longer and narrower ; the cells are practically the same ; the costal cell is almost 
 obsolete. Measurements in p : length of marginal cell, 705 ; its breadth, 225 ; its 
 breadth at the truncate end, 135 ; length of stigma, 420 ; its breadth, 105 ; length 
 of s. m., 760 ; length of first discoidal cell, also 750 ; length of second discoidal, 
 540; r. n. joining s. m. at middle ; t. m. a short distance basad of b. n. 
 
 Type. No. 2015, Mus. Cornp. Zool. Florissant, Col. (No. 3200, S. H. 
 Seudder Coll.). I use the name proposed by Morawitz (1866) for this genus, 
 because Brachymerus Dahlbom, 1845, though earlier, is a homonym. The 
 genus has not hitherto been recognized in America, and it may be that if all 
 the parts (e. g. of the mouth, etc.) of the extinct form could be examined, it 
 would be found genetically separable. At present, however, I can find no 
 grounds for separation. 
 
 PEMPHREDONIDAE. 
 Passaloecus scudderi, sp. nov. 
 
 Length 6 mm. ; black, with a large, broad (width about 2 mm.) head, globose 
 thorax, and narrow sessile abdomen ; breadth of thorax 2^, of abdomen about 1 
 mm., hind margins of abdominal segments broadly rather pale brown; ocelli 
 normal; anterior wing about 3^ mm. long, venation rather pale brown; stigma 
 rather large, about 135 /t broad ; marginal cell normal ; first s. m. about 780 /* 
 long; second s. m. 255 long and about 375 high, its sides parallel; b. n. strongly 
 curved ; first r. n. entering first s. m. about 135 ju from its end ; second r. n. joining 
 second s. m. very slightly beyond the middle ; apical corner of first discoidal 
 rather more elongated and pointed than usual; b. n. falling a little short of t. m.; 
 second discoidal oblique, slanting downwards apicad. 
 
 Type. No. 2016, Mus. Comp. Zool. Florissant, Col. (No. 8758, S. H. 
 Seudder Coll.). Closely allied to existing United States species. 
 
 PHILANTHIDAE. 
 Prophilanthus, gen. nov. 
 
 Large and robust, with a sessile abdomen ; stigma little developed ; marginal 
 cell narrowly but very obtusely rounded at apex, the apical point away from costa, 
 and quite without an appendix ; third submarginal cell very broad, and equally 
 broad above and below ; basal nervure joining subcostal a long way basad of 
 stigma. Compared with Philanthus albifrons Cresson, the fossil insect showed the 
 following differences : 
 
 (1) Portion of stigma in marginal cell shorter. 
 
 (2) Marginal cell bulging basally, i. e., in the direction of the first t. c. 
 
 (3) Marginal cell with apex rounded, the apical point not on costa. 
 
 (4) Third t. c. arched, with more or less of a double curve (a character of 
 Philoponus). 
 
 (5) Second s. m. broader, and receiving first r. n. more distinctly before middle.
 
 COCKERELL: FOSSIL HYMENOPTEKA FROM FLORISSANT. 47 
 
 (6) Third a. m. very much broader above, being equally broad (1050/*) above 
 and below. 
 
 (7) Prothorax, mesothorax, and metathorax longitudinally striate, especially 
 the protliorax. (Very faint striation of the protliorax is visible in P. albifrons.) 
 
 Other characters are : width of third discoidal at base somewhat less than 
 greatest width of first discoidal ; first t. c. not angulated at basal third ; second 
 discoidal cell more than twice as long as its width at apex ; cubital nervure not 
 bent, but slightly curved downwards, at end of first discoidal ; lower section of 
 b. n. (bordering first discoidal) about twice as long as upper. 
 
 Prophilanthus destructus, sp. nov. 
 
 Length about 20 mm., robust, with a thick sessile abdomen, which appears to 
 have had very broad black bands alternating with narrower yellow ones ; antennae 
 6 mm. long, ordinary, the scape thick, flagellum black ; length of anterior wing 
 12 mm. ; costa, up to base of marginal cell, broadly and very deeply infuscated, 
 apex also clouded; marginal cell 3 mm. long, only just surpassing apex of third 
 s. m. ; first s. m. 3 mm. long ; second and third submarginal cells combined, on 
 cubital nervure, 3 mm. long; first discoidal cell 4 mm. long, cr nearly; second 
 r. n. nearly straight, slightly bowed outwards ; second s. m. very broad below ; 
 origin of first t. c. to insertion of first r. n., 450 /a; insertion of first r. n. to origin 
 of second t. c., 976 /* ; origin of second t. c. to insertion of second r. n., less than 
 
 Type. No. 2017, Mas. Comp. Zool. Florissant, Col. (No. 7762, S. H. 
 Scudder Coll.). 
 
 NYSSONIDAE. 
 Hoplisidia, gen. nov. 
 
 Size rather large ; thorax hairy, the hairs long and quite simple ; abdomen sub- 
 cylindrical, broadest at the apex of second segment ; first segment comparatively 
 small, but not petiolate ; apex pointed; the form of the insect like Gorytes or 
 Hoplisus ; stigma very narrow, almost obsolete ; three submarginal cells, the first 
 at least as long as the other two combined, the b. n. joining subcostal very far 
 basad of the stigma ; second s. m. broad, receiving the recurrent nervures near the 
 end of the first and second thirds, the second r. n. curved and bent backwards to 
 its point of insertion ; third s. m. about twice as broad below as above ; marginal 
 cell narrowly rounded at apex ; first discoidal conspicuously longer than marginal ; 
 b. n. sharply bent at origin of cubital ; b. n. meeting t. m., or practically so; hind 
 wings with cubital nervure exactly meeting t. m., which goes downwards for a 
 short distance, and is then bent, finishing its course very obliquely. Among the 
 Gorytinae, this falls closest to Hoplisus by the venation of the hind wings. It is 
 peculiar for the very long first s. m., the b. n. meeting t. m., the reduced stigma, 
 and the long first discoidal, the whole combination seeming to exclude it from the 
 modern genera. The upper apical corner of the second discoidal is obtuse, as in 
 Gorytes mystaceus, not acute as in G. (Hoplisus) quadrifasciatus. The second t. c. 
 is much less oblique than in either of the species just cited, being very nearly 
 vertical, and not at all parallel with the third t. c.
 
 48 BULLETIN: MUSEUM OF COMPAKATIVE ZOOLOGY. 
 
 Hoplisidia kohliana, sp. nov. 
 
 Length (but in a somewhat disintegrated condition) about 20 mm.; anterior 
 wings about 11 mm., with a dusky cloud in the second s. m., and suffusedly 
 below; length of abdomen about 10 mm., its width 4; measurements in /t: 
 length of marginal cell about 2250 ; its breadth 525 ; length of first s. m. about 
 2475; length of second s. m. on marginal, about 600; on cubital nervure about 
 1050; of third s. m. on marginal 600, and on cubital 1140 ; insertion of first r. n. 
 from first t. c. 420; insertion of second r. n. from second t. c., 300 ; length of first 
 discoidal, 3300. 
 
 Type. No. 2018, Mus. Comp. Zool. Florissant, Col. (No. 742, S. H. 
 Scudder Coll.). Named after the author of the most useful work on the 
 genera of fossorial wasps. 
 
 Hoplisus sepultus, sp. nov. 
 
 Probable length about 10 mm. (head, much of thorax, and base of wings missing 
 in the type) ; abdomen sessile, apparently normal, curved downwards as though 
 in the act of stinging ; hind tibia with tarsus about 4| mm. ; wings with a dark 
 cloud in base of marginal cell, filling second s. m., and extending suffusedly below, 
 still showing brilliant iridescent colors, especially in second s. m. ; nervures more 
 slender than in Hoplisidia kohliana ; stigma large, its breadth about 170 (this and 
 all following measurements in /*) ; marginal cell about 1500 long and 300 broad, 
 pointed on costa; b. n. beginning at or very near base of stigma, and going a little 
 basad of t. m. ; length of first s. m. about 1275; second s. m. hexagonal, its length 
 on marginal 300, on oblique apex of first discoidal 170 to 270 (this variation in the 
 opposite wings of the same individual) ; first recurrent nervure to second (both 
 received by second s. m.) 300 ; second r. n. to origin of second t. c. 150 ; third s. m. 
 oblique, its greatest length (from upper basal to lower apical corners) 1125, its length 
 on marginal 460, its length on cubital nervure 750, its tip surpassing marginal cell 
 a little ; insertion of third t. c. to tip of marginal cell 570 ; length of first discoidal 
 about 2100 ; second r. n. very strongly bowed outwards. Hind wing reversed in 
 the specimen ; cubital nervure meeting t. m. ; distance from t. m. to t. c. 1800. 
 
 Type. No. 2019, Mus. Comp. Zool. Florissant, Col. (No. 980, S. H. 
 Scudder Coll.). 
 
 No. 2710, S. H. Scudder Coll., is a wing of Hoplisus sepultus, with a small 
 portion of the body. The measurements are in part greater than in the type 
 indicating perhaps the opposite sex or a larger individual, but evidently not 
 another species. Length of marginal cell about 1500, its breadth about 300; 
 length of first submarginal about 1350, the b. n. hardly going so near stigma 
 as in type ; second s. m. on marginal 240 ; distance between first and second 
 recurrent nervures at insertion 300; length of first r. n. 750 (same in type); 
 insertion of second r. n. to origin of second t. c. 150 ; greatest length of third 
 s. m. about 1200 ; third s. m. on marginal 570 ; length of third t. c. 795 ; 
 insertion of third t. c. to apex of marginal cell 600. Cloud in second s. m., 
 etc., as in type.
 
 COCKERELL: FOSSIL HYMENOPTERA FROM FLORISSANT. 49 
 
 SPHECIDAB. 
 Ammophila antiquella, sp. nov. 
 
 Head and thorax black ; abdomen all light ; hind legs apparently light, with the 
 tarsi black, in strong contrast ; form slender. Length 12 mm. ; abdomen 7-J- mm., 
 of which 3 mm. is petiole ; width of thorax between wings If mm., of head per- 
 haps a trifle less ; hind tibia 2f mm. ; scape rather stout, as usual in the genus, 
 metathorax transversely striated ; wings not preserved. Anteriorly to the trans- 
 versely striate area on thorax, some longitudinal striae can be seen. The petiole of 
 abdomen is two jointed, the first joint scarcely over one third the length of the sec- 
 ond ; the breadth of the apical part of the abdomen is H mm. 
 
 Type. No. 2020, Mus. Comp. Zool. Florissant, Col. (No. 5974, S. H. 
 Scudder Coll.). The specimen is poorly preserved, but as its relationships are 
 evident, it is described. 
 
 YESPOIDEA. 
 SCOLIIDAB. 
 
 I here use this family name in a rather broad sense, including the Myziniclae 
 and Tiphiidae of Ashinead. The two extinct genera here introduced are evi- 
 dently related to the Tiphiid series, though not without features suggestive of 
 the other groups. So far as I am able to judge, their affinity is closest with 
 the rare and apparently primitive genus Engycystis Fox, found in Texas and 
 Lower California. Curiously, however, a new genus from Australia is also re- 
 lated, and for purposes of comparison is herewith described. The following 
 table separates the four genera from each other : 
 
 Basal nervure entering subcostal at a distance from stigma much greater than 
 length of stigma Awtrotiphia, gen. nov. 
 
 Basal nervure entering subcostal at a distance from stigma less than length 
 of stigma 1 
 
 1. T. m. strongly oblique ; stigma very narrow, not nearly filling the large 
 
 stigmatic cell ; marginal cell not surpassing third s. m 
 
 Lithotiphia, gen. nov. 
 T. m. slightly oblique 2 
 
 2. Marginal cell surpassing third 8. m., its apex rounded ; costal cell large. 
 
 Geotiphia, gen. nov. 
 
 Third s. m. surpassing marginal cell ; apex of marginal pointed, on costa ; 
 costal cell small or rudimentary Engycystis Fox. 
 
 In Engycystis, the ventral constriction between the first and second abdom- 
 inal segments is not nearly so marked as in Tiphia and Paratiphia, but in Aus- 
 trotiphia the constriction is still less evident, being hardly appreciable. One 
 would almost hesitate to place the latter genus in the Scoliids, were it not so 
 obviously a Tiphiid in every other feature. The otherwise different Australian 
 VOL. L. No. 2 4
 
 50 BULLETIN: MUSEUM OF COMPARATIVE ZOOLOGY. 
 
 genus Dimorphoptera Smith, appears to share the same character. Unfortu- 
 nately this character cannot be determined in the fossil genera, owing to the 
 position of the specimens. 
 
 ? Having exactly the appearance of a Tiphia, but related to Engycystis, from 
 which it differs thus : basal nervure joining subcostal much more remote from 
 stigma ; first s. m. long, broken by a false vein which passes from near the origin of 
 the first t. c. to near the base of the stigma (the same is found in the Myzinid Ple- 
 sia) ; marginal cell broadly rounded, almost truncate, at apex ; second s. m. 
 extremely broad below, the first and second s. ms. exceedingly oblique ; first dis- 
 coidal cell at base narrower than first submarginal ; t. m. very oblique (a character 
 of the fossil Lithotiphia) ; hind wings with t. c. oblique (its upper end more basad), 
 and cubital nervure ending about as far basad of upper end of t. m., as half the 
 length of the latter. Stigma well-developed. 
 
 Austrotiphia Mrbyi, sp. nov. 9 . Length about 13 mm., entirely black, looking 
 like an ordinary Tiphia ; eyes and mandibles as in Tiphia, simple ; punctures nearly 
 as usual in Tiphia ; hind margin of prothorax straight, or rather gently concave, 
 with no median lobe; anterior part of mesothorax smooth ; scutellum shining, with 
 very sparse and small punctures ; parapsidal grooves very strong ; tegulae small ; 
 abdomen with small punctures, closer and much smaller and more regular on basal 
 part of segments ; apical ventral plate not greatly surpassing dorsal ; legs much as 
 in Tiphia, but hind femora much broadened, sharply keeled below ; hind tibiae 
 short, with five or six rather broad teeth on outer edge ; basal joint of hind tarsus 
 tuberculate on outer side, not spined ; middle and hind tibiae each with two white 
 spurs; last joint of hind tarsi normal. Shoalhaven, Australia (W. W. Froggatt, 
 186). Captured in 1895, and now the property of the British Museum. Named 
 after Mr. W. F. Kirby, in recognition of his work on Scoliidae. 
 
 In Geotiphia the teeth on the outer edge of the hind tibia are very broad, not 
 spine-like as they are in Tiphia and Engycystis. In Austrotiphia they are 
 comparatively broad and short, and the last one, in particular, recalls that of 
 Geotiphia. In Tiphia the suture between the first two abdominal segments is 
 evidently depressed at the sides, the abdomen being viewed from above ; this 
 is not the case in Austrotiphia. In this particular, so far as can be seen, Geo- 
 tiphia and Lithotiphia resemble Austrotiphia. Geotiphia has some appearance of 
 having had emarginate eyes, a character of the true Scoliids, but it is impossi- 
 ble to be sure about it. The spotted abdomen is also suggestive of the Scoliids, 
 but not so the venation and the large stigma. When one uses the compound 
 microscope to examine the eyes, the appearance of einargination disappears, and 
 so far as can be seen, they look normal for the Tiphiidae. On the hind leg of the 
 ?, the tibial spurs are very short in Engycystis, very much less than half the 
 length of the first tarsal joint ; in Austrotiphia these spurs are very long, the long- 
 est (the hind one) being fully three quarters the length of the first tarsal joint ; 
 in the two fossil genera their character has not been determined. Tiphia has 
 them long, like Austrotiphia. The second antonnal joint in Austrotiphia is con- 
 spicuously smaller than in Tiphia. In Geotiphia the middle joints of flagellum 
 are broader than long ; in Austrotiphia ( 9 ) they are about as long as broad ;
 
 COCKEKELL: FOSSIL HYMENOPTERA FROM FLORISSANT. 51 
 
 in Tiphia ( 9 ) they are conspicuously longer than broad. In Geotiphia and 
 Tiphia the claws are bifid; in Austrotiphia they appear at first sight to be 
 simple, because the inner tooth is flattened, shortened, and directed somewhat 
 inwards. In Tiphia the middle coxae are very widely separated by a bilobed 
 projection of the mesosternum ; in Austrotiphia they are considerably closer 
 together, the bilobed projection, although present, being much smaller. In 
 Engycystis they are more aa in Austrotiphia. In Tiphia the middle tibiae 
 have only one spur, in Austrotiphia and Engycystis (as in the Myzinids) 
 there are two. Tiphia has an open marginal cell and only two submarginals ; 
 Austrotiphia, Eugycystis, Geotiphia, and Lithotiphia have a closed marginal 
 and three submarginals. 
 
 Geotiphia, gen. nov. foxiana, sp. nov. 
 
 Length about 11 mm. ; black, with light markings on abdomen ; femora black, 
 tibiae and tarsi light, probably red in life ; the abdominal markings, presumably 
 yellow in life, consist of a broad transverse spot or patch on the first segment, 
 rounded at sides and deeply emarginate posteriorly, a couple of transversely oval 
 spots on second segment, and a pair of smaller and rounder ones on third ; first 
 segment broad, broadly rounded in front ; widtli of abdomen about 3 mm. ; head 
 round, width about 2 mm. ; middle joints of flagellum about 180 ft long and 
 225 broad; middle tibia apparently short, broad (breadth about 300 /*), abruptly 
 truncate ; middle tarsi slender, first joint about 900 /* long, its outer edge straight 
 on first half and convex on second, third and fourth joints each about 225 /t 
 long, and quite slender; claw joint (excluding claws) about 300 fi long; claws 
 bifid, the two teeth about equally long ; hind femora stout but not at all subglobose, 
 about 1500 fi long ; hind tibia about as long, about 600 /* broad ; basal joint of hind 
 tarsus about 250 Abroad. Outer apical edge of hind tibiae with very broad teeth, 
 witli points directed apicad, and long straight or nearly straight upper edges. Wings 
 with a large stigma (solid, filling stigmal cell), its breadth about 300 (this and all 
 following measurements in fj.) ; costal cell very distinct, the costal and subcostal 
 nervures very heavy, and quite wide apart; marginal cell complete, broadly 
 rounded at end, the actual apex not on costa, length of the cell 1500 /*, its tip sur- 
 passing apical point of third a. m. by about 100, although the distance from inser- 
 tion of third t. c. to apex of marginal is about 450 ; three submarginal cells, the first 
 1200 long ; second very broad, 450 long on marginal, and ^00 on cubital nervure, 
 receiving the first r. n. exactly at the middle ; third s. m. 600 long on marginal, its 
 outer side strongly bulging ; first (upper) section of b. n. less than 300 long, bulg- 
 ing at its lower end, just before the origin of cubital nervure; second (lower) sec- 
 tion slightly over 300 long, meeting t. m., which is a little oblique ; first discoidal 
 1500 long ; second r. n. joining third a. m. not far from the middle. 
 
 Type. No. 2021, Mus. Comp. Zool. Florissant, Col. (No. 14,292. S. H. 
 Scudder Coll.). Named after Mr. W. J. Fox, in recognition of his work on 
 Engycystis.
 
 52 BULLETIN: MUSEUM OF COMPARATIVE ZOOLOGY. 
 
 Lithotiphia, gen. nov. scudderi, sp. DOT. 
 
 Length about 12J mm., anterior wing about 8J ; black, the abdomen without 
 light spots ; head round, its width 2 mm. ; width of thorax (doubtless increased by 
 crushing) about 4 mm. ; abdomen about 6f mm. long and 3 broad; hind femora 
 cylindrical, rather stout, with a little concavity followed by a prominence at apex 
 beneath, as in other T iphiids ; hind tibiae greatly swollen, apparently not dentate ; 
 hind tarsi very slender. Wings with apparently a very large stigma, but the 
 microscope shows that this is the stigmal cell, not nearly filled by the long and 
 slender true stigma, which is brown, as long as stigmal cell, but only about 105 
 (this and all following measurements in /*) broad ; the stigmal cell is about 675 long, 
 and 255 broad, broadly truncate posteriorly, and with a rudimentary cross-nervure 
 before the middle, almost meeting the radial nervure (in a modern Scolia I can 
 detect such a cross-nervure, but still more rudimentary) ; marginal cell entire, 
 exceedingly broadly rounded apically, the actual tip not on costa, length of the 
 cell 1650; three submarginals, the second very broad, 750 long on marginal, 
 receiving the first r. n. nearly 600 from its beginning, and 450 from its end; third 
 s. m. broadly bulging apically, not surpassed by the marginal; length of third 
 s. m. on marginal 750 (same as second s. m.) ; second r. n. joining third s. m. 375 
 from its base, and 450 from its lower apical corner ; the second r. n. joins the cubi- 
 tal nervure in such a way that the outer angle formed is less than a right angle. 
 B. n. going very slightly basad of t. m., which is strongly oblique. 
 
 Type. No. 2022, MUB. Comp. Zool. Florissant, Col. (No. 2440, S. H. 
 Scudder Coll.). 
 
 POMPILIDAB. 
 Hemipogonius florissantensis, sp. nov. 
 
 Length nearly 15 mm., anterior wing about 10 mm. ; anterior wings with a 
 transverse dark cloud or suffused band at about the end of the basal third, a very 
 large dark roundish patch in and below the marginal cell, and the tip dusky, 
 a round area between the dusky tip and the large dark region appearing white. 
 General structure of body normal, the abdomen sessile, with the first segment, seen 
 in lateral profile, ascending and convex ; spurs large; antennae about 7 mm. long, 
 one curled under body reaching middle coxae, apparently not curled, or little 
 curled, at tip ; hind coxae long, hind femora about 3 mm. 
 
 Length of stigma 900 (this and the following measurements in /u), its breadth 
 about 285 ; costal cell distinct ; length of marginal cell about 2400, long and nar- 
 row, its breadth about 525, its apex pointed and on costa ; distance from insertion 
 of third t. c. to tip of marginal, 765 ; length of first s. m. nearly 2250 ; first t. c. 
 bowed inwards (basad), its length about 625 ; stigma to insertion of first t. c., 270 ; 
 second s. m. on marginal, 750 ; third s. m. on marginal, 795 ; outer angle formed by 
 insertion of third t. c. on marginal less than a right angle ; distance from insertion 
 of second r. n. to lower apical corner of third s. m., 825 (the second r. n. enters 
 third s. m. toward the base) ; b. n. very far basad of stigma, and its origin 450 
 basad of t. m., its lower section about 600 long, the upper one (bounding first s. m.) 
 considerably less ; lower edge of second discoidal 1275 long, its breadth at apex
 
 COCKERELL: FOSSIL HYMENOPTERA FROM FLORISSANT. 53 
 
 800, at base 300. Hind wings with cubital nerrure inserted a short distance 
 beyond (apicad of) t. m. 
 
 Type. No. 2023, Mus. Comp. Zool. Florissant, Col. (No. 8647, S. H. 
 
 Scudder Coll.). Easily known from H. scudderi by its larger size. The spec- 
 imen shows several venational characters which were not preserved in H. scud- 
 deri, and they confirm the generic reference. 
 
 Hemipogonius scudderi, sp. nor. 
 
 Slender, length about 10 ram. ; antennae 6} mm., the scape thickened ; width of 
 head 2 mm.; length of anterior wing about 6 ; length of thorax almost 3; of 
 hiiid tibia and tarsus about 7 ; hind spurs large. Wings hairy ; stigma distinct ; 
 marginal cell sharply pointed on costa, its length and that of first s. m. the same, 
 1575 n ; greatest width of marginal cell only 405 p. ; second s. m. pentagonal, broad, 
 receiving first r. n. a little beyond its middle; length of second s. m. on marginal 
 450 /* ; third s. m. larger than second, shaped as usual ; b. n. about 120 /* basad of 
 t. m. ; t. m. 255 /j. long, not at all oblique. Stigma dark, and a dark cloud in the region 
 of b. n. and below ; also a diffused brown cloud occupying marginal cell, the 
 second and third submarginals, and the third discoidal; this region still shows 
 bright iridescent colors. Somewhat allied to the living H. alienatus (Smith), but 
 larger. The wings are shorter, and much more strongly clouded, than in H.fortis 
 (Cresson) The sutures of the antennal joints are black. 
 
 Type. No. 2024, Mns. Comp. Zool. Florissant, Col. (No. 8640, S. H. 
 Scudder Coll.). A beautifully preserved specimen. No. 10,813, S. H. Scudder 
 Coll., is the reverse of the same example. 
 
 Ceropalites, gen. nor. 
 
 Abdomen very convex, the first point narrowed to a distinct petiole ; stigma very 
 well-developed, elongate, lanceolate ; subcostal nervure quite widely separated 
 from costa ; first discoidal cell very narrow, the part of the basal nervure bounding 
 it being less than half as long as the part bounding first submarginal ; b. n. passing 
 only just basad of t. m. ; marginal cell large and elongate, probably pointed; 
 antennae long. 
 
 Ceropalites infelix, sp. nov. 
 
 Length 13 mm. ; as preserved, entirely light reddish-brown, probably red in life ; 
 wings hyaline, with the apical margin broadly dusky ; stigma dark, with a brown- 
 ish spot immediately below it ; length of anterior wing 10 mm. ; of abdomen 
 7; height of abdomen (dorso-ventral) 3; length of thorax anterior to wings 1} 
 mm. ; length of marginal cell over 4 mm. (the apex gone) ; of wing anterior to 
 stigma about 5$ mm. ; length of stigma about 1500 /t; its breadth about 375; first 
 section of radial nervure, passing almost straight down from stigma to junction of 
 first t. c., about 450 p long; width of marginal cell 1095 /* ; from first to second 
 t. c. on radial nervure about 1650 /j.; beginning of b. n. from stigma, 600 n; dis- 
 tance of subcostal nervure from costa at this point, 300; first (upper) part of b. n. 
 about 1110 /x, second (lower) part 390; hind part of metathorax with some trans- 
 verse keels.
 
 54 BULLETIN: MUSEUM OF COMPARATIVE ZOOLOGY. 
 
 Type. No. 2025, Mus. Comp. Zool. Florissant, Col. (No. 6013, S. H. 
 Scudder Coll.). The first abdominal segment having a distinct though short 
 (less than 1 mm. long) petiole is suggestive of Sphecidae, but the insect does 
 not otherwise agree with that group. The first abdominal segment really 
 recalls the winged Mutillidae allied to Photopsis, as much as anything ; but 
 the ventral surface of the abdomen is perfectly straight (or rather, gently con- 
 vex), without the least sign of a depression between the first and second seg- 
 ments. The well-developed stigma is suggestive of Ceropales, but the venation 
 differs from that of any modern genus known to me. I cannot see the third 
 submarginal cell distinctly, but it appears to have been present. The dark 
 spot below the stigma is still slightly iridescent. 
 
 VESPIDAE. 
 Palaeovespa, gen. nov. 
 
 With the general form of Vespa, the thorax broadly rounded, and the abdomen 
 sessile and broad at base ; the first segment of transverse form, yet by no means 
 so broad as in true Vespa. Venation more like Polistes, the marginal cell being 
 pointed, the apex of first discoidal oblique, and the recurrent nervures joining the 
 second s. m. far apart, not both entering the basal half of the cell, as in Vespa. 
 The b. n. joins the subcostal nervure nearer to the stigma than is usual in Vespa, 
 but not at its base, as in Polistes. It is impossible to see whether the hind wings 
 have an anal lobe or not. This is a very interesting genus, having the appearance 
 rather of Vespa (it would never occur to any one to refer the specimens to Polistes), 
 but retaining the apparently more primitive venation of Polistes, or a close approx- 
 imation to it. P.florissantia, the largest species, is taken as the type, but the char- 
 acters of the genus are not all ascertainable from the single specimen of that insect. 
 
 Palaeovespa florissantia, sp. nov. 
 
 Very large and robust ; length to beyond middle of fifth abdominal segment, 
 25 mm. ; length from base of abdomen to apex of fourth segment, 14 mm. ; thorax 
 narrow for the size of the insect, its width between wings about 7 mm. ; breadth of 
 abdominal segments in mm. (1) 6, (2) 8|, (3) 8|; color dark, evidently black in 
 life, with the hind margins of the abdominal segments broadly but suffusedly pal- 
 lid ; no distinct abdominal markings ; wings apparently reddish. The venation is 
 obscure, but the wings appear to be folded, and the very long first discoidal cell 
 of the Vespidae is plainly visible, its length about 10 mm., while its breadth is only 
 about 1-J ; the lower part of the basal nervure is about 6 mm. long, and the first 
 s. m. on cubital nervure is about 3J ; the second s. m., very faintly indicated, 
 appears triangular, the first t. c. oblique, its upper end most distad, the acute 
 angles formed being of about 45 ; the apex of the first discoidal, between the first 
 t. c. and the insertion of the first r. n., is obliquely truncate, considerably more 
 obliquely than in a modern Vespa examined. 
 
 Type. No. 2026, Mus. Comp. Zool. Florissant, Col. (No. 11,741, S. H. 
 Scudder Coll.). This is the largest by far of the Florissant Hymenoptera seen
 
 COCKERELL: FOSSIL HYMENOPTERA FROM FLORISSANT. 55 
 
 by me. It gives one the impression, at first, of a krge Scoliid, but it ia 
 unquestionably a member of the Vespidae. 
 
 Palaeovespa scudderi, sp. nov. 
 
 Length of anterior wing about 13 mm.; of head, including mandibles, 6; of 
 thorax, 8 ; of middle femur and trochanter 4i ; of middle tibia and tarsus, 7 ; of 
 first discoidal cell, 7 ; of marginal cell, which ends in a sharp point on costa, 3} ; 
 eyes deeply emarginate, as usual in Vespa, but contrary to what obtains in the 
 modern species, the part of the eye above the emargination is almost if not quite 
 as large as that below it ; mandibles shaped as usual in the genus ; the large lateral 
 lobes of prothorax are strongly vertically striate, the striation resembling that 
 found in the same region in species of Myzine and Ammophila ; pleura without 
 such striation ; head and thorax, dark, doubtless black in life ; the middle leg seems 
 to have been black as far as the beginning of the apical third of the femur, or there- 
 abouts, and beyond that yellow or red ; apex of first discoidal cell about as in 
 modern Vespa, but narrow ; marginal cell much more pointed than in the modern 
 forms, but venation otherwise normal ; abdomen missing. Lateral ocellus about 
 270 n broad, and 300 from eye ; width of marginal cell about 900 /x ; of oblique 
 nervure terminating first discoidal, 225 /*. 
 
 Type. No. 2027, Mus. Comp. Zodl. Florissant, Col. (No. 9065, S. H. 
 Scudder Coll.). 
 
 No. 7738, S. H. Scudder Coll., badly preserved, appears to be a second 
 example of P. scudderi, as it shows well the striation of the prothoracic lobes ; 
 a feature which is not to be seen in any of the specimens of P. gillettei, though it 
 may not really be absent. This specimen has the abdomen, and indicates that 
 P. scudderi was about 18 mm. long. The hind margins of the last two abdom- 
 inal segments were broadly light (probably also the two before these), and the 
 light color (no doubt yellow in life) sent a rounded lobe upwards on each side 
 of the last segment, these markings being of the same type as in modern Vespa. 
 The antennate are normal. 
 
 Palaeovespa gillettei, sp. nov. 
 
 Length about 14| mm. ; of anterior wing about 10 mm., with the nervures more 
 delicate than those of P. scudderi; breadth between wings slightly over 4 mm., of 
 abdomen 5 ; black, with indications on the mesothorax of what appear to have 
 been two longitudinal yellow stripes ; venation as in P. scudderi, with the same 
 sharply pointed marginal cell ; length of first discoidal a little over 5 mm. ; widtli 
 of marginal cell 600 jt. The abdomen is not so broad basally as in modern Vespa ; 
 it is nearly parallel-sided, with the broadest part beyond the middle. The species 
 is allied to P. scudderi, but smaller in every way, with more delicate venation. 
 
 Type. No. 2028, Mus. Comp. Zool. Florissant, Col. (No. 16,325, S. H. 
 Scndder Coll.). No. 2029, M. C. Z. (No. 11,920, S. H. Scudder Coll.), No. 
 5986, S. H. Scudder Coll., and No. 2030, M. C. Z. (No. 14,305, S. H. 
 Scudder Coll.,) are also Palaeovespa, and presumably the present species, but 
 they do not show the venation so well. The first two show very distinctly
 
 56 BULLETIN: MUSEUM OF COMPARATIVE ZOOLOGY. 
 
 two light lines or narrow bars on mesotborax, not reaching the anterior or 
 posterior margins, and 5986 also shows a narrow light anterolateral margin, 
 probably really on the prothorax, as is common in living forms. The antennae 
 appear to be as in Vespa. Nos. 2031, 2032, M. C. Z. (Nos. 18,382 and 7868, 
 S. H. Scudder Coll.), are two isolated anterior wings of Palaeovespa. They 
 exhibit a good deal of diiference in small details, but are, I think, certainly 
 referable to P. gillettei. From them it is possible to ascertain several characters 
 not clearly discernible in the type. Both show a dark cloud in the apical 
 part of the costal cell, such as occurs in modern Vespa. The junction of b. n. 
 to subcostal, which in the type is some 450 /i from base of stigma, is only about 
 300 from it in No. 18,362, and 345 in No. 8981, but it is difficult to say exactly 
 where the basal n. leaves off, and where the stigma begins, the fusion being 
 gradual. The size is throughout too small for P. scudderi. The following 
 measurements are in p. : 
 
 Length of second 8. m. Length of third s. m. 
 
 on marginal. on marginal. 
 
 P. scudderi, Type No. 2027, M. C. Z. 450 825 
 
 No. 2032, M. C. Z. 345 675 
 
 No. 2031, M. C. Z. 225 675 
 
 First r. n. from First r. n. (on Second r. n. 
 beginning of cubital n.) from end of 
 
 second s. m. from second r. n. second s. m. 
 
 P. scudderi, Type No. 2027, M. C .Z. 225 665 1 375 ? 
 
 P. gillettei, Type No. 2028, M. C. Z. ? ? 225 ? 
 
 No. 2032, M. C. Z. 150 535 375 
 
 No. 2031, M. C. Z. 150 375 300 
 
 Vespa (modern) 195 240 750 
 
 P. gillettei is named after Professor C. P. Gillette, in recognition of his 
 work on the entomology of Colorado. 
 
 EUMBNIDAB. 
 Odynerus palaeophilus, sp. nov. 
 
 $ Rather slender, length 9 mm., anterior wing 8 mm. ; black, the wings dusky ; 
 first abdominal segment in lateral profile (i. e. seen from the side) presenting a 
 curve which is uniform, not abruptly bent at any point, and is equal to about 
 a quarter of a circle; abdomen broad and convex, with the apical part separated, 
 doubtless originally marked off by a suture, as in some living forms ; marginal 
 cell very broad, in the form of an elongated triangle, the apex downwards; apex 
 of first discoidal only moderately oblique (much less so than in a modern species 
 compared) ; second submarginal cell narrowed almost to a point above, its length 
 on marginal being only 60 fi, while its length on cubital nervure is 655 n; width of 
 marginal cell, 600 /i ; width of third s. m. on marginal, 525 /*. It would be easy to 
 misinterpret the venation of this insect (as also of the species of Palaeovespa), 
 owing to the folding; but it is easily understood when compared with modern
 
 COCKERELL: FOSSIL HYMENOPTERA FROM FLORISSANT, 57 
 
 examples similarly folded. This insect has the closest possible resemblance to a 
 species still living in Colorado, but it differs in the venation in two respects : (1) 
 the second submarginal cell is more contracted above, (2) the cubital nervure is 
 abruptly bent at the end or the first discoidal cell, as in Vespa, whereas in the 
 modern species it is straight. 
 
 Type. No. 2033, Mus. Comp. Zobl. Florissant, Col. (No. 10,657, S. H. 
 Scudder Coll.) 
 
 Odynerus praesepultus, sp. nov. 
 
 ? Black, apparently with two light longitudinal bars on mesothorax ; length 
 nearly 11 mm., head and thorax about 4 mm., anterior wing 7 mm. ; wings folded, 
 somewhat reddish; flagellum thick, (the end tapering, not clavate,) dark above, 
 light below ; abdomen sessile, second segment not swollen, dorsally or ventrally ; 
 in lateral profile, the dorsum of abdomen is gently curved, the venter nearly 
 straight, no segment markedly different from the one before it. Stigma large, its 
 width (short diameter) about 200 (this and other measurements in p) ; marginal 
 cell subtriangular, 1350 long, about 450 broad, narrowly obliquely truncate, the 
 truncation about 150 broad ; tip of marginal cell about level with apex of third 
 s. m. ; b. n. inserted at base of stigma, its upper section about 450 long; first s. m. 
 1425 long ; stigma to insertion of first t. c., 450 ; second s. m. much narrowed above 
 (150 long on marginal), and receiving both recurrent nervures; first r. n. from 
 origin of first t. c., 250 (lower basal corner of second s. m. very acute) ; distance 
 between insertion of first and second r. n., 300 ; cubital nervure not at all bent at 
 end of first discoidal ; insertion of second r. n. to origin of second t. c., 105 ; lower 
 margin of third s. m., 600; third s. m. on marginal, 450; insertion of third t. c. to 
 apex of marginal cell, 450. 
 
 Type. No. 2034, Mus., Comp. Zool. Florissant, Col. (No. 11,944, S. H. 
 Scudder Coll.). This is readily known from Palaeovespa by (1) marginal cell 
 obliquely truncate at end, the tip not on costa ; (2) cubital nervure not at all 
 bent at end of first discoidal, (3) b. n. originating at base of the very large 
 stigma. It appears to be one of the. Eumenidae, the venation agreeing with 
 that group better than with the Vespidae. Among the Eumenidae, from the 
 venation and structure of the abdomen , it can go only in Odynerini, and it is 
 referred to Odynerus in the old, broad sense. The modern genera of Odynerini 
 are separated mainly on characters which are not discernible in the fossil. 
 
 ICHNEUMONOIDEA. 
 
 STEPHANIDAE. 
 Protostephanus, gen. nov. 
 
 Head rounded or subquadrate, rugose or tuberculate; prothorax broad but 
 produced, with a median longitudinal groove, and fine lateral oblique striae ; 
 abdomen sessile ; hind coxae elongated, about f the length of their femora; hind 
 femora moderately stout, not toothed ; stigma rather large ; costal cell very dis- 
 tinct ; terminal part of subcostal nervure, for a distance nearly equal to the length
 
 58 BULLETIN: MUSEUM OF COMPARATIVE ZOOLOGY. 
 
 of the stigma, much thickened and appearing black ; marginal cell long and quite 
 narrow ; b. n. strongly bent at beginning of cubital nervure ; t. m. opposite b. n., 
 the latter very slightly more basad ; only one a. m., which is considerably broader 
 than the first discoidal ; first r. n. meeting first t. c., which is continued in a straight 
 line with it, making, with the cubital nervure, a large X. This interesting genus 
 differs from those hitherto known by the combination of an elongated prothorax, 
 unarmed hind femora, and sessile abdomen. 
 
 Protostephanus ashmeadi, sp. nov. 
 
 Length about 9$ mm. ; anterior wings clear, with brassy iridescence still show- 
 ing, their length about 6 mm. ; hind coxae transversely striate ; hind tibiae 
 claviform, swollen apically, the hind tibiae and tarsi about as long as hind coxae 
 and femora, but the tibiae somewhat longer than the femora ; pleura finely striate ; 
 measurements in /*: width of head about 1500; length of thorax anterior to 
 wings about 1200 ; length of hind coxae about 1200 ; width of hind tibiae at apex 
 about 450 ; breadth of hind femora about the same ; extension of abdomen beyond 
 apex of hind femora perhaps 1500; breadth of stigma about 180; breadth of 
 marginal cell about 300 ; length of first (and only) s. m. 975 ; length of first dis- 
 coidal, 900 ; length of the quadrate second discoidal about 795 ; length of t. m. not 
 quite 300 ; length of second section of b. n. (bounding first discoidal) about 450. 
 
 Type, No. 2035, Mus. Corap. Zool. Florissant, Col. (No. 13913, S. H. 
 Scudder Coll.). Named after Mr. W. H. Ashmead, whose writings were most 
 useful in determining the affinities of the insect. 
 
 
 
 
 $13
 
 DATE DUE
 
 
 
 
 
 PUBLICATIONS 
 
 OK TIIK 
 
 MUSEUM OF COMPARATIVE ZOOLOGY 
 
 AT HARVARD COLLIDE. 
 
 There have been published of the BULLETIN Vols. I. to XLII., 
 and also Vols. XLIV. to XLVII ; of the MKMOIKS; Vols. I. to 
 XXIV., and also Vols. XXVIII., XXIX., XXXI. to XXXIII. 
 
 Vols. XLIII.. XLVIII., XLIX.. and L. of the BULLETIN, and 
 Vols. XXV., XXVI., XXVII. , XXX., XXXIV., and XXXV. of 
 the MKMOIKS, are now in course of publication. 
 
 The BULLETIN and MEMOIRS are devoted to the publication of 
 original work by the Professors and Assistants of the Museum, 
 of investigations carried on by students and others in the different 
 Laboratories of Natural History, and of work by specialists based 
 upon the Museum Collections and Explorations. 
 
 The following publications are in preparation : 
 
 Reports on the Results of Dredging Operations from 1877 to 1880, in charge of 
 Alexander Agassiz, by the U. S. Coast Survey Steamer " Blake," Lieut. 
 Commander 0. D. Sigsbee, U. S. X., and Commander.!. R. Bartlett, U. S. N., 
 Commanding. 
 
 Reports on the Results of the Expedition of 1S91 of the U. S. Fish Commission 
 Steamer " Albatross," Lieut. Commander Z. L. Tanner, U. S. X., Command- 
 ing, in charge of Alexander Agassiz. 
 
 Reports on the Scientific Results of the Expedition to the Tropical Pacific, in 
 charge of Alexander Agassiz, on the U. S. Fish Commission Steamer 
 "Albatross," from August, 1899, to March, 1900, Commander Jefferson F. 
 Moser, U. S. X., Commanding. 
 
 Reports on the Scientific Results of the Expedition to the Eastern Pacific, in 
 charge of Alexander Agassiz, on the U. S. Fish Commission Steamer 
 "Albatross," from October, 1904, to April, 1905, Lieut. Commander L. M. 
 Garrett. U. S. X., Commanding. 
 
 Contributions from the Zoological Laboratory, Professor E. L. Mark, Director. 
 
 Contributions from the Geological Laboratory. 
 
 These publications are issued in numbers at irregular intervals ; 
 one volume of the Bulletin (Xvo) and half a volume of the Memoirs 
 t-lto) usually appear annually. Each number of the Bulletin and 
 of the Memoirs is sold separately. A price list of the publications 
 of the Museum will be sent on application to the Librarian of the 
 Museum of Comparative Zoology, Cambridge, Mass.