'- New York State Education Department NEW YORK STATE MUSEUM SSth^ANNUAL REPORT \>90,'' TRANSMITTED TO THE LEGISLATURE FEBRUARY i^^^, 1905 ALBANY NEW YORK STATE EDUCATION DEPARTMENT 1906 STATE OF NEW YORK EDUCATION DEPARTMENT Regents of the University I 904 With years when terms expire 1913 Whitelaw Reid M.A. LL.D. Chancellor . New York 1906 St Clair McKelway M.A. L.H.D. LL.D. D.C.L. Vice Chancellor . . . ./ Brooklyn 1908 Daniel Beach Ph.D. LL.D. .' Watkins 1914 Pliny T. Sexton LL.B. LL.D Palmyra 191 2 T. Guilford Smith M.A. C.E. LL.D. . . . Buffalo 1907 William Nottingham M.A. Ph.D. LL.D. . . Syracuse 1910 Charles A. Gardiner Ph.D. L.H.D. LL.D. D.C.L. New York 1 91 5 Charles S. Franc^ B. S Troy 191 1 Edward Lauterbach M.A. LL.D New York 1909 Eugene A. Philbin LL.B. LL.D New York 1916 Lucian L. Shedden VjLjB Plattsburg Commissioner of Education Andrew S. Draper LL.D. ^sistant CommisslonerB Howard J. Rogers M.A. LL.D. First Assistant Commissioner Edward J. Goodwin Lit.D. h.H .D . Second Assistant Commissioner Augustus S. Downing M.A. Third Assistant Commissioner Secretary to the Commissioner Harlan H. Horner B.A^ Director of State Library Melvil Dewey LL.D.' Director of Science and State Museum John M. Clarke Ph.D. LL.D. Chiefs of Divisions Accounts, William Mason Attendance, James D. Sullivan \ Examinations, Charles F. Wheelock B.S. LL.D. Inspections, Fpank H. Wood M.A. Law, Thomas E. Finegan M.A. Records, Charles E. Fitch L.H.D. Statistics, Hiram C. Case Visual Instruction, DeLancey M. Ellis State of New York No. 12. / IN SKNATH, February 15, 1905. 58th ANNUAL REPORT OF THE NEW YORK STATE MUSEUM To the Legislature of the l^tate of New Yoi\ We bav<3' the honor to submit, pur.siiaut to law, the ^th annual report 0/ the New York State Museum. WiiiTELAw Reid GhanceUor of the Universil A. S. Drai'er Commissioner- of Education \ Appendix 7 Entomology 23, 24 Museum bulletins 86, 97 23 May Flies and Midges of New York 24 20th Report of the State Entomologist 1904 Published monthly by the New York State Education Department Bui.LEriN 343 New York State Museum John M. Clarke Director Ephraim Porter Felt State Entomologist Bulletin 86 ENTOMOLOGY 23 MAY FLIES AND MIDGES OF NEW YORK THIRD REPORT ON AQUATIC INSECTS A study conducted at the entomologic Held station, Ithaca N. Y. under the direction of EPHRAIM PORTER FELT D.Sc. JAMES G. NEEDHAM Ph.D. Professor of biology, Lake Forest College KENNETH J. MORTON F.E.S.L. Edinburgh, Scotland O. A. JOHANNSEN M.S. Instructor in civil engineering, Cornell University PAGE Preface 4 Introduction 7 Summer Food of the Bullfrog at Saranac Inn. J. G. Needh.^m g Ephemeridae. J. G. Needham. . 17 North American Hydroptilidae. K.J, Morton 63 P.\GE Aquatic Nematocerous Diptera II. O. A. JOHANNSEN 76 Explanation to the plates 316 Legenda to text figures 331 Plates 1-37 face 332 Index ZT,z state Museum, Albany N. Y. Oct. 17, 1904 Hon. Andrew S. Draper Commissioner of Education, Capitol iSiR : I beg to transmit herewith, for publication as a bulletin of this division, a third report on aquatic insects, entitled Map Flies and Midges of jSfeio York by Dr J. G. Needham, Special Assistant to the State Entomologist. Very respectfully John M. Clarke Director State of New York Education Department Coinimissioxer's Room Approved for puMication Oct. 24, 1904 Commissioner of Education PREFACE This, the third report upon work begun in 1900, like its predecessors, marks an important advance in knowledge. The first report. State Museum Bulletin 47, consisting of 230 pages and 36 plates, gave the life histories of about one hundred aquatic forms and characterized ten species and two new genera. The most important portion of- this work was the monographic ac- count of the larger dragon flies (Odonata Anisoptera). There were also valuable additions to our knowledge of the stone flies (Pie copter a) and the May flies (E ph emer i d ae), and the admirable account of the Caddis flies (T r i c h o rp t e r a) , by Mr Betten, deserves special mention because of its careful bio- logic treatment of a heretofore much neglected group. The second report. State Museum Bulletin 68, comprised 419 pages and 52 plates and was a continuation of the preceding. The monograph of the Odonata is completed by an exhaustive account of the smaller dragon flies (Zygoptera). Among the important contributions may be mentioned-: The key to Coleopterous larvae with an account of some aquatic C h r y s o m e 1 i d a e by Dr MacGillivray, the discussion of cer- tain aquatic nematocerous D i p t e r a by Dr Johannsen, and a monograph on the S i a 1 i d i d a e of the Western Hemisphere. The present report is a continuation of the work, and among its valuable features should be noted the monographic account of our May flies, a group of great importance as food for fish. The small midges, belonging to the C h i r o n o m i d a e, are very important as fish-food and have been treated exhaustively by Mr Johannsen. These three publications mark a most decided ad- vance in our knowledge of aquatic forms and, with the publica- tion of the monograph on stone flies now in preparation, a large fund of information will be availa,ble for the student of aquatic forms. This study, as was pointed out in the introduction to the first report, has been made upon broad lines with the avowed purpose of producing something of value to the fish culturist, who must first of all be aible to identify aquatic forms, something well-nigh MAY FLIES AND MIDGES OF NEW YORK O impossible, before these reports were made public. The investi- gations of Dr S. A. Foi^bes of Illinois convinced him that nearly one-fifth of the entire amount of food consumed by all adult fishes examined by him consisted of aquatic neuropteroid larvae, the greater pai-t of them being tlie young of ]May flies. It may never be possible to rear aquatic insects for the purpose of feeding fish, but it certainly is feasible in some instances to provide conditions adapted to multiplication of aquatic insects, and therefore valu- able as feeding grounds for fish. The history of the shellfish in- dustry gives a little idea of the possibilities along this line. A number of years ago it was at a very low ebb, owing to unscien- tific methods in vogue and the lack of individual control. This has been changed and we now have a thriving industry producing over two million dollars ($2,809,758) worth of products, accord- ing to the report of the United States Fish Commission for 1900. It is exceedingly difficult to obtain figures relating to the value of our fresh-water fishes, but a compilation from the report of the United States Fish Commission for the year 1900 gives the total value of fresh-water fish in the Hudson river valley and Long Island at over one million dollars (.fl,192,544), and the report for 1901 places the value of fresh-water fish obtained in the State from the Great Lakes at nearly one-fourth a million (|241,91G). These figures, it will be observed, give no idea of the value of fresh-water fish taken in various lakes and streams throughout the State, aside from the areas mentioned above. Comparing the water areas available for shellfish culture and those suitable for the development of fresh-water fish, it will be seen that there is a considerable discrepancy in favor of the latter and yet the value of the product is much smaller. It is stated that a lai'ge propor- tion of the market fish of China are grown in ponds, and that carp culture is an imtportant industry not only in China but in Germany, and that formerly carp were extensively reared in Eng- land, Germany and Sweden, and lately France, have also done considerable along this line. It is hardly likely that this country will adopt Chinese methods, because the great difiference in the price of labor makes it imprac- ticable; still the proper knowledge of the conditions suitable 6 NEW YORK STATE MUSEUM for the growth and multiplication of fish may put it within the po^^er of many to make substantial additions to the productivity of areas under control, without great increase in the cost of man- agement. These investigations have been conducted primarily to ascertain tlie relations existing between fish and insects they feed upon, and the conditions necessary for the development of large amounts of fish-food. Much of the preliminary work has been jiccoraplished, and the data already obtained should prove of great service to parties interested in fish culture, especially in making heretofore barren waters productive. E. P. Felt State Entomologist New York State Education Department New York State Museum John M. Clarke Director Bulletin 86 ENTOMOLOGY 23 MAY FLIES AND MIDGES OF NEW YORK I. INTRODUCTION. BY JAMES G. NEEDHAM This bulletin includes further results of the study of material gathered under the auspices of the New York entomologic field station, and is therefore complementary to bulletins 47 and 68 of this same series. Bulletin 47 contains the more general re- sults of the first field season spent at Saranac Inn, introductory keys to aquatic insect larvae, numerous life histories, and a de- tailed report of the dragonflies (O d o n a t a-A n i s 0 p t e r a) of New York State. Bulletin 68 contains the main results of the second field season spent at Ithaca, further life histories, detailed reports on the damselflies (O d o n a t a - Z y g o p t e r a) of the state, on aquatic plant-beetles (O h r y s o m e 11 d a e), on certain families of nematocerous diptera, and on American S i a 1 i d i d a e; also, an account of the food of the brook trout in Bone pond. This bulletin contains the work of three collaborators who have labored apart on the remaining material gathered for the station. Mr O. A. Johannsen furnishes the major part, in the • form of a completed review of the O h i r o n o m i d a e . Not- withstanding that these little gnats are enormously abundant everywhere and are of first importance among insects affecting fish culture, this is the first American monograph we have had dealing with the family to which they belong. It is a generic treatment of the world fauna, together with detailed descriptions and life histories (mostly new) of our known species. It is a 8 NEW YORK STATE MUSEUM work of first importance, and will doubtless serve as a basis for future studies in this long-neglected family. Mr K. J. Morton of Edinburgh contributes a paper on the micro-caddisflies of the family H y d r o p t i 1 i d a e of T r i - choptera, which is practically the beginning of the study of this group in America. M}' own part in this bulletin is a second contribution to the knowledge of our may-flies. Because of the great economic im- portance of this group also, I have thought it worth while to attempt to provide American students with a better introduc- tion to the study of the group than has hitherto been generally available. Hence, in addition to new life histories, I have pre- pared new generic keys to both nymphs and adults, which, with the detailed explanations and figures, should enable even a novice to take up the study of this neglected group with some hope of success. I have also prepared a brief report on the summer food of the bullfrog (R a n a c a t a s b i a n a Shaw) at Saranac Inn, and in the discussion of that food have included a number of ecological nnd systematic notes, among which is a new key to our genera of Hemerobiidae. I planned also to include herein a report on the stoneflies (Per lid ae) and did much work to that end: but the station collections are large, and much material has come to me from friends outside, and my manuscript has grown until it now seems better not to include it herein, but to make a separate bulletin of it. I am therefore continuing the work with the purpose of making the next station bulletin a monograph of North American l*erlidae. I should be greatly obliged if American collectois who have even a few specimens would send me them for study. In this place I may add a note supplementary to bulletin 68. The '' unknown tipulid larva from a spring " described on pp.285-. 286 and figured in pi. 10, figs. 4-5, is Pedicia albivitta Walker. Had Beling's third paper on Tipulid larvae (Verb, zool.-bol. (ies. Wiel, vol. I'6) been available to me when I was studying this larvae, I should have been able to determine it from his keys and description. The " unknown leptid larva from rapid streams" of p.28G and pi. 10, fig.l, is doubtless a MAY PLIES AND MIDGES OP NEW YORK 9 species of Atherix, as has been kindly indicated to me in correspondence by both Professor A. Giard of Paris and Dr R. Lauterborn of Ludwigshafen. THE SUMMER FOOD OF THE BULLFROG (RANA CATES- BIANA SHAW) AT S ARAN AC INN (Witti plate 1) BY JAMES G. NEEDHAM Bullfrogs are common at Sarauac Inn. Any warm evening their sonorous notes may be heard reverberating through the tamarack swamps, echoing and reechoing across Little Clear pond between Green hill and the outlet, or rising with a startling crescendo near at hand from the shallows of the reedy creek, setting the thread-rushes trembling, and fretting the face of the water with infinitestimal wavelets^ striking with wonder and admiration the ears of the stranger accustomed only to the vocal powers of the lesser civilized frogs, By day they sit in the edge of the water, stolidly basking in the sunshine, picking a straying bee or dragonfly out of the air, or lapping a floating ant or an emerging caddisfly from the surface of the water, eating- much or little according to the bestowal of Providence, and when alarmed by our too close approach, plunging away with a single dilatory and awkward leap into deeper water. Their tadpoles, likewise of phenomenal size, are to be seen about the submerged timbers in Little Clear pond and creek. They are oftenest observed resting upon the logs in the sunshine. Frequently, when crossing the bridge over Big Clear creek on the Otisville road during our first field season, I stopped to watch them sun- ning themselves on the submerged bridge timbers, and often dropped pebbles upon them to see them swim away. They would wriggle and sidle and slide off the timbers^ and then with a motion that appeared most deliberate strike a straight course obliquely downward far away across the clear deep waters of the stream, moving slowly forward by sculling undulations of the enormous banner-like tail. During July and August, 1900, I preserved the food of a number of adult bullfrogs from Little Clear creek, taking the stomachs of chance specimens that were killed for food and preserving and 10 NEW YORK STATE MUSEUM cleaning the contents. Most of the specimens were obtained fur me bj m}- friend l)r O. tS. W'estcott of Chicago, who was visitmg the station at that time. 1 suggested that he test the efficiency of a hook and line baited with a little piece of red silk flirted near the bullfrogs' heads. He reported the capture of every specimen properly approached; said that bullfrogs are abject idiots; said that if one is not hooked at his first dash for the dangling cloth, but gets his mouth snagged, he will go for the bait again and again as eagerly as at first. It is indeed remarkable how the predatory reflexes incited by the sight of the dangling red cloth prevail over the effects of the wounds. There now remain in the New York State collection the pre- served contents of the stomachs of fifteen of these frogs, and I have studied this material, with the aid of Mr W. H. Ferguson, and report on it here. The following table is largely the work of Mr Ferguson. I have added to it the single record published in bulletin 47 p.401, making 16 in all. The traditional account of the manner of the bullfrog's feeding pictures him sitting immobile on a bank, watching for insects passing through the air, and, when these approach, capturing them by flirting out his long, bifurcated, sticky tongue and striking them. The picture is incomplete. Doubtless he cap- tures some of the bees and hover flies and others of the fleetest insects in just this way, but the larger, heavier and slower ones he endeavors to meet half way. For instance, on the approach of a big caddisfly or a blackwing damselfly, he becomes greatly ex- cited, especially after an unsuccessful stroke at it, and leaps and plunges toward it with tongue and jaws both reaching for it. Some of the larger of his captives would not be held by the adhesiveness of his tongue without the immediate assistance of his jaws. Moreover, the greater part of his food is not obtained from the air at nil, but from plants, from the ground, and from the water, and doubtless, by more deliberate methods. The cater- pillars and sawfly larvae of the table were probably picked from plants; the beetles and millipedes from the ground; the water striders, floating dead insects, soldierfly larvae, gnat pupae, and transforming caddisflies from the surface of the water; and the mayfly nymph, gnat larvae and some of the snails probably from beneath the water. MAY FLIES AND MIDGES OF NEW YORK 11 Miscellaneous T5 CO o Ml a a CO lock leaves Hemlock leaves Hemlock leaves and J) 'c3 '73 o 4 pine leaves Birch bark Hemlock leaves Hemlock leaves and algae Hemlock leaves and algae -t-3 a be i 1—1 1—1 -^ : : CO t?j8;doii;.io 1—1 tH 1—1 1—1 tH ^ o ^3a^:^dT^I^H o CO 1—1 ^ TH ^ ■ 1—1 CI 1—1 < < o P o SSS8 puB sqdoiA'^j 1—1 © ^ 05 1—1 1—1 s^inpV T-I 1-< tH O-l 1-1 0? 1—1 puBS • 1—1 1—1 tH ^ ^ s;inpi; I15.X8U8X 1-1 ^ 1—1 1—1 tH QO i-< ■- LO < K Eh 8BA.n;i C7 1-1 1-1 e? lO 0? 8Bdn\ 1-1 tH 1-1 1—1 Tf' lO tH o H 12 NEW YORK STATE MUSEUM Notes on the Food General. Leaving aside the plant fragments eaten, which were of considerable number and variety, which were obtained both from the water and the air (as shown by the presence of filamen- tous algae and a broken flower cluster in the same stomach), but which were probably all obtained accidentally along with animal food, there were present the remains of 164 animals. Of these the largest number, 139, were insects, 18 were snails, 3 were Crustacea, 3 were spiders, and 2 were vertebrates. The most im- portant part of the food is doubtless insects and snails; the former in great variety', the latter consisting of a single species. Leaving aside frog no. 16, whose stomach contained only a large meadow mouse, the other 15 had eaten on an average 9 insects and 1,2 snails apiece. Of the insects eaten two were millipedes (apparently J u 1 u s , but not in condition to identify with certainty) and the remainder were hexapods. The ten orders present had the following numer- ical representation : D i p t e r a . 12 ; H y m e n o p t e r a . 22; Hemiptera, 19; Goleoptera, 16; Trichoptera, 15 (not including 4 whose presence Avas evidenced only by sand sup- posed to have been derived from larval cases) ; O d o n a t a , 11, and a large mass of eggs of Tetragoneuria; Orthop- t e r a , 6 ; N e u r o p t e r a , 3 ; L e p i d o p t e r a , 2 ( larvae) ; Ephemeridae, 1 (nymph). Of these the six orders first named were present in fairly equivalent proportions, and these, with the snail, Physa heterostropha, may be said to constitute the staple food of the bullfrog in summer at Saranac Inn, The bulk of the snails eaten was certainly greater than that of the insects of any single order. The largest animal eaten was the meadow mouse, and next in size were the two craw- fishes. Vertebrates. There were two vertebrates eaten; frog no. 16 had eaten nothing but a short-tailed meadow mouse ( A r v i c o 1 a pennsylvanicus) of large size ; that was enough to fill his stomach to its full capacity, Ho\v he came by this sumptuous morsel I am unable to understand unless he found it dead and floating down the creek. Frog no, 15 had swallowed a yearling tadpole of his own species. MAY FLIES AND MIDGES OF NEW YORK 13 Crustaceans. Frogs nos. 7 and 12 had each eaten a crawfish, of which there remained as evidence only the chelipeds. These indi- cated half-grown individuals of the genus C a m b a r u s. Frog no. 15 had eaten, probably by accident, a minute and undeter- mined copepod. Hymenoptera. These collections were made during the season of flight of the winged males and females of the big carpenter ant (Camponotus pennsylvanicus) remains of which were found in nine stomachs. Thus this species occui'red a greater number of times than any other. Stranded specimens were frequently seen floating down the creek, and the frogs may as well have obtained them from the surface as from the air. Worker bumble bees (Bomb us ternarius Say and B. c o n s 1 m i 1 i s Cr.) were found in five stomachs, and these were doubtless obtained alive. The bullfrog would seem to be, like the brook trout, immune to bee poison. The other hymenoptera were but three ; a wasp ( V e s p a d i a b o 1 i c a Sauss.) in frog no. 12, a snwfly larva in frog no. 1, and a minute parasitic hymenopter in frog No, 11. Coleoptera. Of the 16 specimens of this order eaten 12 were Carabidae (11 adults and one larva) , and there were single adults of Scarabaeidae,Ghrysomelidae, and C u r - culionidae, and a single larva of Elateridae. Diptera. This order was represented by the largest number of individuals, but many of them were very small. Six families were represented : T i p u I i d a e , C h i r o n o m i d a e , S t r a t i o - niyidae, Syrphidae,Tabanidae, and Tachinidae. A single adult Tabanid was eaten, two adult T a c h i n i d s , four adult S y r p h i d s , the better preserved appearing to belong to the genus E r i s t a 1 i s , five adult T i p u 1 i d a e , all belong- ing to moderate sized species of the genus T i p u 1 a . There was a single adult Chironomid, but there were eleven pupae, ten of them from frog no. 14, all belonging to the genus C h i r o n o - ra u s and one larva from the same frog belonging to the same genus and one belonging in Ceratopogon. A sixth family, Stratiomyiidae. was represented by twelve larvae of Stratiomyia badius? from frog no. 1. In bulletin 47, p.576, I have recorded that I could find but a single specimen 14 NEW YORK STATE MUSEUM of this Species during the season. Of the total of 42 D i p t e r a eaten 27 were larvae and pupae, and these must have been ob- tained from the water, Trichoptera. With the single exception of the large N e u - ronia postica eaten by frog no. S, all the other caddisflies were teneral images, captured probablj as they came to the sur- face in transformation. This was evidenced by the puipal skins still hanging to many of the specimens. All were in bad con- dition in consequence, and in determining them I placed chief reliance on the characters of the pupal skins. I was able to assure myself that about nine of the specimens belonged to the genus II a 1 e s u s and another to Hydro psyche. The sand found in four of the stomachs seemed to indicate that larvae in their cases had been eaten earlier and entirely digested. Larvae of P o 1 3^ c e n t r o p u s 1 u c i d u s and Molanna cinerea are sufficiently available in Little Clear creek. I have shown in bulletin 68 that the brook trout in Bone pond swallow the larvae of another species ease and all. Odonata. Drangonfiies constituted as large a part of the food as any other single group of insects. Although the number was but eleven, the size of the individuals was relatively large, the adult A e s c h n a and the nymph of A n a x being among the largest insects eaten. Four adult and apparently fully colored blackwings, O a 1 o p t e r y x m a c u 1 a t a, two adults of A r g i a v i o 1 a c e a and single undetermined specimens of L e s t e s , E n a 1 1 a g m a and .^ s c h n a make up the list, together with a nymph of A n a x Junius and an undetermined n^-mph of the subfamily A g r i o n i n a e. Tlio adults, so far as might be determined, were all females and might have been obtained while ovipositing. Frog no. 4 had swallowed a considerable mass of eggs of Tetragoneuria. In bulletin 47, pp.490-492 (with fig.19) I have given an account of these eggs. The frog probably found a cluster unusually close in shore, Hemiptera. The water skaters (llydrotrechus sp?) constitute an important and fairly constant element of the food, 16 of the 19 specimens found being of this genus. Orthoptera. Five grasshoppers were found singly, the one in condition fit for determination being Melanoplus femo- MAY PLIES AND MIDGES OF NEW YORK 15 r a t u s and one grouse locust. Considering the abundance of these about the edges of the creek, I was somewhat surprised that more had not been eaten. As many as this may easily have been picked from the surface of the water. Lepidoptera. Two moth larvae only. Ephemeridae. A single nymph of S i p h 1 u r u s alter- n a t u s Say was eaten by frog no. 14. It must have been taken beneatli the surface of the water as these nj-imphs do not come to the surface, so far as I have observed, except to transform, and this one was not ready for transformation. I have given an ac- count of the habits of the nymph of this species in bulletin 47 p.424. It was a surprise to me that no adult May flies were eaten. Neuroptera. Amphibian stomachs offer a new field for collect- ing representatives of this order, a field in which I have made some of imy Ibest finds, and that in a very little material. I found Sisyra umbrata Ndm. first in the stomach of a tree frog, as recorded in Psyche vol.10, p.29, and these bullfrog stomachs contained specimens of a new species of Micro mus, and of C 1 i m a c i a d i c t y o n a Ndm. and Hemerobius a m i c u 1 u s Fitch, — single specimens of each. SYSTEMATIC NOTES ON HEMEROBIIDAE Micromus jonas sp.no v. Allied to M . a n g u 1 a t u s , but smaller ; expanse 10mm. Known only from its wings, but these alone will distinguish it (pl.3, fig.2). The fore wing is 4.7mm. long and 2mm. wide, Avith front and hind margins nearly parallel in their middle third. Their color is rich fulvous, with darker fuscous oblique streaks along the line of both the gradate series, and less distinct, more transverse marmorate lines between, which become arcuate where they traverse the bases of the apical forks beyond the second gradate series; hind wings pale fulvous about margins, the disc transparent, and the veins traversing it very angulate in their course with crossveins incomplete. Gradate veins in fore wing; inner series 5, outer series 4-5 : in hind wing; inner series, 3-4, cmter series 3-4. Saranac Inn, N. Y. Taken from bullfrog stomnch (no.9 of table) in July, 1900. 16 NEW YORK STATE MUSEUM I have in hand a study of the venation of the H e m e r o - biidae. This material, especially He me robins amicu- 1 u s Fitch, and another of Fitch's rare species, H . o c c i - den talis from Illinois (which I have recently received from Wisconsin), together with other species of Heme robins col- lected at Saranac Inn, Ithaca and in Hlinois, have thrown some light on the evolution of the peculiar Hemerobian type of venation. My study will in due time be published elsewhere when it is com- pleted; and the results to be noted here are merely that H . a m i - cuius Fitch and H. occidentalis Fitch represent two stages in the evolution of the t^ipe which should be marked by generic rank. I therefore characterize them here and in the form of a key, because the key to H e m e r o b i i d a e in bulletin 47 was not made complete for our genera : KEY TO THE GENERA OF HEMER03IIDAE a Branches of the radial sector arising {i. e., sepa- rating from vein R,) by a common stalk h With three ocelli D i 1 a r fi& With no ocelli c Humeral crossvein (the basal costal cross- vein) simple and not recurrent d Some of the" branches of vein Cuj forked. . S i s y r a (Id All of the branches of vein Cui simple. ... C 1 i m a c i a cc Humeral crossvein recurrent and bearing a number of branches on its outer side d Subcosta and radius separate at the tips. . Polystoechotes dd Subcosta and radius conjoined at the tips. . B e r o t h a Art Branches of the radial sector appearing to arise separately from vein R, h Humeral crossvein unbranched and not recur- rent (pi. 3, figs. 1 and 2) M i c r o m u s hi) Humei'al crossvein recurrent and with branches on its outer side c First division of the radial sector arising before or opposite the basal subcostal crossvein ; in the hind wing the vein ]\r,+2 is well separated from the base of the I'adial sector, with a distinct crossvein between d A closed cell in the first fork of the radius before the base of the second division of the sector (pl.2, fig.2); front coxae longer than the femora Spadobius n. gen. type H. occidentalis Fitch MAY FLIES AND MIDGES OF NEW YORK 17 del No closed cell iu the first fork of the radial sector before the base of the second divi- sion of the sector (pl.3, fig.3); coxae of fore legs shorter than femora Palmobius u. gen. type H . a m i c u 1 u s Fitch cc First of the three or four divisions of the radial sector arising well beyond the basal sub- costal crossvein (pi. 2, fig.l); in the hind wing vein M1+2 is more or less confluent with the base of the radial sector, elimi- nating or reducing the crossvein be- tween Hemerobius EPHEMERIDAE BY JAMES G. NEEDHAM Since the iDublication of Museum Bulletin 47 little attention lias been given by the workers at the Entomologic Field Station to the collection and rearing of mayHies. Incidentally, however, a number of new and most interesting forms have been brought together, and nine additional species representing as many addi- tional genera have been reared — mostly by Mr Betten and myself during the summer of 1901 at Ithaca. It is the purpose of this paper to give the results of new life history studies, and also new keys for both adults and nymphs, that shall serve as a better introduction to the study of this interesting group. That the group is of great economic importance in water cultuie there can be no doubt. Past food studies have demonstrated this; and every aquatic collector has found the waters teeming with the immature stages. There are mayfly nymphs for every ."•ort of situation in fresh water, and they are almost everywhere abundant. These are perhaps the dominant insect herbivores of fresh water. Notwithstanding their ecological interest, the won- derful ways in which they have adapted themselves to diverse modes of life in different sorts of places, and their singular, though fragile, beauty, their study is very much neglected among us. It is in the hope of interesting more of our field workers in them that I have added to the life histories and descriptions, the keys and text figures of the present paper. Few life histories of American species, whose nymphs have been positively determined by rearing, have as yet been written. The singular nymph of Baetisca obesa Say has long been 18 NEW YORK STATE MUSEUM known, having been described by Walsh, its discoverer, and by Vayssiere and Eaton. In bulletin 47 1 described the nymphs (having in each case bred the species) of Heptagenia p u 1 c h e 1 1 a Walsh, Baetis pygmaea Hagen, S 1 p h 1 u - rns alter natus Say, Ephemerella excrucians Walsh, Caenis diminuta Walker, Hexagenia varia- bilis Eaton, and Ephemera varia Eaton. In the Ameri- can Natui^alist for 1903, pp.25-31 of vol.37, Mr Edward W\ Berry described the nymphs of ?Habrophlebia americana Banks, Bl a stums cupidus Say and C a 1 1 i b a e t i s f e r- ruginea Walsh, and in Bulletin 68 I described the nymph of CaUi baetis skokiana Needham. That is all the bred species that have hitherto been described in Aimerica, so far as I know. In the following pages I describe the nymphs of the following eight bred species, representing as many genera : Chiroten- etes albomanicatus sp. nov. (Jhoroterpes basa- 1 i s Banks, Leptophlebia praepedita Eaton, Caenis a 1 1 e c t a sp. nov., Ameletus ludenssp. nov. Ephem- erella b i s p i n a sp. nov., Heptagenia inter punctata Say, and Ecdyurus maculipennis Walsh, Mr W. E. Howard furnishing an account of the life history of P o 1 y m i - tarcys albus Say, which he has studied at Ottawa 111., but which I have not seen at large. I add thereto descriptions of five additional species which have not been bred, but to which the names of native genera are assigned tentatively. Some of the above descriptions are generic rather than specific : the study of the nymphs in some genera has hardly gotten down to the species as jet. Representatives of all these genera are de- scribed and figured in Eaton's Monograph of Recent Ephemeridae, at least two of them being tentatively referred to the wrong gen- era, however. But the excellent and copious figures of that work make it possible to refer the five species of unbred nymphs to their o-enera with some de£?ree of assurance. I have published directions for collecting and rearing nymphs of mayflies elsewhere,^ but while speaking of life histories I would not omit to mention how easy it is to get life-history material in iPart 0 of Bull. 3f>. U. S. National Mngonm. MAY FLIES AND MIDGES OF NEW YORK 19 this group. As is well known, there is with majflies one moult during adult life. The nymph, transforming, leaves the water as a subimago, and later moults again and becomes the imago. The subimago stage lasts but a little wliile — ^but a few minutes with the most ephemeral species, about a day with the majority of species, two days with S i p h 1 u r u s a 1 1 e r n a t u s kept indoors — ^being much more brief than is the period of transformation of even those species that are most concerted in time of appearance on the wing. It follows from this that when one finds subimagos flying, he can go to the water whence they came and be rather sure of finding, with projier searching, the full-grown nymphs. The subimagos may be recognized by their generally duller coloration, and the possession of fringes of hairs around the wing border (present in the imago of C a e n i s only among our forms) . Grown nymphs may be placed in any sort of a dish of water near a window out of the direct sunlight to transform. The subimagos picked from the window later may be put in paper bags and left to moult again. All stages are best preserved directly in alcohol of about 80 per cent strength. Besides the material for this paper collected by myself and Mr Betten at Ithaca N. Y. and Lake Forest 111., and that furnished me from the State Museum collection by Dr Felt, I have received material used herein from Professor T. D. A. Oockerell collected at Pecos New Mexico, from the late Mr R. J. Weith, collected at Elkhart Indiana, from Mr ChaunccA- Juday, collected at Twin Lakes Colorado, and from Mrs Mary Rogers Miller, collected at Thousand Island Park N. Y., for all of which I return grateful acknowledgment. For the use of the following keys a little more knowledge of mayfly structure is likely to be required than the average text- book of entomology affords. A knowledge of the names of the parts of the body and legs of the typical insect will be assumed ; also, of the principal mouth parts and antennae. It should be known that the male is readil}-' distinguished from the female by the possession of much larger compound e)"«s, these always being remote from each other in the female, and by the possession of a pair of jointed appendages called forceps that project backward from beneath the penultimate segment of the abdomen. The two 20 NEW YORK STATE MUSEUM or three filiform appendages whieli terminate the abdomen are here called caudal setae. The tarsi are typically five-jointed in the adult, though one or t\^^o basal joints show a marked tendency to fuse with the end of the tibia, and the last joint bears two claws of vari- able form (Fig.5) ; in the nymph the tarsus is one- jointed and bears a single claw (Plate 6, fig. 7 and 8). The mouth parts in the adult are atrophied add functionless, while in the nymph they are highly developed. But one feature of them needs mention here, however ; that is the armature of the mandible. By comparing pl.6, fig.4, and pl.8, fig.6, it will be seen Fig-. 1 Venation of the wings of Siphlurus; lettering explained in text that each mandible bears on its inner side a broad more or less corrugated molar surface^ and at its antero-lateral angle several variable canines?- To the venation of the wings the student who aspires to an .acquaintance with mayflies would do well to pay special heed. This is of chief importance because 1) the venation is perfectly definite and easily observed; 2) it suffers least distortion in pre- served specimens; 3) it remains the same through the different developmental stages, and 4) the wings are better retained than the other appendages, and progress is better in using a key if the structures mentioned in it have not been lost. The main features 1 Following the torniinology of Vayssiero for these parts. Organisation des larves cles Ephemerines: Ann. Sci. Nat. (G) vol.13. 1882. MAY FLIES AND MIDGES OP NEW YORK 21 of the venation are easily learned, and afford a ready clue to the relationships. Eaton says, " Unstable in minutiae, so closely is the essential plan of the neuration adhered to by nearly related mayflies that the general facies of the wing is an imiportant aid to their classification, affording characteristics as easily recog- nizable as the style of branching in the case of trees." By reference to figs.l, 2 or 3, or an}' of the wing figures of the plates, it will be observed that there are three nearly parallel veins extending along the front or costal margin of the wing, eosta (C), subcosta {8c), and radius {R^). These three are fol- lowed by three forking veins that occupy the greater part of the wing area, the radial sector {Rs), the media (M) and the cubitus (Cu). The middle one of these, the media, forking usually far- Fiu'. 3 Wings of Callibaetis ther outward than the others and being more constant in form, is one of the best landmarks of the wing. All that lies between it and vein R^ is radial sector, which, in the fore wings of may- flies, is entirely detached from the radius and functions as a separate vein. The only place in the series where there is likely to be any difficulty in recognizing the media is in the few genera closely allied to Baetis (see fig.2) in which both the media and the cubitus are apparently simple; but it will be readily observed by carefully noting the number and relation of the longitudinal veins that the liindcr branch of the fork of these two veins is detached, and appears as an independent sector standing on the hinder side; the relative lengths of these veins enable one to recog- nize them all, even when detached, or when, through shifting of cross veins at their bases, they appear to have formed attach- ments of a contradictory sort (see vein Cuo in pl.8, fig.9). These 22 NEW YORK STATE MUSEUM three forked veins are followed by three typically simple veins, the first, second and third anal veins, which occupy the smaller area of the hind angle of the wing. There is much variability in this region in the different genera, and it is highly important that these three veins be certainly recognized; to do this it is only necessary to count off the three longitudinal veins of the cubitus — the two branches {Cu^^ and Cuo) and the bisector of the cubital fork — ^back of the media, and these three will be the three best developed veins remaining. In the keys the short, incon- stant interpolated longitudinal veins are called intercalaries , and that whether they become attached to principal veins or branches or remain independent; and the irregular veins about the margin Fig. 3 Venation of the fore wing of E p h e m cr a of the wing are called veinlets. The length of the media is meas- ured on vein M.^. The fore wing is meant in the key except where the hind wing is specified. Fig.3 shows the unilateral forking of the cubital vein and the divergence of the cubital and first anal veins at base, characteristic of the subfamily Ephemerinae. KEY TO THE GENERA OF MAYFLIES OF NORTH AMERICA Imngos a The cubital and first anal veins strongly divorgent at the base (fig.3). Venation never greatly reduced Ephemerinae h The fork of the median vein very deep, almost reaching the wing base; two long simple intercalaries between the first and second anal veins. In the hind wing the vein R5 separates from vein R, close beside and therefore is little longer than the next branch of the radial sector C a m p s u r u s hh The median vein forked for not more than three fourths of its length ; in the hind wing the vein R5 arises much in advance of other branches of the sector, being much longer than any of them MAY FLIES AND MIDGES OP NEW YORK 28 c Between tbe first and second, anal veins is a bunch of 3-4 long, straight intercalaries, conjoined basally before their attachment to the principal veins; the second anal vein nearly straight and iinbranched Polymitarcys CO Between the first and second anal veins are only shorter, sinuate, and sometimes forking intercalaries, attached directly to the first anal ; the second anal vein sinuate and often branched (fig.3) d The median vein forked % to % its length ; vein Cu^ not more strongly bent at base than the first anal Euthyplocia dd The fork of tlae median vein occupying not more than half its length ; vein CUj more strongly bent at base than is the first anal (fig.3) c The third anal vein simple, but attached to the hind margin by a number of crossveins ; in the narrow first fork of the median vein there are one or more crossveins before the origin of the vein M, ; male forceps four-jointed f Caudal setae 3, c? and 5 5 ^ore tarsus of female imago % as long as the tibia Ephemera ff Oaudal setae 2 in c? and 3 in 5 ; fore tarsus of 2 % as long as the tibia Pentagenia fff Caudal setae 2, (^ and ? ; fore tarsus of $ as long as the tibia Hexagenia ce The third anal vein with a simple terminal fork and unattached to the hind margin, although a few isolated short intercala- ries lie between ; in the wider first fork of the median vein there is no crossvein before the origin of vein M^ ; male for- ceps 3-jointed P o t a m a n t h u s aa The cubital and first anal veins parallel at base (in a few forms with reduced and scanty venation, appearing a little divergent) b Eyes of the male simple and remote; hind tarsi with 5 freely movable segments ; venation never greatly re mm. additional; wing, S mm. Color obscure 'brownish, paler on the sutures and below; antennae darker toward the tip; incomplete dark -brownish rings about the ocelli ; on the vertex a pair of longitudinal bla/ck- ish marks, confluent in the middle ; a broad median whitish tra^ct upon the mesothorax, produced behind and dilated at the sides; subjipical paler bands on the femora, the tips again darker; wings uniformly pale fumose, the venation is shown in pl.8, fig.9; brown marks on the ventral ganglia, becoming more evident posteriorly. MAY PLIES AND ISIIDGES OF NEW YORK 37 The accompanying text figures will facilitate tlie recognition of this species when more and better specimens are at hand. This species is a typical representative of A m e 1 e t u s , agreeing in close detail with the generic characters set forth in Eaton's Monograph p.210, but it is smaller than any of its con- geners. Its nymph is a.pparently the one figured by Eaton on pi. 49 of his Monograph, and referred to Chirotenetes. The nymph. (P1.7, fig.l.) Length, 9.5 mm.; antennae, 1 mm. and setae, 4 mm. additional. Body elongate, with vertical face, arched thorax, depressed and tapering abdomen. Antennae short, Fig-. 6 Parts of nymph of Amcletus ludens sp. nov. ; y, maxilla; z, single gill lamella from one of the middle abdominal segments taipering, bare; ocelli in front; labrum quadrangular, a little longer than wide, emarginate in front, where fringed with fine plumose hairs. ]\[andibles stout, triangular beyond the molar surface, ^bearing the canines upon the prominent apex, outer canine more than twice as large as the inner, the latter preceded by a slender subulate sjjine on the distal margin. Maxilla with a very we;ik and slender and obscurely three-jointed palpus. The comibined lacinia and galea obscurely trapezoidal, the tip of the fonuer indicated by a short, slender and sharp spine, the distal border of the galea fringed densely with a series of strongly arched, regularly graduated and beautifully pectinated hooks (fig.6y). Labium with better developed, three-jointed palpi, 38 NEW YORK STATE MUSEUM cultriform galeae, fringed with spiuules externally, and broadly triangular laciniae, separate to the base. Prothorax rather short, closely applied to the front of the large niesothorax ; Aving cases reaching the apex of the second abdominal segment. Legs rather short, stout, pale, with darker lines upon the sutures, the darkest one at the base of the claw. Abdomen gradually tapering, gracefully upcurviug in the rear. Gill laminae on segments 1-7, similar in form on all the seg- ments (fig.62;) ; smallest on segment 1, largest on segment G, obo- vate, with a somewhat thickened front margin, and a longitudinal dorsal chitinous ridge. There are no fi-ee gill filaments attached to lamellae. Lateral spines on segments 4-9 straight, sharp, increas- ing in size posteriorly. Setae rather short and stout, equal, fringed copiously Avithin, traversed by a broad distinct band of brown which occupies their middle third, and slightly Avashed Avith brown again at the extreme tips. This nymph differs from the one figured by Eaton (pi. 49, Monograph) in having the middle lobe of the tongue (hypop- harynx) bilobed. This genus differs from all others as yet known except T li 11 a u 1 u s in the possession of a pectinated fringe on the distal border of the galea of the maxilla. Choroterpes basalis Banks This species I have studied in the Fall creek gorge beside the Cornell Insectary at Ithaca. It is a very common species there. The nymph is found among the smaller stones in the side cur- rents of the creek in the bottom of the gorge, associated Avith other nymphs of E c d y u r u s m a c u 1 i p e n u i s , B a e t i s , Caenis etc. It clambers about under these stones, and when they are lifted out of the water it is easily picked off by hand. The form of the gill tips (Plate 8, fig.8) Avill instantly distinguish it from all others in the stream. Imagos were abundant about the middle of July. ]\Iy bred specimens are dated July 14, 1901. Not many imagos Avere ob- served at large excej)! on early afternoons. Avhen the sunshine Avas Avarm and bright. Then (hey would swarm out in the open- ing of the gorge, and dance high up in the air between the banks of green in myriads. Rising and falling in rapid undulations, moving in large companies up and down the gorge, they rarely descended low enough to bring the lowei-most Avithin the reach MAY FLIES AND MIDGES OF NEAV YORK 30 of the net; aud when by climbing- on a big rock in tlie opening I captured a netfull of them I found they were all males. About the same time also subimagos swarmed into my trap lanterns that overhung Fall creek, and a few imagos with them. The nymph. (PI. 7. flg.2.) Length, 7 mm.; antennae, 3 mm., and setae, 7.5 num. additional. Body strongly depressed, widest across the rather prominent mesothorax. Head flattened above; eyes round, prominent, situated just before the hind margin. Antennae situated midway the length of the head, which before them is pilot shaped, dilated at the sides and sharp-edged. Ocelli three, rather large, situated in a nearly straight transverse row in the male, in a triangle in the female. Labrum half as long as broad, widened anteriorly, rounded on the anterior angles and deeply emarginated in front, where fringed with short stiff bris- tles (pl.8, fig.o). Mandible (])1.8, flg.B) stout, its two canines each tridentate on tip, its ipalp deeply bifid ; on the inner margin just before the molar surface is a low conic tubercle. Maxilla (pl.8, fig.4) short and stout, the ipalpus two-jointed, the consolidated galea and lacinia squarish, the tip of the former ending in a long and distinctly pectinated spine, the inner and distal margins densely fringed with slender hairs. Labium (pl.8, flg.3) with three jointed palpi, the broad galeae and the narrow laciniae with their tips on a level, and densely fringed with spinules, the spin- ules on the laciniae being stouter. Thorax depressed, increasing in width to the bases of the wings. The wing cases reach the base of the fifth abdominal segment. The legs are rather short and stout, with flattened and dilated femiora and slender tibiae, pale with a more or less complete brownish ring beyond the middle of the femora and some fainter markings at the knees. Abdomen depressed, regularly tapering from the third seg- ment to the end, segments slightly increasing in length to the ninth, the tenth somewhat more than half as long as the ninth, produced above in a rounded lobe with a narrow blackish border that is interrupted by paler in the middle of the margin. There are sharp, triangular lateral spines on segments 4-9, increasing in length and sharpness on the succeeding segments, represented on segments 2 and 3 by mere angles of the flat margin, on 8 one fourth as long as the segment. Gills very peculiar; on segment 1 a simple linear or slightly tapering filament (pl.8, flg.7) that is fully as long as the succeeding lamellae; on 2-7 double, lamelli- form, Avith pinnately branching tracheae; each of the pair of lamellae is typicall}' three-lobed ; the middle lobe of the upper- most lamella is itself lamelliform, oval or oblong, separated by 40 NEW YORK STATE MUSEUM niargiual notches from the two otlier lesser lobes (pl.8, fig'.S). The middle lol)e of the lower lamella is likewise flat, but narrow, linear, and with a ibetter development of the two other lobes at its base. There is a slight decrease in length on segments 2-7; and on 2, and again on 7, the anterior of the three lobes of the njiiper lamella is scarcely developed. Setae three, fragile, slender, with minute apical whorls of sipinules on the segments. Color olivaceous brown albove, with a variaible middle pale line, fenestrate upon the dorsum of the abdomen with paler olivaceous. Below, with a ,broad pale median area. Several of my nymiphs from Fall creek have colonial Vorticel- lidae attached promiscuously about the dorsum, or aggregated aibout the bases of the setae. Pl.8, fig.l, shows the venation and fig.2 of the same plate shows the form of the a.ppendages of the male imago in this species. Baetis pygmaea Hagen This dainty little mayfly, which I described in bulletin 47 (pp. 421-423, pl.15, fig.l3 and 14), I bred also from nymphs obtained in Fall creek with those of the preceding species, and 1 took a few specimens of the imagos in trap lanterns hung about the., creek during July 1901. Callibaetis skokiana Needham I wish to record here concerning this species that I have made a careful examination of microscopic mounts of the stomach con- tents of ten Avell-grown nymphs taken from the Gym pond on the campus of I^ke Forest College in Illinois, and have found them containing no recognizable animal remains whatever, but only remains of plant tissues, chiefly the disintegrating fragments of the dead leaves of the higher plants, such as litter from the pond bottom, with a scanty sprinkling of algae — Cyanophy- ceae and stalked diatoms. Blasturus cupidus Say I have found his species common in Six Mile creek at Ithaca, where I bred it in 1897. I have apparently identical nymphs in my collection from Elkhart, Indiana, and Raleigh, North Caro- lina. The imagos of this genus appear in late spring. As be- fore remarked, Berry has described the nymph in the American MAY PLIES AND MIDGES OP NEW YORK 41 Naturalist vol. 37, pp.27-29, 1903. It will be at once distin- guished from all other genera by the form of the gill lamellae, Fig. 7 Gill lamellae of the nymph of Blasturus cupidus Say; e, fiom the 1st segment ; f, from the 4th segment ; g/trom the,7thlsegment a figure of which is herewith given (fig.7). There are well-de- veloped lateral spines present on abdominal segments 8 and 9 only. Ephemerella This is one of the genera of E p h e m e r i d a e that shows great nymphal specialization independently of adult life. The nj'inphs are obviously very diverse in form and structure; the images very much alike, or else their differences are easily over- looked. Eaton pointed out in his Monograph the remarkable differences between the nymph which I have since bred and shown in bulletin 47 to be that of E. e x c r u c i a n s , and that of the European E. i g n i t a , the only bred species with which he was acquainted. He referred to this nymph as a new un- named genus allied to Ephemerella; but it is the nymph of the typical species. I describe herein the nymphs of two native species closely allied to E. i g n i t a . I have compared both nymphs and adults with E. excrucians. I have not found differences that would seem to justify the generic separa- tion of the images; and notwithstanding the evident differences of the nymphs, I think they may as well, for the present, at least, remain associated together under the one name. The nymphal differences are chiefly in the number and arrangement of the gill lamellae, and these things are perhaps most subject to the influence of environment. 42 NEW YORK STATE MUSEUM Among the other four North American nymphs described by Eaton are two that will doubtless represent good and distinct genera; and one of these I have been able to identify; for it I erect the new genus DruneUa. The structural relations between the American nymphs of the E p h e m e r e 1 1 a alli- ance described by Eaton and those I have since obtained may be set forth by means of the following key: a Antennae inserted in deep angular notches in the front margin of the frons ; dorsal hooks of abdomen wanting ; nymph from Colorado, imago unknown^ aa Antennae inserted upon the upper surface of the frons ; dorsal hooks hooks more or less developeerculate anterior lamella of the 4th abdominal segment covers closely all succeeding lamellae, only their extreme apical margins visible e Body hardly more than twice as long as wide; Ephemer- ella sp? from Pecos N. Mex. ee Body more than three times as long as wide Ephemerella excruclans Drunella gen. nov.* I have determined the nymph of this genus by means of the venation of the developing wing. Professor Cockerell sent me two nymphs from Pecos New Mexico, one of which, a male ^Eatonno. III. Monograph, p.l32, pl.39, 22 figs. -Monograph, p.l31, pl.3S, figs. 1-10. 'Monograph, p.l33, pi. 40, 17 figs. (Colorado) *To my friend, Professor Theodore Dru Alison Cockerell. MAY PLIES AND MIDGES OP NEW YORK 43 specimen, is in perfect condition for showing tlie venation. It shows the basal fusion of veins CUo and 1st A that Eaton long since described and figured as characteristic of Ephemerella grand is Etn (Monograph, pi. 14, fig.24&). This character, together with the rather strong joinings together of the other anal veins basally, readily distinguished this large species from Ephemerella proper. The figures of venation I give herewith (pi. 10, figs. 1 and 2) are drawn from the nymphal wing, which shows the venation better than does the single female imago I have seen. I have another identical nymph collected at Twin Lakes Colorado, by Mr Chauncey Juday. Since the type of E . g r a n d i s is from Colorado, it seems very probable that the nymph belongs to this species. In pi. 10, figs.3, 4 and 6 I present figures of the male nymph, which differs slightly from the female, figured by Eaton. Ephemerella bispina sp. nov. The six specimens of this species that I have seen were sent me in the last lot of material received from the late Mr R. J. Weith. They were collected at Elkhart Indiana, shortly before June 18th — the date on which they reached me at Lake Forest. There were among them single male and female images, a male subimago, and three nymphs. The species is apparently near to E. walkeri Eaton from Albany river near Hudson's bay — still so insufficiently known— and to E . i g n i t a Pol. of Europe. Imago. Length, 9 mm.; Aving, 9 mm.; setae of 5, 10 mm. (of (^ wanting) ; of J^ subimago, G.5 mm. Male imago deep brown, varied with olive green. Antennae brown ; a Avhitish ring around their bases. Thorax rich dark brown above and on all carinae, greenish in the sutures and fur- rows, excepting the median longitudinal furrow. Beside the median prolongation of the hinder lobe of the mesothorax is a pair of acute spines, each decurved at tip and about as long as the space between them is wide. Wings subhyaline; veins pale brownish, as is also the subcostal space. Legs brown, the femora sprinkled with distinct blackish dots; fore leg dark, becoming gradually lighter toward the tip; middle and hind legs paler and tinged with greenish ; claws all brown, the obtuse one of each pair darker than, the other. Abdomen pale brown, except the 10th segment Avhicli is yellow- ish, paler on the sutures and thereby appearing ringed; an in- 44 NEW YORK STATE MUSEUM distinct middorsal row of minute brown longitudinal dashes. Appendages all brown, the inner ones angulated and thickened in the middle and bent upward thereafter to the tip (this appear- ing only in lateral view; hence not shown in the figure) ; forceps (pi. 10, fig.lO) strongly directed downward, the basal segment distinctly differentiated, the apical segment unusually long and slender. The female imago is greenish yellow, with pale whitish legs and setae. The basal segments of the antennae are brown and there is a pale brownish tinge to the dorsum of the thorax and the lateral margins of the abdomen. Wings hj^aline, veins whitish. The ventral apical lobe of the 9th abdominal segment surpasses the tip of the 10th segment and is obtusely rounded apically. The male subimago is dark greenish brown, darker on the head, the top of the thorax and the apex of the abdomen; the abdominal sutures, however, are distinctly paler. I^gs pale yellowish or greenish, the fore tarsus pale brownish. Wings smoky brown. The two dorsal apines are paler in 'the J^ subimago and absent in the ?. The nymph. Length, 9 mm; seta, 4.5 mm, additional. Body elongated rather slender, depressed, thinly hairy, widest across the mesothorax. Head short ; face oblique. Antennae hardly longer than the head, almost bare. Labrum quadrangular, one fourth wider than long, emarginate' in front and hairy on the front border, the hairs being longest on the outer angles. Mandibles short and thick, with the outer canine very broad, 3-toothed at apex, the inner canine of equal length but slenderer; molar sur- face narrow. Maxillary palpus hardly half as long as the lacinia. Third joint of the labial palpus a conic rudiment. Thorax fiat below, well rounded above; legs short and thinly hairy; claws (pi. 10, fig.5) with inferior row of about 10 denticles. Abdomen depressed, its lateral margins serrate by reason of the flat lateral spines in which the side margins of segments 3-9 ter- minate. There are two rows of dorsal spines on segments 3-8, erect laiterally, flattened, almost cultriform. Gill lamellae present on segments 3-7, double; anterior lamina thickened, covering the delicate posterior one, trapezoidal, obtusely pointed at its inner apical angle, palmately veined ; posterior lamina shorter, thinner, its margins cut into a small number of fingerlike filaments. The lamellae regularly overlap, each anterior lamina covering the basal fifth of the one behind it, that of segment 7 shorter. Setae closely parallel, slender, fragile, sparingly pilose. Their two proximal articulations faintly ringed with brown. Color olivaceous, with a broad l>and of brown extending from the rear of the eye to tlie base of the lateral caudal seta. There is also a narrow middorsal line of brown on the abdomen. MAY FLIES AND MIDGES OP NEW YORK 45 This species differs in the nymphal stage from the nymph next described, chiefly the presence of well-developed dorsal hooks and the absence of black rings on the base of the setae. Ephemerella unicornis sp, nov. Along with the six specimens of E . b i s p i n a came a single male of another apparently very distinct species, distinguished at a glance from all the others by an erect conic tubercle upon the front margin of the middle lobe of the mesothorax. This species is notably smaller, measuring but 5 mm. in length, with the setae of the same length and the wing hardly longer. The hind wing also is marked with a more distinct basal costal angulation than is common in this genus. The spines beside the backward pro- longation of the middle lobe of the mesothorax are present also in this species but apparently not so large. Unfortunately the specimen, although perfect, is a subimago, and the mature colora- tion can not be given; it will probably be brownish since in the subimago it is greenish as in E . b i s p i n a . The abdominal appendages are well enough developed to show that the end seg- ment of the forceps will be much shorter than inE. bispina, while the inner appendages will probably be of the same type as in that species, though probably relatively shorter. Ephemerella sp?, near ignita This species occurs at Ithaca, but I have thence but a single njTnph. There are two nymphs in the U. S. National Museum labeled " From stream on Mr Chamberlain's farm, Richfield Springs, N. Y., May 13, 1837." It is very closely allied to the European E . i g n i t a , as figured and described by Eaton (Monograph, pl.40 ; whole figure copied in Cambridge Natural History, vol.5, p.436, fig.282). One of the two nymphs from Richfield Springs is apparently grown. It measures in length 8 mm., setae, 3.5 mm. additional. Body rather more elongate than in the typical species; eyes lat- erally prominent; abdomen (pl.lO, fig.7) strongly depressed, the usual submedian double row of dorsal tubercles scarcely indicated. Lateral spines, thin, flat, sharp, on segments 4-9, a mere tooth on 4, increasing in size thereafter to segment 8, broader and less sharp on 9. Gills present on segments 4-7, double, on 4 scarcely 46 NEW YORK STATE MUSEUM operculate^ overlapping the next behind it hardly more than that one overlaps its successor. Setae slender, pale, ringed with dark brown at base, thinly hairy except at base (Plate 10, fig.7). Ephemerella sp? Professor T. D. A. Cockerell has sent me from Pecos, N. Mex., a single nymph of so remarkable form (Plate 9, fig.2). I desire to make it known herewith. Its affinities are obviously with Ephemerella excrucians, and it diifers from all the " allies of Ephemerella" figured by Eaton from western North America. Therefore I briefly characterize it here and present a figure made from a photograph of the single known immature specimen. Body excessively flat and thin, about twice as long as wide, widest across the middle of the abdomen. Head short and much narrower than the prothorax; eyes and ocelli dorsal, remote; antennae short, bare, about as long as the head is wide, composed of only aibout twelve segments, of which the basal one is as usual longest and thickest. All lateral margins very hairy. Prothorax half as long as wide, straight on front and sides with rather acute front angles, somewhat widened posteriorly. Legs short; femora flattened, widest before the middle and fringed on both margins. Abdomen short, about as wide as long, excessively flat, with huge, serrate lateral spines on segments 2-9, increasing in breadth posteriorly, but longest on the middle segments, all strongly curved posteriorly. Segments slightly increasing in length suc- cessively to the 8th, 9 much longer, 10 only about one fifth as long as 9, but slightly produced on the dorsal side. Gills cov- ered by an oblong opercular laimella attached at the apex of segment 4. Of the underlying gills I have made no examination, not wishing to injure the unique specimen. Setae 3, closely paral- lel at base, broken in the specimen. Coloration very obscure, the animal being apparently covered in life by adherent silt, but there is a trace of a brownish ring on the middle of each tibia and another on each tarsus. Pecos, New Mexico, July or August 1903. Professor Cockerell sent me from Pecos also a fine pair of imagos and these may represent the same si)ecies as the nymph above described. I should have felt inclined to refer these to Ephemerella i n e r m i s Eaton but for the conspicuously bifid prolongation of the 9th abdominal sternum in the female; MAY FLIES AND MIDGES OP NEW YORK 47 this Eaton describes as beiug entire. Otherwise, there is close agreement. The length is 7 mm. in male, 8 mm. in female; setae; 10 mm. in male, 6-7 mm. in female. The segments of the male fore tarsus in order of diminishing length are 2, 3, 4, 5, 1, the first segment being one fifth as long as the second. The legs are wholly pale. The setae of the male are strongly ringed with black except at the extreme tip in the male, wholly pale in the female. The head and thorax and basal segments of antennae are brown. The abdomen in the male is rufous, paler on the middle segments, and sufl'used with brownish apically above; in the female abdomen there are broadly triangular transverse basal bauds of paler on the middle segments. The posterior prolonga- tion of the sternum of the 9th segment in the female abdomen is deeply divided by a wide U-shaped notch. The abdominal appen- dages of the male are shown in pi. 10, flg.9. Ephemerella excrucians Walsh In Bulletin 47 I published a description of the nymph of this species (pp.425-426), bred at Saranac Inn. On June 30, 1901, Mr. J. O. Martin gave me a live nymph which he had just col- lected from the shore of Cayuga lake, and I reared this also. Since that time I have received a large number of specimens from dift'erent places in Indiana, notably from Elkhart, sent me by the late Mr R, J. Weith. From some of the latter, selected to show the great variety in depth of color pattern, I have had a new photographic figure made, which I present herewith (pl.9, fig.l). It will serve immediately for comparison with the very different form of nymph found in the species above described. On pi. 10, fig.8 are represented the abdominal appendages of the male imago. ? Caenis allecta, sp. nov. This is the commonest species in Fall creek at Ithaca. It swarmed into trap lanterns hung about the creek during July. Its nymph lives in the pools and side channels of that turbulent stream, w^here the water flows gently among small rock frag- ments over a bottom thinly strewn with silt. Imagos of our smallest species, Oaenis hilaris Say, come to the trap lanterns with this one, but in smaller numbers; its nymph I have not found. 48 NEW YORK STATE MUSEUM Imago. Length, 3.5-4.5 mm.; setae, about 10 mm. additional; expanse of wings, 8 mm.; fore leg of male, 3.5 mm. General color brown, marked with purplish or slaty gray; head and thorax brown, carinae and margins of ocelli blackish. Wings hyaline, with the usual pui^plish streak along the radius for two thirds its length. Abdomen pale yellowish brown on base and apex, the middle two thirds washed with gray; some elongate blackish marks on the lateral margins of the 7th to 9th seg- ments; setae white; antennae, femora and forceps yellowish; tibiae and tarsi, except the terminal joint, white. Venation of the wing and the male forceps as shown in the accompanying figures (figs. 8 and 9) . Fig. 9 Ventral view of Fig. 8 Venation of wing of ?Caenis allecta male abdominal append- sp. nov. ages of yOaenls al- lecta sp. nov., imago. Nymph. Length, 2.5-4 mm.; setae, 1.5 $ to 2 mm,; J* mm. ad- ditional. Color greenish broAvn, obscure on the head, with a transverse broken and obscure line between the paired ocelli, antennae and legs pale, a pair of broAvn submedian dots on the prothorax; ab- dominal segments pale basally and on the sutures; gill covers darker beyond the basal third ; segments 8-10 darker with a mid- dorsal pale line on 8 and 9. Lateral spines on segments 3-9, flat and thin, best developed on the middle segments, becoming less divergent posteriorly and losing their lateral fringes of spinules. Setae stout at base, rapidly tapering; middle one distinctly longer in female and shorter in male than the laterals, all with scanty apical circlets of spinules on the segments. Legs scantily and abdomen copiously beset with short hair that is usually covered with adherent silt. Aside from the not very satisfactory differences of coloration, this nymph differs from that of C. d, i m i n u t a in having the sides of the prothorax parallel; in ddminuta the prothorax is widened anteriorly, and in having a greater part of the abdomen covered by the opercular lamella; in this species that lamella MAY PLIES AND MIDGES OF NEW YORK 49 covers part of the 8tli segment ; in d i in i n ii t a it does not wholly cover the 7th segment. Were it not that these differences of structure of the nymph are so slight I should have thought a separate genus necessary for this new species; for the differences in venation and in the genitalia are certainly as great as usually serve for generic separation. These principal differences may be tabulated as follows : Character Caenis diminuta, hilar is, etc. C. allecta Anal veins disconnected absent conjoined basally present pluriserial three-jointed bilobed at sides and emarginate in the middle Vein M2 Crossveins uniserial Forceps of male .... Basis one-jointed straight edged Among some mayflies that were kindly collected for me bj Mrs Mary Rogers Miller at Thousand Isdand Park, on the St Lawrence river, are a number of typical specimens of our two previously described species, C. diminuta Walker and C. h 11 a r i s Say, that fit the descriptions exactly. In ordet to promote accuracy in the determination of the most' difficult forms, I have prepared the drawings herewith presented (pi. 11, figs.3-6) of the wings and male genitalia of these species. It will be observed by comparing the wings with Eaton's figures that in venational characters ? C. allecta agrees better with the Europaean genus Tricorythus and the South Ameri- can genus Leptophyes, than with Caenis . But there are disagreements also with these, and the (^ genitalia and nymphs of these are as yet not certainly known. Leptophlebia praepedita Etn. ? This species, hitherto known only from New Hampshire and not yet reported from New York State, is common about Lake Forest, Illinois, where I have found it in three quite diverse situations: 1) in the Skokie (north branch of Chicago river), a sluggish creek flowing through open meadows and marshes; 50 NEW YORK STATE MUSEUM "' 2) McGormick ravine, where a puny stream, overhung with witch-hazel and dogwood, flows between deep banks' through a hardwood forest; and 3) in a glacial pothole, grown full of buttonbush (C e p lii a 1 a n t h u s) on the top of a moraine. In all these situations the water is fairly permanent, disappearing only in seasons of extreme drouth. The species appears to be diurnal in its habits. Males may be found in abundance sitting on top of the leaves of shrubs be- side the water, or flitting over them in the bright sunshine, quickly gathering in companies and dancing up and down, and as quickly dispersing and settling again. They fly at low elevation, and are easily taken in large numbers in a net, and are as easily swept when at rest from the witch-hazel leaves. I found the species first in the Skokie May 8, 1901. There were then a very few subimagos on the wing, and a bed of mixed ranunculus and polygonum in the water was fairly ewarming with the nymphs. I took a large number home and placed them in a bowl of water, where they began transforming the next day. The subimago stage lasts about 24 hours. When Eaton desoribed the species he had some doubts as to whether it should go in L e p t o p h 1 e b i a ; but the characters of the nymph are in essential agreement with those of the typi- cal species of Leptophlebia, and thus confirm the refer- ence of the species to that genus. In pl.ll, fig.l, is represented the venation, and in fig.2 the $ abdominal appendages are shown. The nymph. Jjength of body, 0..5 mm.; antennae 2 mm. and setae 6 mm. additional. Body slender, scarcely depressed, widest across the imesothorax, smooth. Face nearly vertical, ocelli in front, eyes rather small situated just before the hind angles of the head ; antennae pale, basal segments rather stout, the follow- ing ones rather tapering to slender and very fragile tips. Mouth parts very similar to those of 0 h o r o t e r p e s , shown on pl.5, the maxillae more oblique on the end of the combined lacinia-galea, and lacking the pectinated s])ine tipping the former; the palpi, however, are three-jointed beyond the basal palpiger, and the palpi of the labium are two-jointed ; thus the conditions of segmentation in these appendages are reversed in the two forms; this segmentation, however, is often very indistinct, and more or less evidence of division of the last segment when there appear to be but two are generally discoverable in all the palpi. MAY FLIES AND MIDGES OF NEW YORK 51 Legs rather short, nearly bare; femora, scarcely flattened, but somewhat concave on the side applied to the body; pale brownish, paler at the sutures. Wing cases reaching posteriorly as far as the apex of the 3d abdominal segment. Abdomen very slightly depressed, regmlarly tapering posteriorly, its segments very slightly increasing in length to the 9th, the 10th a little shorter on the dorsum, where produced backward in a rounded lobe, one half shorter at the sides; short lateral spines on segments 8 and 9, larger on 9, the lateral angles of the pre- ceding segments obtuse. Gills present on segments 1-7, double, similar, or slightly longer on the middle segments, divided in nine tenths of their length into two long, slender, simple tapering filaments, pig- mented with purplish along the tracheae. iSetae 3, equal, nearly bare at base and sparingh^ whorled with spinules ibeyond, grad- ually tapering to long slender tips. General color olivaceous, paler below, with a very narrow median pale line on head and prothorax, a median row of pale spots on the ahdomeu of the female becoming larger posteriorly, and a pair of spots either side on segments 3-9, becoming confluent with the median one on 9 ; male darker and more uiniformly brown. May 8, 27, 30, 31 ; June 13, 14, 18. Heptageninae I deem it necessary to state that I have scarcely entered into the study of this interesting and difficult complex of interrelated forms, having dealt at first hand onl}' with those species in whose life his tories I have become intei'ested. The foregoing keys for this group of genera are based largely on characters culled from Eaton's Monograph, and these are but a few of the many characters therein given, and the value of these few as absolute distinctions of closely allied genera I have not personally tested. This group should furnish a most inviting field for some special student, especially here in North America, where it is so abund antly represented. In this group the independent specialization of the nymphs is extreme. Their life is relatively long, and the conditions under J^g. 10 Ventral view of male abdominal appendages of Bo- dy urns maculipennls Walsh, imago ; f, forceps ; i, in- ner appendages 52 NEW YORK STATE MUSEUM y ^^,xA which they dwell are very diverse. The conditions of adult life are, however, much the same in all; and we find the adults much more alike. The beginner will certainly find them much more (Hfticiilt to distinguish, and would do well to study nymphs and adults together. The critical diagnosis of the species will doubtless rest on the highly individualized genital arma- ture of the male. A suggestion of the strength and definiteness of the characters presented by these parts may be had from reference to the accompanying figure of the male for- ceps and inner appendages of Ecdyurus maculipennis (fig.lO). These project strongly from the ventral side of the apex of the abdomen, and are easily separated therefrom in fresh or Fig-. 11 Tarsal claws of nymphs of Heptageninae; w, of Hepta- genla Interpunctata Say; x, of Rhlthrogena elegantala Etn. ? ; y, of Iron sp ? from Coy Glen, Ithaca; z, of Ecdyurus maculipennis Walsh; hind claws la each case ; middle ones would be similar ; front ones some- times different Fig. 12 liabra of nymphs of Hepta- geninae; h, of Ironsp? from Coy Glen, Ithaca; i, of R h i t h r o g e n a e 1 e g a n - tula Etn. V; j, of Ecdyurus maculi- pennis Walsh; k, of Heptagenia interpunctata Say Fig. 13 Mandibles of nymphs of Hepta- geniuae; c, of Rhithrogena elegan- tula Etn. ?; d, of Iron sp? from Coy Glen, Ithaca; e, of Ecdyurus maculi- pennis Walsh; f, of Heptagenia inlerpunctata Say alcoholic specimens by a longitudinal snip with a pair of fine scissors. They may then be permanently mounted on a slide as microscopic preparations so as to give a square ventral view. It is from preparations so made that all the figures of the male appendages in this paper have been drawn. MAY FLIES AND MIDGES OF NEW YORK 53 The njmplis in this subfamily are recognizable at a glance by their dorsally placed eyes, with the lateral flaring margins of the sides of the head projecting beneath them. They are all strongly depressed also, and have lateral pectinations to the tarsal claws (fig.ll), aiding them doubtless in clinging to their -supporting surfaces! washed by currents of streams or waves of shores. Further than this, however, there is very great diversity among them, and E c d y ui r u s , Iron and R h i t h r o g e n a fur- nish a most interesting illustration of a special adaptation to life in torrents. In Ecdyurus (pl.lO, fig.3) the gill lamellae I 'Fig-. 14 Maxillae of nymphs of Heptageninae ; m, of I r o n sp ? from Coy Glen, Ithaca ;n, of Heptagenia interpunctata Say ; o, of Rhithrogena elegan- tula Etn. V; p, of Ecdyurus maculipennis Walsh are all divergent and the gill filaments are beneath their bases. In Iron (pl.lO, figs.G and 7) and in Rhithrogena (pi. 10, fig.4 and 5) the abdomen is more limpet-shaped, and the gill lamel- lae form a closely overlapping series whose outer border fits the supporting surface to which the nymph clings as closely as do also the flaring lateral and front marginsi of the head; but this is not all, the gills have migrated outward and now lie upon the bases of the lamellae, exposed on the outside to the stream of water which now dashes over, but does not flow beneath the lamellae. Furthermore, by the enlargement and approximation beneath the thorax of the foremost lamellae and by the depression and inward curvature beneath the tip of the abdomen of the hind- moist of them, there is formed beneath the abdomen a disk for 54 NEW YORK STATE MUSEUM adhesion to the surfaces of the stoues, fairly well developed in Iron, very perfect in R h i t h r o g e n a . So complete is its border that when applied to the surface of a stone, any elevation of the abdomen would create a partial vacuum beneath it. It doubtless serves in a different way the same purpose as the row of smaller discs possessed by the larva of the net-winged midge (B 1 e p h a r o c e r a), found in the same situations; and among jinatomical shifts for a living is one of the most remarkable known to me. The three figures of mouth parts of nymphs of the four genera hereinafter described (figs. 12, 13 and 14) show very considerable structural difterences. It is because of the remarkable definite- ness of such minute parts as the canines of , the mandible that I have been able to determine with certainty some of the elements of the food of nymphs of C h i r o t e n e t e s a 1 b o m a n i - c a t u s , its food being all reduced to very minute fragments. Heptagenia interpunctata Say This is the commonest species in Fall creek at Ithaca, with E . m a c u 1 i 1) e n u i s a close second. Both species swarm into trap lanterns set about the creek during July — mostly subimagos just risen from the water. During the daytime images are easily found sitting on the vegetation along the sides of the gorge. H. interpunctata is also a common species on the shore of Lake Michigan near my home in Lake Forest, being very abundant along shore in the latter part of August, associated with n . f 1 a V i p e n n i s Walsh, and a few scattering speci- mens may be taken as late as September. The rather well marked color pattern of the wing of this species is shown in the photograph reproduced in Plate 4, fig.4. The nymphs of this species are found in all the streams about Ithaca in rapid water under large stones. They are distinguish- able at a glance from those of all other species by the black markings of the under side, shown in pl.O, fig.3. The nymph, (ri.6, fig.3) . Length of full grown female nymph, 9.5 mm.; antenna, 2 mm., and setae, 11 mm. additional. Body strongly depressed, widest across the head, but with the sides behind the head parallel to the middle of the abdomen, thence tapering rather raipidly to the base of the setae. MAY PLIES AND MIDGES OF NEW YORK '55 Head flat, almoist orbicularly rounded, limuloid, the infero- lateral margins thin, flat, flaring, fringed with decurved hairs. Eves distinctly dorsal, the lateral margins of the head projecting beneath them. Antennae minute, hardly longer than the head, the basal segments ibrown and the remainder pale. Mouth parts as shown in figs. 12h, 13f and 14n. Prothorax with its declined and flaring lateral margins decur- rent upon the sides of the imesothorax ; legs short ; femora much flattened, and with AYell-developed posterior fringes of hair; simi- lar fringes on middle and hind tibiae, but scarcely developed of fore tibiae. Aibdomen rather short, strongly depressed, and much tapering beyond the 7th segment; lateral spines on segments 2-9, on 3-5 minute, on 6-9 strong, straight and sharp, longest on 7 and 8, the tip of that on 8 reaching the level of the middle of the 9th seg- ment. Segments of the abdomen diminishing slightly in length to the 7th, the 8th and 9th, then successively a little longer ; lOth produced in a rounded posterior lobe. Gills represented on segments 1-7, on 1-6 double, consisting of an anterior protecting lamina and a posterior basal one, mar- gined with respiratory filaments, whose tips are visible at the inner margin of the lamina. Gill lamella on 1 olblong, somewhat oblique, with a small lolie beside the basal attachment on the side next the median line of the body; on 2-6 similar, becoming somewhat more elongate and less oblique ; posterior lamina with its outer two fifths cut into a border of 1-2 branched respiratory filaments. On segment 7 there is a simple linear lanceolate fila- ment (representing the anterior lamina only) whose tip reaches the level of the apex of the 9th abdominal segment. Setae long, slender and very hairy for more than half their length, the hairs distinctly shorter externally; tips pale, whitish, ringed with darker and nearly destitute of hair. Coloration olivaceous or gi-eenish ibrown, darker on head on sides of prothorax and on dorsuan of abdominal segments 6 and 10. On the head there is a pale spot ibefore the middle ocellus, another one Ibetween each lateral ocellus and the eye, and an oblique pale streak extends from the eye to the margin below it. A pale, narrow' middorsal line extends from the rear of the head to the metathorax. The legs are pale, with t^vo broad light-brown bands on each of the' femora. On each of the exposed aibdominal segments is a transverse pale basal area which includes on each segment, except the 6th. a pair of brownish dots; these dots are elongated into longitudinal dashes on segments 8 and 9. Abdo- men beneath conspicuously barred with brown (pl.9, fig.3), one angulated bar on each segment, the bars interrupted in the 56 NEW YORK STATE MUSEUM middle on both basal and apical segments, but best defined apically. Bred at Ithaca 18th July, 1901. Heptagenia sp. no. 8 The nymph of this species was not bred. Like the two preceding it is strongly marked and easily recognized. It occurs in the larger streams^ clinging to rocks in the swiftest currents. The nymph. (P1.6, fig.4.) Length, 10 mm.; antenna, 3 mm., and setae 13 mm. additional. Body rather elongate, scarcely wider across the head than across the mesothorax. Head strongly depressed, evenly rounded in front, with flaring infero-lateral margins, dilated at the sides and distinctly visible outside the ejes. Antennae slender, pale. Prothorax slightly narrowed posteriorly, its margins flaring, dilated. Legs moderate, pale; femora with indistinct median and apical darker bands, and with a scanty development of the usual posterior fringes of hair. Wings reaching the level of the base of the 4th albdominal segment. AJbdomen regularly tapering posteriorly. Segments 8-10 slightly diminishing in length, 10 a little produiced on the dorsal side, and produced in a sharp triangular spine on each lateral margin; lateral spines developed only on segments 7-9, best developed on 8. Gills represented on segments 1-7, on 1-6 double, consisting of an anterior protecting lamella and a posterior respiratory lamella whose border is cut into a long fringe of branching gill filaments; upper lamella obliquely oval, produced at the tip into an acute spine-like point, becoming broader distally (obovate) on 4 and 5, and narrower again on 6. On segment 7 there is a simple linear hairy filament, dbtuse at the apex and hardly reach- ing the level of the base of the lateral s})ine on the 8th segment. iSetae 3, long and hairy, the hairs becoming whorled and finally obsolete toward the tip. The distinctive features of the color pattern are shown in the photographs reproduced in ])1.9, fig.4. These are a broad pale middorsal band divided with brown on the posterior abdominal segments, lateral pale dashes at the sides of the abdominal segments, and a brown longitudinal dash either side of segments 8 and 9 below. MAY FLIES AND MIDGES OF NEW YORK 57 Ecdyurus maculipennis Walsh As remarked under the account of the H e p t a g e n i a inter punctata, this species was found commonly in Fall creek, at Ithaca, Its larvae were more commonly found at the edges of the stream ; those of that species oftener in the current, but both often occurred together. This is a dainty little species with narrow wings, conspicuously marked with black on the crossveins (pi. I, tig.3). My bred specimens bear the dates July 9th, 13th and 14th, 1901. The nymph. (P1.7, fig.3.) Length, 7 mm.; antennae, 2 mm., and setae, 5 num. additional. Body strongly depressed, elongate triangular in outline, widest across the dilated, depressed and squarish frons, and rather regu- larly tapering posteriorly; head rather flat above; paired ocelli larger and more approximate above in the male than in the fe- male; antennae slender and short. Mouth parts as shown in flgs.l2j, 13e and 14p. Legs short, femora flattened, with a thin external fringe of hairs; tibiae slender and somewhat tapering; wing cases reach- ing the level of the apex of the 3d abdominal segment; abdomen rather short and slender, slowly tapering to the apex, middle seg- ments longest, segments 8 and 9 slightly shorter, 10 again longer on the dorsal side, but shorter at the sides and below; lateral spines present on segments 5-9, longest on 6 and 7, straight and sharp ; setae divaricate, the median one in the mature nymph more slender; the apical rings of brown on the segments of the setae are alternately ibroader and narrower, and the apical whorls of setae are excessively short. General color pattern olive brown, mottled with pale greenish, darker on head and prothorax, divided by a median narrow pale line, and varied upon the sides Avith pale hieroglyphics ; abdomen with pale and indistinct fenestrate markings along the sides. In the male there are broad dorsal blotches on the dorsum of seg- ments 4 and 5; in the female, on segments 7, 8 and 9. Iron sp? This species has not been bred. It is found in Coy Glen — a spring-fed stream near Ithaca, possessing a rich and peculiar fauna. Among our forms hitherto made known this species is peculiar in the possession of but two caudal setae in the nymphal stage. I have a number of nymphs collected years ago, from which, unfortunately, the date label has become detached. 58 NEW YORK STATE MUSEUM The nymph. (P1.7, figs.G and 7.) Length, apparently full grown, 9 mm.; antenna, 1.5 mm., and setae, 9 mm. additional. Body elongate, strongly depressed, widest across the front of the head and the meosothorax, these ibeing of about equal width ; head widest across the front well before the eyes, and strongly narrowed posteriorly to the obtuse hind angles; infero-lateral margins of the head thin and flaring as usual, and closely fringed with hairs; antennae short, slender, pale. Mouth parts as shown in flgs.l2h, 13d and 14m. Dorsum of the prothorax a little produced laterally, and angn- late obtusely in the middle of the sides. I^gs moderate, tibiae .and femora, all with well developed external fringes of hair; femora very moderately flattened and dilated, the fore femora most so; tarsal claw pectinate, there being two to four minute teeth at its anterior border before the apex (flg.ll?/). The wing cases reach the base of the 4th abdominal segment. Abdomen regularly tapering posteriorly, its segments increas- ing in length to the middle. Two lateral spines each side of segments 2-6 and one on segment 7, all stout, triangular, and directed outward. Gills represented on segments 1-7; on 1 a very large, broadly and obtusely triangular flap of membrane shaped like the gill scoop of a crawfish, attached by the middle of one of the sides, its front end extending forward and lying against the base of the hind leg, its hind end overlapping the succeeding gill lamella. On segments 2-7 the lamellae are ovoid, dorsally carinate, ob- tusely pointed membranous plates, each with regularly arcuate front margin overlapping the hind margin of the one on the pre- ceding segment, and each bearing at its base a tuft of 7-15 short, finger-like gill filaments. The lamellae diminish in breadth pos- teriorly, and 'become less divaricate in pairs, and the tips of the 7th pair are curved beneath the abdomen. Setae 2, rather short and stout, the median seta being repi"e- sented by a minute triangular rudiment. The tips of the develop- ing male forceps project beyond the apex to the 10th segment. This remarkable nymph dwells in the swiftest parts of the stream, and its whole organization exhibits the most wonderful adaptation to life in such a place; the extra grappling armature appended to its claws and especially its flattened form with thin edges nil the vay around closely applicable to the supporting surface, and admirably adapted to divert the flow of the water. Probably the oval enclosure of the gill lamellae of the ventral side of the abdomen acts as a sort of sucker, and holds the animal ]MAY FLIES AND MIDGES OF NEW YORK 39 securely to the rock surface. The net winged midge B 1 e p h a r - o c e r a is the commonest associate of this species in Coy Glen. This is another genus that has not hitherto been known east- ward of the Rocky mountains. Rhithrogena elegantula Etn ? For the sake of illustrating a still more perfect development of the ventral abdominal disk framed with gill lamellae, as well as illustrating the variety of form in this group, I insert here a figure and a brief description of a nymph from Twin Lakes, Colorado, sent me for study by ISIr Chauncey Juday, collected in the summer of 1902 : The nymph. (Pl.T, figs.4 and 5.) Length of full grown female nymiph, 10 mm.; male, 9 mm.; antennae and setae broken. Body short, stout, flat, narrowly elliptical behind the dilated head; head widest across the eves, semicircular in outline, its thin lateral margins naked; behind the widest portion the sides con- verg-e with very great abruptness to the hind margin; antennae short and stout and bare, the joinings of the segments becoming Oblique apically. Mouth parts as shown on figs.l2i, 13c and 14o. Prothorax three to four times as A^ide as long, produced at the sides in an olbtuse projecting angle; legs rather short and nearly bare, the femora moderately curved and flattened with a fringe of rather stifl:', very short bristles on the curving superior carina; each of the claws with a basal lateral tooth (fig.ll.r). Abdomen short and ovate; gill plates on segments 1-7 mem- branous, white, obtuse, closely superposed at their broadly over- la.pjjing edges, bearing copious tufts of long, simple gill filaments at their bases above. The anterior ends of the lamellae of the 1st segment meet beneath the nietathorax, and the incurved tips of those of the 7th segment meet beneatli the slightly upcurved tip of the abdomen. Setae in male 2, with a rudimentary middle one, in female 3 well-developed, bare, the median paler than the others; extreme bases of setae brown, like the general integimient of the body. Ephemerinae Since the publication of bulletin 47 I have made no new breed- ings in this subfamily, but my friend Mr W. E. Howard of Ottawa, 111., has reared and studied our P o 1 y m i t a r c y s a 1 b u s Say and has prepared at my request the following 60 NEW YORK STATE MUSEUM account of that interesting species, which differs in some respects from the well-known P o 1 v m i t a r c y s Virgo Oliv : Polymitarcys albus Say BY W. E. HOWARD This description was undertaken at the request of Prof. Need- ham, who identified the imago for me. Without his kind assist- ance in this respect, as well as in many others, it would not have been prepared. Nymphs of P. albus are abuindant in both the Illinois and Fox rivers at Ottawa. These rivers flow at this place over bot- toms of solid sandstone, with bars of loose sand accumulated in the eddies. The streams are swift in the main currents, and the nymphs of this species are to be found under flat stones at the edge of swift water when about ready to transform. It was from two such' situations that most of my collections were made, from which I succeeded in breeding a single specimen. I have seen the subimagos emerge and arise from the surface of the water in great numbers, but always just far enough out from the shore, so that the nymph skins were immediately swept into the current, where they disappeared before they could be procured. The difficulty in collecting the skins from the natural breeding places is further heightened by the emergence occurring during the evening twilight. According to my observation, not only this species but all others observed invariably emerge from the nymph skin at the surface of the water and leave the skin afloat. This makes the collecting of the sloughs a much more difficult task than in the case of stoneflies and dragonflies. My collections indicate that this is a midsummer species in northern Illinois. ]\Iy bred specimen is dated June 22. None of the imagos in my collections shows an earlier date than this, but I have nymphs which are evidently near to transforming which were collected the first week of June. Imagos and subimagos of the collections are scattered all through July, but August 5th shows them most abundant. At about this date they were ob- served in swarms. By the end of August they are much less numerous, and I have no collections which are as late as Sep- tember. MAY FLIES AND IMIDGES OF NEW YORK 61 The subimago stage lasts 24 hours, and when the final emergence takes place the subimago alights on some object near the edge of the stream, where it transforms in less than a minute. The skin of the subimago remains attached to the bases of the setae of the imago and in this manner is carried out over the stream by the flying insect, where it is finally released after some minutes. The adult of this species is briefly described in Eaton's Mono- graph, p.47. The habitat given there is Passaic river, Belle- ville, N. J. (Williamson); Winnipeg river (Say); Red river of the north and New York (Hagen). This seems to indicate a rather wide distribution for P . a 1 b u s in the eastern and northern United States, but during the summer of 1903 I made collections from several of the boulder and limestone streams tributary to the Wabash in Indiana without obtaining a single specimen. The nymph. Leng-th, 14-16 mm.; an-tennae, 3.5-4 mm., and setae, 7-8 mm. Body depressed, widest across prothorax where the thin lateral margins project; eyes prominent and lateral; three somewhat crescent-shaped ocelli arranged in the form of a broad-based triangle; antennae many-jointed, bearing a whorl of minute bristles at the apical ends of the joints, the first two joints much stouter and the joints 4-8 decidedly shorter than the others, projecting beyond the mandibular tusks by a little less than half the length of the latter; mandibular tusks about 2.5 mm. long, stout at base, narrowing rather abruptly near the mid- dle, the slender distal half tapering gradually to the acute, slightly out-curved tip; the basal half of tusk is thickly set with stout, acute spines, being less numerous on the slender distal portion, and entirely disappearing at about one third the distance from the tip; a few long hairs are borne on the outer side near the base; mandible stout, bearing two prominent tridentate fangs on its anterior surface nearly parallel with the distal half of the tusk, the middle tooth of each the longest ; the endopodite arising from the base of the inner fang is inclined toward the molar surface, and bears a brush of long hairs on its inner side near the tip; the laibrum is about half as long as broad, slightly emarginate in front, and thickly covered with fine hairs; maxillae somewhat slender, the outer basal portion fringed with stiff hairs; the maxillary palpi three-jointed besides the short pedicel, the second joint the shortest, the third joint about as long as the first and second together; outer side of third joint bearing a few long hairs, the stoutest ones being near the distal extremity, the 62 NEW YORK STATE MUSEUM inner side covered with fincu* ones; the first joint frinjjed witli sliort, stitf liairs on onter side onlr; tlie galea trnncate at the tip, which is densely fringed with line hairs, inner margin with fewer stout hairs, and the inner distal angle bearing a few stout spines ; the labium Avith three-jointed paljii, the large ovate galeae with their tips projecting a little beyond the tips of the small lanceo- late laciniae, both galeae and laciniae pilose. Thorax depressed, widest across prothorax where the pronotum is continued laterally into a wide, thin margin, each margin end- ing anteriorally with a prominent acute process. The wing pads reach to about the base of the third abdominal segment. The legs are rather long, the foreleg being the longest; the fore femora stand almost at right angles to the long axis of the body, the middle femora at about -15°, and the hind femora closely ap- pressed and nearly parallel with the body; the fore femur bears three or four rows of short spines on its anterior side, and a few stout hairs on its posterior side near the distal extremity; the fore tibia is long, and il^s inner side, as well as that of the tarsus, bears a fringe of long hairs ; the former ibears at its inner apical extremity a long, pointed process closely appressed against the inner side of the base of the tarsus; the inner sides of the fore tibia and tarsus both bear a numiber of small, sharp spines, being the most prominent on the tarsus and the apical process of the tibia. Abdomen long and gradually tapering from about the third segment, tenth segment about as long as wide. (Jills present on segments 1-7, each gill inserted on a lateral prominence bearing a minute tooth just in front of the gill base, the lateral promi- nence located just in front of the lateral, hinder angle of the segment, directed outward at an angle of aibout 45° to the long axis of the body; the first pair of gills small, single and spatu- late, with minute fringes, are curved upward against the body and are nearly concealed beneath the edges of the wing pads ; the other gills are double and shaped somewhat like a tuning-fork, the two branches linear acuminate, about equal in length, the outer branch bearing a rounded prominence at its basal end at the outer side; the respiratory filaments long, linear, about one third the length the gill lamina. Setae about half the length of the body, plumose throughout the greater part of their length, and then tail-pointed, the median seta not so stout at the base as the others. Colors of the '])ody chiefly brown, amil>or on legs and thin mar- gins of the body, an amber stripe along the dorsal median line of the abdomen ; a brow^n band on each femur near the distal extremitv, and a small brown blotch near the base of each. MAY FLIES AND MIDGES OF NEW YORK 63 NORTH AMERICAN HYDROPTILIDAE BY KENNETH J. MORTON, F. E. S. ( LONDON) ^>ome time ago Professor Betten sent to me several tubes con- taining a large number of examples of H y d r o p t i 1 i d a e collected at Ithaca, N. Y. The examination of these has re- vealed quite an unexpected wealth of species, so many indeed that they should foi'm a very satisfactory basis from which to start on the investigation of the North American forms com- prised in this Avoiiderful family. With the approval of Profes- sor Needham and Professor Betten, I have here undertaken the attempt to describe the species found at Ithaca, and further ad- ditional material has been received from Illinois (Needham) and from New Mexico (Oockerell). A few specimens from the col- lection of the late lamented Mr. R. Mc-Lachlan, London, have also passed through my hands. The condition of the material, preserved as it is in alcohol, puts a limitation on the character of the descriptions. The ex- ternal facies of the perfect insects is practically lost, most of the hairy clothing having been rubbed off. To a great extent, there- fore, the descriptions will be confined to the details of the J" genitalia, and even with regard to these I may add that much better preparations could have been made from dried specimens. After all the external facies affords very little aid in the determi- nation of the species, although in collecting these insects in a restricted area one learns to know them by their appearance. In the meantime only the males are dealt with. The difflcnlty of associating with the respective males the true females of say six species of H y d r o p t i 1 a found in the same tube is one that will appeal to any one who ha® ever made the attempt. Eaton's memoir on the family (Trans. Ent. Soc, Lon., 1873, pp.125-150) was the first which, dealt with these minute insect forms after the newer methods. McLachlan in his Monographic Revision and ^Si/nopsis of the Triehoptera of the European Fauna (London 1874-1880) gives a full general account of the family and good descriptions of the European species which were then known to him. As the best general account existing, McLacli- lan's work forms the basis foir the present paper as regards 64 NEW YORK STATE MUSEUM genera, etc. Siuce McLaehlan's work appeared, much lias been done towards obtaining a better knowledge of the family from different points of view, and our information is now miuh more detailed and exact with respect to the structure and morphologi- cal value of the so-called appendages, this improvement being largely due to the more extended use of microscopical prepara- tions in studying these insects; and something has also been learned concei-ning the early stages of the lives of these tiny creatures whicli, as larvae, construct curious habitations of most diverse forms. As contributors to this knowledge may be mentioned the names of Klapalek^ Ris, and the writer of the present notice. Quoting from McLachlan (op. cit. p. 503): " The minute insects comprised in this famih^ may be justly termed Micro-Trichoptera, for the largest European form expands to no more than 10 mm., the smallest to only 3^ mm. Where they occur they usually swarm in great numbers, running with extreme rapidity, and very diflQcult to capture. Although they occasionally appear to delight in warm sunshine, they are more especially crepuscular or even nocturnal, and are attracted by light to such an extent that the walls and ceilings of rooms near water are often dotted with these dark atoms which have entered by open windows." The largest measurements are attained in A g r a y 1 e a and A 1 1 0 t r i c h i a , no species of which is here described. The average expanse may be set down as 5-7 mm., the females as a rule being slightly larger than the males in a given locality. The following table of the genera is taken from McLaehlan's Revision and Sj/nnpsis of European Trichopiera, pp. 504-5, with the addition of a new genus which is described on p.72. A Wings broader, subobtuse, the posterior with no costal elevation or ex- cision. (Neuration tolerably complete; spurs 0, 3, 4 ; ocelli present and distinct) Agraylea B Wings narrower, often acute or subacute; the posterior with a more or less decided costal elevation followed by an excision c Spurs 0, 3, 4 h Ocelli present and distinct c Neuration tolerably complete Allot richia CG Neuration less complete d Wings scarcely acuminate I t h y t r i c h i a (Id Wings strongly acuminate O x y e t h i r a 66 Ocelli absent (wings acuminate) Orthotrichia MAY FLIES AND MIDGES OF NEW YORK 65 aa Spurs 0, 2, 4 (ocelli absent; wings scarcely acuminate; head ^Yitll elevated lobes posteriorly) Hydroptila aaa Spurs 1, 2, 4 (ocelli present, wings acuminate, but scarcely acute) Stactobia aaaa Spurs 0, 2, 3 (ocelli present ; wings acuminate) . .N e o t r i c h i a In ignoring the three species referred to in Hagen's iSynopsis, pp.27-45, I have followed a course far from satisfactory to myself, but the only one possible in the circumstances. Whether these species can now be satisfactorily elucidated depends altogether on the condition of the types. Mr. IMcLachlan informs me that the C y 1 1 e n e m i n u t i s - s i m e 1 1 a of Chambers i>ro\ed to be lepidopterous. Unless otherwise stated the locality is Ithaca, N. Y. ! Hydroptila The genus Hydroptila, as at present constituted, con- tains many Eurojiean species, and it is highly probaible that the North American forms will prove to be more numerous. A glance at the figures of the appendages will however serve to show that the genus is, in respect of these important parts, far from homo- geneous, and it will no doubt sooner or later be split up into two or more genera. The typical form may be taken as that represented by II y d r o p t i 1 a spars a Curt., the nearest American form being H . c o n s i m i 1 i s . In this group there is a large somewhat campanulate dorsal i)late, regarded by McLachlan as the united superior appendages. 1. Hydroptila consimilis u. sp. The (^ appendages as far as they can be described from the available specimens, are as follows : A large dorsal plate, convex above, notched on its outer margin, and the angles of the plate seen from above appear to be acutely produced, viewed from the side they are seen to be rounded. Side pieces of the last segment produced into somewhat curved subacute processes. Inferior appendages nearly parallel, blades slightly outturned at the tip which is pointed and blackened; the inner margin is oblique near the apex; a small wart on the inner margin before the apex ; the outer or upper margin bears a row of short spinous hairs. The penis considerably below the apex has a strong curved acute process. Ventral Jamina short. 66 NEW YORK STATE MUSEUM This Species, as has been indicated, is close to H . s p a r s a , but differs from that species in minor details^ and particularly in the process on the penis, this process in H , s p a r s a being smaller and nearer the tip of the organ. Ithaca, N. Y., and Belfrage, Texas. 2. Hydroptila delineatus n. sp. This si)ecies has the lobes on the head enormously developed. The antennae are about 31-jointed in the J*. The last dorsal segment appears to be deeply excised, the excision being followed by a large semitrausparent plate difficult to sep- arate from the basal portion of the appendages, but appiu'ently having a small submarginal projection on either side of a larger median one, but the outline of the plate is rather uncertain. The inferior appendages are nearly parallel, the apex outturned and upturned; the apical portion seen from the side has the lower angle somewhat produced while the upi)er part of the apex forms a kind of knob which in some positions shows a slight angle on its anterior side. Arising from the upper side of these append- ages is a long spirally curved acute process which winds itself round the under side of the apical lobe. On the outer (or lower) edge of the appendages, near their middle, is a slightly raised part bearing two or three spinous hairs. Ventral lamina short. This spOL'ies can be determined without difficulty from the ventral aspect of the appendages. It has no very near known ally. 3. Hydroptila spatulata n. sp. The dorsal plate in the ^ narrower than in consimilis, deeply notched in the posterior margin. Inferior appendages api)roximated at the base, slightly curved outwards and down- Avards, superior edge with a few spines; apex apparently scabrous. Side pieces of the last segment produced into rather long sub- acute processes. The penis, usually much exserted, Avith a flat- tened rounded apex, immediately below which is a small acute process placed at right angles. Ventral lamina very long and slender, slightly swollen at the apex when viewed from the side, the outer margin obliquely truncate, blackish and roughened. No verv near allv known to me. MAY FLIES AND MIDGES OF NE\y YOEK 67 4. Hydroptila haniata n. sp. The dorsal jjlate in the male is small, broadest in the middle, apex excised. Beneath it there is a small rounded penis-cover. The appendages are narrow, finger-shaped, aristate and widely divergent. Two strong hooks visible at the posterior margin of the last ventral plate or segment. The penis is very slender, curved at the apex and accompanied by a slender sheath of equal length. Ventral lamina very long; in lateral outline somewhat club-shaped, but lower margins oblique and minutely serrate. This species is certainly near to the European f e m o r a 1 i s , but it differs especially in the form of the dorsal plate which in f e m 0 r a 1 i s is not notched and is boat-shaped in outline. 5. Hydroptila perdita n. sp. Antenme about 31-jointed in the J^. The dorsal plate is large, rounded posteriorly, gradually becoming wider, the margins slightly retracted at the base, a small hollow looking patch near the apex. The inferior appendages subparallel blades, twisted outwards at the apex, which is obscurely bifid; scattered spinous hairs, particularly on external (superior) margin. Penis broad towards the base; in the exserted part tapering to a point, below which is a very strong curved acute process standing out strongly from the stem. Ventral lamina small. Ithytrichia The typical forms of this genus are rather robust-looking insects bearing considerable resemblance to the species of Hydroptila, but they are at once distinguishable on account of their distinct ocelli and the absence of the elevated lobes on the posterior part of the head. 6. Ithytrichia clavata n. sp. In the (^ the antennae are about 24-jointed. The appendages may be described as follows: There appears to be a transparent dorsal plate with outer margin nearly semicircular in outline, but the plate is very difiicult to separate from the other parts. There are very large side pieces of subtriangular form in the lateral aspect, blade-like if viewed from above, the apex down- turned and outturned rather acute and blackened at the extreme 68 NEW YORK STATE MUSEUM tip. What may be termed the inferior appendages are not dis- tinctly separated from the side pieces, and consist of a ventral plate apparently deeply slit; seen from the side the separate divisions of the plate appear as a rather long appendage slightly upturned and blackened at the point. The penis is club-shaped at the apex in one aspect; in another it is slightly excised at the apical margin and the club is hollow with a slender rod-like process lying within it. The penis seems to be formed of two joints, the apical jirobably capable of being partly retracted within the other; the latter joint is broad at the proximal part, but becomes constricted before the apex; a spiral sheath arises from about the middle of the organ. Beneath the penis is a transparent process out of which proceed two or more spines, and on either side of this central process are sometimes visible two minor ones. There is some uncertainty about the form of the dorsal plate, but the true form of this will be readily ascertained from prepara- ations made from fresh or dried specimens. Equally there is uncertainty about the cleavage of the ventral plate. These trans- parent menifbranous plates appear to suffer in form from immer- sion in alcohol. A species closely allied to I. lamellaris of Europe, of whose appendages no adequate figures have so far, been published. Tn I. lamellaris in the part corresponding to that which is called above " inferior appendages " there is, I believe, no slit, although a long, narrow part where the membrane is thinner sometimes gives nn illusory idea of a slit. In lamellaris the dorsal plate is produced in the centre with a blunt, slightly rounded lobe, while the process underneath the penis is about equal in breadth to the above-mentioned lobe, and instead of l)eing simple as in c 1 a v n t a it is subdivided Ivy an excision. The larva referred to in P.^j/rhr, vol. ix, pp.37.5-8, is almost cer- tainly that of the species just described. Whether the views ex- pressed in that paper are well founded or not can only be determined by a more complete knoAvledge of the life history of the creature, the working out of which sliould be sufficiently at- tractive even if it does not result in the verifiration of Professor Needham's views. MAY FLIES AND MIDGES OF NEW YORK 69 The larva of I t h y t r i c li i a was originally discovered by a Mr Bolton of Birmingham who formerly supplied living microscopical material for students, and it was noticed by me in Ent. Mon. Mag., 1st series, vol. xxiv, p. 171 (1888) ; it has also recently been described by Ulmer (Stett. Eutomol. Zeit. 1902, p.304). 7. Ithytrichia confusa, n. s]>. Antennae about L'8-jointed in the J^, long and slender, none of the joints really submoniliform, entirely fuscous, save a few basal joints, which are pale yellow. The last segment is open dorsall.y and within it lies a compli- cated organ represented in fig. 16. The appendages are closely approximated ventrally; in the side aspect they consist of a broad basal part, from the upper portion of which springs a long blade- like j)rocess. with rounded outei- margin. l>eset within with numerous spines or spinous hairs. The outer margin of the basal part is beset with strong incurved teeth. This species," which is very easy of identification, is referred l»rovisionally to 1 1 h y t r i c h i a , but the form of the append- ages and the long slender antennae isolate it from the typical species of 1 1 h y t r i c h i a . It is almost certain to be ultimately regarded as the ty})e of a new genus. Oktpiotrichia The iMiropean species frequent both standing waters and rivers. They are insects which measure from 6-8 mm. in expanse of wing, and McLachlan says of O. a n gu s t e 1 1 a that its antennae have about 39 joints in the J" and about 31 in the $. The most promi- nent features in the q genitalia are as follows: Large dorsal plate which is more or less asymmetrical, usually notched in its outer margin and with stronger rhitinized parts, ^^•hich assume the form of hooks or strong teeth. The penis is of very great length, very slender, apparently divided into two parts, the apical part which probably forms about ^ of the whole, being probably retractile (according to McLachlan, and I think he is correct). In Orthotrichia tetensii Kolbe. the side pieces of the last segment are asymmetrical, on one side being produced into a rather long, slightly curved sub-acute process, the other 70 NEW YORK STATE MUSEUM into quite a short process, while the inferior appendages are much incurved, also asymmetrical, one of the appendages having a large ante-apical swelling or blunt tooth. In 0 r t h o t r i c h i a angustella McL. the side pieces are apparently on much the same plan, but tlie appendages are strongly divergent. 8. Orthotrichia brachiata, n. sp. Although but one poor specimen existed in the collection, I have no hesitation in referring this to the genus Orthtrichia on its general characters. As a species it is critically near O . angustella and the two are almost certainly representative of each other on the two continents. In 0 . b r a c h i a t a the side pieces appear to be asymmetrical, but the specimen is some- what damaged in this respect. The penis is typical of the genus and is notched at the apex. Alongside of it is a very strong acute sheath and under it a cover with rounded apex. The inferior appendages are very strongly divergent, aristate. Above these (according to what is found in O. angustella) there should be two spiniform processes slightly divergent and terminating in a hair, but in the specimen before me these can not be made out satisfactorily. The margin of the last ventral segment has a small chitinized median plate broader anteriorly and with excised fore and hind margins, the posterior projections each bearing a short bristle. The dorsal plate is as usual unequal on its two sides, but I can see no trace of teeth, although portions of the margin are more or less strongly chitinized. It is to be regretted that the material is so unsatisfactory, but the figures should serve for identification. OXYBTIITRA This genus, conspicuous on account of its narrow wings, has proved one of the most prolific of species in the European fauna. No estimates can be placed on the number of species that may eventually be found in North America ; the possibilities are boundless. 9. Oxyethira coercens, n. sj). Antennfe in tlie rT about 31-jointed. The last segment dorsally is semicircularly excised when viewed from above; seen from the side the edges are strongly MAY PLIES AND MIDGES OF NEW YORK 71 toothed. The ventral plate, if viewed from the side, is in the form of a strong claw with serrate margin; from beneath, its margin is excised rather deeply, and each of the side lobes has about three teeth. On either side the apex of a blackish rod is usually visible. The penis has the apex much dilated and armed with two or three teeth. This species, which belongs to the same group as the European O. f r i c i Klap., should be very easily recognized from the ventral plate. 10. Oxyethira viminalis, n. sp. The margin of the last dorsal segment is slightly and simply excised. Beneath, the margins of the segment recede and form a deep excision. The ventral plate has the margin nearly straight^ and above it are two blackish appendages^ — no doubt the homologues of the hooks which lie above the ventral plate in several European forms. The penis is large, and with its sheaths forms a very complicate structure; the apex, however, is simple. The only other appendages visible are a pair of lateral rods. Described with great hesitation from a single (^ from Ithaca, but afterwards fully verified from further material from Prof. Needham (Lake Forest, 15 Oct. 1902). 11. Oxyethira dualis In this species the appendages are very simple; seen from above the penultimate segment is deeply excised; from the same point of view is visible a quadrate penis cover, the posterior margin of which has two slight excisions. From beneath is seen the simple ventral plate, almost straight in its posterior outline, the hind angles slightly oblique; the plate is rather broader at its base and on each lateral margin is a faint angle. The apex of the penis has an elongate swollen part, probably exaggerated in figs. 37 and 38, through the presence of some extraneous matter; after preparation this part had the appear- ance of having two slightly chitinized straight rods and a sickle- shaped one. 1 (^ Las Vegas, New Mexico. (Cockerell.) <2 NEW YORK STATE MUSEUM Anotlier larger female, probably also an O x j e t h i r a , was in the sending from JSew Mexico, but I can say nothing definite about it in tlie absence of the J'. NeOTRICHIA, NOV. GEN. Bpurs 0, 2^ 3. Ocelli present. Head posteriorly with two large pyriform warts; disc elevated with a median longitudinal line. Antenme with 18-19 joints (probably in both sexes), joints shorter and more submoniliform in the $; rather stout, basal joint moderately long and slightly curved, second joint some- what shorter, the four following subequal and cylindrical, the remainder submoniliform. (Palpi uncertain from the material in hand.) Legs long and slender. Neuration apparently simpler than in any of the other known genera. (Compare the figures.) 12. Neotrichia collata u. sp. In the (^ the apex of the abdomen may be described as fol- lows : Above there is a semitransparent transverse plate, at either side of which are two rather longer thin processes which seem to be slightly out-turned at the apex. There are also semitransparent side pieces rather slender and slightly curved. The inferior appendages consist of two parallel contiguous proc- esses which at the apex on their inner edges are obliquely truncate and bear one or two teeth. The penis in its apical portion consists of two closely lying parts of nearly equal length, slightly hooked at the tips. A minute and exceedingly interesting form. The figures were rather difficult to obtain from the material received, but the important characters are set forth and will serve sufficiently for identification. From New INIexico a large number of interesting cases be- longing to tAVO or probably three species have been sent. One is a veritable O x y e t h i r a , while the others belong probably to H y d r o p t i 1 a , but I am not sure whether the differences shown by some of the cases are due to individual variation or point to the presence of two species. MAY FLIES AND MIDGES OF NEW YORK 73 SUPPLEMENTARY NOTE In the interval that has elapsed since the above paper was written, Mr. Nathan Banks has published two papers which con- tain references to North American IJ y d r o p t i 1 i d a e (two species of Hydroptilida?, Ent. News, April '04, p.ll6; and A List of Neuropteroid insects, exclusive of Odonata, from the vicinity of Washington, D. C, Proc. Ent. Soc. Wash. Vol. vi. No. 4, pp.215-6). Tliese may give rise to questions of synonymy, but at t*r Fig. 15 Hydroptilid structures: A g r a y 1 e a in u It i p u n c t a t a ; 40, apex of abdomen from beneath, 41, same from side, 42, triangular pieces from side, 43, same when much exserted, 44, ventral lamina; Ortliotrichia americana (?) 45, apex of abdomen from beneath. the moment I am unable to deal with the subject which I think may be safely left to be cleared up by American students. In the same interval some further material has come to hand from Professor Needham and Professor Betten, but this has been only partially examined. It includes an A g r a y 1 e a which ap- pears to be the same as the European A. m u 1 1 i p u n c t a t a , and an Ortliotrichia which I had described under the name of O. c r i s t a t a but which may be O. a m e r i c a n a , Banks. There is also another good species of H y d r o p t i 1 a , but the material is hardly sufficient for description. 74 NEW YORK STATE MUSEUM Professor Needham bred the A g r a y 1 e a and sent me the larva and cases. The metamorphoses of A g r a y 1 e a and many other European forms have been described recently in an admira- ble way by Herr A. J. Silfvenius of Helsingfors. Agraylea multipunctata Curt. McLachlan, Rev. & Synopsis, p.SUG, describes the species as follows: Antennae blackish fuscous. Body blackish fuscous; abdomen greenish in life; blackish in dry examples with pale lateral lines; the ventral surface clothed with silky yellowish hairs. Head and pronotum clothed with greyish yellow hairs. Legs subtestaceous with fuscous femora; clothed with pale hairs. Anterior wings ordinarily blackish, with numerous distinct golden-j'ellow mark- ings, mostly forming spots, but usually there is also a long and broad space near the apex of the costal margin, about two elon- gate spaces on the inner margin, and two or three apical spots extending into the fringes which are otherwise dark (but these markings are very variable; individuals occur in which they are entirely absent, the wings then being wholly blackish, or in which they invade the whole wing, obliterating the dark ground and causing the insect to appear pale cinereous). Posterior wings cinereous grey, with concolorous, slightly iridescent fringes. In the (^ there appears to be a bilobed shining yellowish plate under the margin of the last dorsal segment. Superior (interme- diate?) appendages in the form of two contiguous yellowish bands, very strongly curved downwards; from between them escapes the penis, which is dark, straight, iipdirected, its apex dilated and truncate. Inferior appendages, viewed ventrally, very distant, upcurved, yellowish, subcylindrical, but they are apparently connected with two upcurved contiguous median processes, seen from the middle of the ventral margin, more slen- der than the appendages and blackish at the tips; internally, on either side of these, is a triangular piece. Lobe of the ante- penultimate ventral segment long, fiatlcncd and adpressed dilated gradually to the apex, w^hich is shallowly excised and narrowly blackish ; the colour otherwise testaceous ; at the base of this lobe is a subtriangular blunt tooth. In the $ the abdomen ends in a stout upcurved ovipositor, and there is the usual small sharp tooth on the antepenultimate ven- tral segment. Expanse 71/2-9 mm. MAY FLIES AND MIDGES OF NEW YORK 75 The foregoing description was made no doubt entirely from dry examples. The words italicized and the stouter form of the in- ferior appendages are almost the only characters which served to separate the closely allied A. cognatella, a very critical species regarding which I have some doubt. The superior (intermediate?) appendages of McLaehlan, de- scribed by him as in the form of two contiguous yellowish bands very strongly curved downward, appear to be separate at the base only, the downturned tip being rounded. Between the divided base escapes the penis, which is not shown in the figures here. The construction of the socalled triangular pieces is very peculiar and their appearance depends much on the degree to which they are exserted. The North American insect is practically identical with the European; the only difference appears to be in the upcurved con- tiguous median processes which are probably rather shorter and stouter in the American form. The latter was bred by Needham at Lake Forest in June, and he sent the same species from Chic- opee Mass. (23 April '03). Tt will probably prove to be common in the Northern States. Orthotrichia americana Banks ? (P]nt. News 1904, p.ll6) 0. cristata Morton, U. S. The dorsal plate in the ^ is almost entire in its outer edge; from tlie one side, on the upper surface, arises a short hook, while on the other side, nearer the middle, arises a long, slender, much- curved acute process bent strongly inwards under the plate itself. The last ventral segment terminates in two separate blackish plates which are broadest where they approach each other, nar- rowing outwards, their apices bearing a spine; between these plates is a black appendage which seen from the side is hook- shaped and bears two spines. The seventh ventral segment with a process covered with strong thick blunt hairs ; the sixth segment with a small tooth. Lake Forest, 111. (Needham). The process on the seventh ventral segment is very similar to that found in the European 0. t e t e n s i i . 76 NEW YORK STATE MUSEUM AQUATIC NEMATOCEROUS DIPTERA II Chironomidae BY OSKAR AUGUSTUS JOIIANNSEN The present work forms a continuation of the paper entitled "Aquatic Neuiatocerous Diptera," published in bulletin 68 of the New "York State Museum (1903). In that paper the Blepharoceridae, Simuliidae, Culicidae, and the D i X i d a e were treated. In this paper the Chironomi- dae will be considered, the classification reviewed, the chirono- mid genera of the world, described and discussed, and finally de- scriptions given of the imagoes of all known North American species except those belonging to the group Ceratopogon (sens. lat.). Descriptions will also be given of all larvae and pupae as far as known. The bulk of the material studied was collected in New York, but many specimens were obtained from the Mississippi valley. Rocky mountain and Pacific coast states. In drawing up the descriptions of the species upward of 5000 pinned specimens and much alcoholic material was studied. In order to save space the references to works of authors are abbre- viated, only the author's name followed by a date is given, the complete reference being gi^en in the bibliography in the back of the book. The stud}' upon this family of flies was begun in the spring of 1901 and was continued throughout four sea- sons. The work was done in the entomological laboratory of Cornell university under the direction of Professor J. H. Com- stock, to whom I wish to express my thanks for advice in the preparation of this work. I am also under obligations to Pro- fessor Kellogg of Leland Stanford jr. university. Professor C. O. Houghton of Delaware agricultural college. Professor Aldrich of Idaho, Mr. A. L. INIelander of Washington State, and Doctors Mac- Gillivray and Riley of the instructing staff of Cornell university; and especially to Professor J. G. Needham of Lnke Forest uni- versity for specimens and many favors. The family of the Chironomidae or midges comprises a large numter of very delicate, and often minute flies, of which over 800 species are known throughout the world. They resemble mosquitoes in some respects, but are usually more delicate, and MAY FLIES AND MIDGES OF NEW YORK T7 may be distinguislied from them by their wing venation. These midges are often seen, especially in early spring or autumn, in immense swarms dancing in the air, and are frequently to be found at these seasons upon the windoAvs of dwellings where they are often, perhaps usually, mistaken for mosquitoes. Professor Williston relates (1806) that over meadows in the Rocky mountains he has seen them rise at nightfall in most incredible numbers, producing a humming noise like that of a distant waterfall, and audible for a considerable distance. Most of the species are inoffensive or actually beneficial as scavengers. The group C e r a t o p o g o n , however, forms an exception, some members of which, known as sandflies, or punkies, have the power of sucking blood, and are particularly troublesome in the mountains, along streams, and at the seashore. The Chiro- nomidae are very widely distributed, being ap]>arently as prevalent and as numerous in the frigid as in the torrid zone. There are about 500 Eiu'opean species, many of which were described by Zetterstedt, from Sweden and Lapland. Of the species hitherto described from North America over one third are from Alaska, Gi'^eenland and Hudson bay territory. A surprisingly large num- ber of species are common to Jioth Eimape and North America. When the fauna of Asia, Africa and South America is as well known the total number of species will doubtless be increased many fold. Geological distribution One would scarcely expect the delicate, minute flies of this family to be preserved from mesozoic times, yet they seem to be not altogether unknown. Two species referred to Macropeza are figured, one by Geinitz from the Lias of Dobbertin and one by Erodie from the English Purbecks ; two other obscure forms from the English Purbecks are figured under the name of C h i r o n o - m u s ; and Corethrium per tin ax and Cecidomium g r a n d a e V u m of Westwood, from the same beds, appear to belong to this family rather than to the C e c i d o m y i i d ae or C u 1 i c i d a e . R h y p h u s p r i s c u s Erodie, from the r^nglish Purbecks, also probably belongs here and not to the Rhyphidae. • 78 NEW YORK STATE MUSEUM The family is very abundant in amber, Loew having found seven species of Tanypus, more than forty of Chironomus and twenty-six of Ceratopogon. Giebel also describes two species of Chironomus and one of Ceratopogon in amber, and these genera had previously been recognized as occurring there by Burmeister, Erickson and others. Duisberg also records a peculiar genus, S e n d e 1 i a , from the same. But the occurrence of the family in a fossil state is not confined to amber ; thirteen species of Chironomus have been described from Rott, Oeningen, Rodoboj, and Utah, and the genus has been recognized also in Wyoming, while numerous pupae distinguishable as belonging to several species are recorded by Heyden from Rott. Ceratopogon has also a species at Rott, and it has been recognized at Aix and in Sicilian amber. Numerous specimens of the family occur at Florissant, but they are usually in very poor condition; they have also occurred in the British Columbian tertiaries. Scudder (1886). Economic importance The only function of the imago of the Chiromomid, at least in the group Chironomus, and perhaps Tanypus also, is that of reproduction. Miall and Hammond (1900) say, " It is evident that Chironomus does not feed in the winged state. The mouth parts, though of elaborate structure, are never used in feeding, and the alimentary canal of the fly is empty, except for a greenish fluid, which fills the stomach of the pupa and newly hatched fly. . . . The larvae of Chironomus feed on dead leaves and other vegetable refuse. Microscopic examina- tion of the contents of the stomach reveals a blackish mass of vegetable fragments, besides Diatoms, Infusoria, eggs of other aquatic animals and grains of sand." Some species of the group Ceratopogo n are blood suckers and their mouth parts as figured by Professor Kellogg (1899) seem admirably adapted to this function. The larvae and pupae of the Chironomidae are of much importance as fish-food. Professor Needham (190.3, p.204) men- tions the fact that large numbers of the larvae of a species of Chironomus were taken from the stomachs of brook trout, MAY FLIES AND MIDGES OP NEW YORK 79 proving that these fish live almost exclusively upon " blood- wonus/' at certain seasons at least. Garman (1888) says : '' Probably no other one genus of insect constitutes as important an item in the food of as large a number of fishes." While Forbes (1877) in giving a list of the organisms which form food of fishes records (J h i r o n o m i d a e as occurring in the stomachs of many species. Enemies Besides the fish which devour vast numbers of Chironomid larvae, the nymphs of dragonflies, caddis worms, Perla, Sialis, beetle and other predaiceous larvae constantly prey upon them; while the adults are eaten by dragonflies, by the net-winged midges (Blepharoceridae) and other predaceous insects. In a swarm of these midges very often one also sees a number of dancetiies (E m p i d s) constantly seeking victims. Methods of capturing, rearing land mounting^ Sweeping the low brush, rank grass, and herbage along the banks of ponds and streams is the usual way of capturing these flies, and often in a favorable location hundreds may be caught in a few hours; but the most satisfactory manner of catching is by means of a cyanide traplantern. such as is described by Professor Needham (1901, p.oO^)/ By means of it several thousand specimens may be taken in a single night. The most favorable time for setting the lantern is a sultry, cloudy night, during the summer or spring; and the most favorable location is near the bank of a pond or creek. The larvae and pupae and sometimes the eggs also may be scooped from the bottom of the pond by means of a small coffee- strainer net; or swept by means of a brush into a cloth sagnet from the surface of the rocks at the bottom of the shallow creek as described by Professor Needham (1899, p.5). From thence they are transferred to jelly tumblers, or for those forms which require rapidly flowing water to a jar from which the water is drawn as rapidlj^ as it enters by means of a continuous siphon as described by Professor Oom-stock in " Insect Life," p.330. If the larva is nearly full fed, but a short time will be required for it to transform. If the specimens are still quite small, some 80 NEW YORK STATE MUSEUM dead leaves and rubbish may be put in the glass for them to f e( d ni)ou. The larvae of C h i r o u o m u s usually hide them- selves from view, and in the mud and debris form tubes which open at the surface. When placed in a jar their chief anxiety is to bury themselves in the mud, and very soon they will gather bits of dead leaves and particles of sand about them, binding them together Avith viscid threads passed out of the mouth, and in a short time will be completely concealed in a rough tube. These tubes are frequently seen upon the surface of dead leaves, on stones, sticks, etc. One species is known to be a leaf miner (Pettit, 1900). The larvae of species belonging to the groups Oeratopogon and T a n y pus usually do not form tubes, but remain free. Specimens captured in the fall may live all winter and not transform until spring. Some specimens of Oeratopogon taken by the writer in September lived until the following May in the larval stage, and it is probable that they live thus nearly a year. The larvae may be found all the year around, while the adults are common excepting in the dead of winter, and a few specimens may be found even at that season. The larvae are best preserved in alcohol, either first killing them in hot water, or placing directly in the alcohol. If any peculiarity of color marking is observed it should first be noted, as the s])irits soon remove much of the pigment. The adult should be mounted with great care, either upon an elbow pin (^lacGil- livray, 1903b), or upon a minuticn nadel, a short and very slender headless pin, thrust through a small piece of cork or pith and then into the under side of the thorax of the fly. Through the other end of the cork an ordinary insect pin is placed, and the specimen is then ready for the cabinet. It is also veiy desirable to preserve some specimens of each species in alcohol, or better still, in a mixture of formaldehyde and glycerine. The latter preserves the original colors quite well, especially if kc]it in the dark. These specimens should be put in very snmll vials and should, of course, bear the same number label as the pinned specimen. The fore tarsi of the genus C h i r o n o m n s are very easily broken oflf, and therefore it is quite necessary that great care should be observed in preserving them since their presence is necessary in Jlie determination of MAY FLIES AND MIDGES OP NEW YORK 81 some of the species. It is the practice of tlie writer to remove oue wing and all the legs from one side of at least one specimen of each species and to mount them (preferably dry) upon a slide; this method allows of ready measurement and comparison. Characteristics of egg, larva, pupa, and adult The adults may be characterized as follows: More or less mosquito like in form, seldom reaching ten millimeters in length. The head is small, somewhat compressed, palpi usually four-joiuted; proboscis short; antennae of variable length, from six to fifteen jointed; the first joint disk-like, the last one elongated, the male antennae usually plumose. Eyes reniform or oval; ocelli rudimentary or wanting. Thorax highl}' arched, frequently projecting over the head, without transverse suture; scutellum small and hemispherical; metathorax well-developed. Abdomen long and slender, eight- jointed, the hypopygium pro- jecting forceps-like; ovipositor but little developed. Legs usually long and slender; coxae moderately long; tarsi frequently very long. Wings either bare or hairy; the veins of the costal margin being stout and in marked contrast with those in the other ])art of the wing, which apjear to be fading out. Venation variable. The larvae usually have blood or tracheal gills, and are soft- skinned and worm-like. The pupae are free, some are active and resemble C u 1 e x , others float upon the surface of the water and still others remain at the bottom of the pond until ready to emerge. The flies with which they might be confused are the crane flies and the mosquitoes. The former (T i p u 1 i d s) are usually larger, have proportionately longer legs, have more numerous and more distinct wing-veiuis, and have a V-shaped suture upon the dorsum of the thorax. The latter ( C u 1 i c i d s) have scales upon the wing, and all, excepting the subfamily C o r e - thrinae, have an elongate biting proboscis. The adults of the genus C h i r o n o m u s have a peculiar habit of holding their fore legs high above the surface upon which they stand, while the mosquitoes usually hold up their hind legs. There is one other family of flies, the Stenoxenidae, which must be distinguished from the Chironomidae. This 82 NEW YORK STATE MUSEUM peculiar family lias but a single genus, a single species, and is represented by but a single specimen, which is now in the United States National Museum. This family will fall in the couplet with the Chironomidae in the key given by Oomstock (1895), and by Williston (1896), but dill'ers from all the members of this family by its peculiar wing venation (pl.oS, fig.29). The description of the family was tirst given by Mr. Coquillett (1899a). The eggs The eggs of most of the Chironomidae are deposited in water; some species in swift flowing water, others in sea or lake water, while most of them lay them in stagnant pools or ponds, or in slow flowing streams. The eggs of a few species are deposited in bark, in manure, and in debris. Some species lay them in strings resembling somewhat a minature string of toad's eggs; while others lay them in clumps. The eggs themselves are elongate, cigar-shaped, usually pointed at each end. They hatch in a few days. The larvae The larvae are worm-like, but vary somewhat in form with the different genera. Most of them are aquatic, while a few live in the earth, in manure, or under bark (pl.KJ, tig.4 ; pi. 17, figs.l and 4;pl.l9,fig.9). Excepting some of the C e r a t o p o g o n they are provided with both thoracic and anal prolegs, and move by creeping in a manner somewhat like a geometer larva, without, however, such regularity, nor does the middle section hump up, but loops irregularly to one side or the other. Many species are blood-red in color, and hence are frequently known as blood-worms. They have a distinct head with well-formed labrum, labium, epipharynx, hypopharynx, mandibles and maxillae, the mandibles moving in oblique planes. The body is distinctly segmented, usually with twelve joints besides the head, the three thoracic segments being but little thicker than those of the abdomen. The twelfth seg- ment is provided with a pair of prolegs, some caudal setae and blood gills; sometimes there are two pairs of consjiicuous blood gills upon the ventral surface of the eleventh segment also. The MAY FLIES AND MIDGES OF NEW YORK 83 terrestrial larvae of Ceratopogon have prominent spines and setae upon the bod}-, vihile the aquatic forms are nearly devoid of them (pi. 17, figs.l and 4) . The aquatic larvae of C e r a- t o p o g o n have no prolegs and the body is very slender and snake-like. The larvae can exist at great depths, and have been hshed up from the bottom of deep lakes. They have been found in salt water (Packard, '70a). The pupae The pupae of C h i r o n o m u s are frequently found in the old larval cases; others swim very freely near or at the surface after the fashion of a pollywog. The pupa of T a n y p u s is active and resembles that of 0 u 1 e x in form and habit. The pupa of the aquatic Ceratopogon is more elongate than T a n y - pus, is not active, and floats, nearly motionless, in a vertical position. All of the pupae have an enlarged, thorax and usually a pair of respiratory tubes or filaments, while the caudal end is somewhat broadened and paddle-like or prolonged into two- pointed lobes, with ciliate margin. The imagines To the imaginal characters of the family already given the fol- lowing may be appended : The head is small, spheroidal, flattened where it joins the thorax, in some genera somewhat hollowed out between the eyes. The compound eyes are large, with conspicuous facets and dis- tinctly separated from each other. They are kidney-shaped (reni- form) ; that is, hollowed out around the base of the antennae. The ocelli are wanting. The front, the space between the eyes, is limited by the upper margin of the head and a line drawn through the root of the antennae. The vertex is the uppermost part of the front, near the margin of the occiput. The face is the portion below the antennae, which is prolonged more or less downward to form the proboscis. The oral margin and an indefinite space immediately contiguous to it is called the epistoma or peristoma. The e])istoma is usually convex, provided with setae or sensory hairs. The maxillary palpi are the slender, usually four-jointed appendages, the most conspicuous of the mouth parts. The labrum, hypopharynx and labium differ with 84 NEW YORK STATE MUSEUM the different genera. In C li i r o n o m u s and allied genera there is no trace of mandibles. In C e r a t o p o g o n the mouth jiarts are fitted for piercing. For homologies of the mouth parts, see Kellogg's papers in l*syclie, li^VA). The antennae or feelers are variable in form and number of joints ; the hrst visible joint (called L*d joint b^' Miall and Hammond liJUO) is usually enlarged, fol- lowed by a second which is sometimes also somewhat enlarged, these two being called the scape. These are always more or less differentiated from the remainder, which constitute the flagellum. In the male the joints of the flagellum are usually provided with long hairs. The first joint of Miall and Hammond (1900) is the extremely short hidden one, which is sunk in the head, and almost entirely occupied by the muscles which move the antennae to and fro. The next joint, the large one, exhibits a peculiar structure, which is believed to serve for the perception of sound, (M. and H. 1900, and Mayer 1874.) The head is connected with the thorax by a neck, Avhose cuticle is membranous. The thorax is composed of three parts, the prothorax, the meso- thorax and metathorax. The prothorax is (piite narrow, forming a rounded collar back of the neck, within which are the muscles of the foreleg. On the dorsal surface it appears as a narrow band with a median incisure and suture. The humerus or humeral callus belongs also to' the prothorax according to Miall and Ham- mond (1900). it is called. the ])aratreme by Lowne. The meso- thorax is ver}- large; it is highly arched, and in some it projects somewhat over the head. On its fore edge is the anterior thoracic spiracle. The ui)i)er or dorsal surface of the mesothorax is often called the mesonotum. and it has attached to it at its posterior margin, and cut off" from it by an impressed line, the scutellum, a small, semioval body, which really belongs to the mesothorax (see 1)1.31, fig. 10, C h a s m a t o n o t u s). The wings are attached to each side of and just below the scutellum. Behind and beneath the scutellum is a smooth and rather prominent oval-arched portion, the metanotum or upper portion of the metathorax (or post scutel- lum of Miall and Hammond 1900) . Below and between the fore and middle legs is a very prominent hemispherical part (especially in Chiro nonius), the mesosternum. The sides of the body in front of the wings are called the pleura, and the under surface of MAY FLIES AND MIDGES OF NEW YORK 85 the thorax as a whole is called the sternum or pectus. The inter- niediate legs are attached to the hind part of the mesosternum by oval sockets. The metathorax is much smaller than the meso- thorax. Its dorsal surface, called the metanotum, has already been mentioned. On the side is the posterior thoracic spiracle, and above it is the haltere (balancer or poiser) the rudimentary hind wing, a slender organ with a dilated hetid. The ventral sur- face of the metathorax is short and narrow and is largely occupied by the insertion of the hind legs. The abdomen is comijosed of nine segments more or less closely fused together. In the male especially it is long and slender and terminates with the genitalia. The genitalia varies greatly with the different genera (pis. '.'2. oo). The anus opens on the dorsal surface of the ninth segment. The under surface of abdomen is sometimes called the venter. The three pairs of legs are long and slender, especially so in Chironomus, are attached to the jirothorax, mesothorax and metathorax, and are called respectively the front, middle and hind pairs. The older writers who used the Latin terminology spoke of the forelegs as pedes antici, the middle legs as pedes medii, and the hind legs, pedes i>ostici. ^Vlieu they spoke of the fore and middle legs together they called them pedes anteriores. Some writers still use the term anterior legs for fore and middle pairs, and posterior legs for the middle and hind pairs. The coxa is the part attaching the leg proper to the thorax; while the trochanter is the short, small, ring-like portion between the femur and coxa. The femur or thigh is the stoutest portion of the leg; the tibia is the next part succet^ding the femur. The tarsus is the distal division of the leg and is coanposed of five joints, of which the first, that next the tibia, is called the metatarsus. The ungues or claws are two booklets on the underside of the last tarsal joint. In most genera these are simple, but a few have uniserrate or bifid claws. The pulvilli, two pad-like fleshy cushions attached to the last joint of the tarsus below the claws, are often })resent. The empodium is a median appendage between the claws, and is usually present also. The usual shape in this family is that of a sickle-shaped process, pectinate on the convex side. 86 NEW YORK STATE MUSEUM The wings are usually rather slender and delicate, with the anterior veins (those nearest the costal border) rather stout, while the posterior veins are usually very delicate and indistinct. The surface of the wing is delicately hairy in a number of species, though the majority have bare wings. Below is given the Comstock-Needham terminology of wing venation as used in this paper, together with the equivalent terms of the Schinerian system as applied with but slight modifications to the Nematocera generally : Comstock-Needham Schinerian Costa (C) = Costa Subcosta (Sc) = Auxiliary Radius (RJ= First longitudinal R.^,3 = Second *' R,+3 = Third Media (M) = Fourth " Cubitus (Cu) = Fifth Anal (A) = Sixth The costa ends at or before the tip of the wing in all the genera; the subcosta, though sometimes rather indistinct,- is usually present; the radius is stout and well developed, and usually with two or three branches, R, in some genera appear- ing like a crossvein; the media usually present and always sim- ple; the cubitus is nearly always two-branched; anal vein usually present though delicate. Compare pi. 17, figs.lo-lG, pls.27 to 31. The halteres, the slender organs with knobbed ends which are supposed to be the rudimentary second i>air of wings, are rarely wanting. For a description of the internal anatomy of both larva and imago the reader is referred to Miall and Hammond's work on the harlequin fly (1900). A large number of genera have been erected to contain the species of the world. Of these some may be placed as the synonpus of others, leaving still over 40 valid genera. In order to facilitate identification a key to the North American genera is offered besides the more general one for the genera of the world. MAY FLIES AND MIDGES OF NEW YORK 87 KEYS TO GENERA OF THE CHIRONOMIDAE Larvae a Abdomen with prominent rounded elevations or cushions, with rows of teeth on the inferior (anterior) angles of the segments 13. P s a m a t h i o m y i a aa Abdominal segments without these cushions b Aquatic footless snake-like larva, or terrestrial larva with thoracic and anal feet, and many setae and bristles on body segments, pi. 17, figs. 1 and 4 ( Group C e r a t o p o g o n) hb Not as above c With retractile antennae, the hitter often quite long, long stilt-like legs, the caudal tufts of hair mounted on cylindrical processes, pl.l9, fig.9 ( Group T any pu s) cc Not with all the above characters d With the two caudal hair tufts mounted on cylindrical projec- tions r With six seta-like processes on each of the caudal projections, three long aud three short. Eyes prominent, round, on ante- rior angles of the liead, pi. 34, figs. 21, 22, 23 (European) 26. W u 1 p i e 1 1 a cc Eyes not on anterior angles of head f With blood gills on venter of eleventh segment 31. H y d r 0 b a e n u s ff With blood gills only at end of twelfth segment 44. M e t r i o c n e m u s dd Caudal tufts on small rounded papillae e Antennae elongate, at least one half and often as long or longer than the head ; compare also pi. 20, fig. 10 / With two anal blood gills, pl.3G, figs. 1, 2, 3 2.5. C o r y n o n e u r a ( 1 e m n a ) ff With four anal blood gills ; antennae mounted on basal promi- nence, pl.26, figs. 5, 8. 42. Tanytarsus ee Antennae short f Larvae usually blood red ; eleventh body segment with two pairs of blood gills, pi. 15, fig.4. .38. Chironomus (pt.) ff Larvae greenish, yellowish, or whitish g The maxiiiary palpus usually noticeably longer than broad. Larva in pools, pond water, or slow streams, pi. 23, figs. 3 and IG 38. Chironomus (pt.) gg Palpus about as long as broad, pl.24, figs. 5, 12, 20 h Full-grown larva not over G mm. long, green or bluish- gi-een in color. Anterior abdominal segments of greater diameter than the posterior ones. Mandibles often transversely wrinkled ; the anterior prolegs usually with pectinate setae [39. Cricotopus 41. O r t h o c 1 a d i u s y 88 NEW YORK STATE MUSEUM lih Full-grown larva over 6mm. in length; mandible not transversely wrinkled i Labium with its teeth rounded, pl.20, fig.9 35. D i a m e s a waltUi.' a Labium with its middle tooth broadly truncate 37. T h a 1 a s s o m y i a f u s c a Note. — See addenda for .several anomalous species. Pupae a Pupa floats nearly motionless in a vertical position at the surface of the water, pi. 17, fig.ll. {Group C e.r at o p o g on) aa Pupa not as above h Active pupa swimming with a jerking motion in the water like a culex pupa. pi. 19, fig.8 (Group T a n y p us) hb Not culexlike c With long setae or filaments at caudal end (/ Caudal filaments very numerous and forming a caudal paddle, pl.22, fig.l4, and pl.26, fig.15 V Thoracic respiratory organs a tuft of filaments, pi. 10, fig.2 38. C h i r o n o m u s ce Respiratory organs consisting of a main shaft with lateral hairs or setae. Abdomen with setae and bristles. 42. T a n y t a r s u s dfl Caudal appendage with long setae c With eight long setae on each side of caudal appendage, pi. 30. figs. 4 and 5 25. C o r y n o n e u r a 1 e m n a e cc With a tuft of long setae on each side; thoracic respiratory organ a simple fingor-likp jirocess. pi. 84. figs. 5 and 8 31. H y d r o b a e n u s cc With three or four pairs of short setae, a plate-like sucker or with a paddle d Plate-like sucker at caudal end, pi. 34, figs. 14 and 15 29. T e 1 m a t o g e t o n ( St Paulii ) rJd With two or three pairs of short setae or with paddle c Without thoracic respiratory tubes. Abdominal segments with a fringe of conspicuous spines or setae or projections, pi. 48, fig.13, and pi. 50, fig.lO in Bui. 68 N. Y. State Museum, 1903 35. D i a m e s a 37. T h a 1 a s s 0 m y i a 44. M e t r i o c n e m u s cc With respiratory tubes. AixjiMiiinal segments usually without a fringe of ].romineiit setae, pi. 21, fig.24, also pl.24, figs. 13, 14, 15 39. C r i c 0 t o p u s 41. Orthocladius 'The larva of T h a 1 a s s o m y i a c o n g r e g a t a (an European species) has a labium like Diamesa waltlii. MAY FLIES AND MIDGES OP NEW YORK 89 Imagines Note. — In counting the antemial joints the large basal joint is included, but not the hidden first joint. a Wings absent or rudimentary h Wings reduced to mere vestiges, legs short and not slender, antenna with seven joints, mouth parts rudimentai*y, pl.36, fig.l3, female 14. C 1 u n i o 66 Rudimentary wings reaching at least to the end of the first abdominal segment c Halteres wanting, wings in the form of a flattened racket, tarsal claws with a little subapical tooth. Palpi four-jointed ; antenna five ( ?) jointed 10. B e 1 g i c a cc Halteres distinct (I Palpi four-jointed ; anteima of the female four-jointed, of the male six-jointed, pl..35, figs. 16, 21, 22 11. E r e t m o p t e r a dd Palpi two-jointed ; antenna of the female six-jointed c Second joint of antenna lilce those following, monilliform, pi. 37, figs. 5, 6, 7 12. H a 1 i r y t u s cc Second joint longer than those following ; male antenna also with six joints, pi. 35, figs. 4 to 9. ..13. P s a m a t h i o ra y i a a a Wings present 6 The M-Cu crossvein present (i. e., cell M closed by a vein), pi. 37, fig.24 c Antenna with twelve or more joints d Antenna with fourteen joints usually plumose ; fourth tarsal joint usually shorter than the fifth ; wing bare, pl.30, flg.l3, male 35. D i a m e s a dd Not as above c Antennae with fifteen joints both in male and female ; plinnose in the former ; the vein M simple, pi. 27, figs. 1 to 15 ( Group T any p u s ) f Wing bare g Fork of the cubitus petiolate 15. P r o c 1 a d i u s gg Fork of cubitus proximad of crossvein 16. A n a t o p y n i a n. gen. ff Wing pubescent g Fork of cubitus proximad of crossvein h Antennae of male and female each with fifteen joints, Rj and Ra distinct 17. A b 1 a b e s m y i a n. gen. hh Antennae of female with twelve joints, the male with fif- teen joints, Ra indistinct (Australian genus) 18. I s o p 1 a s t u s gg Fork of cubitus petiolate 19. T a n y p u s ee Antennae with twelve or fourteen joints ; wings hairy ; cubitus not forked ; anterior crossvein long and very oblique, pi. 37, fig.16 20. P e n t a n e u r a cc Antennae with less than ten joints d Crossvein near basal third of wing, pi. 37, fig.ll. Antennae about eight-jointed (Chile) 21. Podonomus 90 NEW YORK STATE MUSEUM d(l Crossveins near middle of wing e Palpi six-jointed; antennae seven-jointed; wing venation as shown on pl.37, fig.21 (Chile) 22. Heptagyia ee Palpi four-jointed f Antennae of female seven or eight jointed; male antennae fourteen-jointed ; wing bare 35. D i a m e s a ff Antennae of female with eight joints, the male with nine, short-haired 36. E ii t a n y p u s According to the description the female of E u t a n y p n s does not seem to differ from the female of D i a m e s a . hi) Tlie M-Cu crossvein absent c Wing with four or five very indistinct longitudinal veins ; wing club- shaped, the anterior margin with a long curved seta, the antemia with about twelve joints 23. C o r y n o c e r a cc Wing margin without a long curved seta d Probocis and palpi rudimentary ; abdomen shorter than the thorax ( female apterous) 14. C 1 u n i o (Id Palpi not rudimentary e Antennae with not more than ten joints f Antennae six-jointed g The R-M crossvein, if present, at the basal quarter of the wing h The R-M crossvein near basal quarter of the wing, pl.37, fig.l3 (Chile) 24. Spaniotoma hh The R-M crossvein coalescent with the longitudinal veins ; wing club-shaped ; the anterior cells thickened, pi. 36, fig.7 25. C 0 r y u 0 n e u r a gg The R-M crossvein near the middle of the wing h Wings hairy ; antennae with the four intermediate joints verticillate with very long hairs ; male unknown (Europe) , pl.34, fig.20 26. W u 1 p i e 1 1 a hh Wings with margin ciliated; antennae with sparse ver- ticils of spreading hairs (Kerguelen Island), pl.37, fig.l 27. L i m n 0 p h y e s ff Antennae \Vith seven to ten joints g Antennae with ten joints, not plumose ; costal cell thickened, pi. 36, fig.7. 25. O o r y n 0 n e u r a (male) gg Antennae with seven or eight joints h Thorax with a longitiulinal fissure; wings black with white markings, pl.31, fig.lG, and pL27, flg.16 28. Chasmatonotus h7i Thorax without this fissure i Claws cleft, venation as figured ; antennae seven-jointed in male and female, pl.34, fig.l6 29. Telmatogeton MAY FLIES AND MIDGES OF NEW YORK 91 a Claws simple ;■ Very small species ; black, iiicliuliug its legs, wiugs and lialteres ; male with fourteen nearly bare joints, female with seven joints ; legs with woolly hairs ; metatarsi somewhat elongated ; claws distinct, pl.34, figs. 6-11 31. Hydrobaenus // Not such flies k Antenna of male with eight joints; female like O r t h o c 1 a d i u s , ( Australia ) , pl.36, fig.26 32. Doloplastus kk Antenna of male not eight jointed ; the female with seven joints I Wings very short, in the female scarcely reach- ing the middle of the abdomen ; the abdomen with a seta-like tubercle near the tip on each side; fore tibia longer than the metatarsus ( Spitzbergen) 33. S m i 1 1 i a // Wings of moderate length m Thorax prolonged and bent downward ; hal- teres hanamer-llke, pl.35, figs. 26, 27, 28 (Ar- gentina). Synonym of Chironomus? Si. B u r m e i s t e r 1 a mm Thorax highly arched; halteres with knob ( Group G h i r on 0 m u s ) 11 The fourth tarsal joint obcordate, shorter ■ than the fifth. J^^' Thalassomyia [45. Scopelodromus nn Tarsal joint linear 0 Wings bare p Front metatarsi as long or longer than the tibiae 38. Chironomus pp Front metatarsi distinctly shorter than their tibiae q I^egs black and white annulate, at least the fore pair . . .39. C r i c o t o p u s qq Legs not so banded r Posterior branch of cubitus sinuous, pl.30, figs. 1 to 4 40. C a m p t o c 1 a d i u s rr This branch straight, gently arched, pl.30, figs. 5 to 10 41. O r t h o c 1 a d i u 3 00 Wings hairy p Front metatarsi longer than their tibiae 42. T a n y t a r s u s 92 XEW YOrtK STATE MUSEUM pp Front Dietatarsi shorter than their tibiae q Thorax produced eonically in front over the head ; hind tibiae dilated and hairy, pi. 34, fig.24 43. E u r y c n e m u s