3 / rns , ” pelyieks ‘ ‘ be J 4 Mire * } | ries \\ . ; a ponte i i hotties Surette 4 " * 1 ie “i ‘a shelate } ; sem shh ord | ea ah bd sss oe stale > “ 4, ) ‘ mae at ana = ’ AL baphie ey nett! tr 9 Sia iets ~ bale ‘ ony eo phy ag os edhe “ Orel eat sg ’ { me URE AC HSH histor eta af Troetinssttee ieasraacy H ieee ay Tek: i 4 ith eet ‘ ae Hyak Pet etherats RE umes ait Vai sideline PEE eet tstee ; ' j " ee ‘ ATE Habeel ps vRaral of tribe Leer Y : 4 rire ” PES rherter vy! ebtereate hens 2 mart ata Ms 4s) : wits ts 1b) Clestatesed I belie jana aoa atees ' la! RAY ag ; vase Peranen teen ’ He ie) ey Hee ‘ Dike | 6 whe feat . wasbriat sea} la /apit Ric's he Cacti any ; ; belie sephe? ANE phe eree rsdn) SBP tye " } > SONI eoecater ti pet 4 ys 7 it pmet th spheres a Me Hy bbb tropa Gas ‘ heh Pe “ ayayt 2 OTe ay many isteto ehh 4 iT) a ' eet po hie dy oh tee é j , ‘ Pret ety dh abl Geral, pari’ Her astae peak ’ 2 Disgeaet tebte absetiah 4 ; ; 5 crys patel os bra lehd fentath rege Seren hoes engl hips obo bs i . vee PUNT HEMEL E B60 CERT ENE bpd. ries bah bait ay * ' | ¥) 8 Lines penton ve gtee be eedue eat ily ster eyh Thshteea si sletse Gia iimar ad thy: } at ; A Ware meyer toe err eriratr is ia mia : 1) 4 ae sab phpdastanehaceesin ea poet bho VByatap b¢ v4 bow itetheona ty > teidiiee Pune pl bese ee ahhaes ei vee stad ey Fave er bo aha ty any sit Eh ie ht: np | ob-t Okey sth Lehi reed idet oxi Ma dtinias asda bonaned teeters ' Tt) pa) “; Hirid Th be ear binelh) Dende! Mt bbsens ups be AES 4 he tetet Seepe tpt j . in 1) be baw hed ai a 4 si a i Rn tit to piper ie d ’ ‘ , UL byte be et oe had py PT ePeiel FOU ARE) bee, saan eral Tine weone ae lateobae Bia leey lhe Sibe cht ret bn -} i tity . We : Hor rts ees souaeat aenlyy ast pees soeaereaeesy ij M | i top tlait wh vie see “is Or ereirLStapstanwenetes vats ’ dene st oie + rs Hs dhe ‘ }} Patient Beit ta tis As rab ate tH 4 Peat oh Ane aon rs an Praca tiie! tetris kiets Meehan} Ave sie + t 4 veg! Wiles ; vk en “ecm endtahisary stra i\ hive deaae etched ‘ater reopens * strony hrs rea eh atl sat ir “¥ dup) seit rh eirk of ms 1 : Sb9 008 ee ienstaste i Ma) ‘t ne ooh ay phe oy Sart) 4 ‘ > shut eee \ ‘ Linire pestis } if artist its ’ aH A ifn pate bhp Re oP i fee u pi hetded ni. riety’ Bi His 4 ; Tees dhe Sweets : ner peat dugthinesny ihr oe Serco Hisitehatie od tai ah a EE Cd = , hes bsphar roses Antlers sited, ret x oes Wensaepeneah$ S/DEMNETON RTD Fact ee! woe. yee a Feet eters Pesci treabtiasee ” it AE len tela ie pereovers husteatatt Hert en where byeadeeune ovina A) Ha esen vi SMe ORPL eli y telewererehevered De tuter itieae Wire mereerare fa hew, eure ‘ay rare viria-Y eeees r * Sc beets (bh ys AY pee eave " SSAA LM eeaGE aia lareth rest Ned Ply Rs bare he} EME a er hate oie lad KBP ban inte vey be Doreen ies enrarhren ike ua ste aphon aise ene ‘irae TEMP Te © viNubserery a , retiyra bas ip bevpaamass aerate rane eee u hobs bass eed bd ih here's Ties e snaeie ; *dayt tia eee FAs ty ‘ Mv vere telt 71 kIb teield pjepery Vely le ® bn teres > eee Agigers 4 yp Bory) bribe ees Salto weber sae then aParet 18 py eae Chane hal oF HA cas ps fb “th abibiek ie pH ealed * slespsyee Shh SPL seat ry ie 55 he plaid ey ieee gett peeve, eit ome, Deir eee eget 14 a FRM eet reer. eet re ior Coad eae ae eae rar Me vopter day ate int wae tine sages Neate +4 4k ory ties LPbregai gh tit tirttey 1 toa te} sa abi aatuaer essed tesa ttetey Ti yee ye brio Steegtate tat aytirer sien odin dy avele 1" vinnie amet hits ii Mepe etasieepiey Penh ah ety ae rer rdi ed 0 ene beajectt yet od ait fades eu, Hide 4eFed 9h os 0 saree) seth: eae pe veebay ee a s ' PER abedemetephpes syohaprae: bier \ i rani ’ ST Shae Deve tba rie bos ples Siar ry Pept & H He i Hie? ut at ote) thy er bent , i vegan hr we FOR THE REOPLE FOR EDVCATION FOR SCIENCE LIBRARY | OF THE AMERICAN MUSEUM OF NATURAL HISTORY UNIVERSITY OF MICHIGAN. MUSEUM OF ZOOLOGY. Miscellaneous Publications No. 1. —— | / Directions for Collecting and Preserving Specimens of Dragonflies for Museum Purposes BY EK. B. WILLIAMSON. ANN ARBOR, MICHIGAN. PUBLISHED BY THE UNIVERSITY. DECEMBER I, 1916. — | G24 ADVERTISEMENT. The publications of the Museum of Zoology, University of Michigan, consist of two series—the Occasional Papers and the Miscellaneous Publica- tions. Both series were founded and are being supported by Dr. Bryant Walker, Mr. Bradshaw H. Swales and Dr. W. W. Newcomb, except that the cost of the illustrations and of distribution are borne by the University. The Occasional Papers, publication of which was begun in 1913, serve as a medium for the publication of brief original papers based principally upon the collections in the Museum. The papers are issued separately to libraries and specialists, and when a sufficient number of pages have been printed to make a volume a title page and table of contents are supplied to libraries and individuals on the mailing list for the entire series. The Miscellaneous Publications include papers on field and museum technique, monographic studies and other papers not within the scope of the Occasional Papers. ‘The papers are published separately, and, as it is not intended that they shall be grouped into volumes, each number will have its own title page and table of contents. 2 b- Heed o 2 ALENANDER G.) RUTHVEN, Director of the Museum of Zoology, University of Michigan. CONTENTS. PeemGoliceunn and Preservingvor UNASOES: .... 2.» sensaeeenaa Pek oer c eden anees ae TCM MUL LVAR AES, cet sec e-em «, scare te AeSeI gays. 3 e's ob ge ORR elo vos ecw card 5 MCGEE INCE nat Mireles tna Merete + o's s Win weelttatine Ie omene weak aite «kia wiP screws 6 REM RRELSS Was act ar. ox ONG eae Cg ae Se ee an eegemta near. AoCayias co's esas. 6 SURG AINE CASE <3 ic crash ots cerivastatin KR LEME Ore Biel See neta Sie win weed 6 MST SPECIMENS « ssleta slates: ctcage Wie oe Pe arare ger Sa Wea b wen 5 ea slid warts 9 Mee SPE MMEN Sc. ics Mosc > ebediae oh ot PU apsS A ee a Gdns a EA > Seeley ae II fie backine of Papered, Specimens << .62< 4 lth oe ore gees Bolg © ois wae « in alls 13 iter PCCIMEIIS yates Sauce Mees = ooo wee De Aepoclste - Svea « zag AAI ae-tSitats < 13 Bpecervringsopecimens. in AlcolMol §. <2 sive sa@senelele as sett winlom est «ante arepemtamtayeas a Aigiete 13 AAG R ET VALICNEGONE IN GRILD ITS. oi ars foetus scones, oS ke cranaeenmice ovis, oD Ree ee Sh elomraeatier atten oes 14 ‘The Preservation of Nymphs and Imagoes for Histological and Cytological Studies 14 CES Sn TSE ES Oe. Ce Pn ne Se 3 15 ILLUSTRATIONS. Renee element BON «, «cet «vie Goad 6 veins ot tau lala eS Une nana bin siso's waders +09 7 Piet 2 seriverapfector Papered Spectinerts> «cian mesiied nce smmeieauaraverer s+ «, aoe vee tole 9 er EURMIER BELSON iat dias Soltis ssc « sate natetpeme en oda sale ae eee hs os 0 oe ove < ars,cxmpats 12 DIRECTIONS FOR COLLECTING AND PRESERVING DRAGON- FLIES FOR MUSEUM PURPOSES. BY E. B. WILLIAMSON. The methods of collecting and preserving dragonflies described in this paper are for the most part those which have been worked out by the writer in the course of several years of field work. No attempt has been made to try out the many kinds of collecting appliances on the market nor to ex- periment with all of the described methods, but an effort has been made to discover simple and yet adequate appliances and methods for the collecting and proper preservation of specimens of this most interesting group. These directions should, then, be considered as suggestions rather than as dogmatic rules. THE COLLECTING AND PRESERVING OF IMAGOES. Dragonfly habitats —Years ago the writer heard Dr. Kellicott say that the way to get all of the rotifiers was to look in all the unlikely places,—the likely places would supply a great many, the unlikely places would complete the list. The same is true of all collecting. Visiting a stream at one point does not determine its Odonata fauna on that day. One may follow it with almost barren results for miles, till a sudden turn brings one on a rocky ripple, grass margined, with clumps of lizard-tail, and broken with patches of water-willow where untold numbers of multicolored and iridescent wings spring into view, a score or more species inviting one to the chase. Or a sunny meadow may rise, brush and forest girded, up the hillside from the brook you are following; along the edges of this meadow more than one rare dragonfly may be basking on the leaves of some bush or low tree. What has been said about visiting streams is also true of lakes. If the lake visited is small the entire circumference should be explored by the collector. For various reasons, such as temperature and soil differences, the inlets and out- lets of the lake may differ more or less in their dragonfly species. Prevailing winds through centuries modify the shores of lakes, and one side may be found to be boggy and without a beach while the opposite shore may have a well defined gravel and sand beach, with consequent differences in both fauna and flora. If the lake is large, consideration of the above points should re- ceive the collector’s attention when selecting places to work. If one can spend several days along one stream or about some small marsh, the first day or two may profitably be spent in a general survey and in paying atten- tion to the larger, free flying dragonflies, leaving for a later date the close searching of low vegetation and concealed nooks for the smaller, less agile and less readily discerned species. The last day or more may be given to same particularly interesting find of the preceeding days. Remembering, in conclusion, that it has been truly said of the dragonfly collector, “all things come to him who wades.” 6 E. B. WILLIAMSON The collecting net—The writer has tried only a few kinds of nets. For several years he has used what is known as “Our Own” patent landing net, a steel collapsing ring, which can be obtained from entomological supply companies. ‘The frame used has the two steel sides each about 22 inches long. ‘The bag is made of light bobinet (15 or 16 meshes to the inch) and when flat is about 22 inches long (deep). The edge of the open end con- sists of two narrow sleeves of stout muslin which slip over the steel sides which make the ring of the frame. This landing net is provided with a three jointed bamboo-handle, two joints of which make a handle of suit-— able length for insect collecting. ‘This insect net is light, strong, and when not in use, may be conveniently packed or carried. Bags of several kinds, suitable for beating and dredging, for example, can be carried in the field and changes from one bag to another, or the replacing of a ripped bag, can be effected in a moment’s time. Killing bottles—Of several ways of making killing bottles the most satis- factory has been to place potassium cyanide crushed to about the size of grains of corn over the bottom of the bottle selected; this is then covered with about a quarter of an inch of sawdust, over which is poured plaster of Paris mixed with water to the consistency of cream. The whole need not occupy over an inch and a quarter of the bottle,—less if one is fortunate in pouring in the plaster. The most convenient bottle I have found is the ounce morphine bottle. Four such bottles can be conveniently carried in the double side pockets of a coat or blouse. Smaller bottles can be made in the same way of straight glass vials four inches high and seven-eights of an inch in diameter. Such small bottles are valuable for killing pairs of smaller dragon- flies or for particularly small and fragile species. Such vials are also of a convenient size for carrying alcohol or other preservatives in the field. It goes without saying that miscellaneous collections of insects should never be dumped in the same killing bottle. The scales of Lepidoptera and the juices of various Orthoptera are especially objectionable, and none of these insects should go into the bottles containing dragonflies, or in which dragonflies will be subsequently placed unless the bottles are thoroughly cleaned in the meantime. A killing bottle made as above described will sometimes “sweat.” This moisture is ruinous to dragonflies. For this reason bottles should be made several weeks before they are intended for field use and all moisture should be guarded against by airing the bottle in the sun and packing the bottom with scraps of paper or cloth if necessary. Specimens should be removed from the killing bottle as soon as they are dead or at frequent intervals. If the bottle is “strong” a short exposure may completely destroy the colors of many species. They must be removed to some convenient and safe receptacle in which to carry them till the day’s collecting is over. The collecting case-—The box here described has been used satisfactorily for several years in various climates and under varying collecting conditions. The pattern for this leather box is shown in the accompanying diagram (Fig. 1). COLLECTING AND PRESERVING DRAGONFLIES 7 Fic. 1.—Collecting box; 1, side, about 8% x 11 inches; 2, bottom, about 7%4 x I1 inches ; 3, end, about 714 x 8% inches; 4, top, about 814 x 11 inches; 5, buckles in which two straps (9) on the top fasten; 6, rings for shoulder strap; 7, ring on strap, so it can slip up or down, in which to fasten snap attached to back of a belt; 8, handle on top of box, to carry when it is not over the shoulder; 9, short straps on flap (11) on top, to fasten in buckles (5) on side; 10, straps on end of box, to turn in and be covered by the top (4) when box is closed; 11, flap on top (4) of box; 12, compart- ments in the lower or bottom tin pan in the collecting box; 13, suggested compartments in the shallow pan (1 inch deep, 7 inches square) for the collecting box. 8 E.. B. WILLIAMSON The shoulder strap should be about an inch and a half wide over the shoulder and at either end should have short straps with spring snaps for the rings shown on the ends of the box. The strap at one end of the shoulder strap should provide for varying adjustments in length. In the back of a wide belt, fasten a small spring snap to attach to the ring shown in the dia- gram on the side next the top. This ring is on a small strap, attached at its ends. When the box is on one’s back and the snap on the belt is in the ring on the back (side) of the collecting box, the latter is held securely in place without at the same time interfering with one’s movements. One may run or stoop over at will without having the box come tumbling in his way. This leather box which stands 8% inches high, 11 inches long and 7% inches wide is now ready for its fittings. Have the tinner make a pan 4 inches deep which will just force down in it. Depending on how the harness makér has made the leather box, this will vary a little but the pan will be about 1034 x 7% inches and should fit tightly in the leather box. This pan should have two partitions dividing it into three equal compartments as shown in the diagram. Make another pan about seven inches square and one inch deep divided by partitions as you wish. I have mine divided in halves, and one-half with two cross partitions dividing it into one larger and two smaller compartments. This pan must be made to slip easily in and out of the leather box. Make a third pan also about seven inches square and deep enough to just fill the leather box when placed on top of the shallow pan, that is about three inches deep. A fourth pan is about the same size as one of the com- partments in the first made pan. It will be about 4 inches deep and 7 inches long by 3% inches wide. ‘These four pans will just fill the leather box, the one in the bottom being firmly wedged in, the others fitting loosely so they can be readily removed. A piece of tin with turned edges, measuring about 714 x 105% inches, will serve as a cover for the box. Over this the leather flaps on the ends are turned and the leather lid brought down and strapped, holding everything securely. In the three bottom compartments and in the other pan of the same size, carry ordinary envelopes (354 x 6% inches) with their flaps removed, set on edge. In these envelopes place your specimens with the wings folded back and the abdomens straight. Several specimens, depending on size, may be placed in the same envelope. The end specimens should have their heads toward the ends of the envelope. In the envelopes in these four compart- ments a large day’s collecting can be carried home in perfect condition and in admirable shape for future papering or pinning. Some other form of collecting box may be used to suit the collector, but no better method I think can be devised for carrying home specimens than to place them in envelopes which are carried on edge. ‘The time required to place them in the envelopes is more than saved by the admirable condition of the specimens when one comes to paper them. In the shallow pan in the leather box the writer carries a pair of small pliers, fine tweezers for handling specimens, some vials of pins and alcohol, string, lead pencils, memorandum book and a small box or two for exuvie. In the larger square pan is carried the lunch, some insect bags, etc. On the COLLECTING AND PRESERVING DRAGONFLIES 9 return in the evening the lunch is usually replaced by a small turtle, a few small snakes, a set of bird’s eggs, some seeds, leaves or flowers, or whatever of interest the day may have yielded. Pairing dragonflies, if it is desired to keep them together, are killed in smaller bottles and the pair put in an envelope by themselves, or, as taken from the net, the pair are impaled on a No. I insect pin, and later unpinned and placed together in an envelope. Large dragonflies such as the larger libellulines, Macromia, the aeschnines and gomphines may be carried home alive. Simply rotate the head once till the neck snaps, and place them in the envelopes. They will live a day or more in a comatose condition, expelling the contents of the alimentary canal. In the evening at home they can be killed in a cyanide bottle and papered Fic. 2. Envelope for papered specimens. at once, or they may be allowed to die naturally and then papered. Twisting the heads of large species will prevent their biting and possibly injuring the wings of other specimens if they are dropped into the cyanide bottle in the field. With the specimens safely home, a piece of white oilcloth spread on a table, makes an ideal place for sorting, arranging and papering specimens. Personally the writer prefers to paper all of the material, later selecting from the dried specimens such as it is desirable to pin. As the conventional methods of pinning and expanding specimens have been so well explained by lepidopterists little will be said on the subject (see page 13). Papering specimens.—The most convenient envelope is the usual three- cornered one made by folding a rectangular piece of paper as indicated, the folds to be made in the order shown, a and b being the inside of the envelope. Do not fold with a and b outside as the resulting envelope will IO E.. B. WILLAAMSON be different at the corners, and is less readily opened and closed. Nothing is more annoying in its way than to examine large series of specimens in papers folded some one way and some the other. Envelopes may be obtained at any newspaper office. Have light news- paper stock cut in rectangles 2 x 33@ inches and 234 x 4% inches, and have heavier newspaper stock cut in rectangles 354 x 6 inches, 414 x 6% inches, 5 x 7% inches, and so on to larger sizes if desired. The last named will paper a dragonfly nearly 5 inches long. Fold your envelopes before the collecting season or before a collecting trip. Cut strawboard corners the size of the folded envelopes, and bunch the envelopes in fifties, with a straw- board corner on either side, all held together by a small rubber band. Put these bundles under a weight or in a letter press for a few days. Nothing is more wasteful of time and energy than to make slipshod envelopes from day to day when the collecting should receive all one’s energy, using for material old magazines, newspapers, discarded correspondence, etc., etc. The method here suggested provides uniform envelopes, well folded, of clean paper, unmarred by writing or printing, on which intelligible data can be written or stamped, leaving space for brief field notes or references to other fuller notes in a suitable note book. And they represent less loss of time and energy than the heterogeneous collection of half-folded waste paper in which too many collectors entomb objects of once animate nature for which they might well show more respect. With the envelopes prepared nothing will save more time and energy than a collection of suitable rubber stamps or a small stamp in which the desired type can be set, and a small line dater. The data should include the place, the collector’s name, and the date, to which may be added in writing any field observations. The ink used for the stamp should be one that will not fade or “run” if the envelope is moistened. In papering material the first consideration is to prevent distortion of the specimen due to pressure by the envelope. The abdomen and especially the appendages are most likely to suffer. For specimens with dilated abdomens or wide spreading appendages, nothing is better than a bit of cord of suit- able diameter fastened with a drop of glue on the inside of the envelope along its folded edge near one corner, the head of the insect resting at the other corner. Since dragonfly thoraces are usually larger than the abdomen and are relatively firm it is often well to paper two specimens of the same species in one paper, the heads at the corners and the abdomen of one above the other, not crossed, if the abdomens are long. So papered the thoraces will bear the pressure of the envelope and the specimens will have the abdo- mens and their appendages preserved without any distortion. Keep the abdo- mens straight and wings properly folded back together. In the case of libellulines especially see to it that the legs do not cover and conceal the accessory genitalia. When the specimen is in the envelope and the envelope has been carefully closed, fasten the corners down with a bit of gummed paper M4c-%52 of an inch wide and about % inch long fastened to the flap, at each acute corner of the envelope and around to the opposite side of the envelope. Avoid any pressure on the specimen in attaching these gummed CoLLECTING AND PRESERVING. DRAGONFLIES II strips. This method of sealing the envelope is infinitely superior to the time-honored method of turning the corners. It requires less time, en- dangers less the contents of the envelope, is more secure, and facilitates the ultimate packing of the dried material. After papering, the envelopes should not be tumbled about till the specimens are thoroughly dry. Otherwise abdomens may be thrown up between the wings, may be crossed and constricted or bent at the point of crossing, or other damage may be done. Drying Specimens.—As has been stated, in papering material the first consideration is to prevent distortion. The second point is to preserve as fully as possible, and that is often far too little at the best, the structures and the colors of the insect in life, or at least to dry it rapidly and thoroughly, and prevent a slow decomposition which would render the preserved speci- men unduly fragile. In this connection it may be stated that it is unnecessary and undesirable to make a general practice of passing bristles or other sup- ports through dragonflies’ bodies if they have been properly dried. Bristles may be necessary in repairing broken specimens, but such specimens should be considered surgical cases, and this occasional heroic treatment should not be inflicted on all. Another reason for rapid drying is facility in handling a large number of specimens which once dried can be safely packed away,. but which if undried and accumulating from day to day would form a serious obstacle to the movements of the collector. If the amount of material collected is not large, if the collecting days. are not continuous, and if frequent changes of locality are not made, papered dragonflies can be dried very well under ordinary conditions without recourse to artificial heat. At his home the writer arranges the papered material on large sheets of cardboard which are placed on tables in a dry room where the full sunlight can fall on them. As stated above, while drying the enve- lopes containing specimens must not be tumbled about. Rapidity of drying is essential to good color preservation, and is more readily obtained in some climates than others. In a humid climate artificial drying is necessary if good preservation is to be obtained. Artificial drying facilitates handling large quantities of material rapidly, prevents large accumulations of un- dried material which require the collector’s attention, and seriously burden him if he wishes to make frequent moves to new localities. After several experiments with differently constructed artificial driers, I have made one which seems to answer all the requirements. This is a tin box 15 inches high and 10 inches square, open at the top and bottom. Three sides are of tin without openings. The fourth side is open except for about 2 inches at the top and 1 inch at the bottom (where the tin of the adjacent sides is continuous), with the proper supports for 4 interchangeable drawers, each 3 inches high. The supports for these drawers are placed so the first drawer is just above the inch at the bottom of the box. Each drawer is about 3 inches deep and 10 inches square; the sides are of tin and the bottonr of wire mesh, about 6 meshes to the inch; the top is open. These 4 drawers occupy the space between the inch of tin at the bottom and the tin about 2 inches wide at the top. A shallow pan of similar construction to the drawers. is made to drop into the top of the box. When all the drawers are in place 12 E.. B. WILLIAMSON the four sides of the tin box are enclosed. To the lower end of this box a tapering, heavy canvas funnel or chimney about 3 feet long is fastened. This must fit tightly over the bottom of the tin box and at the opposite end be tapered to go neatly over a small lantern using a 5Zinch wick. In use the tin box is hung at the proper height against the wall by cords attached to rings at the top of the box. So hung, with the lantern burning in the cloth funnel or chimney, and the drawers in the box, a warm current of air passes through the box and out the top. The bottom drawers are the warmest and I6 SLSRETLS INTENSE NNN CECE NS: esi 4 1] aig - BPTI NY AKAN YY ON IGT Fic. 3.—Drying box; 14, two rings by which drying box is suspended by cords against the wall; 15, four interchangeable drawers; 16 heavy canvas chimney, fitted closely around box; 17, short, sharpened, upturned wires fastened at each lower corner -of the box, over which the canvas chimney (16) is hooked, to facilitate holding the -chimney securely in place. COLLECTING AND PRESERVING DRAGONFLIES 13. this should be borne in mind, larger species, or the most recently captured specimens being placed in the lower drawers. Place envelopes in the drawers upright or inclining, and resting on their long edge, but not piled up flat, as in the latter case the circulation of warm air about each envelope will be interfered with. Do not put all of one day’s catch in one drawer, but, on the basis of size, distribute the papered specimens in two or more drawers. A little experience will teach one when the specimens can be safely removed for permanent packing. In actual practice the writer empties the drawers only as the acquisition of new material requires that room be made for it. Specimens are not injured if left for several days in the drier. The length of time required to dry specimens is dependent on the size (bulkiness) of the specimens themselves, and on the atmospheric conditions. Small agrionines such as Argia and Enallagma will sometimes be dry at the end of 12 hours, and I have found bulkier things, such as the libellulines, dry in 24 hours. The packing of papered specimens.—Cigar boxes make as convenient and safe receptacles for packing dried material as any other. If cigar boxes are dipped or otherwise thoroughly saturated with a mixture of equal parts of crude creosote and creolin their contents, if taken directly from the drying box, will be free from the attacks of ants and other pests. A small quantity of napthaline flakes should be placed in the bottom of the box and covered with a piece of paper or cotton wadding. Boxes should be treated with the creosote-creolin mixture several weeks or even months before specimens are put in them. In packing material in boxes, put in pieces of cotton wadding at frequent intervals cut to the size of the box, building the box up in this way in layers. This will give a certain elasticity to the mass, and prevent shaking about, and the sheets of cotton so interspersed will prevent the envelopes packing and jamming together at one end of the box. It is an easy matter to accumulate a lot of cigar boxes of the same size, and if this is done one can arrange a space for them in a trunk or suit case more advantageously than if boxes of various sizes and shapes are used. Pinning specimens.—As has been said, this subject will not be discussed in detail, but one remark on pinning specimens may be permitted. The writer has long since ceased to relax specimens for pinning except in ex- ceptional cases, such as those having highly or densely colored wings. In the great majority of cases the specimens are taken directly from the enve- lopes and pinned without any treatment, the Zygoptera vertically between the front wings with a smooth No. 1 pin, the Anisoptera laterally through the bases of the hind legs, close up to the body, with a No. 1 or No. 2 or rarely a No. 3 pin for the largest species. Specimens so pinned require less room and are in better condition for critical study. Of course this method is not recommended for material designed largely or solely for public dis- play purposes. : Preserving specimens in alcohol_—The colors of dragonflies may be well or almost perfectly preserved in full strength alcohol. Such specimens should be placed in alcohol while alive or at once after being killed and the vials 14 E. B. WILLIAMSON or bottles of specimens kept in the dark. After a few days the alcohol in which they are preserved should be replaced with clean, full-strength alcohol. THE PRESERVATION OF NYMPHS. Mr. R. J. Tillyard recommends the following formula for preserving larve where preservation of internal parts is desired. 15 parts 98% alcohol. 6 parts formal. 2 parts glacial acetic acid. 30 parts distilled water. Place larvze in this alive. As soon as possible make ventral incision opposite mid-gut. After 24 hours preserve in 70% alcohol. Professor James G. Needham, in reply to my inquiry about any special methods, wrote, “I simply use alcohol and, when particularly careful, change it a time or two within a few days after putting specimens in and endeavor to have it, when osmosis is complete, of a strength of 70-80%. Preserva- tion of the nymphs is then entirely satisfactory. I have found formalin an abomination, as it does not penetrate and does make brittle. The glycerine mixtures make things greasy, so I have ended up by using the method that is simplified.” THE PRESERVATION OF NYMPHS AND IMAGOES FOR HISTOLOGICAL AND CYTOLOGICAL STUDIES. The following notes on preservation of nymphs and imagoes for his- tological and cytological work have been kindly furnished by Dr. Philip P. Calvert of the University of Pennsylvania. Larve and imagoes of Odonata may be fixed and preserved for study of internal organs as follows. Plunge the living insect into hot water, hot alcohol (30-50% ) or hot Gilson’s fixing mixture. The temperature should be 80°-90°C. Gilson’s mixture is composed of nitric acid (46° strength) 78 cc., glacial acetic acid 22 cc., corrosive sublimate 95 grams, 60% alcohol 500 cc., distilled water 4400 cc. When the entire larva or imago is to be preserved for dissection, and not for histological or cytological work, hot water or hot alcohol (as above) will suffice. The insect should be left in them only until muscular movements cease, then withdrawn from the liquid, one or more slits cut through the chitin (according to the size of the speci- men) in places where internal organs will not be injured, always taking care to cut no deeper than through the chitin and underlying hypodermis, and immediately placed in alcohol of greater strength than that employed for fixing. After one or more hours this alcohol should be replaced with stronger and so on until a strength of 70 or 75% is reached. When Gilson’s mixture is used, the specimen should be (after cutting the slits) washed in water for an hour or more to remove the mercury salts. When internal organs are desired for histological or cytological work, section-cutting, etc., the insect should be cut in pieces as it is allowed to fall into the killing fluid. This is for the purpose of affording more rapid COLLECTING AND PRESERVING DRAGONFLIES 15 penetration of the fluid into the body cavity and the organs lying in it. In passing such material through alcohols of increasing strength as directed above, the transfer should be carried to 90 or 95% strength. Permanent preservation of material for dissection should be in 70% alcohol, for histological purposes 85-95% alcohol. Strickland (Biol. Bull., XXI, 1911, 306), has recommended Kahle’s fluid (30 parts water, 15 parts 960% alcohol, 6 parts 40% formalin, 1 part glacial acetic acid) as superior to Gilson’s mixture for fixing larve of Simulium and it may perhaps be superior for Odonata. Schulze (Deut. ent. Zeitschr., 204, 1915) suggests a mixture of 200 cc. glycerine, 200 cc. distilled water and I gram crystallized carbolic acid for preservation of larvze for dissection, especially in the tropics. Further directions for preparation of histological material must be sought in such works as Bolles Lee’s “Microscopists Vade Mecum,” 7th edition, Philadelphia, Blakiston, 1913.. COLLECTING MANUALS. For suggestions and directions on collecting and preserving dragon- flies see: 1. Directions for collecting and rearing dragonflies, stone flies and May flies, by James G. Needham. Part O of Bulletin No. 39, U. S. Nat. Mus., 1899. 2. Directions for collecting and preserving insects, by Nathan Banks. Bulletin 67, U. 5. Nat. Mus., 1909. UNIVERSITY OF MICHIGAN, MUSEUM OF ZOOLOGY. Miscellaneous Publications No. 2. An Annotated List of the Odonata of Indiana BY E. BPWILLIAMSON. ANN PUBLISHED-BY THE UNIVERSITY. JUNE 5, 1917. ARBOR, MICHIGAN. ADVERTISEMENT. The publications of the Museum of Zoology, University of Michigan, consist of two series—the Occasional Papers and the Miscellaneous Publica- tions. Both series were founded and are being supported by Dr. Bryant Walker, Mr. Bradshaw H. Swales and Dr. W. W. Newcomb, except that the cost of the illustrations and of distribution are borne by the University. The Occasional Papers, publication of which was begun in 1913, serve as a medium for the publication of brief original papers based principally upon the collections in the Museum. The papers are issued separately to libraries and specialists, and, when a sufficient number of pages have been orinted to make a volume, a title page and table of contents are supplied to libraries and individuals on the mailing list for the entire series. The Miscellaneous Publications include papers on field and museun* technique, monographic studies and other papers not within the scope of the Occasional Papers. The papers are published separately, and, as it is not intended that they shall be grouped into volumes, each number has a title page and table of contents. ALEXANDER G. RUTHVEN, Director of the Museum of Zoology, University of Michigan. CONTENTS. —_ BE pate yn es eee Sete Ee ore Sa 3) Jed 6 % (opiiphicnl Notes: 2) wiv. e.n 43. 44. 49. 50. Sir E. B. WILLIAMSON Ischnura posita Hagen..—Allen, De Kalb, Elkhart, Kosciusko, Lagrange, Noble, Steuben, Wells and Whitley Counties. May; June; July; August; September 1-20. Ischnura prognata Hagen.1°—Wells County. April 21-30; June 21-30. Ischnura verticalis Say—Allen, Blackford, Crawford, De Kalb, Elkhart, Fountain, Kosciusko, Lagrange, Marshall, Noble, Steuben, Wells and Whitley Counties. May; June; July; August 1-20; September 1-20. Genus Anomalagrion Selys. Anomalagrion hastatwin Say—RBlackford, Elkhart and Wells Counties. May 1- 10; June 21-30; July. Suborder ANISOPTERA. Family AESHNIDAE. Subfamily Petalurinae. Genus Tachopteryx Selys. Tachopteryx thoreyi Hagen—Montgomery County (C. C. Deam). July 21-31. Subfamily Cordulegasterinae. Genus Cordulegaster Leach. Cordulegaster diastatops Selys..1—St. Joseph County. May 21-31. Cordulegaster maculatus Selys.12—Noble County. June I-10. Cordulegaster obliquus Say.—‘Inhabits Indiana” (Say). Subfamily Gomphinae. Genus Progomphus Selys. Progomphus obscurus Rambur.—Ailen, Dearborn, Elkhart, Huntington, Koscius- ko, Putnam, Steuben, Wells and Whitley Counties. June; July. Genus Hagenius Selys. Hagenius brevistylus Selys.—Allen, Elkhart. Kosciusko, Noble, Putnam, Shelby and Wells Counties. June; July 1-10; August 11-31; September I-1c. Genus Ophiogomphus Selys. Ophiogomphus rupinsulensis Walsh—Allen, Elkhart, Kosciusko and Wells Coun- ties.—May I-10, 21-31; June. Genus Erpetogomphus Selys. Erpetogomphus designatus Wagen.43—Dearborn and Franklin Counties. July 21- 31; August 21-31. Genus Gomphus Leach. Gomphus crassus Hagen.44—Wabash and Wells Counties. May 11-31; June 1-20; July 11-31. Gomphus dilatatus Rambur—-Crawford and Kosciusko Counties, May 21-31; June 21-31. Gomphus exilis Selys—Elkhart, Kosciusko, Noble, Steuben, Wells and Whitley Counties. May 21-31; June: July 1-20. Gomphus fraternus Say.—Allen, Blackford, Dearborn, De Kalb, Elkhart. Hunt- ington, Noble, Putnam, St. Joseph, Wabash, Wells and Whitley Counties. May 11-31; June; July 1-20. Gomphus furcifer Hagen—Allen and Lake Counties, June 1-20. Gomphus graslinellus Walsh—Adams, Blackford, Huntington, Marshall, Putnam, Wabash, Wells and Whitley Counties. May 11-31; June; July. 59. 60. 70. ei: DRAGONFLIES OF INDIANA 9 Gomphus lividus Selys.45—Allen, Elkhart, Lagrange, Noble, Steuben and Wells Counties. May 21-31; June 1-20; July 1-10. Gomphus quadricolor Walsh.—Allen, Elkhart and Putnam Counties. May 21-31; June 11-30. Gomphus spicatus Hagen.—Allen, Elkhart, Noble, Steuben and Whitley Counties. May 11-31; June 1-20; July 1-10. P Gomphus spiniceps Walsh.18—Allen and Wells Counties, July 11-20; August £-10, Gomphus submedianus Williamson.1“-—Elkhart County. May 21-31; June £-10. Gomphus vastus Walsh.—Elkhart, Putnam and St. Joseph Counties. May 11-31; June 21-31. Gomphus ventricosus Walsh—Allen, Lagrange and St. Joseph Counties. May 21-31; June 11-20; July 1-10. Gomphus villosipes Selys—Allen, Kosciusko, Putnam and Wells Counties, May 21-31; June 11-30; July I-Io. Gomphus williamsoni Muttkowski.17—Wells County. June 1-r1o. Genus Dromogomphus Selys. Dromogomphus spinosus Selys——Allen, Hikhart, Kosciusko, Putnam, Steuben, Wells and Whitley Counties. May 21-31; June; July 1-20; August I-10. Dromogomphus spoliatus Hagen.—Allen and Wells Counties. July; August; Sep- tember I1-20. Subfamily Aeshninae. Genus Boyeria MacLachlan. Boyeria vinosa Say—Allen, Delaware, Elkhart, Kosciusko, Lagrange, Noble, Putnam, Steuben and Wells Counties. June 11-30; July 11-20; August I-10, 21-31; September I-10. Genus Basiaeschna Selys. Basiaeschna janata Say—Allen, Elkhart. Kosciusko, Noble, St. Joseph, Wells and Whitley Counties. April 21-31; May; June. Genus Anax Leach. Anax junius Drury.—Allen, Blackford, De Kalb, Elihart, Fountain, Kosciusko, Marshall, Noble, Steuben, Wells and Whitley Counties. April 21-31; May; June; July; August; September 1-20; October I-10, Genus Aeshna Fabricius.19 Aeshna canadensis Walker.19—Steuben County. August I1-31. Aeshna clepsydra Say.—Steuben and Whitley Counties.. August 11-31; Septem- ber I-10. Aeshna constricta Say.—Allen, Grant, Lagrange, Steuben and Wells Counties. August 11-31; September; October I-10. Aeshna mutata Hagen.19—Lagrange, Noble, Steuben and Wells Counties. June; July 1-20. Aeshna tuberculifera Walker.°—Lagrange County (C. C. Deam). August 11-20. Aeshna wmbrosa Walker29—Allen, Grant, Lagrange, Steuben and Wells Counties. July 11-20; August 11-31; September 1-20. Aeshna verticalis Hagen.—Allen, Steuben and Wells Counties. August 21-31; September 1-20. Genus Nasiaeschna Selys.2° Nasiaeschna pentacantha Rambur.21—Allen, Elkhart, Wells and Whitley Coun- ties. June 1-20; July 1-20. Genus Epiaeschna Hagen. Epiaeschna heros Fabricius.—Allen, Elkhart and Wells Counties. March 21-31; April 11-30; May 1-10, 21-31; June; July 1-20; August 21-31; September 11-20. IO 99. 100. Io]. E.. B. WILLIAMSON Family LieeLLuLipar. Subfamily Cordulinae. Genus Didymops Rambur. Didymops transversa Say—Allen, Elichart, Noble, Steuben, Wells and Whitley Counties. May 11-31; June 1-20. Genus Macromia Ram) ur.2? Macromia illinotensis Walsh—-Allen, Elkhart, Kosciusko, Putnam, and Wells Counties. May 21-31; June 11-30; July: August 11-20. Macromia pacifica Hagen.—Huntington and Wells Counties. June 11-30; July: August. Macromia taeniolata Rambur.—-Allen, Elikhart, Noble and Wells Counties. May 21-31; June 11-20; July 11-20; August I-20. Macromia wabashensis Williamson.22—Wells County. June 11-30; July; August. Genus Epicordulia Selys. Epicordulia princeps Hagen.—Allen, Elkhart, Kosciusko, Noble, Wells and Whit- ley Counties. May 21-31; June; July; August 1-20. Genus Neurocordulia Selys. Neurocordulia obsoleta Sey.—‘‘Inhabits Indiana” (Say). Genus Tetragoneuria Hagen.?* Tetragoneuria cynosura Say.—Allen, St. Joseph, Wells and Whitley Counties. May 21-31; June. Tetragoneuria simulans Muttkowski.23—Allen, De Kalb, Noble, St. Joseph, Steu- ben, Wells and Whitley Counties. Mav 21-31; June. Telragoneuria spinigera Selys—Steuben County (Mr. and Mrs. C. C. Deam). June II-20. Genus Dorocordulia Needham.?! Dorocordulia libera Selys.2°—Allen and St. Joseph Counties. May 21-31; June II-20. Genus Somatochlora Selys. Somatochlora charadraca Williamson.2°—Wells County. July 1-10. Somatochlora linearis Hagen.—Fountain and Wells Counties. June 1-10; July I-10. Somatochlora tenebrosa Say—‘“Inhabits Indiana” (Say). Subfamily Libellulinae. Genus Libellula Linne. Libellula cyanea Fabricius — Allen, Flkhart, Kosciusko. Noble, Steuben, Wells and Whitley Counties. May 21-31; June; July; August 11-20. Libellula exusta Say—Ekhart, Lake and Wells Counties. May; June 1-10; Sep- tember I-10. Libellula incesta Hagen—Allen, Kosciusko, Lagrange, Noble, Steuben, Wells and Whitley Counties. June; July; August; September 1-10. Libellula luctuosa Burmeister.27—Allen, Blackford, Elkhart, Kasciusko, Noble, Steuben, Wells and Whitley Counties. May 21-31; June; July; August 1-20; September 1-10. Libellula pulchella Drury.—Allen, Blackford. Crawford, De Kalb, Elkhart, Hunt- ington, Kosciusko, Marion, Noble, Steuben, Wells and Whitley Counties. May 21-31; June; July; August I-20. Libellula quadrimaculata Linne—Plackford, Elkhart and Wells Counties. May 21-31; July 11-20; September 21-31. IIo. Tir mas 114. IT9. 120. DRAGONFLIES OF INDIANA II Libellula semifasciata Burmeister—Blackford, Elkhart, Kosciusko, Steuben and Wells Counties. April 21-31; May; June 1-20; July 1-20. Libellula vibrans Fabricius—Blackford, Lake and Wells Counties, June: July I-10. Genus Plathemis lagen. Plathemis lydia Drury—Allen, Blackford, Crawford, Elkhart, Kosciusko, Mar- shall, Wells and Whitley Counties. May 11-31; June; July 1-10, 21-31: Au- gust 11-31; September 21-31. Genus Perithemis Hagen. Perithemis tenera Say.28—Allen, Blackford, Elkhart, Jay, Kosciusko, Lagrange, Noble, Steuben, Wells and Whitley Counties. May 21-31; June; July, Au- gust I-I0; September I-Ic. Genus Nannothemis Brauer. Nannothemis bella Ubler—lkhart and Iagrange Counties. June 11-30; July I-10. Genus Erythemis Hagen.29 Erythemis simplicicollis Say—Allen, Blackford, Elkhart, Kosciusko, Noble, Steu- ben, Wells and Whitley Counties. May 21-31: June; July 1-20; August 1-10. Genus Sympetrum Newman. Sympetrum anbiguum Rambur.2°—Blackford and Wells Counties. June 21-30; August; September I-20. Sympetrum corruptun, Hagen.—Kosciusko and Wells Counties. July 11-31; Au- gust I-I0. Sympetrum decisum Hagen.21—Wells County. September 21-30. Sympetrum obtrusum Hagen.—Allen, Blackford, Kosciusko, Lagrange. Noble. Steuben, Wells and Whitley Counties. May 21-31: June; July; August; Sep- tember; October 1-10. Sympetrum rubicundulun Say.-—Allen, Blackford, Iikhart, Huntington, Koscius- ko, Lagrange, Noble, Steuben, Wells and Whitley Counties. June 11-20; July; August 11-31; September 1-20. Sympetrum senucinctum Say.— Huntington, Lagrange and Whitley Counties. July 1-10, 21-31; September I-io. Sympetrum vicinum Iagen.—Allen, Elkhart, Grant, Kosciusko, Lagrange, Noble, Steuben, Wells and Whitley Counties. June 21-30; July 11-31; August; Sep- tember; October 1-10; November I-1¢, 21-30. Genus Pachydiplax Brauer. Pachydiplax longipennis Burmeister—Adams, Allen, Blackford, Elkhart, Kos- ciusko, Noble, Steuben, Wells and Whitley Counties. April 21-30; May: June; July; August 1-20; September 1-10. Genus Leucorhinia Brittinger. Cn St. Joseph, Steuben and Wells Counties. May Leucorhinia frigida Hagen.*? 21-31; July I-10. Leucorhinia intacta Hagen.—Allen, Blackford, Elkhart, Marshall, Noble St. Joseph, Steuben, Wells and Whitley Counties. May 11-31; June; July, Genus Celithemis Hagen. Celithemis elisa Hagen.—Allen, Blackford, Elkhart, Kosciusko, Lagrange, Noble, Steuben, Wells and Whitley Counties. April 21-31; May 21-31; June; July: August; September 1-10. Celithemts eponina Drury.—-Allen, Elkhart, Kosciusko, Lagrange, Noble, Steu- ben, Wells and Whitley Counties. June 11-30; July; August 1-20; September I-10. Celithemis monomelaena Williamson.?’—Allen, Kosciusko, Lagrange, Noble, Steu- ben, Wells and Whitley Counties. June i1-31; July; August 1-20. 12 E. B. WILLIAMSON Genus Pantala Flagen. 121. Pantala flavescens Fabricius-—lIake and Wells Counties. June 1-10; August I1-20. 122. Pantala hymenaea Say.—‘“Inhabits Indiana” (Say). Wells County, July 11-20; August I-10. Genus Tramea Hagen. 123. Tramea carolina Linne.—Blackford and Wells Counties. April 21-31; May. 124. Tramea lacerata Hagen.—Adams, Allen, Biackford, Kosciusko, Lake, Marshall, Noble, Wells and Whitley Counties. April 21-31; May 11-31; June; July; August 11-20; September 1-J0, 21-30. 125. Tramea onusta Hagen—Wells County. July 1-10. BIBLIOGRAPHICAL NovtEs.* 1. Catalogue of the Odonata of North America. Eulletin of the Public Mus., Milwaukee, Vol. I, Article 1, May, 1970, R. A. Muttkowski. 2. Calopteryx of earlier authors must give way to Agrion, 3. Putrida is a synonym. 4. Additions to the Indiana List of Dragonflies. II, Proc. Ind. Acad. of Sci., 1901, E. B. Williamson, p. 120. 5. Additions to the Indiana List of Dragonflies. Proc. Ind. Acad. of Sci., 1900, E. B. Williamson, p. 173. 6. Syn. Neur. N. Am., 1861, Hagen, p. 87 (as annexum) ; see 4, p. 121. 7. Psyche, 7, 1895, A. P. Motse, p. 274. 8. Bulletin 48, N. Y. State Museum, 1902, J. G. Needhamni, p. 246. 9. Under Nehalennia in Dragoniies of Indiana. To. wee 6, p. 83. ile SECT aph 70: 12. See 6, p. II5. 13. See 6, p. 99, and 4, p. 121. 14, nt. News,-12, 1901, P. P. Calvert, p. 65. 15. G. sordidus is a synonym. 16. Ent. News, 25, 1914, KE. B. Williamson, pp. 49-58. 17. See 1, p. 99; Ent. News, 14, 1903, E. B. Williamson, pp. 253-258. 18. G. segregans Needham is a synonym. 19. The North American Dragon-flies of the Genus Aeshna, Univ. of Toronto Studies, Biol. Series, 1912, E. M. Walker. 20. See I, p. 116. 21. Under Aeschna in Drf. of Ind. 22. Proc. U. S. Nat. Mus., 37, 1909, KE. B Williamson, pp. 368-308. 23. Bulietin Wisconsin Nat. Hist. Soc. 9, 1911, R. A. Muttkowski, pp. 91-134. 24. Bulletin 47, N. Y. State Museum, roor, J. G. Needham, pp. 485, 504. 25. Under Somatochlora in Drf. of Ind. 26.~ Ent. News, 18, 1907, E. B. Williamson, p. 5. 27. As L. basalis in Drf. of Ind. 28. As P. donutia in Drf. of Ind. 29. As Mesothemis in Drf. of Ind.: see 1, p. I=5. 30. As albifrons in Drf. of Ind. 31. U.S. Survey Terr. Colo., 1874, Hagen, p. 588. Not a synonym of obtrusiwmn as stated in I, p. 162. Boy branss Amt Manit SOC. 17) Too0" hlacens pa23i- 33. Ohio Naturalist, 10, 1910, EK. B. Williamson, p. 155; as fasciata in Drf. of Ind. 34. As tricolor in Drf. of Indiana. * This list is in no sense complete. and is designed merely as an aid to students of Indiana dragonflies. For a fuller bibliography see 1 above, and the American entomolog- ical journals, especially the Canadian ntomologist and the Entomological News, since IQIO. SS ome eee ee er oe 7 | ! "St Joseph | enemart =: Lasrange « Steuben Laporte! | ee eee LGB ‘ ! Marshall | = “Starke , Kosciusko , <= apes -=-—-- \ | oa Pulaski Fulton “L-_.—.— Rewton t —— ae 4 \ | | white ae Gass: Miami, is mee — e - - _—- « | i Fe) = ao “” = Mp 1, a aay | ! : ; 1 I ! Benton | Carrol | a pane ete Warren 4 Tippecanoe i Clinton Tipton = 4 ! et BEES Bs OS amen. ' Delaware | os: ‘Madison , * Randolph : Boone =; Hamilton ! ‘ec ane — ee ( ( Fountain Vermillion | eee | Putnam bee Oe. jae seen | \ | | ; : : / Parke} ‘Hendricks Marion | Toe Tet \ Sia aa baer Fe i | i Rush i : Fayetic | r | ; ; i | - eae Morgan | Johnson | Shelby ees I ! cay | | ! r VT Franklin; an oi era acaltet| Paes ! | (x = oa o a) fod \ e+ 4 ; | Dwen ! oe i Cot ieee ! Brown laoonctt tee i ES. » Monroe ! | : s : | Ripley ee * Jenaings i 7. \% 7 es Ne = i Shee 2 ey, ‘ise ies: ; Jackson | ct ~~" Switzer“ pale ~ Lawrence j ior? setferson 1 tan L Jf { - ‘ ; “+ : N Knox x Daviess | Martin t Se ee 7 lie, tt Tee & Lf : ! j Washington | a) ~7 ' | i ' | ‘ : ee eaten 1 __j Orange MAP OF INDIANA SHOWING LOCATION OF COUNTIES ‘ a hs Oe ee Oe Nee Te UNIVERSITY OF MICHIGAN MUSEUM OF ZOOLOGY Miscellaneous Publications No. 3 A Collecting Trip to Colombia, South America BY E, B. WILLIAMSON ANN ARBOR, MICHIGAN PUBLISHED BY THE UNIVERSITY FEBRUARY 22, 1918 ADVERTISEMENT The publications of the Museum of Zoology, University of Michigan, consist of two series—the Occasional Papers and the Miscellaneous Publica- tions. Both series were founded and are being supported by Dr. Bryant Walker, Mr. Bradshaw H. Swales and Dr. W. W. Newcomb, except that the cost of the illustrations and of distribution are borne by the University. The Occasional Papers, publication of which was begun in 1913, serve as a medium for the publication of brief original papers based principally upon the collections in the Museum. The papers are issued separately to libraries and specialists, and, when a sufficient number of pages have been printed to make a volume, a title page and table of contents are supplied to libraries and individuals on the mailing list for the entire series. The Miscellaneous Publications include papers on field and museum technique, monographic studies and other papers not within the scope of the Occasional Papers. ‘The papers are published separately, and, as it is not intended that they shall be grouped into volumes, each number has a title page and table of contents. ALEXANDER G. RUTHVEN, Director of the Museum of Zoology, University of Michigan CONTENTS PAGE Mame ETON ere rey 0 Oy oases Oh a Sortie: oe bake dase br cia stains aoe abe Mania 5 Mase rte Ne dete ernst cs ates aly lela sls = vies wa ta eg sw een 6 (OE Cl tino ae ht i i ape 21 Mares on Equipment and Collecting Methods ~....2202....¢.0006.- 2h Seed eae PINTS ota c ine pt ew aero s eee gat sees 21 MeN S mi grec dir. Sats wre s RA hn cinder Maen sah e.e ee ie Be 21 Mieco NNGLWOS,. oss... Whe ne Rana ttes ch Male eet oe 6 hace eee 22 aati iMote AINES (Oe WGAPONNIES ce 5 42 ao ecusce aves Sree wiets Sree @ epee the eros 22 RO EMO CMCC pes ceca Spiess kd Sete as od ain, os Gas, ounce wala Bele ars 24 & wins J : : ¢ A COULECTING TRIP-TO COLOMBIA; SOUTH AMERICA BY EK. B. WILLIAMSON INTRODUCTION This trip to Colombia was made in the winter of 1916-1917 for the purpose of collecting dragonflies. The expenses were met by the Museum of Zoology and by the members of the collecting party, Jesse H. and E. B. Williamson, and the expedition will be known as the University of Michi- gan-Williamson Expedition. Dragonfly collecting had been done in Central America and in British Guiana by E. B. Williamson, and Colombia was se- lected for the present trip as offering a readily accessible, intermediate and relatively desirable region.. Enthusiastic accounts of Dr. Figenmann and Dr. Ruthven increased the writer’s interest in the possibilities of Colombia as a collecting ground for dragonflies. The time of year chosen for our trip was largely determined by in- formation furnished us by Mr. M. A. Carriker, Jr., the well known bird student and collector, whose long residence in Colombia enabled him to give us accurately the data which we needed. This referred primarily to Santa Marta, which we had selected as our landing place, and to the adjacent coun- try. At Santa Marta there are only two seasons, a dry season from Decem- ber to March, inclusive, and a rainy season from April to November, in- clusive. The first four months of the rainy season, April to July, inclusive, are bright and showery. September, October and November are cloudy and misty. The maximum drought is in February and March. The highest wa- ter stage in rivers is reached in September to November, and the lowest in March and April. Most birds nest in May, June and July, and most plants flower in April, May and December. These conditions prevail more or less up the Magdalena River to Honda, but in the interior there are two rainy and two dry seasons, the two dry seasons being December to February, in- clusive, and June to August, inclusive. Other months are more or less rainy. In some localities rains may occur at any time throughout the year hut espe- cially in October. These Colombian seasons do not correspond with the sea- sons at Georgetown, British Guiana, where Mr. Rodway gave the writer the following data: short rainy season, November 15-February 15; short dry season, February 15-May 15; long rainy reason, May 15-August 15; long dry season, August 15-November 15. From other sources it was learned that in British Guiana most birds nest in May and June; showers may be- gin in April; highwater stage is reached June 30-July 15; October and No- vember are low water stages ; September is one of the most pleasant months, and the short rainy season of November 15-February 15 is the most disagree- able time of the year, being misty or drizzly. 6 EF. B. WILLIAMSON With Mr. Carriker’s data before us, we should have selected February or March to May or June as the most favorable four months for our work, but as other engagements prevented this we selected December-March as the best months at our disposal. In selecting a time or season for zoological collecting in the tropics one must of course take into consideration other factors than those of rain-fall. The most essential of these factors are the group or groups of animals especially sought, and the elevation at which collecting is contemplated. In my case, experience in tropical collecting covers only the first six months of the year and regions north of the equator. Without presuming to gen- eralize from such limited experience, I believe that insect collecting generally will prove profitable throughout the dry season in well watered lowlands up to an elevation of five hundred and possibly in some cases one thousand meters ; that high elevations tend during the dry season to become progres- sively less productive as the elevation increases and as the season becomes drier; that the most quiescent state of tropical animal life is attained as the end of the dry season (April to June) is reached, and that at this time many rarer lowland species which are to be found at no cther season appear or attain their maturity ; and that the appearance of similar rarer species in the higher altitudes awaits the showery rains (April to June) which re- juvenate the fauna of both the high mountain valleys and the lowland swamps with the sudden appearance of many species, often in great numbers. ITINERARY Jesse H. Williamson left Seattle, Washington, in October. I left my home at Bluffton, Indiana, in November, and we met at Panama December 5. On December 6 and 7 Charles T. Tribolet and I collected at two small streams in the Canal Zone. Several stream species were teneral ; of Perilestes and Psaironeura remissa only tenerals were seen ; Coryphaeschna adnexa was pairing and ovipositing at a small back-water pool along the railroad track. On December 7 it began raining before noon and rained hard the balance of the day. December 8 was also rainy and no collecting was attempted. The two days, December 6 and 7, mark the beginning and end of Mr. Tribolet’s career as a dragonfly hunter. While in the Canal Zone, we were lhospitably entertained at the home of Mr. and Mrs. Raymond Shady, former 1esidents of Wells County, Indiana. Enroute from Colon to Santa Marta, we collected about Puerto Co- jombia, Colombia, on December 10 and 11. The country was very dry and sandy, and back for three or four miles we found no running streams, only small pools in creek beds. The so-called Arroyo Grande is a muddy, brack- ish backwater at this season. Some of the coastal dragonflies which con- gregate in great soaring flocks were abundant here. These flocks, ustially rather loosely organized, may consist of as many as six species, and hun- dreds or thousands of individuals may be in view at one time. We landed at Santa Marta December 12. The surrounding hills are dry and support a scanty vegetation at this season, cacti producing most of A CoLLEctING Trip To CoLoMBIA 7 the green color. The hospitality of the United Fruit Company was extended to us through the kindness of the local manager, Mr. Sinners, and we made the bachelor apartments our headquarters while at Santa Marta. December 13 to 19 were spent collecting in the vicinity, one day being spent with Mr. James Ryan on a gasoline motor locating favorable points along the rail- road between Santa Marta and Aracataca. The railroad is in operation from Santa Marta to Fundacion, a few miles beyond Aracataca. ‘here is a distinct change from the semi-arid conditions about Santa Marta to a more humid condition about Aracataca and Fundacion. Near the residential prop- erty of the United Fruit Company in Santa Marta is the Manzanares River, which about a mile above here has a left affluent, the Tamacal. These streams can also be reached by following the railroad track. Along the railroad track, before reaching the Manzanares, there were some good pools, and just beyond the bridge over the Manzanares was a pond, on the upper side of the track, of possibly two or three acres. At the lower end of this pond were large areas of a three-parted leaved plant which, in habit, much re- sembled spatterdock, and near the upper end were areas of cat-tails. Fol- lowing the railroad track beyond the Manzanares we came to a low forest of large trees with many small muddy streams, and, near the farther edge of this forest, to the Tamacal River. The gullies leading back from the Manzanares and ‘I'amacal were dry. Near the coast were extensive shallow salt marshes, known as salinas, where Ischnuras abounded and where the coastal libellulines flew by thousands. The days at Santa Marta were sunshiny, rarely hazy, with but one after- noon’s rain. It was generally windy and at night blew so hard that we had to close the doors and windows in order to paper our dragonflies. We spent one day on the Rio Donjaca, a small, clear rocky stream ten to fifteen feet wide about twenty-five kilometers from Santa Marta. We found little of interest here, and walked back to a small sandy stream between kilometers 17 and 18. This also we found disappointing. The seven days about Santa Marta yielded 60 species of dragonflies and 1,208 specimens. Bolivar, located about five miles out of Santa Marta, was a residence of Mr. O. L. Flye, general manager of the Cincinnati Coffee Company. Through the hospitality of Mr. Flye and his secretary, Mr. Robert Sargent, we were entertained there from December 20 to 26. The road from Santa Marta to Bolivar had been extended beyond Bolivar five miles to La ‘Tigrera, following the Tamacal most of the way.. Along the T’amacal and its few tributaries were medium growth forests; the hills were covered with grasses except where thickets or rank herbaceous growths predominated. Ipomoeas or related plants were not so conspicuous as they were about Santa Marta and along the railroad, where they frequently formed beautiful curtains of bloom. About Bolivar the country was largely pasture land of guinea grass with extensive areas of native xerophytic plants. The altitude of Bolivar is probably about fifteen metres, that of La Tigrera about ninety- one metres. The Tamacal and its few permanent tributaries between Bolivar and La Tigrera were clear, rapid and rocky, and for the most part in shade. 8 E. B. WILLIAMSON West and north of the house at Bolivar on the east side of the road from Mamatoca to Lavilla and south of Mamatoca was a large and varied swamp. It was in both sun and shade and contained rank growths of bushes, sedges, cat-tails, ferns often ten feet high, duck-weed, the spatter-dock-like plant seen at ponds near Santa Marta, and a common thick-leaved, floating rosette plant in which Coryphaeschna virens usually oviposited. Here dragonflies were very numerous. Smaller pools occurred elsewhere in bush, mostly a small spiny palm, but they were less varied and few dragonflies were found about them. Along the Bolivar-La Tigrera road and in sheltered sunny nooks Macrothemis flew in great numbers, and at sunset Gynacanthas in lim- ited numbers came out of the brush to patrol the edges of the road. Where this road crosses the T'amacal on a plank bridge one can leave the wagon 10ad and follow the old mule trail from that point to the house »t Bolivar. We found this trail shadier and more interesting than the main road, and usually followed it in our tramps up and down the Tamacal. About four miles from Bolivar, on the road to La Tigrera, an obscure foot path crossed the T'amacal and passed back through the brush and over the hills to a small banana planting. A beautiful little stream, a tributary of the Tamacal, flowed past this planting. On low vegetation in sunny spots in this field we took about twenty-five specimens of a small Progomphus found nowhere else on the trip. Our work about Bolivar brought our collection up to 70 species and 2,257 specimens. . On December 27, we went by mule from Bolivar to Cincinnati, where, through the kindness of Mr. Flye, we were received as guests in his home. Cincinnati is situated at an elevation of 1,371 metres. The distance by mule trail from the end of the wagon road at La Tigrera to Cincinnati was ten miles. Above La Tigrera the mountains were for the most part wooded with only occasional clearings. Near the many permanent streams tree and climbing ferns were numerous and conspicuous, and on many of the banks were great expanses of flowering begonias. Coffee was being picked at the time of our visit, and the many flowering Compositae gave one the impression of autumn at home. We collected for five days about Cincinnati, assisted at times by William Flye and W. L. Laux. Collections were made along various streams from an elevation of 762 metres to 1,371 metres, ihe lower elevations proving more productive. About Cincinnati we found for the first time the peculiar dragonfly Allopodagrion, one of the coena- grionines, which nevertheless rests on leaves, stones, logs or sand with wings as flatly spread as any gomphine. Here also we took at the smallest streams a beautiful and erratic Aeshna, seen nowhere else. While there was no rain during our stay at Cincinnati, great masses of clouds frequently obscured the sun, often for long intervals, and at such times odonate life simply dis- appeared and did not immediately reappear with the sunshine. Moreover the deep canyons in which the streams flowed served to keep them shaded except where some were exposed to the stn a few hours during the middle of the day. Under such conditions all dragonflies became exceedingly wary and sensitive to the slightest movements, and, by rising among the trees to lofty perches far out of reach, sought to escape in the very direction in which A CouLEctiInc Trip to CoLloMBIA 9 pursuit was impossible. Even the usually rather stolid Argias showed this wariness. One day at Agua Dulce, at an elevation.of about 762 metres, we took seven species of this genus. No Gynacanthas were seen about Cincin- nati though they were sought in several likely places. Below the coffee mill we found a peculiar small catfish which climbed waterfalls by clinging to the vertical rocks by means of spines on the under side of the head, and by making occasional flips upward of an inch or less at atime. At rest it looked like a whisp of algae or a bit of rag attached to the rock face and washing about in the falling water. We left Cincinnati January 2 with our collection totalling 80 species and 2,916 specimens. ‘That evening we looked without success tor Gyna- canthas near the pools and in the woods up the railroad above Santa Marta. The Manzanares was carrying more water than when we saw it in De- cember. January 3 was spent along the Tamacal, now scarcely occupying its flat bed of fine sand, the stream being only six to ten feet wide. Our total collection to date was 81 species and 3,196 specimens. On January 4 we went to Rio Frio. We selected this place because of the fine stream, forty to sixty feet wide,—the Rio Frio,—the close prox- imity of the hills, and the accommodations kindly offered us by the United Fruit Company through the Rio Frio manager, Mr. Pepin. About this place there were extensive banana fields. Where not irrigated the soil was dry and sandy and the vegetation stunted. A stream six to ten feet wide with a good flow of water in the hills was found to disappear when it came out on the flat sandy country. Many other streams were diverted for ir- rigation, and the flow of the Rio Frio was largely used for the same pur- pose. There were a number of small marshes in the neighborhood. Weather conditions were good during our stay at Rio Frio. It was usually cloudy in the early morning and again about four p. m. In the afternoon we fre- quently encountered flocks of Macrothemis which numbered thousands of individuals. They were mostly of one species, the females greatly predom- inating, the males very rare, and mature adults of either sex being still rarer or wanting. ‘The only interest of the mdividuals of these flocks was the catching and eating of small insects, and their life at this time was a rather definite transitional stage between the larva and the sexually mature insect. When we left Rio Frio on the morning of January 9 our collection totalled 94 species and 4,135 specimens. Enroute to Fundacion we collected an hour or two about some small marshes and along irrigating ditches at Aracataca. Up to this time the only Neoneura taken had been Neoneura esthera, but at Fundacion we found Neoneura bilinearis associated with Neoneura esthera, numbers of both species ovipositing in dead grassy debris caught in fences and similar obstructions in irrigating ditches. Neoneura bilinearis was so much paler than Brazilian and Guianan specimens, with the areas of green and yellow so much larger, that it was mistaken for a new species. Later along the Rio Cesar, the same species was seen ovipositing at the water’s edge in the solid mud banks. Both species, Neoneura esthera and Neoneura bilinearis, were found together as far inland as Cristalina, near Puerto Berrio. IO E.. B. WiLLIAMSON At three-thirty p. m. we arrived at Fundacion, the end of the railroad. Smaller areas were under cultivation here than at most towns along the rail- road. Along the grade of the railroad Y at the west edge of the town were a few dirty shallow pools about which numbers of Gynacanthas flew in the evenings. West of this Y and visible from it was a large marsh situated in a great pasture field. This marsh was shallow and the open area did not exceed an acre. ‘The principal vegetation was grass. Other plants which attracted attention were a conspicuous Juncus; a water-lily with crinkled margined leaves and white elevated flowers ; a low growing, beautiful purple flower; and clumps of a large plant resembling a Heliconia, but with branched flower stalk and small purplish flowers. Over the low hills west of this marsh was a similar but smaller one. Here were growing another species of water-lily with entire leaves, and two large clumps of a giant Cyperus. Still farther west we encountered a series of long pools or lagoons, running in a northerly direction. One may reach these lagoons also by going west of town along the grade made for the extension of the railroad, this grade having been extended beyond the point where the lagoons are crossed. These lagoons represent an old river bed, probably that of the Rio Fundacion. Near where the railroad grade crosses them they were deep and treacherous and filled with floating grass and debris. Several times in collecting about them I found myself floundering in the water over my head. The adjacent country was semi-arid and brushy and was generally over-pastured by half- starved cattle. ‘Trees on the immediate banks of the lagoons were literally filled with iguanas, which hunted about on the ground, and, when disiurbed, ran for the trees, reminding us, strange as it may seem, of fox squirrels at home. The lagoons themselves were tecming with crocodiles, whose grunt- ing sounded much like that of dogs. From the southwest part of Fundacion a well worn trail led off through the woods in a southerly direction. A mile or less from town smaller trails led off from the main trail. These, as far as explored, led to smaii farms of sugar cane, yuca and bananas, or pastures cleared in the forest. The main trail one and a half or two miles out from town crossed a large, nearly dry creek or arroyo in which remained pools of varying lengths, eight to ten feet wide. This very muddy creek bed, followed down stream for about half a mile, brought us to the: Rio Fundacion. Just before the main trail crossed the arroyo, a trail led off to the left. This itself immediately crossed the arroyo, and, after passing through a small growth wood and a pasture, came out on the bank of the Rio Fundacion above the mouth of the arroyo and then followed up the river. Where the path followed the river closely, we took a number of specimens of a mest peculiar small gomphine. This gomphine was most common in a growth of Heliconias and mixed palm and bamboo forest, where the path passed between the river and a large lagoon into which the river flows during high waters. This lagoon was not discovered until our second visit. Doubtless many such lagoons are scattered through the forest, unknown and inaccessible because of the spiny and thorny jungle. ‘The path ended m a small farm where a number of products were cultivated, and in the fields, resting on grass and low plants, A CoLLEcrInc Trip to CoLoMBIA II another gomphine was taken. Going back to the main trail and crossing the arroyo, we found another trail brarching to the left. This trail passed through the wood and by some small clearings and plantings, including a small banana field on the left, to end at a shelter hut in a yuca and sugar cane field two or three miles from the main trail. Resting on twig tips in the wood along this trail a third gomphine was taken. This gomphine was usually found in moist locations where Heliconias of gigantic size grew in clumps and where the forest had a dense growth of dwarf bamboo one to two feet high. They also occurred, but in fewer numbers, wherever a fallen tree top furnished some dead twigs for perches. They were found in both sun and shade but preferred mixed or patchy sunlight. Fast of town was an irrigating ditch. ‘The trail on the south side of the cemetery led to it through brushy pastures. Just above the juncture of the ditch and the trail were some low woods densely carpeted with selaginel- las and similar low plants. In this low vegetation and in adjacent brush, consisting of spiny palms, hooked bamboos, Heliconias and other mostly spiny plants, we found large numbers of Metaleptobasis. Most of these we caught by hand as we wriggled through the brush. Any other mode of progress was impossible, and an insect net was useless under such condi- tions. The days at Fundacion were generally bright. One afternoon was cloudy ; and it rained most of one night and till about ten a. m. the next day, remaining cloudy till noon. The great flocks of Macrothemis with which we had grown familiar about Santa Marta, Bolivar and Rio Frio, were absent here. We left Fundacion the morning of January 15, the collection to date numbering 107 species and 4,945 specimens. We took a night ferry from Cienaga to Barranquilla which we reached about six-thirty a. m. January 16. We left Barranquilla on the river steamer Manizales after sundown Jan- uary 18. The boat spent the early part of the night ramming first one bank and then the,other, and tearing souvenirs from sundry boats and barges tied up in the canal leading from the city to the Magdalena River. We ar- rived at Calamar about sundown January 19. The night of January 20 we tied up at Yati where a lot of cattle were to be loaded on board next day. We took advantage of this delay and spent the forenoon of January 21 col- jecting up river to Magangue where our boat arrived about two p.m. The country about Yati and Magangue was fairly typical of a hundred miles or more of the country adjacent to the lower Magdalena. Great pastures suc- ceeded one another, with frequent lagoons a short distance back from the river and paralleling it. Heavy forests were rare. The hills are far back from the river, frequently none being seen in a day’s travel. Walking from Yati to Magangue the road followed closely along the river with extensive lagcons on the other hand, and passed over two small tributaries. We saw Brachymesia and Miathyria literally by thousands. El Banco, 86 leagues up the river from Barranquilla, had been chosen as our first stop. We arrived there about five p. m., January 22, and made ourselves at home at the Central Hoiel. Cattle raising was the principal T2 FE. B. WILLIAMSON business at El Banco. For several hours before reaching the town we were passing, on the left bank of the river, the finest and most extensive pasture seen. El Banco is well situated on high ground at the juncture of the Rio Cesar with the Magdalena. Above El Banco was extensive fishing in the Cesar which was deep and swift with bare washed mud banks. The fish, taken in nets, were dried on racks or on the ground. We followed a wide road, the Camina Chimichague Y Chiriguana, leading out from El] Banco, which at first passed through some slightly rolling country. The stream beds here were all dry, with the banks and immediately adjacent country wooded with bamboo and palms. About three miles from town two trails left this road. The trail to the right apparently followed the Rio Cesar; the other, to the left, was probably the Mompos road. These trails immediately crossed a perfectly flat, scantily grassed) sun-baked plain over which termite nests were scattered with suiticient regularity and with just suf- ficient variation in size to give the impression of an old and very large ceme- tery. ‘That we were suffering from thirst and our heads fairly rocking with the heat when this funeral plain came into view, detracted nothing from the vividness of the impression. Both trails led to a scarcely perceptible de- pression, marked by adjoining forest which traversed the plain like a broad ribbon. ‘The shallow creek bed in this forest was almost dry, but about the long stagnant pools Psaironeura remissa flew by hundreds. One of these pools was possibly one hundred yards long. All were shaded. Large numbers of small brown monkeys took considerable interest in our work and showed some offense at our presence. About half a mile out from town, on the right hand side of the main road, are several marshes. We attempted to a these and then cut across in an easterly direction to the Rio Cesar, but were unable to do this because of low impenetrable jungle. ‘The marshes themselves yielded but poor re- sults. For about two miles which we explored above El Banco the right bank of the Rio Cesar was largely pasture, the original forest having Sees almost entirely destroyed. At this point a large affluent from the right comes into the Rio Cesar. Like the Magdalena, the Cesar had lagoons, only on a smaller scale, and in some of these, which were stagnant and muddy and heavily shaded with brush, we took numbers of shade-frequenting coena- grionines. Sometimes we secured these where the dense gloom made it al- most impossible for us to distinguish the insects. It is rather remarkable that the great forested areas of the north, at least of the Nearctic region, have no such shade-frequenting dragonflies as occur about both ponds and streams in the tropics. Possibly our more northern species are more ancient than the environment in which they now find themselves. We left El Banco the afternoon cf January 26, bound up river for Barranca Bermeja, 139% leagues above Barranquilla. Our collection now totalled 113 species and 5,407 specimens. The days that followed differed but little from the days spent between Barranquilla and El Banco. At night the boat was tied up to the bank. During the day there were the same frequent stops to take on wood. The A CoLLEcTING TriP To CoLomBIA 13 food was good, sleeping on deck at night was pleasant, our traveling com- panions were friendly and agreeable, and the frequent delays cost us nothing as our tickets included meals. Consequently when a sand bar held us for a few hours or a smashed wheel delayed our progress for a day, the calmest philosophy could prevail, at least among the passengers. It would be possible for a collector interested in insects attracted by light to rig his outfit on the upper deck of one of the slower river steamers and to obtain thus on a round trip from Barranquilla to La Dorado, a distance of 198 leagues, a fair representation of the night-flying insects of the whole lower Magdalena. The time required would be about three weeks, and the round trip trans- portation, including meals, was, when we were there, $30 on one line of steamers and $60 on another; the service was about the same on both. We approached Gamarra January 27. The country is flat with moun- tains in the distance. At Gamarra the hills seemed relatively near at hand, but we learned that Aguachica, about twelve miles from Gamarra on the road to Ocana, is in the river plain, and that Pie de la Cuesta, about twenty-four miles from Gamarra, on the same road, is in the hills. Though a small town, Pie de la Cuesta had ample accommodations for travellers. Ocana is about forty-two miles from Gamarra, a journey of a day and a half by saddle horse, costing, with a pack mule, $6 for the trip. It was our intention to go to Ocana on our return down the river but lack of time prevented this. The topography above Gamarra is the same as that below the town. The country is flat with mountains in the distance. There were numerous sand bars in the river where we observed many crocodiles. As on preceding days small collections were made at several points where the boat tied up to take on wood. During the afternoon of January 29 we reached Barranca Bermeja and put our baggage ashore. As no hills were in sight, and the country was very dry and unpromising with no small streams, we re-embarked and paid our fare to La Dorado, the end of lower river navigation. Above El Banco the forests were larger, the bluffs more frequent, and the cienagas (lagoons) rarer. Below Puerto Berrio the Magdalena cuts through a bluff or ridge fifty to seventy-five feet high. January 30 about seven p. m. we arrived at Puerto Berrio, 16334 leagues above Barranquilla. On both sides of the town are frequent low hills. Through the kindness of Mr. Lope M. Montoya G., of Medellin, to whom we are indebted for other data, we learned that the elevation of Puerto Berrio is about 130 meters. Our boat was scheduled to leave Puerto Berrio January 31 at four p. m., so we decided to spend the day collecting and set out at seven-thirty a.m. About three and one-third kilometers from town a small sluggish stream, flowing north, crossed the railroad track. Near the track this stream was in a grass marsh condition, but it soon passed below into a dense over- flow forest of tall trees, spiny palms and lower shrubs and vines, through which we could pick our way with slight difficulty. At this season the wood was moist but there was no standing water. The creek was from four to 14 E. B. W1LLIAMSON ten feet in width and had a maximum depth of three feet and a generally firm mud bottom. Dragonflies were numerous in the varied habitats formed by forest, stream, sun and shade, and such diverse things as Perithemis and Metaleptobasis were found in numbers. ‘This forest was rapidly being de- stroyed to furnish fuel for the railroad. We were returning from our collecting about three p. m. and were near the Magdalena about a kilometer above town, when we heard a river boat chugging its way up stream. A hurried dash to the river’s bank showed us our boat in midstream, carrying away, as we knew, all our baggage and supplies, while we stood in our wet clothes in a bamboo thicket on the rivers bank and marked her passing. We yelled and waved our nets but attracted no attention. Fortunately J. W. thought of his revolver, and three shots attracted the attention of the captain, who most obligingly ran into shore and picked us up. From the other passengers, apparently as excited and delighted as ourselves, we learned that the boat had discharged its cargo, and, after whistling for us in vain for about an hour, had left Puerto Berrio between two and three p. m. This incident is related for two purposes,— to show the obliging friendliness of the captain in picking us up, and to record an instance in Latin-American affairs, when, for the only time known, events moved ahead of their schedule. The statement may be received in- credulously by other travelers. Twelve to sixteen leagues below La Dorado the forest was not quite so large as that below Puerto Berrio. The dirt banks were generally slop- ing, instead of vertical and caving as they usually were up to El Banco and some distance above it; and frequently the shores were rocky or gravelly instead of sandy as below. Both above and below Puerto Berrio were a few rock exposures. We arrived at La Dorado, 198 leagues above Bar- ranquilla, on February 1. Contrary to what one might have expected from its commercial importance, La Dorado was a poor and insignificant town. The character of the country along the railroad between La Dorado and Honda has changed greatly since Professor Bingham described it as a dense tropical jungle. We found it to be almost continuous pasture. There were two or three nice streams and some marshes within five miles of La Dorado, but from that point on to Honda the country was almost worthless to a dragonfly collector. We arrived at Honda about three p. m., February 2. Enroute we had made up our minds not to stop there, as we had originally planned to do. Observations and inquiries, which we made while the train lay over at Honda for nearly two hours, strengthened our decision and we entrained for Maraquita. There were small streams near kilometer post 37, between posts 38 and 39, and between posts 43 and 44. From about kilo- meter post 44 to Maraquita the country was a continuous flat pasture with high rocky hills in the background. At Maraquita we found a small satis- factory hotel, and Mr. A. G. Le Clercq, general manager of the railroad, to whom we carried a letter of introduction, gave us valuable assistance and advice in reaching desirable collecting grounds. The elevation at Maraquita is about 457 metres. The San Juan River near Maraquita furnished the station and town A CoLLEcTING Trip To COLOMBIA 15 with its water supply. Near where the trail leading from town to the intake joined the river there was a large waterfall. Above this fall the San Juan was a fine rocky stream five to ten feet wide with rapid drop and with some high waterfalls. The sides of the valley through which it flows were high and rough and wooded with tall trees and bamboos, four to five inches in diameter, beneath which grew many Heliconias and some aquatic plants. Below the waterfall to its mouth in the Cuamo, it was a beautiful rapid stream six to fifteen feet wide, with some cliffs and generally high banks, but with no falls. The Cuamo carried about three or four times as much water as the San Juan, and, where we explored it for a short distance above the mouth of the San Juan, it was similar in character to the lower San Juan. It was one of the most beautiful rivers I have ever seen. About three kilometers south of town the railroad crossed a very small stream which, when we were there, had almost ceased to flow. This we followed to its mouth in the Poquera. The Poquera was about five to fifteen feet wide and had a wide, relatively shallow, rocky and sandy bed. It is said to become nearly dry in the dry season. As might be expected, species of Argia and Hetaerina were the con- spicuous dragonflies where we collected about Maraquita. Along the upper San Juan we saw for the first time specimens of a new Thorine genus, Miocora, since described by Dr. Calvert from a Costa Rican male. Near the same place we took a female of an undescribed Erpetogomphus, a genus hitherto not known south of Costa Rica. On the same stretch of river we saw Heteragrions for the first time since leaving the Canal Zone, and Al- lopodagrion, previously taken at Cincinnati near Santa Marta, was common. Along the Poquera and its small tributary we found a small protoneurine, a much more inconspicuous insect than the notoriously inconspicuous Psaironeura remissa. Along the bank of the lower San Juan, near the wa- ter’s edge, where a yellowish seepage formed a little swampy spot about one foot wide and six feet long, we took several specimens of a small Argia which was seen nowhere else. Damp places about the taps of the water pipes in the town were frequented during the day by the tropically omnipresent Orthemis ferruginea and after sundown by a few Gynacanthas. The weather at Maraquita was not the most favorable for collecting as generally it was very cloudy until about ten a. m.; and during the rest of the day the sun usually shone through a haze. A light rain fell during the night of Feb- ruary 5. On February 6 we took train from Maraquita and returned to La Dorado. Our collection now numbered 126 species and 6,097 specimens. Among the five river steamers tied up at La Dorado we found the General Cordoba, on which we had travelled from El Banco to La Dorado, and we at once engaged passage to Puerto Berrio. We left La Dorado about five a. m., February 7, and reached Puerto Berrio about ten a. m., February 8, having been delayed by a smashed wheel and stranded on a sand bar. The remaining hours of the day were spent along the same stream where we collected January 31. At some small grassy marshes along the railroad grade, enroute to this stream, we saw Coryphaeschna virens, a 16 E. B. WILLIAMSON Tramea, and several other typical tropical swamp species of wide distribu- tion. At six a. m., February 9, we took the train from Puerto Berrio to Cisneros. For about twenty miles the track ran through forest and fields and then began to climb into higher hills. Cisneros, the end of the railroad, at an elevation of 1,060 metres, was surrounded by country almost stripped of its forests, the remnants consisting of scattered clumps on the high hill tops. ‘The Nus was yellow and dirty, due to the activities of gold diggers. The hotel at Cisneros was good but the devastation of the adjoining country made it an unfavorable collecting point. Sofia, near kilometer post 101, had a hotel, and was probably a much better collecting point: than Cisneros. The valley of the Rio Santa Getrudis opposite Cisneros, across the Nus, seemed to have more trees than usual in this region and we collected there February 10. The Rio Santa Getrudis was a clear stream ten to thirty feet wide, rapid throughout, and with many large waterfalls and rock masses. Had time permitted, we probably should have found the many small tribu- taries more productive than the main stream, but a little time spent on these was not encouraging. Along the main stream were only small clumps of forest, but the tree ferns, some thirty feet high, gave some hint of the glory that had been. From its mouth in the Nus we followed the stream up through the hills for several miles. ‘Throughout this distance it was largely exposed to the sun, and the abandoned hill sides were covered with rank im- penetrable grass and dewberry briars. Apparently the agricultural method in use consisted in clearing the forest, planting to crops, usually corn or sugar cane, until the soil was exhausted, abandoning it, and making another encroachment on the forest. The result was that the farmers lived far away from town, up the valleys of the numerous streams. The paths from the farms to town were conspicuous as yellow streaks on the devastated hills. The days collecting netted us only four dragonflies out of six seen, an almost unbelievably poor day. We needed no further evidence, and the next day, February 11, we took train back towards Puerto Berrio, leav- ing it at Cristalina, about twenty-eight kilometers above Puerto Berrio. Several good points were accessible along the railroad between Puerto Berrio and Cisneros. There was a good hotel at Puerto Berrio, and the train, leav- ing there early in the morning and returning in the evening, would have made it possible to collect over a wide variety of country with Puetro Berrio as headquarters. Cristalina was a little native village with no posada (inn), but we were able to rent a room in a hut from one family and to obtain board with another family, that of Lino Zapata, an intelligent native who acted as our mozo during our stay at Cristalina. The elevation of Cristalina is about 320 metres. It lay in a densely wooded country abounding in beautiful small streams. Conditions were humid, vegetation was rank and small tree ferns were noted. A small stream, the Quebrada Cristalina, flowed directly through Cristaiina. The water of this stream was brought from the hills to the village and railroad through an iron pipe. During our first day at Cristalina, February 12, we followed this pipe back to the intake. Between the intake and the town the A CoLLEcTING Trip To COLOMBIA i stream flowed largely through pasture and brush land, but above the intake it flowed in forest. Here it was only one to three feet wide and was fre- quently lost in the stony gravelly bed. It had its origin in the hills about a mile above the intake. The richness and peculiarity of its dragonfly fauna may be realized from the fact that our first day’s collecting yielded a Miocora, two species of Palaemnema, a Perilestes, three protoneurines, three Heteragrions, a Philogenia, an Allopodagrion, an Acanthagrion, a Megalopre- pus, two Mecistogasters, and numbers of Argias, Hetaerinas, and libellulines. At a later date, collecting along the quebrada in the brush between the town and the forest we found two species of gomphines. Below town the rail- road followed the Quebrada Cristalina more or less closely. About midway between kilometer posts 26 and 27 was a path to the left, past a single native hut in a field. This path brought us to the Rio Diez-y-ses, where it was bridged by a large log. Just above this foot bridge was the mouth of ‘Juebrada Cristalina. The Rio Diez-y-ses above the foot bridge was largely in forest. It was a fine stream fifteen to thirty feet wide, of varied character, with mud, sand, rocks, long pools and some rapids. Beyond the foot bridge over the Diez-y-ses was an abandoned railroad spur, running back at right angles to the stream. Following this spur to its end and taking the trail which skirted the forest, with brushy pasture on the right hand, we came to the Quebrada Sabaleticus, about three-fourths of a mile from the Diez-y-ses. This quebrada, in the forest, was a wonderfully fine stream six to twelve feet wide with frequent long pools and rapids. The firm footing, the beau- tiful and varied scenery, and the rich fauna made collecting here almost ideal. Here for the first time we saw Cora and observed its peculiar habit of ovipositing, which it shares with Miocora. The eggs are inserted in com- paratively solid but barkless horizontal tree trunks or pieces of logs over water. In some cases the logs were scarcely damp, and the eggs were piaced as high as six feet above the water. At least three species of gomphines lived along Sabaleticus, and no less than four Heteragrions. Following the Sabaleticus up stream one and a half or two miles from where it emerges from the forest into the brushy pasture we came to a waterfall between four and five feet high—the highest waterfail we dis- covered on this quebrada. About a quarter of a mile above this waterfall a very small tributary from the right enters the quebrada. We followed this tributary to its source in the hills. About a half mile back from its mouth, standing at an angle of 60°-g0°, is a broken rock face over which ' water dripped. Here and here only we found Mesagrion, hitherto known from a single male specimen, which, while closely related to Heteragrion, has the peculiar habit of resting with wings closed, and not half opened as one would expect from its relationships. Below kilometer post 25 and on the left side of the railroad track was a cattle chute, from which a trail led to a clump of farm houses in a brushy pasture on the right bank of the Rio Diez-y-ses. A short distance below the cattle chute was a gate and another more distinct trail which brought us to the same destination. If we crossed the Rio Diez-y-ses at this point 18 E. B. WILLIAMSON and then kept to the right, following the cattle paths through the brushy pasture, we came to the Quebrada Camelia. By keeping to the left instead of to the right, we would come to the Quebrada Sabaleticus, but this route to the latter stream was longer than the one described above. The Quebrada Camelia is another beautiful little forest stream about twice as large as the Cristalina and two-thirds or three-fourths as large as the Sabaieticus. Here we took the fifth known specimen of Cyanogomphus, a new species, no two specimens of the same species being known. These three quebradas are all tributaries of the Rio Diez-y-ses. Their jocations have been described at some length because of the wonderful collecting conditions found about Cristalina. In the town and along the edges of the brushy pastures Gynacanthas flew in numbers after sundown. The edges of the forest where tree tops and logs lay in confusion were alive with a great variety.of libellulines. In sunny nooks several species of Macrothemis hawked in numbers. Butterflies and beetles were numerous and varied, and many birds noticed nowhere else were seen here. ‘The lo- cality was readily accessible, being reached by train from Puerto Berrio, where all the river steamers from Barranquilla stopped. The people were kind and helpful in every way, and altogether this was the most ideal col- lecting locality I have found in the tropics. Expenses were at a minimum. Our room cost us about fifty cents a day. Arrangements could be made doubtless at half this price. Our food with the Zapata family was good, though unvaried, and cost fifty cents a day. We paid Lina one dollar a day for acting as mozo. The weather conditions at Cristalina were not the best when we were there as the following brief notations for several consecutive days will show: February 12, sunshine most of the day; 13, cloudy most all day; 14, cloudy until noon, afternoon sunny; 15, cloudy all day, a few minutes rain in the afternoon ; 16, forenoon cloudy threatening rain, afternoon cloudy with short intervals of sunshine, rained during the night; 17, forenoon fairly clear, afternoon mostly cloudy; 18, rained about six a. m. but the day was gen- erally sunny; 19, a drizzling rain for fifteen to twenty minutes about eleven a. m., weather conditions unfavorable all day. February 20 we returned to Puerto Berrio with our collection number- ing 148 species and 7,993 specimens. The next day we collected once more in the wood near Puerto Berrio. Along a wood cutter’s trail in the forest near the stream J. W. saw a red or reddish aeshnine (?) almost as large as a Staurophlebia. February 22 we left Puerto Berrio on the steamer Ayapel and passed Barranca Bermeja about one p.m. Here great schools of fish were swim- ming up stream near the bank, forming a wide silvery ribbon in the water. Crocodiles were very numerous in the river. On February 23 we saw twenty-eight lying closely together on one sandbar and forty-six on another. A herd of cattle were in the water near the second group but the crocodiles and cattle paid no attention to each other. Our boat tied up for the night at a wood pile a short distance above El Banco, where.we arrived about six- thirty a. m., February 24. We arrived at Magangue about five-thirty p. m. A CoLLEctTING TRIP To CoLOMBIA 19 and spent a pleasant hour with Mr. Carlos Nieto, a commission merchant with whom we had formed an acquaintance enroute up the river. That night our boat tied up at a cattle chute below Magangue. About daybreak they began loading seventy-five cattle on board, and, this accomplished, we reached Calamar about three p.m. We remained at Calamar until about two a. m., February 26, when we started for Barranquilla. The long delay we ex- perienced in getting out of the canal leading to Barranquilla was repeated on our return. About eight a. m. the yacht Ideal, running as a ferry be- tween Barranquilla and Cienaga, passed us in the canal, and, after prolonged shoutings and wavings on our part, returned, picked us up, and landed us at Cienaga about four p.m. We caught an extra mail train from Cienaga about seven p. m. and slept that night in our old rooms at the bachelors’ quarters of the United Fruit Company at Santa Marta. While ascending and descending the Rio Magdalena the objects probably most interesting both to the other passengers and ourselves were the large number of flowering trees in the forest. Other trees were also interesting out for other reasons. Frequently along the water’s edge and just back from it were dense stands of small or moderately sized trees with very light colored trunks and branches and large leaves, known as the guarumo. Overtopping the forest, and sometimes standing out alone in beautiful ma- jesty, were the bonga trees, at times with unbranched trunks rising one hun- dred feet, the perfect mushroom-shaped tops adding another fifty feet to their heights. The trunks exude a latex widely used by the natives in the treatment of boils. The wood is soft and useless. A number of flowering irees, as seen from the river steamer, seemed to rise to about the same height in the forest. One rare kind was a snowy white; another commoner kind was a bright dandelion yellow; another was blue; another dark blood red; and a very common and very beautiful one was a peach pink. ‘There was little uniformity among the names given these trees by various pas- sengers on the boats. A small broad tree which was very common had orange yellow flowers and orange brown buds in great profusion. Another common small tree had flowers in clusters, some of them red and some white or greenish. What we at first thought were the dull reddish bronze flowers of a tree, we found, on a closer view, to be the flowers of a vine which in some cases almost concealed the tree top over which it spread. Once at Maraquita we saw an ant trail where the petals of one of the pink trees were carried in a continuous narrow ribbon of color; a short distance away another ribbon of color, in this case bright yellow, moved slowly but uninterruptedly along. It must be understood that the flowering trees mentioned above were not dominant in the forests along the Magdalena at the season we saw them. These forests were, like all lowland tropical forests we have seen, rather sombre masses of varying shades of green only rarely relieved by other and brighter colors. That enraptured vision which beholds tropical rivers flowing through ever-changing vistas of brilliant flowers of every conceivable hue has never been vouchsafed us. Neither have we been gifted with those supernatural powers of sight and hearing which some travellers possess and 20 E.. B. WILLIAMSON which reveal to them in their voyages some of the rarest and most secluded animal denizens of the densest jungles. In fact nothing is more difficult to reconcile than the varying accounts of travelers in the ¢ropics. For example, one celebrated author has reported a dense tropical jungle where we found broad pastures, and on the Magdalena he never saw a single crocodile where we observed thousands and where, by actual count, forty-six were seen on one small sandbar. February 27 we collected in the low forest along the Tamacal and along the railroad about three miles out from Santa Marta. The leaves under the trees were very numerous and dry, but the trees were bright green, and more in flower than when we previously collected here. The ground was still moist under the leaves, and a small flow of water was still in the little streams in the wood. Dragonflies were rarer than formerly and no additional species were seen. ‘The vegetation on the adjoining hills was quite brown and dead in appearance except for an occasional bush and numerous cacti. A yellow flowered prickly pear cactus was in bloom. The abundant, large, coarse, bushy, pink flowered Ipomoea (or related genus) was about done flowering and was becoming leafless, its seed pods burst- ing and revealing the hairy or woolly seeds. The flowering trees, which were numerous, had, so far as noticed, white blossoms. Fires were burn- ing on the hills at some places. Coming down the Magdalena River we had seen many fires in the mountains. Our collection, including the speci- mens taken at the Canal Zone, Cartegena, Puerto Colombia and at various stops up the Magdalena River, not hitherto counted in the total, now num- bered 150 species and 8,553 specimens. The evening of February 27 we spent with Mr. and Mrs. Flye, Mr. and Mrs. Ryan, Robert Sargent and other friends; and the morning of February 28 about five o’clock we left Santa Marta on the United Fruit Company steamer of the same name bound for Colon. We arrived at Colon March 1, and left there March 4. On March 5 we came at five p. m. to Bocas del Toro, Panama, and about six p. m. to Almirante. Here we looked without success for Gynacanthas. . March 6 was spent pleasantly with the English entomologist, Mr. C. B. Williams, who was studying the local frog-hoppers. We went on the railroad to Guabita and spent a few hours collecting there. Along the rail- road from Almirante to Guabita there is some good collecting country, especially about twenty-three miles out from Almirante. Almirante would provide good headquarters for one wishing to work in southern Central America, as a varied country is accessible from here, and the expenses would be much less than in the Canal Zone. The evening of March 6 we left Almirante, and arrived at Havana about ten a. m., March 10. We took a short country drive and at favorable looking streams searched in vain for Neoneuras. The boat sailed about four- thirty p. m. and reached New Orleans about six p. m. March 12, where we passed customs without friction or expense. A CoLLectinc Trip To COLOMBIA 21 COST Ob TRIE Our expenses were about $650 each, but, as this included a number of unnecessary purchases, $600 may be considered a fair estimate of the cost. The round trip fare from New York or New Orleans to Santa Marta on the United Fruit Company steamers, including meals, was $180. Hotel rates at Santa Marta and Barranquilla were $2 a day; at El Banco the rate was $1.20 a day. The first class charge from Cienaga to Barranquilla was $3.50, with stateroom $1 extra. The round trip first class fare from Barranquilla to La Dorado without stateroom was $30 on one line of steam- ers and $60 on another line. On the first line the stateroom adds about forty per cent to the cost; on the second line about twenty per cent—that is, stateroom charges are about the same on both lines. We had state- rooms from Barranquilla to El Banco, but at no time thereafter, as we really preferred sleeping on déck and the shower bath rooms sufficed for shaving and changing clothing. Sleeping on deck, however, requires retir- ing at a certain hour, when all the cots are brought out and distributed over the deck, and arising at a certain hour in order that the deck may be cleared for the day. Travelling on the river with frequent stopovers adds very lit- tle if any to the fare, as this is equitably figured on the basis of leagues between points. Mozos such as we employed for carrying our collecting out- fits and lunches charged from sixty cents to one dollar a day. Carriages in cities for ourselves and several pieces of baggage cost us usually about $1 aotrip. NOTES ON EQUIPMENT AND COLLECTING METHODS Extra Equipment.—At the hotels at Barranquilla one can arrange for the purchase of a mosquito net, a light blanket and a pillow for use on the river boats. We carried these necessary articles in an old burlap sack. Mr. Carriker recommended folding cots, but we were not able to purchase these in Barranquilla and were able to obtain only one cot at Cristalina. As a consequence during our stay there, J. W. slept on a narrow bench up- holstered with old shirts and miscellaneous clothing. On the river steamers cots are furnished. Medicines ——We made it a practice to take four grains of quinine a day in four doses. J. W. had no fever and I had only one light attack which put me in bed one day at Rio Frio, and from which I recovered after increasing the quinine for a few days to about fifteen grains a day. Amoebic dysentery was common along the river, and I had an attack at Maraquita, contracted probably at El Banco. No physics were taken but paregoric in frequent doses and alcresta ipecac were used from the start. Paregoric was discontinued when it was no longer needed, and ipecac, equivalent to 400 grains, was taken sixty grains a day. Recovery was rapid and I was incapacitated for field work only two or three days. For red bugs (harvest mites) we used the liquid known as cresol compound. This is known also by a great many trade names, and is sold everywhere as a disinfectant and for a dip or wash for live stock. It makes a milky mixture with water. We used it about 22 FE. B. WILLIAMSON one part to ten parts of water, dipping a small rag in the mixture and moist- ening the entire body, morning and evening, and sometimes once during the night; in each case we allowed the fluid to dry on the body. It acts as a local anesthetic, and allays the irritation caused by the bites of other in- sects. It is also a splendid antiseptic. Its use enables one to collect where continuous field work would otherwise be impossible. It may be remarked that the country about Santa Marta and more especially the country about Kl Banco have more than their share of biting and burrowing pests. At Bolivar excessive swelling of the hands, caused by insect bites, compelled J. W. to stay indoors a day or two. He counted one hundred and sixteen fresh bites on one arm one day. At El Banco we entertained red bugs by the thousands. Aqua ammonia was also useful in allaying the irritation caused by insect bites, and a scorpion sting J. W. received at Maraquita gave him little trouble when so treated. ‘The prickles of a large leaved spiny plant, called ortiga or pringamosa, which were very irritating, invariably produced pustules in spite of all our remedies. Among so many spiny plants scratches and cuts result almost daily and in wading among logs and rocks one’s shins are frequently barked. The wounds so caused were thoroughly cleansed and rubbed with a bit of gauze or cotton soaked in a solution of one hundred and twenty grains of resorcin and twenty grains of salicylic acid in eight ounces of fifty per cent alcohol. During the entire trip no wound became infected. Constant wading softens the feet and often the toes about the nails become sore and irritated. Iodine is the best treatment for this. It was found helpful to dust in our dry socks, after the day’s collecting, a mixture of talcum powder, powered alum and sulphur. An attack of “dobie itch” was quickly cured with balsam of Peru after treating it without success with iodine and later with boracic acid. Of course we carried the other paraphenalia usually found in first-aid outfits, but the above suggestions may prove useful to one planning his first collecting trip in the tropics. It should also be stated that salol is very valuable as an intestinal antiseptic; a missionary nurse told us she cured many cases of dysentery with it. J. W. had occasion to use it at one time and it gave almost immediate relief. Collecting Methods—The methods employed were essentially those which I have previously described. However, a few additional suggestions may be made. In papering it is well to hold the envelope flat and leave the abdomen of the insect, especially of coenagrionines, a little above and parallel to the long fold in the envelope, and thus avoid pinching the speci- men in the fold of the paper when the envelope is closed. At Cristalina our supply of smaller envelopes became exhausted, and it was necessary to paper several specimens in one envelope. This was accomplished by keeping the abdomens parallel and at right angles to the long fold in the envelope, arranging the insects so that the pressure of the envelope fell entirely on thoraces. ‘he envelope was so placed in the drier that the abdomens hung 1“Directions for Collecting and Preserving Specimens of Dragonflies for Museum Purposes.” Misc. Publ., Mus. of Zoology, Univ. of Mich., No. 1. A CoLLEcTING TRIP To CoLOMBIA 23 straight or extended parallel to the long fold. Of course careful handling ‘of the envelopes was necessary until the insects were dry, but when, on our return, we opened these envelopes, sometimes containing six insects, we found the specimens beautifully preserved with a minimum amount of dis- tortion. In papering aeshnines and gomphines, instead of using the bit of glued cord at one end of the envelope, as is our custom, we usually used some com- mon libelluline with a thorax sufficiently bulky to insure no pressure of the envelope on the abdomen of the other specimen. This worked as well as the glued cord and reduced the number of envelopes required to paper the day’s catch. In papering libellulines, in which it is desirable that the legs should not cover nor obstruct the view of the accessory genitalia, it was found that by taking the opened envelope in the left hand and by holding the wings of the insect with tweezers in the right hand, it was possible, without loss of time in papering, to brush the legs of the insect forward, over the edge of the envelope, thus entirely clearing the genitalia. This may seem a small matter except to one who has had occasion to examine the genitalia of a few hundred or thousand libellulines. During our trip some of our killing bottles became “weak,” and in several instances inert but not dead specimens were placed in the ordinary letter envelopes in the collecting box for transportation during the day. Such specimens, when they revived, chewed and tore other specimens placed in the envelopes with them. This was guarded against by using “strong” bot- tles for killing and weaker bottles for carrying material till a sufficient num- ber had accumulated to justify opening the collecting box. Specimens with broken necks, caused by twisting tlie head, will not damage other specimens. While papering material at night it was found that if the letter en- velopes were white or light colored one could more readily see whether or not they were emptied of all specimens, since, in poor light, any insect clinging to the inside of an envelope is more readily seen in or through a white envelope. It was found desirable to letter or number these envelopes, desig- nating three or four groups. For example, each one-fourth of the envelopes would be plainly labelled A, B, C, and D, respectively. In the field it is ad- visable that the specimens gathered in the same habitats or localities be placed in the envelopes similarly lettered. Such a procedure will facilitate paper- ing the material in properly designated envelopes when the day’s collecting is over. Other conditions also arise where separation of material during the day is desirable. Of course variously colored envelopes are even better than lettered or numbered envelopes, except for the reason given above for pre- ferring white envelopes. COMMON NAMES OF DRAGONFLIES Mr. H. L. Tyrer, of Barranquiila, told us that so far as he knew caballito was the only name in common use in Colombia. At any rate in our limited experience we always found this name readily understood. Young men travelling with us on one of the river steamers told us that at Medellin 24 E. B. WILLIAMSON dragonflies were commonly called senoritas, and at Bogota they were called tuatapiojo. The first part of the latter name means a killer, and the last part an insect the same as or similar to the red bug,—in other words, the red bug killer. A well educated gentleman at Puerto Berrio said that in the Department of Antioquia dragonflies were called by the common people chilcaqua, from chiquear, to reduce, to make less and less, hence to fiip or dip out, and from aqua, water, the name doubtless tracing back to some early observer of an ovipositing libelluline, very probably Orthemis ferruginea. One evening at FE] Banco when we had an audience of several persons while putting up our day’s catch, we asked if the insects were called caballitos. To this we received an affirmative reply from everyone but a smali boy who pointed to a zygopterous specimen and said “ceritongo” (cieratonga). The day’s catch included libellulines, large and small, aeshnines, gomphines, and a mixture of zygopteras including Mecistogasters. We at once mixed up the lot and asked him to separate the caballitos and the ceritongos. As rapidly as he could pick them up with the tweezers he separated the anisopteras from the zygopteras without a mistake. Several persons ques- tioned were unable to give us the meaning of the word, but one gentleman thought that it was probably derived from cerar, to shut or block off, and tonga, a current of water. ACKNOWLEDGEMENTS In concluding this narrative we wish to thank many friends and ac- quaintances not mentioned by name, who gave us advice and assistance, and added to the pleasure and effectiveness of our trip. At all times we were shown the greatest courtesy and kindness by those with whom we came in contact in Colombia. UNIVERSITY OF MICHIGAN MUSEUM OF ZOOLOGY Miscellaneous Publications No. 4 Contributions to the Botany of Michigan BY C. K. DODGE ANN ARBOR, MICHIGAN PUBLISHED BY THE UNIVERSITY FEBRUARY 23, I918 ADVERTISEMENT The publications of the Museum of Zoology, University of Michigan, consist of two series—the Occasional Papers and the Miscellaneous Publica- tions. Both series were founded and are being supported by Dr. Bryant Walker, Mr. Bradshaw H. Swales and Dr. W. W. Newcomb, except that the cost of the illustrations and of distribution are borne by the University. The Occasional Papers, publication of which was begun in 1913. serve as a medium for the publication of brief original papers based principally upon the collections in the Museum. The papers are issued separately to libraries and specialists, and, when a sufficient number of pages have been printed to make a volume, a title page and table of contents are supplied to libraries and individuals on the mailing list for the entire series. The Miscellaneous Publications include papers on field and museum technique, monographic studies and other papers not within the scope of the Occasional Papers. The papers are published separately, and, as it is not intended that they shall be grouped into volumes, each number has a title page and table of contents. ALEXANDER G. RUTHVEN, Director of the Museum of Zoology, University of Michigan CONTENTS PAGE mr REND etm ty ata Sos 4 io, 62 dd SEAR CIAO oe oe ae a 5 LO tial OS en oh chet, 25's Fost 3 Vee a aa a os oodles 5 NEEM ATI DP ALON CRI ors arc aren ascot S's oe vce cin awoke oe 6 Polystichum Braunu ........ Rae RS nae me SRE COR OS CP Ae RE 6 MEMEG EM INIL-ITISTECHYWM Se. Soe cose dls sai sa sacee wy Boe oa we Pale erie a A Sporganwum diversifoluum var. acaule...0.... 0.00. ce cee cece ee naees 7 RPT OSTLIL I= OC OF CUMS 2 iss Saco ry. ov fea dons ove sig ahd Mee a een os a. SE UNOS IETS TOROS a aaa et 7 Dna eaNEtS MYPCTDONEO, © op oe oe cas Da nl SR te ve Rena het wow ee 7 Dea meOSTIS TCE PONS, Sree ees © oh kk« Ye sins ¢ oattiia’ Sata dg « coieeevs ee cc 8 1 TELE PONE ITE GSC RAG ee Ree eRe a a eto See a 8 Spartina patens........ SARE C ache Ahpein, hee EAS, ih ne Cl a 8 ear N DORIS my as ae cies) 4 UR RN Orica eg a ohieeatc keene wv ap be 8 ea ORME LAMCIICIE SOM cc y.c't one > eet sae oe RE Aig Oe on ees os ee 9 Juncus Gerardt........ eee cp clwsal Beier Pad WS chs tae en. sh) 9 Ene ha UES 09 SEG SERPS Sa Sat SR a a ee 9 (Se (FELIU or Is Ge SS a Ec 9 MEAD TINA UGSIL MOVIE» whos. 503s «Sols le tes die. eee Blevins @, ons Shiaye ave 0a Goa addon de 10 re ICT Ore "Fd edaisea ssh Rea Poa Mike Ss Bates a eee 10 Reatb este GMG ERICK aah erm cr aes Gc badge oR eau hes 32's Rb owe SS 10 RNC OG IECET MIU IE? A era tet ie ge So cya a Se. Seo don Leake tia. o dle myava iw ace eer Sa eee A hore ds hh Wes edie ae stds. dled ba, ne RN TI RR Fy DIL SORME OI Rens Son ato as “cea Pers xen Raa so Na ela Ne 12 MIME OPC CLOS chen ete eed et ic dees eis, . ORM Were. 12 Gentiana linearis var. latifolia.............. aS, Sct, a ee ee 12 REE MONON sr. <5 x <2 < 6 chove'= Se id CP eS aul dle « Le Woon 13 Mamulus moschatus ......... 600000000 Be Aye a ee: Ber: eee 13 LO FADS GR a NPS i a a RP Ae en | ae 13 Solidago lepidd var. molind. . 2+... tvs oes Me ces. Sara tae 14 ES eMOCRUIOMADIGOIOT Six 006 ax Ds.n bee can a Ree Cian. oe oN ee Gt se Ske CONTRIBUTIONS TO THE BOTANY OF MICHIGAN Bye Co 1c, DODGE INTRODUCTION A considerable amount of data on the plants of the state has accumulated since the appearance of the last edition of W. J. Beal’s Flora of Michigan.t Some of this data has been published but a large part is in the form of herbarium specimens, notes, and manuscripts. In his work on the phan- erogamic flora of the state, carried on during the past ten years for the Michigan Geological and Biological Survey and the Museum of Zoology, the writer has been able to add a number of species to the known flora and to obtain much additional information on the ranges of many of the species. Some of this data has been published in local lists, but a large number of miscellaneous notes have not been appurtenant to the local lists and there has been in Michigan no suitable medium of publication for them. It is now purposed to publish such notes in the Miscellaneous Publications of the Mu- seum of Zoology, under the general title Contributions to the Botany of Michigan, and to number these consecutively for ease in reference. The present paper records a number of species new to the Michigan flora and extends the range of several forms. Unless otherwise stated speci- mens are in the writer’s herbarium at Port Huron, Michigan. Pellaea glabella Mett. Smooth Cliff-brake The smooth cliff-brake is perhaps very close to P. atropurpurea (L.) Link. It is not recognized in the last edition of Gray,? and it is given as a synonym of P. atropurpurea in Britton and Brown. It is not referred to in the last Michigan Flora. In 1905, while examining peat beds in the Upper Peninsula, the late Prof. C. A. Davis made a large collection from dry rocks near Norway, Dickinson County, of a fern that was then supposed to be P. atropurpurea but which has since been referred to P. glabella. *sth Ann. Rept. Mich. Acad. Sci., 1904. *The manuals referred to in this paper are Gray, New Manual of Botany, 7th edition, 1908; Britton and Brown, An Iliustrated Flora of the Northern U. S., Canada and the British Possessions, 2nd Ed., 1913; and W. J. Beal, Michigan Flora, Fifth Annual Report Mich. Acad. Sci., 1¢04. 6 Ci'Ki Dopck Cryptogramma Stelleri (Gmel.) Prantl Slender Cliff-brake As far as known to the writer this beautiful little fern is seldom seen in the state. The Michigan Flora reports it as found on Louse Isle and as seen by the late G. H. Hicks at the Pictured Rocks on the south shore of Lake Superior. In 1916 a visit was made to Grand Marais, Grand Island and Munising, and an attempt was made to reach the Pictured Rocks which lie between these two places. The old lumber roads and trails to the latter locality were found to be impassable, but at Munising conditions were more favorable and a search for the species was made at that place. The city is situated in a depression or valley surrounded by high abrupt hills covered with hardwoods. In the hills there are several small streams with over- hanging rocky banks and precipitous falls. Under these overhanging rocks, in crevices or on little benches or projections, the slender cliff-brake was found in abundance, growing with Cystopteris fragilis, the brittle fern. A further search in similar places, at least in the Northern Peninsula, will undoubtedly bring to light other stations for this delicate species. Polystichum Braunii (Spenner) Fee Braun’s Holly-fern The Michigan Flora only mentions this fern as noted by O. A. Farwell in Keweenaw County. This would seem to indicate that it is scarce and has seldom been seen by our botanists, for, as compared with other ferns in the state, it is certainly very distinct. The writer observed it first in 1916, in the rich woods on Grand Island and near Peter White Camp in Alger County. In June, 1917, it was found to be very plentiful on the alluvial flood plain of the Salmon Trout River near the Huron Mountain Club on the south shore of Lake Superior, northwest of Marquette,- Marquette County. Lycopodium tristachyum Pursh Clustered Club Moss, Ground Pine The range of this club moss is given by Gray as “Dry sandy soil N. Me to Del. and L. Superior,’ and by Britton and Brown as “Maine to Minne- sota, etc.” The Michigan Flora mentions it under the name of L. Chamaecy- parissus A. Br. and says that it has been located in three places—Clare County, Keweenaw County, and near Manistee. In 1908, while engaged in a botanical survey of the Saginaw Bay sand dune district, the writer found it in abundance in Huron County but at the time supposed it to be L. complanatum L,. and so listed it. Since then it has been noted as quite common from Mackinac to Houghton and Ke- weenaw Counties inclusive. ConTRIBUTIONS TO THE Borany oF MICHIGAN 7 Sparganium diversifolium Graebner var. acaule (Beeby) Fernald and Eames Stemless Bur-reed The stemless bur-reed, given as a variety in Gray and as a species in Britton and Brown, has apparently been generally overlooked by Michigan botanists. There seems to be no published record, but the writer has found the species in several localities. In 1911 it was discovered on Harsen’s Is- land in the southern part of St. Clair County, growing in abundance in muddy places and shallow water. In 1914 it was noted as abundant about Vermillion in Chippewa County, and in 1915 it was found to be common in Schoolcraft County. The plant very probably occurs throughout the state. Anthoxanthum odoratum L,. Sweet Vernal Grass This grass seems to have been naturalized from Europe. ‘The habits and range are given by Gray as “Meadows, pastures and waste places, espe- cially eastw.” ; and by Britton and Brown as “In fields and meadows through- out the whole of North America.” The distribution given by the Michi- gan Flora is “Ionia; Grand Rapids, Coleman Cat. ; well established in the col- lege lawn. Rare.” This would indicate that it is not at all common in the state. The writer saw it in abundance on and near the bank of Echo Lake, a small body of water in the woods of Grand Island, near Munising, Alger County, in 1914. It had the appearance of a native wild plant, but there are a number of buildings on the bank, and the lake is often visited. Aristida oligantha Michx. Few-flowered Aristida, Ant-rice The habitat and range of this grass are given in Gray as “Dry sterile soil N. J. to Neb. and southw.” ; in Britton and Brown as “Dry soil; New Jersey to Nebraska, and Texas.” It is not listed in the last Michigan Flora. To the writer’s knowledge it has persisted in dry open ground near the city of Port Huron for about ten years. Calamagrostis hyperborea Lange Northern Reed-grass The habitat and range of the northern reed-grass is given in Gray as “Moist meadows and calcareous cliffs, Greenl. to Alaska s. to e. Oue., ay Vie, ‘Pa.’, Minn., and the Rocky Mts.” ; in Britton and Brown as “Meadows and swamps, Greenland to Alaska, south to Pennsylvania, Colorado and Cali- fornia.” It is not listed in the Michigan Flora. In 1916 the writer observed the species to be very abundant in a meadow-like opening on the high bank of Lake Superior near Sable Banks, west of Grand Marais, in Alger County. 8 C.K; Dopck Calamagrostis inexpansa Gray Bog Reed-grass It is to be inferred that both Gray and Britton and Brown include Michi- gan in the range of this grass, although the state is not specifically men- tioned. ‘The Michigan Flora states in its additions that Dr. H. 5. Pepoon saw it near Keeler, Van Buren County. For over five years it has persisted in open wet ground along the Grand Trunk Railway near the city of Port Huron. It has perhaps been introduced at this point. Danthonia compressa Aust. Flattened Wild Oat-grass This species is new to Michigan flora. The habitat and range are given by Gray as “Dry woods, Me. to N. Y. and southw.”, and in Britton and Brown as “In woods, Maine to New York, south to: North Carolina and Tenessee.” The writer found it plentiful in several dry, open or partially shaded places in the northeastern part of Schoolcraft County in 1916. Spartina patens (Ait.) Muhl. Salt-Meadow Grass The habitat of this grass is “salt meadows and sandy beaches”’ along the Atlantic coast. It grows in salty ground, and with “Juncus Gerardi, the ‘black grass,’ furnishes most of the salt meadow hay of the Atlantic coast.” Both manuals substantially agree as to its habitat and range. It is not in- cluded in the Michigan Flora. The species has become established and has persisted for over ten years in the Grand Trunk Railway freight yards above the Tunnel Station near the city of Port Huron. One mound in particular, about thirty feet long and fifteen feet wide, is snugly sodded over with salt-meadow grass to the exclusion of almost every other plant. A few rods distant Juncus Gerardi is very abundant. Appropriate soil conditions are furnished by salt water from a large ice house and many salt vats where refrigerator cars are iced before going farther east. The salt used with the ice is often scattered about in substantial quantities. Glyceria borealis (Nash) Batchelder Northern-manna Grass Michigan is apparently included in the range of this plant by Gray and Britton and Brown, although the state is not specifically mentioned. In the latter it bears the scientific name Panicularia borealis Nash, and the same common name as P. lava Scribn, a grass whose range appears to be farther east. It is not listed in the Michigan Flora. It has been observed by the writer as plentiful in damp open ground near Fostoria, Tuscola County, 1910, and near L,’Anse, Baraga County, 1916. CoNn’TRIBUTIONS To THE Botany of MICHIGAN 9 Eriophorum tenellum Nutt. Rough Cotton-grass It is stated in Gray that the habitat and range of this cotton-grass is “Swamps and bogs. Nfd. to Ont. s. to N. J. and Ill.”, and about the same distribution is given by Britton and Brown. It is not mentioned in the Michi- gan Flora. In 1914 the species was noticed as frequent in swampy ground near Vermillion and at the lower falls of the Tahquamenon River, Chippewa County, and again near Newberry in Luce County. The writer believes this plant to be common throughout the eastern half of the Northern Peninsula. Juncus Gerardi Loisel Black Grass The habitat of the black grass is given by Gray and Britton and Brown as “Salt marshes” and the range in part as “Vicinity of the Great Lakes.” In the Michigan Flora it is listed as “Rare about the Great Lakes, Gray’s Manual,” which would indicate that it had been seldom seen in Michigan by our local botanists. As noted under Spartina patens it is found near Port Huron on soil which has become impregnated with salt in the icing of re- frigerator cars. It has persisted here for more than ten years. Streptopus longipes Fernald Fernald’s Streptopus This plant was first described in 1g06, and the description in Gray men- tions the character, “rootstock slender and wide creeping” and gives in the range “woods Marquette Co., Mich.” In the last edition of Britton and Brown it is thrown into synonymy. The writer is familar with the other two species in our territory, S. roseus and S. amplexifolius, and believes that S. longipes is a valid species. The only other Michigan records, so far as the writer has been able to ascertain, are those of Frank C. Gates*® for Douglas Lake, Cheboygan County, Michigan. In 1914 it was noted as common in Chippewa County, especially in the woods about Vermillion. It was again seen in [uce, Schoolcraft, Mackinac and Alger Counties in 1915, and was abundant in the hardwoods of Mar- quette County in 1916 and 1917. S. roseus was not observed in the last named county. Salix pellita. Anders. Satiny Willow The satiny willow apparently has not often been seen in Michigan. The manuals do not mention Michigan in its range and the Michigan Flora does not refer to it. W. S. Cooper‘ noted it on Isle Royale, and in 1915 the *Rhodora, XIII (1orr), p. 237; 14th Ann. Rept. Mich. Acad. Sci., 1912,p. 88. *16th Ann. Rept. Mich. Acad. Sci., 1914, p. 119. IO C. K. DopcE writer found it in abundance along the low banks of the Tahquamenon River in Luce County. It was later located on swampy ground near Munising, Alger County, and on the margin of Pine Lake near the Huron Mountain Club, Marquette County. The under sides of the leaves are usually velvety- white, presenting a beautiful sight when turned up by a breeze. It is prob- ably frequent throughout the Northern Peninsula. Arceuthobium pusillum Peck Dwarf Mistletoe This small parasitic plant, known as Arceuthobium pusillum Peck in Gray’s Mannual and as Rozoumofskyan pusilla (Peck) Kuntze, small mistle- toe, in Britton and Brown is attributed to Michigan in Gray. The Michigan Flora mentions it as observed by the late Prof. C. F. Wheeler near Chatham, Alger County; by B. Barlow at Turin, Marquette County; and by the late Prof. C. A. Davis near Cadillac, Wexford County. In 1908, the writer saw it in abundance on black spruce in Huron County ; in 1913 on white spruce, Bois Blanc Island, Mackinac County, and in 1916 on white spruce near Peter White Camp in.the western part of Alger County. It is now probably well established throughout Michigan. Spergularia media (L.) C. Presl. Large Flowered Sand Spurry This plant is described in Gray, and it is included in the description of Tissa marina (1,.) Britton, salt-marsh sand spurry, in Britton and Brown. It is not mentioned in the Michigan Flora. The range given in Gray is “Near Selina, N. Y.; also Cal. (Eu.)”. The plant has been well established and abundant for nearly ten years at the Grand “Trunk freight yards near the city of Port Huron. It is mostly found on wet ground and covers a large area. Amelanchier species Juneberry Gray describes four species and two varieties and Britton and Brown six species of this difficult genus. The Michigan Flora mentions six species. The species are so variable that there has been little unanimity of opinion as to their status, and the identification of specimens has been difficult. For the purpose of this paper it will be assumed that Prof. K. M. Wiegand’s” revision of the eastern species is to be adopted by botanists generally. Hight species are recognized by Wiegand, seven of which occur in Michigan. The known species in the state so far as described by him will be taken up in | order. Amelanchier sanguinea (Pursh) DC. Round-leaved Juneberry.—Prob- ably common throughout Michigan. 5 Rhodora, Vol. 14, 1912, p. 117. CONTRIBUTIONS TO THE BoTANy oF MICHIGAN II Amelanchier humilis Wieg. Small Juneberry—Barren sand in Kent County ; near Huron Mountain Club in Marquette County. Amelanchier florida Lindley. Flowering Juneberry—On Isle Royale; Keweenaw County; Alcona County; Alpena County. Amelanchier stolonifera Wieg. Low Juneberry—Alpena County ; Mar- quette County ; Huron County; Arenac County. Amelanchier canadensis (1,.) Medicus. Common Juneberry.—Probably common throughout the state. Often large trees. Amelanchier laevis Wieg. Smooth-leaved Juneberry.—Probably abund- ant throughout the state. Trees often large. Amelanchier Bartramiana (Tausch.) Roemer. Oblong-fruited June- berry.—Keweenaw County; on Isle Royale ; Chippewa County ; Alger County. To this list should be added the western species: Amelanchier alnifolia Nutt. Northwestern Juneberry—A _ western species. Dry open ground near Norway, Dickinson County, in 1905 by the late Prof. C. A. Davis; Presque Isle County; Keweenaw County, O. A. Farwell. Geranium sanguinium L,. Red-rooted Geranium Some time during the season of 1914, Miss R. M. Kearsley, of Detroit, observed an unfamilar road-side geranium near Birmingham, Oakland County. B. Gladewitz, of Detroit, determined the plant as G. sanguinium, and this identification has been confirmed by K. K. Mackenzie. The species is from one to one and one-half feet high, inclined to be ascending, flowers purple and showy, petals heart-shaped, roots long and red—fully as red as the rootstocks of our common bloodroot, Sanguinaria canadensis. ‘The plants occupy a space by the roadside about ten by twenty feet, extend slightly into an adjoining field and are pau by nearby peo- ple as having been there at least fifteen years. Viola Selkirkii Pursh Great-spurred Violet According to Gray the range of this violet is in part “L. Superior and Northw., rare.” ; and in Britton and Brown the distribution is given in part as “New Brunswick to Pennsylvania and Minnesota.” The Michigan Flora lists the following localities, “Gillman, A. Gray; Keweenaw County, O. A. Farwell. U. P.” This violet is very distinct and easily recognized. It was first noted by the writer in the hardwoods near Manistique, in School- craft County, and in 1916-7 it was observed frequently in rich, shaded ground in Marquette County, especially about the premises of the Huron Mountain Club. It is very probably to be found throughout the Upper Peninsula. 12 C. K. DopcE Opuntia Rafinesquii /ngelm. Western Prickly Pear The above scientific name is given in Gray which specifically mentions Michigan as part of its range. Britton and Brown call it O. humifusa Rat., and the distribution is given as “Ohio to Minnesota, etc.” In the Michigan Flora under the latter name the following localities are mentioned: “Common in Newago County along the Muskegon River, Cedar Creek Tp., Muskegon Co., C. F. Wheeler; and a stunted variety on sand barrens near Greebville, J. Satterlee ; also northward into British Am., Englemann.” The writer first saw the species in 1910, on Point Pelee, Essex County, Ontario. In September, 1916, while making a botanical survey about the Huron Mountain Club in Marquette County, Michigan, Miss Anne Rus- sell, of Detroit, discovered it in rocky places on a near-by granite hill known as Huron Mountain. In August, 1917, the writer noted it as plentiful about New Buffalo, Berrien County. It has been reported on rocky hills about the city of Marquette and on sandy open ground west of Alpena in Alpena County. Osmorhiza species The Sweet Cicelies In Gray and Britton and Brown four species and one variety of sweet cicely are mentioned and described. In Britton and Brown the generic name is Washingtonia, in Gray, Osmorhiza. In this paper the latter name will be used. The manuals apparently do not specifically credit Michigan with more than two species, Osmorhiza Claytoni (Michx.) Clarke, woolly sweet cicely, and O. longistylis (Torr.) DC., smoother sweet cicely. The Michi- gan Flora mentions these two and reports them common throughout the state. The variety O. longistylis villicaulis Fernald, pubescent smoother sweet cicely, has been noted as frequent from St. Clair County north and west to Delta County. In 1907, the writer discovered O. divaricata Nutt., western sweet cicely, in rich shaded ground on Thunder Bay Island, Alpena County,° and again in rich woods near Eckerman, Chippewa County, in 1914. In the same year O. obtusa (Coult. & Rose) Fernald, blunt-fruited sweet cicely, was noted in rich woods in Chippewa County and throughout Alger County. It appears that the last two species may be considered abundant throughout the eastern part of the Upper Peninsula, and perhaps occasional in the north- ern part of the Lower Peninsula. Gentiana linearis Froel. var. latifolia Gray Gray’s Gentian Gray’s gentian is listed in Gray under the above scientific name and the range is given as “Lake Superior; N. B.” Britton and Brown record the species as Dasystephana grayi (Kusnezow) Britton. There is no reference ®° Additions to the Michigan Tlora, by W. J. Beal. toth Ann. Rept. Mich. Acad. Sci., 1908, p. 80. ConTRIBUTIONS TO THE Borany oF MICHIGAN 13 to it in the Michigan Flora. In the course of botanical work in the eastern part of the Northern Peninsula, from 1912 to 1917, the writer found it abundant on damp, open or partially shaded ground on Bois Blanc Island, about St. Ignace, and in all the northern counties from Macinac to Mar- quette inclusive. Collinsia parviflora Lindl. Small-flowered Collinsia This little plant is given in Gray, and its habitat and range are described as “Rich soil and limy gravel, Ont. N. Mich. and westw.”’ In Britton and Brown it goes by the name of C. tenella (Pursh) Piper, and the habitat and range are given as “Moist places, Ontario to British Columbia, Michigan, Colorado, Arizona and Utah.” The Michigan Flora mentions only one Mich- igan record, Keweenaw County, observed by O. A. Farwell. Owing to the fact that it is usually only three to six inches long and grows in obscure places it easily escapes observation. In early June; 1917, it was found by Mr. B. Gladewitz and the writer in Marquette County, where it grew in abund- ance in rock cracks and crevices on the summit and sides of a large granitic hill known as Huron Mountain. It is very probably to be found in like places throughout the county. Mimulus moschatus Dougl. Musk Flower Both Gray and Britton and Brown give Michigan in the range of this plant and intimate that perhaps it has been introduced from the west. In the Michigan Flora it is mentioned as having been observed by O. A. Far- well in Keweenaw County. The writer found it plentiful about Grand Marais in Alger County in 1916, where it occurs in damp places mostly along roads but also in other damp open spots. It may have been introduced. Digitalis purpurea LL. Purple Foxglove Both Gray and Britton and Brown substantially agree that this plant is “naturalized from Furope, sparingly escaped from cultivation.” It is not only cultivated as an ornamental plant but it is a valuable and much used drug sold under the name of digitalis. It is not mentioned in the Michigan Flora as an escape in the state. In 1916, in company with Mr. George W. Howe of Port Huron, Michigan, the writer collected in the territory ad- jacent to the south shore of Lake Superior. Among other places visited was the Peter White Camp, a beautiful place about ten miles from the south shore of Lake Superior, in the western part of Alger County. The purple foxglove was growing wild at this place and also at Howe’s Lake, a small body of water near by. Howe's Lake is surrounded by primitive forests and only a short distance from its margin, but entirely hidden from view from the lake, was discovered a small clearing of perhaps a quarter of an acre in extent with two very old log hunting huts. The camps and clearing 14 CK; Donce had the appearance of having been abandoned for many years. In this clearing the purple foxglove was holding its own with grass and weeds; there were many plants five feet high, and varieties with both white and purple flowers were present. Its companion was the sweet william, Dianthus borbatus L. We were informed that these plants had been growing wild there for over thirty-five years. Solidago lepida DC., var. molina Fernald Woolly Goldenrod In 1916, while engaged in a botanical survey of a part of the south shore of Lake Superior, the writer collected near L’Anse, Baraga County, and also in Houghton County a very woolly stemmed goldenrod, usually in open places not far from the lake shore. Specimens were referred to Mr. K. K. Mackenzie, who identified them as S. lepida Fernald, the original de- scription of which appeared in Rhodora, Vol. 17, p. 8 (1915). Adenocaulon bicolor Hook Adenocaulon As far as can be ascertained Michigan botanists know very little about this plant. The range is given in Gray as “Moist woods, shores of Lake Huron, Lake Superior, and westward,” in Britton and Brown as “In moist woods, northern Michigan and Lake Superior to British Columbia, Mon- tana and California,” and in the Michigan Flora as ‘“Ontonagon River, U. P.” In June, 1916, the writer found it to be plentiful near Peter White Camp in the western part of Alger County, and again, in September of the same year, in open hardwoods at the foot of Huron Mountain in the north- western part of Marquette County. Hieracium aurantiacum L. Orange Hawkweed The orange hawkweed is now well known in the state, especially in the Lower Peninsula, and the manuals give it a wide range as a weed and pest from Europe. Curiously enough, it has only once been recorded for Michi- gan,—in the writer’s Mackinac Island list. In the late Prof. C. F. Wheel- er’s annotated copy of Beal and Wheeler’s Michigan Flora is the note “Hieracium aurantiacum is becoming a weed near Huron (Lenawee County). It is omitted in the last Michigan Flora. In 1912, the writer examined the wild plants on or near Lake Huron shore from near Bay City to Mackinaw City. This species was abundant in fields and on roadsides just north of Alpena, where farmers reported that it had been noted for many years, and it was common even in the open woods from Alpena to Mackinaw City. UNIVERSITY OF MICHIGAN MUSEUM OF ZOOLOGY Miscellaneous Publications No. 5 Contributions to the Botany of Michigan, II BY Cy kK. DODGE ANN ARBOR, MICHIGAN PUBLISHED BY THE UNIVERSITY JULY 13, 1918 ADVERTISEMENT The publications of the Museum of Zoology, University of Michigan, consist of two series—the Occasional Papers and the Miscellaneous Publica- tions: Both series were founded and are being supported by Dr. Bryant Walker, Mr. Bradshaw H. Swales and Dr. W. W. Newcomb, except that the cost of the illustrations and of distribution. are borne by the University. The Occasional Papers, publication of which was begun in 1913, serve as a medium for the publication of brief original papers based principally upon the collections in the Museum. The papers are issued separately to libraries and specialists, and, when a sufficient number of pages have been printed to make a volume, a title page and table of contents are supplied to libraries and individuals on the mailing list for the entire series. The Miscellaneous Publications include papers on field and museum technique, monographic studies and other papers not within the scope of the Occasional Papers. The papers are published separately, and, as it is not intended that they shall be grouped into volumes, each number has a title page and table of contents. ALEXANDER G. RUTHVEN, Director of the Museum of Zoology, University of Michigan. OBSERVATIONS ON THE FLOWERING PLANTS, FERNS AND FERN ALLIES GROWING WILD IN MARQUETTE COUNTY, MICHIGAN, IN 1916 AND 1917, ESPECIALLY IN THE VI- CINITY OF THE HURON MOUNTAIN CLUB By CHARLES KEENE DopckE IntTRopuUCcTORY NOTE The manuscript of this paper was completed by Mr. Dodge shortly before his death. It represents the results of several field trips to Marquette County, the last in the spring of 1917, in the course of an investigation of the phanerogamic flora of the state upon which he had been engaged for the Michigan Geological and Biological Survey. T'wo trips were made to the Huron Mountains, in the summer of 1916 and the early spring of 1917, and on the second trip Mr. Dodge was accompanied by Mr. B. Gladewitz. The mountains are specifically mentioned as a locality in the paper for the convenience of the summer residents, and a map of the part most carefully worked is appended. The paper has been edited for publication by Messrs. C. Billington and B. Gladewitz. ALEXANDER G. RUTHVEN. List oF SPECIES POLYPODIACEAE Polypodium vulgare L. Common Polypody.—Rocky, usually shaded ground, often protruding from cracks and crevices of rocks on Huron Moun- tain, a large granitic hill near Huron Mountain Club, and in rocky places throughout Marquette County. Often abundant. Phegopteris polypodioides Fée. (Dryopteris phegopteris 1,., C. Chr.). Long Beech Fern.—Rich woods and shaded banks of small streams near Huron Mountain Club and throughout Marquette County. Plentiful. Phegopteris Dryopteris (L.) Fée. (Dryopteris dryopteris 1,., Britton). Oak Fern—Rich moist woods near Huron Mountain Club and throughout Marquette County. Often abundant. Adiantum pedatum L,. Maidenhair Fern.—Rich shaded ground through- out. Plentiful along Salmon Trout River near Huron Mountain Club. Pieris aquilina ,. (Pteridium aquilinum L., Kuhn). Common Brake.— Dry, sandy, open or partially shaded ground near Huron Mountain Club. Plentiful throughout. Often in swamps. 4 C.K Dopcr Asplenium Trichomanes L,. Maidenhair Spleenwort——Cracks and crev- ices of perpendicular rocks at Canyon Lake near Huron Mountain Club, and propably on shaded rocks throughout. Noted by B. Gladewitz. Asplenium acrostichoides Sw. (Athyrium thelypteroides Michx., Desv.). Silvery Spleenwort——Rich shaded ground near Negaunee. Perhaps fre- quent throughout. Asplenium Filix-femina (L.) Bernh. (Athyrium filix-foenuina L., Roth.). Lady Fern.—Rich open or shaded ground near Huron Mountain Club. Abundant throughout. Polystichum acrostichoides (Michx.) Schott. Christmas Fern.—Rich woods and shaded rocky ground about Sugar Loaf Mountain near Mar- quette. Very probably to be found throughout the county. Apparently in- frequent. Polystichum Braunii (Spenner) Fée. Braun’s Holly Fern.—Occasional throughout, but plentiful in rich shaded ground along Salmon Trout River near Huron Mountain Club. Aspidium Thelypteris (1,.) Sw. (Dryopteris thelypteris L., A. Gray). Marsh Shield Fern.—Marshy open or partially shaded ground near Huron Mountain Club. Plentiful and often abundant throughout. Aspidium noveboracense (L.) Sw. (Dryopteris noveboracensis L., A. Gray). New York Fern.—Rich woods about Sugar Loaf Mountain near Marquette. Apparently infrequent. Aspidium fragrans (1,.) Sw. (Dryopteris fragrans L., Schott.). Fra- grant Shield Fern.—Rocky granitic hills about Huron Mountain Club. Ap- parently infrequent. Noted by Mrs. Wallace Radcliffe. Aspidium marginale (L.) Sw. (Dryopteris marginalis L., A. Gray). Evergreen Wood Fern.—Rich shaded ground throughout. Observed in particular as plentiful on wooded sides of Huron Mountain near Huron Mountain Club. Aspidium Boottii Tuckerm. (Dryopteris Boottii Tuckerm., Underw.). Boott’s Shield Fern.—Damp shaded ground near Negaunee. Apparently infrequent. ; Aspidium cristatum (1,.) Sw. (Dryopteris cristata L., A. Gray). Crest- ed Shield Fern—Damp open or partially shaded ground near Huron Moun- tain Club. Probably frequent throughout. Aspidium spinulosum (O. F. Miller) Sw. (Dryopteris spinulosa Muell., Kuntze.). Spinulose Shield Fern—In hardwoods near Huron Mountain Club. Common throughout. Aspidium spinulosum intermedium (Muhl.) D. C. Eaton. (Dryopteris intermedia Muhl., Gray). American Shield Fern.—Frequent in damp rich woods near Huron Mountain Club and throughout. Cystopteris bulbifera (L.) Bernh. (Filix bulbifera 1,., Underw.). Bulb- let Cystopteris.—Shaded rocky ravines about Manquetic and Huron Moun- tain Club. Plentiful. Cystopteris fragilis (,.) Bernh. (Filix fragilis 1, Underw.). Brittle Fern.—Rich woods and shaded rocks near Huron Mountain Club and about Marquette. Plentiful. A delicate little fern. PLANTS OF MARQUETTE COUNTY s Woodsia ilvensis (.) R. Br. Rusty Woodsia.—Cracks and crevices of rocks on Huron Mountain near Huron Mountain Club. Plentiful. W oodsia obtusa (Spreng.) Torr. Blunt-lobed Woodsia.—In crecks and crevices of rocks on Sugar Loaf Mountain near Marquette. Abundant. Probably to be found throughout. Onoclea sensibilis 1, Sensitive Fern——Damp open or partially shaded ground near Huron Mountain Club. Plentiful throughout. Onoclea Struthiopteris (L.) Hoffm. (Matteuccia Struthiopteris L., Tod- aro). Ostrich Fern—Near Huron Mountain Club in alluvial ground, along small streams and in damp rich partially shaded ground. Frequent through- out. A fine fern for cultivation. OSIMUNDACEAE Osmunda regalis L. Royal Fern—Wet rich shaded ground near Huron Mountain Club. Plentiful throughout. Osmunda Claytoniana L. Clayton’s Fern, Interrupted Fern. — Damp rich open or partially shaded ground near Huron Mountain Club. A large and beautiful fern. Frequent. Osmunda cinnamomea L. Cinnamon Fern.—Open or partially shaded places near Huron Mountain Club. Frequent throughout. OPHIOGLOSSACEAE Botrychium virginianum (L.) Sw. Rattlesnake Fern, Virginia Grape Fern.—Rich shaded ground near Huron Mountain Club. Frequent through- out. EQUISETACEAE Equisetum arvense L. Common Horsetail—In damp, dry, sandy, open or shaded ground near Huron Mountain Club. Common throughout. Of- ten acting as an efficient sand binder. Equisetum sylvaticum 1, Wood Horstetail—Damp shaded ground near Huron Mountain Club. Frequent throughout. Equisetum fluviatile L. Swamp Horsetail.— Near Huron Mountain Club in very wet places, or still shallow water. Common. Equisetum hvemale intermedium A. A. Eaton. (E. hyemale L.). Scour- ing Rush.—Dry open or shaded ground near Huron Mountain Club. Fre- quent and in spots abundant throughout, often acting as an efficient sand binder. Equisetum variegatum Schleich. Variegated Equisetum.—Lake shores near Huron Mountain Club. Frequent throughout. LYCOPODIACEAE Lycopodium Iucidulum Michx. Shining Club Moss. —Damp_ shaded ground near Huron Mountain Club. Frequent throughout. Lycopodium inundatum 1, Bog Club Moss.—Damp sandy open or par- tially shaded ground near Huron Mountain Club. Frequent throughout. 6 C.K: “Dopce Lycopodium annotinum 1, Stiff Club Moss——Open hardwoods near Huron Mountain Club. Plentiful and often abundant throughout. Lycopodiuin clavatum 1, Common Club Moss, Christmas Green—Dry, shaded ground near Huron Mountain Club. Frequent and often abundant throughout. Lycopodium obscurum 1, Ground Pine.—Rich shaded ground, usually in woods. Apparently frequent near Huron Mountain Club and throughout. It appears to be difficult to distinguish this from the following. Lycopodium obscurum dendroideum (Michx.) D. C. Eaton. Tree-like Club Moss.—Woods about Huron Mountain Club, and plentiful throughout. Included by many botanists in the preceding. Lycopodium tristachyum Pursh. Festoon Ground Pine, Ground Pine. —In sandy ground among pines near Huron Mountain Club and plentiful throughout. SELAGINELLACEAE Selaginella rupestris (1,.) Spring. Rock Selaginella—Dry rocks at the “Fortress” on Huron Mountain, near Huron Mountain Club. Plentiful throughout in similar situations. Selaginella apus (.) Spring. Creeping Selaginella—Damp, mostly shaded ground near Huron Mountain Club. Plentiful throughout. ISOETACEAE Isoétes macrospora Dur. Lake Quillwort.—In shallow water on margin of Rush Lake near Huron Mountain Club. Abundant. Perhaps to be found in most of the small lakes in the county. TAXACEKAR Taxus canadensis Marsh. American Yew, Ground Hemlock.—A low evergreen straggling shrub with diffuse stems in hardwoods near Huron Mountain Club. — Plentiful throughout. PINACEAE Pinus Strobus L. White Pine-—A few large trees left among hard- woods near Huron Mountain Club and throughout the county. Often a very large tree from 70 to 175 feet high with a trunk diameter from one and one-half to ten and one-half feet. Various common names have been given it by land lookers, lumbermen and others. Fat pine is the old resinous part taken from stumps and old logs and used for fuel. Cork pine is applied to very large and old trees with soft wood. A large tree, free from limbs and containing several logs of sixteen feet each, is called by land cruisers sap pine, and one with limbs on the body nearly to the ground, buckwheat pine. Pinus Banksiana Tamb. Jack Pine——Usually in dry sandy ground. Plentiful near Huron Mountain Club and throughout. Pinus resinosa Ait. Red Pine, Norway Pine.—In dry sandy ground near Huron Mountain Club. Plentiful throughout. A tree, with body free from limbs, containing several logs of sixteen feet each is known as yellow pine PLANTS OF MARQUETTE CoUNTY 7 by land cruisers, one with limbs near the ground, buckwheat pine. When logs of this species are or become so heavy as to sink in water many lumber- men call them pig-iron pine. Larix laricina (Du Roi) Koch. Tamarack, American Larch.—In swamps near Huron Mountain Club and frequent throughout. Ficea canadensis (Mill.) BSP. White Spruce—Dry ground often on margins of hardwoods near Huron Mountain Club. Frequent throughout. Picea mariana (Mill.) BSP. Black Spruce—Usually in swamps with tamaracks and noted near Huron Mountain Club. Plentiful throughout in such situations. Abies balsamea (1,.) Mill. Balsam Fir, Balsam.—In and on the margin of hardwoods near Huron Mountain Club. Frequent throughout. Tsuga canadensis (L,.) Carr. Hemlock.—Often large trees on the mar- gin of and in hardwoods near Huron Mountain Club. Plentiful throughout. Large trees are called yellow hemlock, and second growth white hemlock by land cruisers. Thuja occidentalis 1. White Cedar, Arbor Vitae—In damp open hard- woods and swamps near Huron Mountain Club. Plentiful throughout. Formerly abundant in “cedar swamps”, but lumbering and destructive fires have in many places nearly exterminated it. Small trees often grow on or near sandy beaches. As understood by land cruisers, second growth trees and those having bodies covered with limbs to or nearly to the ground are white cedar; those with tall smooth bodies mostly without limbs but with small tops are yellow cedar. Juniperus communis L. Common Juniper.—On granitic hills near Hu- ron Mountain Club. Small and apparently infrequent. The difference be- tween this and the following, where the trees are mere shrubs, does not seem to be so very clear. This species when well developed is quite tree- like. Juniperus communis depressa Pursh. (J. Sibirica Burgsd.) Low Juni- per.—Near Huron Mountain Club along sandy beaches and in rocky places. Plentiful throughout. TYPHACEAE : Typha latifolia 1. Common Cat-tail.—In wet marshy and swampy places near Huron Mountain Club. Plentiful throughout. SPARGANIACEAE Sparganium eurycarpum Engelm. Broad-fruited Bur-reed.—Borders of ponds, lakes and streams near Huron Mountain Club. Plentiful throughout. Sparganum americanum androcladum (Engelm.) Fernald & Eames. (S. androcladum Hngelm., Morong.). Branching Bur-reed.—Third Pine Lake near Huron Mountain Club. Frequent throughout. Sparganium diversifolium acaule (Beeby) Fernald & Eames. (S. acaule Beeby, Rydb.). Stemless Bur-reed—Wet muddy places and in shallow water. Third Pine Lake, at the “Styx” near Huron Mountain Club, and probably frequent throughout. 8 C. K. Dopcr NAJADACEAE Potamogeton natans I,, Common Floating Pondweed.—lIn all the small lakes near Huron Mountain Club. Common throughout. Potamogeton epihydrus Raf. Nuttall’s Pondweed.—In small lakes near Huron Mountain Club. Frequent throughout. Potamogeton amplifolius 'Tuckerm. Large-leaved Pondweed.—Common in small lakes near Huron Mountain Club and throughout. Potamogeton heterophyllus Schreb. Various-leaved Pondweed—Abun- dant in small lakes near Huron Mountain Club, and throughout. Potamogeton Richardsonii (Benn.) Rydb. Richardson’s Pondweed.— Common in the small lakes near Huron Mountain Club, and probably throughout. Potamogeton perfoliatus L. Clasping-leaved Pondweed.—Plentiful in small lakes near Huron Mountain Club and very probably throughout. Potamogeton zosterifoliuns Schumacher. (P. compressa L.). Eel-grass Pondweed.—Abundant in small lakes near Huron Mountain Club and throughout. Potamogeton pusillus 1, Small Pondweed.—On margins of small lakes, in pools, ditches, small slow streams and ponds near Huron Mountain Club. Abundant throughout. Potamogeton dimorphus Raf. Spiral Pondweed—lIn small lakes near Huron Mountain Club and throughout. Plentiful. Najas flexilis (Willd.) Rostk. & Schmidt. Slender Najas—lIn shallow water of small lakes near Huron Mountain Club. Plentiful throughout. JUNCAGINACEAE Triglochin maritima 1, Seaside Arrow Grass.—Marshy open ground near Marquette. Apparently infrequent. ALISMACEAE Sagittaria latifolia Willd. Broad-leaved Arrow-head. —In wet places and shallow water near Huron Mountain Club and Big Bay. Common throughout. * Sagittaria graminea Michx. Grass-leaved Sagittaria—Third Pine Lake in shallow water by the “Styx” near Huron Mountain Club. Occasional throughout. Alisma Plantago-aquatica 1, (Alisma Subcordatum Raf.). American Water Plantain—Shallow water on margins of small lakes, along small streams, in ponds and ditches near Huron Mountain Club. Plentiful throughout. H YDROCH ARTTACEAE Flodea canadensis Michx. (Philotria canadensis Michx., Britton). Water-weed.—Common in water of small lakes, ponds and sluggish streams near Huron Mountain Club, and throughout. Vallisneria spiralis 1. Tape Grass, Wild Celery—Common and often abundant in water of small lakes near Huron Mountain Club and through- out. PLANTS OF MARQUETTE CouNTY 9 GRAMINEAE Andropogon scoparius Michx. (Schizachyrium scoparium Michx., Nash). Broom Beard Grass.—Dry sandy or rocky ground near Huron Mountain Club. Apparently infrequent. Andropogon furcatus Muhl. Forked Beard Grass——Dry open ground near Marquette. Apparently infrequent. Digitaria hwmifusa Pers. (Syntherisma ischaemum Schreb., Nash). Small Crab Grass.—Cultivated and waste grounds in and about Marquette. Frequent. Digitaria sanguinalis (L.) Scop. (Syntherisma Sanguinale 1,., Dulac.). Large Crab Grass, Finger Grass-—A weed in fields, gardens and waste places of cities and villages. Plentiful. Panicum capillare 1,, Old Witch Grass.—In dry sandy ground as a weed in and about cities and villages. Often abundant. Panicum depauperatum Muhl. Starved Panicum.—Dry sandy ground and on granitic rocky hills near Huron Mountain Club. Noted throughout as plentiful. Panicum huachucae Ashe. Hairy Panic Grass. — Abundant on sandy beach of First Pine Lake near Huron Mountain Club. Plentiful throughout Panicum xanthophysum Gray. Slender Panic Grass.—Rocky places on Huron Mountain near Huron Mountain Club. Probably frequent through- out. Echinochloa crusgalli (1) Beauv. Barnyard Grass——A common weed in fields, gardens and waste places. Plentiful throughout. Prefers damp open ground. Variable. Echinochloa muricata (Michx.) Fernald. Prostrate Cockspur Grass.— See Rhodora, 17, 198. June 1915. A grass not yet mentioned in the man- uals. Prostrate, forming large mats in damp sand near Big Bay. Perhaps not always prostrate. Setaria glauca (L.) Beauv. (Chaetochloa glauca L., Secribn.). Yellow Foxtail, Pigeon Grass.—A common weed in fields, gardens and waste places throughout. Setaria viridis (L,.) Beauv. (Chaetochloa viridis L,., Scribn.). Green Foxtail Grass—A common weed in fields, gardens and waste places through- out. Cenchrus carolinianus Walt. Small Bur Grass—In sandy open ground and waste places in and about cities and villages. Apparently infrequent. Zizania palustris 1, Wild Rice.—In shallow water of Third Pine Lake by the “Styx” near Huron Mountain Club. Apparently infrequent. Leersia oryzoides (1,.) Sw. (Homalocenchrus oryzoides L,., Poll.). Rice Cut-grass, Scratch Grass.—Wet, usually open places about Marquette. .Ap- parently infrequent. Phalaris arundinacea 1, Reed Canary Grass.—Open wet ground near Marquette. Frequent. Milium effusum 1, Tall Millet Grass——In hardwoods about Huron Mountain Club. Plentiful throughout. 10 C. K. Dopcr Oryzopsis pungens (TYorr.) Hitche. Slender Mountain Rice. — Dry sandy open ground near Huron Mountain Club. Plentiful throughout. Oryzopsis asperifolia Michx. White-grained Mountain Rice—In par- tially shaded ground, usually woods, near Huron Mountain Club. Frequent throughout. Muhlenbergia sylvatica Torr. (Muhlenbergia umbrosa Scribn.). Wood- land Drop-seed.—Shaded ground near Big Bay. Apparently infrequent. Muhlenbergia racemosa (Michx.) BSP. Satin Grass, Wild Timothy.— Borders of marshy places and in damp meadow-like ground about Mar- quette. Frequent throughout. Brachyelytrum erectum (Schreb.) Beauv. Bearded Short-husk—Damp rich shaded ground near Huron Mountain Club. Frequent throughout. Phleum pratense L. ‘Timothy.—Plentiful throughout as an escape from cultivation to roadsides and open woods. Alopecurus geniculatus aristulatus Torr. (A. Aristulatus Michx.). Short- awned Foxtail—Wet places and often in shallow water. Noted at Big Bay and near Negaunee. Frequent. Agrostis alba L. Red Top—Dry sandy places or in damp meadow-like ground. Frequent throughout. Agrostis hyemalis (Walt.) BSP. Rough Hair Grass, Fool Hay.—Dry or damp, open or partially shaded ground near Huron Mountain Club and throughout. Calamovilfa longifolia (Hook.) Hack. Long-leaved Reed Grass.—Sandy beach of Lake Supericr near Huron Mountain Club and often on sand dunes along the Great Lakes acting as an efficient sand binder against the action of wind. Calamagrostis canadensis (Michx.) Beauv. Blue-joint Grass.——Wet open marshy places near Huron Mountain Club. Plentiful throughout. Ammophila arenaria (L.) Link. Sea Sand-reed.—Sandy beach of Lake Superior near Huron Mountain Club. One of the most efficient sand bind- ers against the action of wind and wave. Common and often abundant along the Great Lakes. Cinna arundinacea L. Wood Reed Grass.—On wooded hillsides of Hu- ron Mountain near Huron Mountain Club. Frequent throughout. Cinna latifolia (Trev.) Griseb—Damp open woods near Huron Moun- tain Club. Apparently infrequent. Trisetum spicatum (L,.) Richter. Narrow False Oat.—Noted in rocky places about Marquette by the late Rev. E. J. Hill and reported in Beal’s Michigan Flora. Deschampsia flexuosa (1,.) Trin. Wavy Hair Grass.—Open dry or par- tially shaded places near Huron Mountain Club. Frequent throughout. Danthonia spicata (1,.) Beauv. Common Wild Oat Grass——Open dry and sterile ground near Huron Mountain Club. Common throughout. Phragmites communis Trin. (Phragmites phragmites L,., Karst.). Com- mon Red Grass.—Borders of tamarack-black spruce swamps near Huron Mountain Club. Frequent throughout. PLANTS OF MARQUETTE CoUNTY II Eragrostis pilosa (L.) Beauv. Small Tufted Love Grass.—Dry open ground in Marquette and other cities and along railroads. Frequent. Eragrostis megastachya (Koeler) Link. (E. Major Host.). Strong- scented Love Grass.—A frequent weed in gardens and waste places through- out. Melica Smithi (Porter) Vasey. (Avena Smithii Porter). Smith’s Oat. —Moist open woods near Huron Mountain Club. Apparently infrequent. Dactylis glomerata L. Orchard Grass.—In cities and villages and along railroads. Frequent. Poe annua I, Dwarf Meadow Grass.—A small grass growing in lawns and waste places throughout. Plentiful. Poa compressa 1,. English Blue Grass.—In dry sterile open ground near Huron Mountain Club. Frequent throughout. Poa triflora:Gilib. Fowl Meadow Grass.—Damp open places near Big Bay. Apparently infrequent. Poa pratensis L.. Kentucky Blue Grass, June Grass.—The common grass of our lawns, fields, roadsides and open woods. Abundant. Poa alsodes Gray. Grove Meadow Grass.—Roadsides near Big Bay. Apparently infrequent. Glyceria canadensis {(Michx.) Trin. (Panicularia canadensis Michx., Kuntze). Rattlesnake Grass——Wet and boggy places near Huron Mountain Club. Plentiful throughout. Glyceria nervata (Willd.) Trin. (Panicularia nervata Willd., Kuntze). Nerved Manna Grass, Meadow Grass. — Moist open or partially shaded ground near Huron Mountain Club. Plentiful throughout. Giyceria grandis Wats. (Panicularia grandis S. Wats., Nash). © Reed Meadow Grass.—Low banks of streams, ditches and wet meadow-like ground near Huron Mountain Club. Frequent throughout. Glyceria septentrionalis Hitche. (Panicularia Septentrionalis Hitche., Bicknell). Floating Manna Grass.—In shallow water near Huron Moun- tain Club. Frequent throughout. Glyceria borealis (Nash) Batchelder. (Panicularia borealis Nash). Northern Manna Grass.—In shallow water on border of Third Pine Lake at the “Styx” near Huron Mountain Club. Apparently infrequent. Festuca octoflora Walt. Slender Fescue Grass.—Dry sterile open ground near Huron Mountain Club. Frequent throughout. Festuca ovina 1, Sheep’s Fescue-—Along sandy beach near Huron Mountain Club. Frequent throughout. Festuca nutans Spreng. Nodding Fescue Grass.—In and on borders of moist woods near Huron Mountain Club. Frequent throughout. Bromus secalinus L. Common Chess.—About cities and villages and along railroads. Frequent. Bromus ciliatus L.. Fringed Brome Grass.—Moist shaded ground near Huron Mountain Club. Frequent throughout. Agropyron repens (L,.) Beauv. Quack Grass. gardens and on roadsides throughout. Plentiful. A vicious weed in fields ? 12 C. K. DopcE Agropyron tenerum Vasey. Slender Wheat Grass. — Sandy beach of Lake Superior near Huron Mountain Club. Plentiful. Hordeum jubatum L, Squirrel-tail Grass.—In cities and villages as a weed. A grass introduced from the west. Elymus virginicus l,. Virginia Wild Rye—Margins of small streams and hardwoods near Huron Mountain Club. Frequent throughout. Elymus canadensis L. Nodding Wild Rye.—Dry sandy ground near Huron Mountain Club. Common throughout. Often plentiful along the sandy beaches of Lake Superior where it-acts as an efficient sand binder. Hystrix patula Moench. (Hystrix hystrix L., Millsp.). Bottle-brush Grass—Moist shaded ground near Huron Mountain Club. Frequent throughout. CYPERACEAE Duhcnium arundinaceum (1) Britton. Dulichium.—Borders of ponds and in wet places near Huron Mountain Club. Frequent throughout. Eleocharis obtusa (Willd.) Schultes. Blunt Spike Rush—Wet open places near Huron Mountain Club. Noted in great abundance in wet sand near Big Bay. 3 Eleocharis palustris (L.) R. & S. Creeping Spike Rush.—In damp open ground and often in shallow water near Huron Mountain Club. Plentiful throughout. Eleocharis palustris vigens Bailey. Large Spike Rush—Usually in shal- low water on borders of ponds and small lakes. Noticed in pastures as abundant on the margin of Rush Lake. Eleocharis acicularis (L.) R. & S. Needle Spike Rush—Wet muddy shores near Huron Mountain Club. Abundant throughout. Eleocharis tenuis (Willd.) Schultes. Slender Spike Rush—Damp open ground and marshy places. Frequent throughout. Scirpus hudsonianus (Michx.) Fernald. (Eriophorum alpinum L.). Al- pine Cotton Grass.—Wet marshy places near Marquette. Plentiful through- out. Scirpus subterminalis Torr. Water Club Rush.—In shallow water on border of Mountain Lake near Huron Mcuntain Club. Probably frequent throughout. Scirpus americanus Pers. ‘Three Square—Borders of ponds, streams and small lakes near Huron Mountain Club. Plentiful throughout. Stems sharply three-angled. Scirpus validus Vahl. American Great Bulrush—Margins of ponds, streams and small lakes near Huron Mountain Club. Plentiful throughout. Scirpus occidentalis (Wats.) Chase. Viscid Great Bulrush.—Borders of lakes and often in 6 or 7 feet of water. Stems 6 to to feet long. Frequent throughout. Scirpus rubrotinctus Fernald. (S. microcarpus Presl.). Small-fruited Bulrush.—Very wet open places near Marquette and Big Bay. Plentiful. Scirpus atrovirens Muhl. Dark-green Bulrush.—Bogs and damp mea- dow-like ground near Huron Mountain Club. Plentiful throughout. PLANTS OF MARQUETTE CouUNTY 13 Scirpus cyperinus pelins Fernald. Wool Grass.—Open marshy or boggy ground near Huron Mountain Club. Abundant throughout. This is sup- posed to be included in S. Cyperinus (L.) Kuntze in Britton & Brown. Scirpus pedicellatus Fernald. Pedicelled Wool Grass.—Marshy open or partially shaded ground near Huron Mountain Club. Frequent. Scirpus atrocinctus Fernald. Fernald’s Bulrush—Damp meadow-like and swampy ground near Huron Mountain Club. Plentiful. Eriophorum callitrix Cham. Sheathed Cotton Grass, Hare’s Tail—Open boggy ground near Huron Mountain Club. Frequent throughout. Eriophorum gracile Roth. Slender Cotton Grass.——Wet open boggy places about Marquette. Frequent. Eriophorum angustifolium Roth. ‘Tall Cotton Grass—Wet marshy open places near Marquette. Frequent. Eriophorum viridi-carinatum (Engelm.) Fernald. ‘Thin-leaved Cotton Grass.—Margin of and in open spots of tamarack-black spruce swamps near Huron Mountain Club. Frequent throughout. Eriophorum virginicum L,. Virginia Cotton Grass——Damp open mead- ow-like or boggy ground near Huron Mountain Club. Plentiful throughout. Rynchospora alba (L..) Vahl. White Beak Rush.—Open boggy ground near Marquette. Plentiful. Carex Crawfordii Fernald. Crawford’s Sedge.—Dryish open ground near Big Bay where it is plentiful. Carex tincta Fernald. (C. mirabilis tincta Fernald). Stained Sedge— Dry open or partially shaded ground near Big Bay. Plentiful. Carex aenea Fernald. Fernald’s Hay Sedge—Dry open ground near Marquette. Plentiful. Carex angustiar Mack. Narrow-fruited Sedge—Open marshy ground near city of Marquette. Perhaps to be found throughout. This sedge has not yet appeared in the manuals. Carex canescens L, Silvery Sedge—Wet open places near Big Bay. Plentiful. Carex brunnescens Poir. Brownish Sedge—Open woods near Huron Mountain Club. Plentiful throughout. Carex Deweyana Schwein. Dewey’s Sedge.—Rich open woods near Hu- ron Mountain Club. Plentiful throughout. Carex trisperma Dewey. ‘Three-fruited Sedge—Boggy shaded ground near Huron Mountain Club. Plentiful throughout. Carex disperma Dewey. (C. tenella Schkuhr.). Soft-leaved Sedge— Damp rich woods and shaded swampy ground. Common throughout. Carex rosea Schkuhr. Stellate Sedge—Dry open woods near Mar- quette. Plentiful. Carex vulpinoidea Michx. Fox Sedge—Mostly damp low ground about Marquette. Plentiful. Carex diandra Schrank. Lesser Panicled Sedge——Open boggy ground near Huron Mountain Club. Frequent throughout. 14 C. 7K DopcE Carex stipata Muhl. Awl-fruited Sedge—Wet open or partial!» shaded places. Common throughout. Carex Sartwellii Dewey. Sartwell’s Sedge—Wet boggy ground, often in shallow water about Marquette. Plentiful. Carex crinita Lam. Fringed Sedge.—Rich wet shaded ground near Hu- ron Mountain Club. Plentiful throughout. Carex aquatilis substricta Kitkenthal. Smaller Water Sedge—Very wet open places near Huron Mountain Club and about Marquette. Plentiful. Carex hassei Bailey. (C. bicolor All.) Hasse’s Sedge-——Damp sandy open ground along shores near Huron Mountain Club. Frequent throughout. Carex leptalea Wahlenb. Bristle-stalked Sedge. — Wet meadow-like ground and in boggy places near Huron Mountain Club. Abundant. Carex polygama Schkuhr. (C. Buxbaumii Wahl.). Brown Sedge.—Wet marshy places near Huron Mountain Club. Frequent throughout. Carex stricta lam. ‘Tussock Sedge—Very wet open or partially shaded places near Huron Mountain Club and about Marquette. Frequent. Carex strictior Dewey. Stiff Sedge—Open marshy or damp meadow- like ground near Huron Mountain Club. Plentiful. This sedge has not yet appeared in the manuals. Carex gracillima Schwein. Graceful Sedge—Woods or any rich par- tially shaded ground near Huron Mountain Club. Frequent throughout. Carex tonsa (Fernald) Bicknell. (C. wmbellata tonsa Fernald). Deep- green Sedge.—On the upper sandy beach of Lake Superior near the Huron Mountain Club. Abundant. Grows in large mats. Carex peckii F.. C. Howe. (C. albicans Willd.). White-tinged Sedge — Open woods and on rocky granitic hills near Huron Mountain Club. Plen- tiful throughout. Carex communis Bailey. Fibrous-rooted Sedge.— Rich dry shaded ground near Huron Mountain Club. Plentiful throughout. Carex pennsylvanica Lam. Pennsylvania Sedge.—Dry open sandy ground and on granitic hills near Huron Mountain Club. Plentiful through- out. Carex pedunculata Muhl. Long-stalked Sedge—Margin of hardwoods near Huron Mountain Club. Frequent. Carex leptonervia Fernald. (C. laxiflora leptonervia Fernald.). ‘Two- edged Sedge.—Rich open or partially shaded ground. Frequent throughout. Carex flava 1,. Yellow Sedge—Damp open places near Huron Moun- tain Club. Plentiful throughout. Carex arctata Boott. Drooping Wood Sedge.—Rich shaded ground near Huron Mountain Club. Frequent throughout. Carex scabrata Schwein. Rough Sedge—In damp open or partially shaded ground near Huron Mountain Club. Plentiful throughout. Carex lasiocarpa Ehrh. (Carex filiformis Good). Slender Sedge—Wet meadow-like ground, boggy places and often in shallow water near Huron Mountain Club. Abundant throughout. Carex lanuginosa Michx. Woolly Sedge. — Low meadow-like ground and wet places near Marquette. Plentiful. ; PLANTS OF MARQUETTE COUNTY 15 Carex Pseudo-Cyperus L, Cyperus-like Sedge-——Wet open ground near Negaunee. Frequent. Carex hystricina Muhl. (C. hystericina Muhl.). Porcupine Sedge — Open wet ground and often in shallow water near Huron Mountain Club. Plentiful throughout. Carex retrorsa Schwein. Retrorse Sedge——Wet open or partially shaded ground near Huron Mountain Club. Plentiful throughout. Carex lupulina Muhl. Hop Sedge——Very wet open or shaded ground near Huron Mountain Club. Frequent throughout. Carex intumescens Rudge. Bladder Sedge.— Damp open or partially shaded ground near Big Bay. Frequent throughout. Carex rostrata Stokes. Beaked Sedge—Wet swampy places, often in shallow water near Marquette. Frequent throughout. Carex Tuckermani Dewey. ‘Tuckerman’s Sedge.—Damp rich open or partially shaded ground near Huron Mountain Club. Plentiful throughout. ARACEAE Arisaema triphyllum (L.) Schott. Jack-in-the-pulpit, Indian Turnip.— Damp rich shaded ground near Huron Mountain Club. Plentiful through- out. Calla palustris L. Wild Calla——Very wet open or partially shaded places near Marquette. Plentiful throughout. LEMNACEAE Lemna minor L. Lesser Duckweed.—Covering still shallow water near Big Bay. Probably frequent throughout. ERIOCAULACEAE Eriocaulon articulatum (Huds.) Morong. (E&. septangulare With.). Seven-angled Pipewort—In damp sand, mud or shallow water on margin of small lakes near Huron Mountain Club. Common throughout. PONTEDERIACEAE Pontederia cordata lL, Pickerel Weed.—In shallow water of small lakes near Huron Mountain Club. Plentiful throughout. JUNCACEAE Juncus bufonius L. Toad Rush.—Damp open ground, often along damp roadsides. Abundant throughout. Juncus tenuis Willd. Slender Rush—Common in fields, paths and on roadsides throughout. Juncus balticus littoralis Engelm. Baltic Rush—On sandy shores or in damp open ground. Frequent throughout. An efficient sand binder. Juncus effusus 1, Common Rush.—Marshy open ground. Common throughout. 16 CUR} Doper . Juncus brachycephalus (FKngelm.) Buchenau. Small-headed Rush—Wet marshy ground near Marquette. Plentiful throughout. Juncus brevicaudatus (Engelm.) Fernald. Narrow-panicled Rush—Wet or muddy places near Marquette. Plentiful. Juncus canadensis J. Gay. Canada Rush.——Open marshy places near Huron Mountain Club. Plentiful throughout. Juncus pelocarpus Mey. Brown-fruited Rush.—Wet swampy places and in damp sand on margin of small lakes near Huron Mountain Club. Plenti- ful throughout. Juncus nodosus L. Knotted Rush—Wet sandy ground and swampy places near Marquette. Frequent throughout. LILIACEAE Uvularia grandiflora Sm. Large-flowered Bellwort—Rich open woods along Salmon Trout River near Huron Mountain Club. Frequent through- out. Allium tricoccum Ait. Wild Leek.—Hardwoods near Huron Mountain Club. Common throughout. Hemerocallis fulva L. Common Day Lily.—Inclined to escape from cultivation and persist in cities and villages. Noticed throughout. Asparagus officinalis L, Garden Asparagus.—Dry open sandy ground throughout. Frequent. Clintonia borealis ( Ait.) Raf. Yellow Clintonia—Mostly in hardwoods near Huron Mountain Club. Plentiful throughout. Smilacina racemosa (L.) Desf. (Vagnera racemosa L,., Morong.). False Spikenard—Moist rich shaded ground near Huron Mountain Club. Plenti- ful throughout. Smilacina stellata (L.) Desf. (Vagnera stellata L., Morong.). Star- flowered Solomon’s Seal—-Dry sandy open, or damp rich shaded ground near Huron Mountain Club. Often on sandy shores. Plentiful throughout. Smilacina trifolia (L.) Desf. (Vagnera trifolia L., Morong.). Three- leaved Solomon’s Seal.—-Very wet or boggy open or partially shaded ground near Huron Mountain Club. Frequent throughout. Maianthemum canadense Desf. (Unifolium canadense Desf., Greene.). Wild Lily of the Valley—Rich shaded ground especially in hardwoods near Huron Mountain Club. Common throughout. Sireptopus amplexifolius (I,.) DC. Clasping-leaved Twisted-stallk.— Moist shaded places especially along small streams near Huron Mountain Club. Frequent throughout. Streptopus roseus Michx. Sessile-leaved Twisted-stalk—A specimen in herbarum of Miss Nellie M. Fairbanks collected on Presque Isle at Mar- quette appears to be this species. Apparently infrequent. Streptopus longipes Fernald. Fernald’s Twisted-stalk—In woods near Huron Mountain Club. Abundant throughout Marquette County. Polygonatum biflorum (Walt.) Ell. Hairy Solomon’s Seal.—Hardwoods near Huron Mountain Club. Plentiful throughout. PLANTS OF MARQUETTE CouUNTY 17 Convallaria majalis 1, Lily of the Valley.—Inclined to escape from cultivation frequently and persist in cities and villages. Trillium sessile 1. Sessile-flowered ‘Wake Robin.—In herbarium of Miss Nellie M. Fairbanks, collected on Presque Isle near city of Marquette. Ap- parently infrequent. Trillium grandiflorum (Michx.) Salisb. Large-flowered Wake Robin.— Rich shaded ground especially in hardwoods. Plentiful throughout. Trillium cernuum 1... Nodding Wake Robin. — Rich shaded ground throughout. Noted in particular along Salmon Trout River near Huron Mountain Club. TRIDACEAE Iris versicolor L. Large Blue Flag—Wet open or partially shaded ground near Huron Mountain Club. Not rare throughout, but not noted as abundant anywhere. Sisyrinchium angustifolinim Mill. Pointed Blue-eyed Grass.—In mead- ew-like open ground and often in damp sandy places about city of Mar- quette. Probably frequent throughout. ORCHIDACEAE Cypripedium parvifloruim Salisb. Smaller Yellow Lady’s Slipper.—Most- ly in swampy or boggy places near Huron Mountain Club. Frequent throughout. Specimen in herbarium of Miss Nellie M. Fairbanks, Mar- quette. Cypripedium parviflorum pubescens (Willd.) Knight. Larger Yellow Lady’s Slipper—Rich shaded ground especially in woods near Huron Moun- tain Club. Frequent throughout. Cypripedium hirsutum Mall. (C. reginae Walt.). Showy Lady’s Slipper.— In open or partially shaded swampy places about Marquette. Said to be frequent. Specimen in herbarium of Miss Nellie M. Fairbanks, Marquette. Cypripedium acaule Ait. (Fissipes acaulis Ait., Small.). Stemless Lady’s Slipper, Moccasin Flower——Dry shaded places about Marquette. Fre- quent, Orchis rotundifolia Banks. Small Round-leaved Orchis.—Noted in an open swampy place near Marquette by Thomas M. Danger of Port Huron, Michigan, and reported in the Michigan Flora. Orchis spectabilis L,. (Galeorchis spectabilis L,., Rydb.). Showy Orchis. —Rich woods about Marquette, and reported as frequent. Specimen in herbarium of Miss Nellie M. Fairbanks, Marquette. Habenaria bracteata ( Willd.) R. Br. (Coeloglosswm bracteatum Willd., Parl.). Long-bracted Orchis—Damp shaded ground near Huron Moun- tain Club. Probably frequent throughout. Habenaria flava (1,.) Gray. (Perularia flava V,., Farwell). Tubercled Orchis.—Wet open or partially shaded places near Huron Mountain Club. Probably frequent throughout. 18 C. K. Dopcr Habernaria hyperborea (.) R. Br. (Limnorchis hyperborea \,., Rydb.). Tall Leafy Green Orchis.—Damp rich woods near Huron Mountain Club. Probably frequent throughout. Habenaria dilatata (Pursh) Gray. (Limnorchis dilatata Pursh, Rydb.). Tall White Bog Orchis.—Damp open or partially shaded ground about Mar- quette. Said to be frequent throughout. Habenaria clavellata (Michx.) Spreng. (Gymnadeniopsis clavellata Michx., Rydb.).. Small Green Wood Orchis.—Damp or boggy open or par- tially shaded ground near Huron Mountain Club. Frequent throughout. Habenaria obtusata (Pursh) Richards. (Lysiella obtusata Pursh, Rich- ards). Small Northern Bog Orchis.—Swampy places and rich damp woods near Huron Mountain Club. Probably frequent throughout. Habenaria Hookeri Torr. (Lysias hookeriana A. Gray, Rydb.). Hook- er’s Orchis.—Damp shaded ground near Sugar Loaf Mountain, a large rocky hill near Marquette. Said to be frequent throughout. FHabenaria orbiculata (Pursh) Torr. (Lysias orbiculata Pursh, Rydb.). Smaller Round-leaved Orchis.—Rich shaded ground especially in hardwoods near Huron Mountain Club. Probably frequent throughout. Flabenaria psycodes (\.) Sw. (Blephariglottis psycodes \., Rydb.). Smaller Purple-fringed Orchis—Wet meadow-like ground and swampy places near Huron Mountain Club. Probably frequent throughout. Pegonia ophioglossoides (1,.) Ker. Rose Pogonia, Snake-mouth.—Bog- gy ground near Huron Mountain Club. Noted by Mrs. Wallace Radcliffe. Said to be frequent. i Arethusa bulbosa L. Arethusa, Dragon’s-mouth.—Wet boggy places about the small lakes near Huron Mountain Club. Probably occasional throughout. Spiranthes cernua (1,.) Richard. (Ibidium cernuum L., House). Nod- ding Ladies’ Tresses.-—Summit of Huron Mountain, a large granitic hill near Huron Mountain Club. Probably frequent throughout. Epipactis repens ophioides (Fernald) A. A. Eaton. (Peramium ophioides Fernald, Rydb.). Lesser Rattlesnake Plantain—Damp hardwoods near Huron Mountain Club. Frequent throughout. Epipactis decipiens (Hook.) Ames. (Peramium decipiens Hook., Piper). Menzie’s Rattlesnake Plantain.—In hardwoods near Huron Mountain Club. Frequent throughout. Listera cordata (1,.) R. Br. (Ohprys cordata L.). Heart-leaved Tway- blade, Double-leaf—In a swampy place about Mountain Lake near Huron Mountain Club. Probably frequent throughout. Corallorrhiza trifida Chatelain. (C. corallorhiza L,., Karst.). Early Coral Root.—Damp shaded ground near Marquette. Probably frequent through- out. Specimen in herbarium of Miss Nellie M. Fairbanks, Marquette. Corallorrhiza maculata Raf. Large Coral Root.—Rich woods near Hu- ron Mountain Club. Probably frequent throughout. Corallorrhiza striata Lindl. Striped Coral Root.—Rich shaded ground near Marquette. Occasional. Specimen in herbarium of Miss Nellie M. Fairbanks of Marquette. PLANTS OF MARQUETTE CouN’tTY 19 SALICACEAE Salix pentandra 1, Bay-leaved Willow.—Along lake shore in Mar- quette. Cultivated, not escaping. Salix lucida Muhl. Shining Willow.—Wet open ground with other wil- lows near Huron Mountain Club. Common throughout. Salix serissima (Bailey) Fernald. Autumn Willow.—Near Marquette. Probably frequent throughout. Salix alba L,. White Willow.—In cities and pee Planted for orna- mental purposes, apparently not escaping. Salix alba vitellina (1...) Koch. Golden Willow.—Occasionally cultivated in cities and villages but not permanently escaping. Noted at Marquette. Salix babylonica L. Weeping Willow.—Cultivated as an ornamental tree in cities and villages but not permanently escaping. Salix longifolia Muhl. (S. interior Rowlee). Sandbar Willow.—Along banks of streams and lake shores near Huron Mountain Club. Plentiful throughout. Salix cordata Muhl. Heart-leaved Willow.—Banks of streams and in damp open places near Huron Mountain Club. Probably frequent through- out. Leaves usually not heart-shaped. Salix glaucophylla Bebb. Broad-leaved Willow.—Lake shores near Hu- ron Mountain Club. Plentiful throughout in similar habitats. Salix balsamifera Barratt. (S. pyrifolia Anders.). Balsam Willow.— Border of First Pine Lake in low ground with other willows near Huron Mountain Club. Probably frequent throughout. Noted by B. Gladewitz. Salix syrticola Fernald. (S. adenophylla Hook.). Furry Willow.—On and near sandy beach of Lake Superior near Huron Mountain Club. Plen- tiful. Salix pedicellaris Pursh. Bog Willow.—TIn wet boggy open places near Huren Mountain Club. Probably frequent throughout. Salix discolor Muhl. Glaucous Willow, Pussy Willow.—Low ground or low banks of streams near Huron Mountain Club. Common throughout. Salix petiolaris Sm. Slender Willow.—Damp open ground especially on borders of smali lakes and along banks of streams near Huron Mountain Club. Common throughout. Salix humilis Marsh. Prairie Willow.—Dry open ground near Huron Mountain Club. Plentiful throughout. Salix rostrata Richards. (S. Bebbiana Sarg.). Bebb’s Willow.—Dry or low damp ground especially on lake borders near Huron Mountain Club. Common throughout. Salix candida Fligge. Hoary Willow, Sage Willow.—Marshy and boggy open ground about Marquette. Salix pellita Anders. Satiny Willow—Damp sandy shore of First Pine Take near Huron Mountain Club. Probably occasional throughout. A rare willow. Populus alba 1... White Poplar.—Occasionally set out as a yard or street ornamental tree in cities and villages, but not escaping. 20 Ck Doper Populus tremuloides Michx. Common Poplar, American Aspen, Quiver- leaf —Dry or damp open ground near Huron Mountain Club. Frequent throughout but not noted as abundant anywhere. Populus grandidentata Michx. Large-toothed Poplar, Large-toothed Aspen. — Bordering hardwoods near Huron Mountain Club. Plentiful throughout. Populus balsamifera L. Balsani Poplar.—Occasional along the sandy beach of Lake Superior near Huron Mountain Club. Frequent throughout. Populus deltoides Marsh. Cotton-wood.— Not native in Marquette County but a form of it is known as the Large-leaved Poplar or Carolina Poplar, often set out as a street tree in cities and villages. Populus nigra italica Du Roi. (P. italica Moench.). Lombardy Poplar.— Occasionally set out in cities and villages as a street tree and often spread- ing by root. Introduced from Europe. MYRICACEAE Myrica Gale L. Sweet Gale.——-A shrub from three to five feet high in damp open ground along streams but mostly bordering the small Jakes near Huron Mountain Club. Abundant throughout. Myrica asplenifolia 1. (Comptonia peregrina L., Coulter). Sweet Fern. —A shrub one to three feet high growing in dry sandy mostly open ground near Huron Mountain Club. Frequent, often abundant throughout. BETULACEAE Corylus rostrata Ait. Beaked Hazelnut.—Rich open or partially shaded ground about Marquette and Big Bay. Plentiful throughout. Ostrya virginiana (Mill.) K. Koch. Ironwood.—In hardwoods near Huron Mountain Club. Plentiful throughout. Betula lutea Michx. f. Yellow Birch—One of the principal trees in the extensive hardwood forests of the county. Lumbermen, woodsmen and land cruisers call the small trees with smooth bark yellow birch; the large ones with rough bark and red hard wood, red birch or cherry birch, the lat- ter being much more valuable. Betula alba papyrifera ( Marsh.) Spach. (B. papyrifera Marsh.). Canoe Birch, White Birch—With other trees near Huron Mountain Club. Fre- quent throughout. Betula pumila glandulifera Regel. (B. glandulifera Regel, Butler). Glandular Low Birch.—Borders of tamarack-black spruce swamps near Hu- ron Mountain Club. Frequent throughout. Alnus crispa (Ait.) Pursh. Green Alder, Mountain Alder—Borders of Rush Lake near Huron Mountain Club. Probably frequent throughout. This and the following species are much alike in general appearance. Alnus mollis Fernald. Downy Green Alder.—Bordering woods near Huron Mountain Club, and near lighthouse at Marquette. Probably fre- quent throughout. Alnus incana (L.) Moench. Speckled Alder—Swamps and borders of streams near Huron Mountain Club. Common throughout. PLANTS OF Margurrre County 2I FAGACEAE Quercus rubra L. Red Oak—A large forest tree sometimes 140 feet high and 7 feet in diameter, the species is dwarfed on Huron Mountain and other rocky hills near Huron Mountain Club. Seldom abundant but to be found throughout with other trees in dry or damp rich ground. Quercus rubra anbigua (Michx. f.) Fernald. (Q. borealis Michx. f=) Gray Oak, Northern Red Oak.—A few trees noted northwest of Marquette were thought to be this variety. According to systematists it is very close to the preceding and cannot be distinguished from it without perfect fruit and even then it is often very doubtful. Also thought to be on Grand Island, Alger County. URTICACEAE Ulmus americana L. American Elm.—With other trees near Huron Mountain Club and throughout the county, along streams and in hardwoods. Cannabis sativa L. Hemp.—Frequent about cities and villages through- out, persisting at least for a few years. Humulus Lupulus 1, Common Hop, Hops.—Common along banks of Salmon Trout River near Huron Mountain Club. Also noted on rocky bluffs at Marquette. Perhaps frequent throughout. Urtica gracilis Ait. Slender Nettle—Damp open or shaded ground about Marquette. Frequent throughout. Laportea canadensis (L.) Gaud. (Urticastrum divaricatum 1,., Kuntze). Wood Nettle—Open rich woods near Huron Mountain Club. Probably freauent throughout. SANTALACEAE Comandra Richardsiana Fernald. Richards’ ‘Toad-flax.—Dry open or partially shaded ground near Huron Mountain Club. Frequent throughout. ARISTOLOCHIACEAE Asarum canadense 1, Wild Ginger—Rich shaded ground near Mar- quette. In herbarium of Miss Nellie M. Fairbanks at Marquette. Said to be frequent. POLYGON ACEAE Rumex Britannica L. Great Water Dock.—Wet and swampy open places near Huron Mountain Club. Frequent throughout. Rume.x crispus L. Yellow Dock, Curled Dock—JIn fields and on road- sides as a weed. Common throughout. Rumex mexicanus Meisn. Willow-leaved Dock.—JIn rich dry sandy ground about Marquette and at Big Bay. Probably frequent throughout. Rume.x obtusifolius . Broad-leaved Dock, Bitter Dock.—In fields, pas- tures, on roadsides and a common weed in waste places. Plentiful through- out. Rumex Acctosella L,. Field Sorrel—A common weed in fields and waste places. 22 C. K. DopcE Polygonum aviculare L,., Knot Grass, Door-weed.—Common as a weed in cities and villages, especially in yards, lawns and waste places. Polygonum erectum 1, Erect Knotweed.—On roadsides and in streets of cities, villages and in waste places, Plentiful throughout. Polygonum tenue Michx. Slender Knotweed.—Dry open ground near Marquette, and on summit of Huron Mountain near Huron Mountain Club. Probably frequent throughout. Polygonum lapathifolium L. (Persicaria lapathifolia L., S. F. Gray). Pale Persicaria-—Wet open places at Marquette and Big Bay. Frequent throughout. Polygonum amphibium L. (Persicaria amphibia L,., S. F. Gray). Water Persicaria——Low banks of streams and small lakes and often in shallow water where it floats like a pondweed. Polygonum Hydropiper L. (Persicaria hydropiper L.., Opiz.). Common Smartweed.—Moist open or partially shaded ground near Marquette and at Big Bay. Probably frequent throughout. Polygonum acre HBK. (Persicaria punctata Ell., Small). Water Smart- weed.—Wet open places or often in shallow water near Marquette and at Big Bay. Frequent throughout. Polygonum Persicaria L,. (Persicaria persicaria 1, Small.). Lady’s Thumb.—Open damp ground, usually a weed in cities, villages and waste places. Common throughout. Polygonum hydropiperoides Michx. (Persicaria hydropiperoides Michx., Small). Mild Water Pepper—Margin of Independence Lake and in shal- low water at Big Bay. Apparently infrequent. Polygonum sagittatum L. (Tracaulon sagittatum L,., Small.). Arrow- leaved Tear-thumb.—J,ow open or partially shaded ground about Marquette. Frequent throughout. Polygonum Convolvulus 1. (Tiniaria convolvulus Webb & Mogq.). Black Bindweed.—A weed in gardens, other cultivated grounds and waste places. Common throughout. Polygonum cilinode Michx. (Tiniaria cilinodis Michx., Small). Fringed Black Bindweed.—Dry sandy ground or rocky hills near Marquette. Fre- quent throughout. . Fagopyrum esculentum Moench. (F. fagopyrum 1, Karst.). » Buck- wheat.—Occasional in waste places of cities and villages throughout. Per- haps not a permanent escape. Polygonella articulata (L.) Meisn. Coast Jointweed.—Dry open sandy ground near Marquette. Probably frequent throughout. CHENOPODIACEAE Cycloloma atriplicifolium (Spreng.) Coult. Winged Pigweed.—Sandy ground in waste places about Marquette. Chenopodium Botrys L. Jerusalem Oak.—In waste places as a weed about cities and villages. Chenopodium capitatum (L.) Asch. (Blitum capitatum L.). Strawberry Blite—Dry open ground near Marquette. Occasional throughout. PLANTS OF MARQUETTE CouNtTY 23 Chenopodium glaucum 1, Oak-leaved Goosefoot.—Only as a weed in gardens, streets and waste places of cities and villages throughout. Chenopodium hybridum 1,. Maple-leaved Goosefoot.—Margins of woods and as a weed in and about cities and villages throughout. Chenopodium album L.. Lamb’s Quarters, Pigweed.—One of our com- mon weeds in cultivated grounds everywhere. Atriplex patula hastata (1,.) Gray. Halberd-leaved Orache.—A com- mon weed in the waste places of cities and villages throughout. Salsola Kali tenuifolia G. F. W. Mey. (S. pestifer A. Nelson). Russian Thistle—Noted as a weed throughout in cities and villages. AMARANTHACEAE Amaranthus retroflexus L. Amaranth Pigweed, Green Amaranth.—A common weed of cultivated grounds and in waste places of cities and vil- lages. Amaranthus graecizans L. Tumble Weed.—Fields, gardens and waste places. Plentiful as a weed throughout. Amaranthus blitoides Wats. Prostrate Amaranth.—A weed in cultivated grounds and waste places of cities and villages throughout. CARYOPHYLLACEAE Arenaria serpyllifolia L. ‘Thyme-leaved Sandwort.—Dry open ground as a weed in and about cities and villages throughout. Stellaria borealis Bigel. (Alsine borealis Bigel, Britton). Northern Stitchwort——Damp open or partially shaded places near Marquette and Ne- gaunee. Probably frequent throughout. Stellaria media (I,.) Cyrill. (Alsine media L.) Common Chickweed.— A common weed in gardens, lawns and waste places of cities and villages throughout. Cerastium arvense 1, Field Chickweed, Meadow Chickweed.—Cracks and crevices of granitic rocks on rocky hills near Huron Mountain Club. Probably in similar situations throughout. Plentiful. Cerastium vulgatum 1, Larger Mouse-ear Chickweed.—Fields, gardens, lawns and waste places throughout as a weed. Common. Agrostemma Githago L. Corn Cockle, Cockle—Noted in this county only as an occasional weed about cities and villages. Frequent in wheat- fields of lower Michigan and known as “Cockle.” Lychnis alba Mill. White Campion.—Fields and waste places as a weed. Throughout. becoming commen. Silene antirrhina L. Sleepy Catchfly. quette. Probably frequent throughout. Silene noctiflora 1.. Night-flowering Catchfly.— Fields, gardens and waste places in and about cities and villages. Frequent throughout. Saponaria officinalis 1. Bouncing Bet—On dry sandy ground in and about Marquette. Probably frequent throughout. Dry open ground about Mar- 24 C. K. Dopcr Saponaria Vaccaria L,. (Vaccaria vaccaria L,., Britton).Cow-herb, Cock- le-—QOccasional along railroads and in depot grounds and freight yards, Marquette, Negaunee and Ishpeming. Dianthus barbatus 1, Sweet William.—lInclined to escape from cultiva- tion and persist near old dwellings. Noted in particular in a small clearing near Huron Mountain Club. PORTULACACEAE Claytonia virginica L. Spring Beauty.—Rich moist shaded grounds, us- ually in woods. Noted near Marquette by Miss Nellie May Fairbanks who has a specimen in her herbarium. Probably frequent throughout, but ap- parently less so than the following. Claytonia caroliniana Michx. Carolina Spring Beauty.—Rich shaded ground near Huron Mountain Club. Apparently abundant throughout. Portulaca oleracea L.. Purslane, Pussley—Gardens and waste places as a weed in and about cities and villages. Often abundant. NYMPHAEACEAE Nymphaea advena variegata (Engelm.) Fernald. Variegated Yellow Pond Lily—Common in still and stagnant water near Huron Mountain Clug, and throughout. Included in N. advenea Soland in Briton & Brown’s new illustrated manual. [Castalia odorata (Ait.) Woodville & Wood. Sweet-scented White Water Lily.—In the opinion of the writer it is very doubtful whether this species exists in Marquette County. No very sweet-scented flowers have been noted. A careful examination of rootstocks will perhaps decide the question. For the present all white water lilies of the county so far as seen and examined are referred to the next species. ] Castalia tuberosa (Paine) Greene. Tuberous White Water Lily—In small lakes and small slow streams near Huron Mountain Club. Plentiful throughout. Brasenia Schreberi Gmel. Water Shield——Noted in a pond or small lake near Negaunee. Probably frequent throughout. RANUNCULACEAE Ranunculus Flammula filiformis (Michx.) Hook. Smaller Creeping Spearwort.—Damp sandy shores of small lakes near Huron Mountain Club. Abundant. Ranunculus sceleratus 1, Celery-leaved Crowfoot.—Very wet places about Marquette, Negaunee and Ishpeming. Probably frequent throughout. Ranunculus abortivus 1, Small-flowered Crowfoot. — Rich shaded ground near Huron Mountain Club. Probably frequent throughout. Ranunculus recurvatus Poir. Hooked Crowfoot.—Dry open or shaded ground near Huron Mountain Club. Plentiful throughout. Ranunculus fascicularis Muhl. Early Crowfoot—Open rocky places on Huron Mountain near Huron Mountain Club. Not noted anywhere before in Northern Peninsula by the writer. PLANTS OF MARQUETTE CouUNTY 25 Ranunculus septentrionalis Poir. Swamp Buttercup—Wet shaded ground at Big Bay, and near Marquette, Negaunee and Ishpeming. Fre- quent throughout. Ranunculus pennsylvanicus L. f. Bristly Crowfoot——Wet open or par- tially shaded places near Huron Mountain Club. Frequent throughout. Ranunculus acris 1, Tall Buttercup.—Abundant throughout the county. Noted in particular at and near Negaunee, where it covers the sides and summits of rocky hills giving them a beautiful golden color when the plant is in bloom. Thalictrum dasycarpum Fisch. & Lall. Tall Meadow Rue.—In rich open or shaded ground bordering woods or along streams near Huron Mountain Club. Frequent throughout. Hepatica triloba Chaix. (H. hepatica L., Karst.). Round-lobed Liver- leaf. — Rich shaded ground near city of Marquette. Probably frequent throughout. Anemone canadensis L. Canada Anemone. — Rich open or partially shaded ground near Big Bay. Probably frequent throughout. Anemone quingquefolia LL. Wind-flower—Rich shaded ground along Salmon Trout River near Huron Mountain Club. Probably frequent throughout. Clematis virginiana 1, Virginia Virgin’s Bower.—Rich shaded ground especially along stream banks near Huron Mountain Club. Plentiful throughout. Caltha palustris L. Marsh Marigold—Wet open or shaded ground. Abundant throughout. Coftis trifolia (L.) Salisb. Goldthread—Damp shaded ground near Huron Mountain Club, Common throughout. Aguileqia canadensis L. Wild Columbine.—Rocky granitic hills near Huron Mountain Club. Frequent throughout. Actaea alba (L.) Mill. White Baneberry—Rich shaded ground near Huron Mountain Club. Plentiful throughout. BERBERIDACEAE Caulophyllum thalictroides (L.) Michx. Blue Cohosh—Rich shaded ground along Salmon Trout River near Huron Mountain Club. Perhaps frequent throughout. PAPAVERACEAE Sanguinaria canadensis L. Bloodroot—Rich shaded ground along Sal- mon Trout River near Huron Mountain Club. Probably frequent through- out. FUMARIACEAE Dicentra Cucullaria (U.) Barnh. (Bicuculla cucullaria L., Muillsp.). Dutchman’s Breeches——Abundant in shaded ground along Salmon Trout River near Huron Mountain Club. Common throughout. Corydalis sempervirens (L.) Pers. (Capnoides sempervirens Borck.). Pale Corydalis.—Granitic rocky hills near Huron Mountain Club, and about Marquette. Probably common throughout. 26 C. K. Dopcs CRUCIFERAE Alyssum alyssoides 1.. Yellow Alyssum.—Dry ground in and about cities and villages. Frequent throughout. Lepidium virginicum L. Wald Peppergrass—A common weed in culti- vated grounds, on roadsides, and in and about cities and villages. Capsella Bursa-pastoris (1,.) Medic. (Bursa bursa-pastoris L., Britton). Shepherd’s Purse—A common weed in cultivated grounds and in and about cities and villages. Cakile edentula (Bigel.) Hook. American Sea Rocket.——On and near sandy beach of Lake Superior near Huron Mountain Club. Common on the sandy beaches of the Great Lakes. Brassica arvensis (L.) Ktze. (Sinapis arvensis L.). Wild Mustard, Common Mustard—A bad weed in grain fields, and frequent about cities and villages. Sisymbrium officinale leiocarpum DC. Smooth-podded Hedge Mustard. —A common weed on roadsides and in waste places of cities and villages. throughout. Included in Erysimum officinale L. in Britton & Brown. Sisymbrium altissimum L. (Norta altissima L., Britton). Tumble Mus- tard.—A weed about cities and villages and along railroads throughout. Plentiful. Erysimum cheiranthoides 1. (Cheirinia cheiranthoides L., Link.). Wormseed Mustard.—A weed along railroads, in fields, and waste places of cities and villages. Frequent. . Radicula Nasturtium-aquaticum (L.) Britton & Rendle. (Sisymbriunt nasturtium-aquaticum L,.). True Water Cress—Creeks and ditches about Marquette and Big Bay. Plentiful. Radicula palustris (L.) Moench. Marsh Water Cress. — Wet open places or in shallow water near Huron Mountain Club. Frequent through- out. Radicula Armoracia (1,.) Robinson. (Armoracia armoracia L,., Britton). Horseradish.—Noted throughout in damp open ground about and in cities and villages. Dentaria diphylla Michx. Two-leaved Toothwort, Crinkle-root—Rich shaded ground near Big Bay. Probably frequent throughout. Dentario laciniata Muhl. Cut-leaved Toothwort—Rich shaded ground along Salmon Trout River near Huron Mountain Club. Probably frequent throughout. Cardamine parviflora L. Small-flowered Bitter Cress—Rocky summit of Huron Mountain. Probably to be found in similar situations throughout. Noted by B. Gladewitz. Cardamine pennsylvanica Muhl. Pennsylvania Bitter Cress. — Along Salmon Trout River near Huron Mountain Club. Apparently infrequent. Arabis lyrata L. Tiyre-leaved Rock Cress.—On sandy beaches and shores, and on rocky hills about Huron Mountain Club. Probably plentiful throughout. PLANTS OF MARQUETTE COUNTY 27 Arabis brachycarpa (T. & G.) Britton. Purple Rock Cress.——On rocky summit of Huron Mountain, and on rocks about the lighthouse in Mar- quette. Probably in like situations throughout. SARRACENIACEAE Sarracenia purpurea L,. Pitcher-plant—Noted in swampy places and even on logs at Canyon Lake near Huron Mountain Club. Frequent throughout. DROSERACEAE Drosera rotundifolia L, Round-leaved Sundew.—Boggy ground and even on old logs at Canyon Lake near Huron Mountain Club. Probably frequent throughout. CRASSULACEAE Sedum purpureum Tausch. (Sedum triphyllum Haw., S. F. Gray). Liveforever.—Roadsides and waste places in cities and villages. Plentiful throughout. SAXIERAGACEAE Saxifraga virgimensis Michx. (Micranthes virginiensis Michx., Small). Early Saxifrage-—Cracks and crevices of rocks on Huron Mountain near Huron Mountain Club. Abundant. Probably to be found in similar situa- tions throughout the county. First noted by Mrs. A. L. Farwell, on June 3, 1917. Tiarella cordifolia L. False Mitrewort—Rich woods at Sugar Loaf Mountain near Marquette. Apparently infrequent. Mitella nuda L. Naked Mitrewort.—Rich woods near Huron Mountain Club. Common throughout. Chrysosplenium americanum Schwein. Golden Saxifrage—In shaded cool wet and springy places near Huron Mountain Club. Plentiful through- out. Ribes Cynosbati L.. (Grossularia Cynosbati L., Mill.). Prickly Goose- berry.—Open or shaded rocky ground near Sugar Loaf Mountain. Appar- ently infrequent. Ribes setosa (Lindl.) Cav. & Britt. Bristly Gooseberry.—Rich partially shaded ground along Salmon Trout River and about Mountain Lake near Huron Mountain Club. Apparently frequent. See Britton and Brown, 2 ene Vol. 2, 240. Ribes prostratum L/Her. (R. glandulosum Grauer.). Fetid Currant.— Rich shaded ground near Huron Mountain Club. Frequent throughout. Ribes prostratum L?Her. (R. glandulosum Grauer.). Fetid Currant— Abundant along Salmon Trout River near Huron Mountain Club. Probably plentiful throughout. Ribes triste albinerviwm (Michx.) Fernald. Smooth-leaved Swamp Red Currant——Along Salmon Trout River near Huron Mountain Club. Plentiful. 28 C. K. Dopcr ROSACEAE Physocarpus opulifolius (1,.) Maxim. (Opulaster opulifolius 1,., Kuntze). Nine-bark.—Banks of streams and rocky hills near Huron Moun- tain Club. Plentiful throughout. Spiraea salicifolia 1, Common Meadow-sweet—Low damp open ground near Huron Mountain Club. Frequent throughout. Pyrus Malus 1. (Malus malus L., Britton). Common Apple—Noted throughout along roads, on margins of woods, and about cities and villages. Pyrus americana (Marsh.) DC. (Sorbus americana Marsh.). American Mountain Ash.—Borders of hardwoods throughout. Frequent. Amelanchier humilis Wieg. Small Juneberry—Rocky summit of Hu- ron Mountain near Huron Mountain Club. Apparently infrequent. See Rhodora, Vol. 14, 117, July, 1912. Amelanchier canadensis (1,.) Medic. Juneberry—Dry ground bordering hardwoods near Huron Mountain Club. Frequent throughout. Amelanchier laewis Wieg. Smooth-leaved Juneberry.—Dry open or par- tially shaded ground near Huron Mountain Club. Probably frequent throughout. Amelanchier bartramiana (‘Tausch) Roemer. Oblong-fruited Juneberry. —Dry open ground near Lake Superior beach at Huron Mountain Club. Probably frequent throughout. See Rhodora, Vol. 17, 158, July, 1912. Crataegus Oxyacantha 1. English Hawthorn.—Cultivated in cities and villages throughout but not noted as escaping. Crataegus punctaia Jacq. Large-fruited Thorn.—Rich open or partially shaded ground near Marquette. Probably frequent throughout. Crataegus pruinosa (‘Wendl.) C. Koch. Waxy-fruited Thorn —Foot of bluffs in and about Marquette. Plentiful. Fragaria virginiana Duchesne. Common Strawberry.—Common through- out in damp or dry open or partially shaded ground. Fragaria vesca americana Porter. (F. americana Porter, Britton). Amer- ican Wood Strawberry—Usually in rich shaded ground. Frequent throughout. I aldsteinia- fragarioides (Michx.) Trattinick. Barren Strawberry. — Dry woods about Marquette. Specimen in herbarium of Miss Nellie M. Fairbanks, Marquette. Said to be plentiful. Potentilla monspeliensis 1, Rough Cinquefoil.—Dry open ground throughout mostly as a weed. Plentiful. Potentilla argentea 1.. Silvery Cinquefoil—Dry barren open ground near Marquette. Probably frequent throughout. Potentilla palustris (1,.) Scop. (Comarum palustris L.). Marsh Cinque- foil—Wet marshy and boggy places near Huron Mountain Club. Plenti- ful throughout. Potentilla fruticosa L. (Dasiphora fruticosa L., Rydb.). Shrubby Cin- quefoil—Wet or dry open ground often on sandy shores near Huron Moun- tain Club. Probably plentiful throughout. PLANTS OF MARQUETTE CouNTY 29 Potentilla tridentata Ait. (Sibbaldiopsis tridentata Soland, Rydb.). Three-toothed Cinquefoil.—Dry open ground and on rocky hills near Hu- ron Mountain Club. Plentiful throughout. Potentilla Anserina 1. (Argentina Anserina L., Rydb.). Silver-weed. —Usually in dry open ground not far from Lake Superior shore. It is a common plant in the Great Lakes region. Potentilla canadensis L. Common Cinquefoil—Dry sandy open ground near Huron Mountain Club. Frequent throughout. Geum canadense Jacq. White Avens.—Woods near Marquette. Prob- ably frequent throughout. Geum macrophyllum Willd. Large-leaved Avens.—Rich shaded acer at Sugar Loaf Mountain, a large rocky hill near Marquette. Probably plen- ~ tiful throughout. Geum strictum Ait. Yellow Avens.—Moist open ground near Huron Mountain Club. Common throughout. Geum rivale L. Purple Avens.—-Wet, usually partially shaded places near Marquette. Frequent throughout. Rubus idaeus aculeatissimus (C. A. Mey.). Regel & Tiling. (R. strigo- sus Michx.). Wild Red teks —Danp or dry, open or partially shaded ground. Common and often abundant throughout. Rubus parviflorus Nutt. White-flowering Raspberry. — Mostly open places in woods or rocky ground near Huron Mountain Club. Common and often abundant throughout. Rubus triflorus Richards. Dwarf Red Raspberry.—Wet or damp shaded places near Huron Mountain Club. Common throughout. Rubus allegheniensis Porter. High-bush Blackberry—Dry open ground or in thickets. Plentiful throughout. Scientific name taken from last edi- tion of Gray, page 489. Rubus villosus Ait. Dewberry.—Sandy open or partially shaded ground near Huron Mountain Club. Plentiful throughout. See Gray’s New Man- ual of Botany, 7 ed., 402. Agrimonia striata Michx. Britton’s Agrimony.— Damp rich shaded ground along small streams near Marquette. Probably frequent through- out. Rosa acicularis Lind). Prickly Rose-—Dry partially shaded ground near Marauette. Plentiful throughout. Rosa blanda Ait. Meadow Rose—Dry open or partially shaded ground and often on and near sandy shores. Common throughout. Rosa carolina L.. Swamp Rose.—Borders of swamps and streams near Huron Mountain Club. Frequent throughout. Prunus virginiana L. (Padus nana Du Roi, Roemer). Choke Cherry.— In damp or dry, open or partially shaded ground near Huron Mountain Club. Common throughout. Prunus pennsylvanica L. f. Wild*Red Cherry, Bird Cherry—Rocky hills near Huron Mountain Club. Usually grows as a tree from fifteen to twenty-five feet high, but on the summit of Huron Mountain and other 30 C. K. DopcEr rocky hills it is a mere shrub two to three feet high, although blooming and bearing fruit abundantly. Prunus pumila L. Sand Cherry——Dry open or damp ground often on shores near Huron Mountain Club. Plentiful throughout. LEGUMINOSAE Trifolium pratense L. Red Clover.— Roadsides, fields and in open woods. Plentiful throughout. An escape from cultivation. Trifolium repens L. White Clover.—Along roads, in pastures and open woods. Abundant throughout. Individual plants large and vigorous. Trifolium hybridum L. Alsike Clover—Roadsides, and in waste places of cities and villages throughout. Melilotus officinalis (1,.) Lam. Yellow Sweet Clover—Occasional in waste places of cities and villages throughout. Melilotus alba Desr. White Sweet Clover—Occasional along roads and in waste places of cities and villages throughout. Medicago sativa 1, Alfalfa——Occasional in waste places about Mar- quette. Medicago lupulina 1,. Black Medic, Hop Clover.—Frequent in pastures, lawns, and waste places of cities and villages. Vicia angustifolia (1,.) Reichard. Smaller Common Vetch.—Along rail- roads, in cultivated ground and especially about cities and villages. Plenti- ful throughout. Vicia angustifolia segetalis (Thuillier) Koch. Common Field Vetch.— Noted in one place on shaded ground near Huron Mountain Club. Appar- ently infrequent. Vicia Cracca 1. Tufted Vetch—Along railroads near Marquette. Ap- parently infrequent. Lathyrus maritimus (1,.) Bigel. Beach Pea——-Common on and near sandy beach of Lake Superior. Often plentiful near Huron Mountain Club. Lathyrus palustris L. Marsh Vetchling—Damp open places through- out. Lathyrus palustris linearifolius Ser.—Margins of woods and near base of rocky hills about Marquette. Probably frequent throughout. Lathyrus ochroleucus Hook. Cream-colored Vetchling—Rocky places about the lighthouse in Marquette. Plentiful. Probably to be found through- out. LINACEAE Linum usitatissimum 1. Common Flax.—Occasional along railroads and about freight yards and depot grounds in cities and villages throughout. OX ALIDACEAE Oxalis Acetosella 1, Common Wood Sorrel.—In hardwoods often car- peting the ground. Abundant near Huron Mountain Club. O-xalis corniculata 1, Lady’s Sorrel—A weed in gardens, lawns and waste places of cities and villages. Scientific name taken from Gray’s New Manual of Botany, 7 ed., page 534. PLANTS OF MARQUETTE COUNTY 31 GERANIACEAE Geranium Bicknellit Britton. Bicknell’s Cranesbill—Open rocky ground in and about Marquette. Probably frequent throughout. POLYGALACEAE Polygala paucifolia Willd. Fringed Milkwort—Dry or rocky ground near Marquette. Specimen in herbarium of .Miss Nellie M. Fairbanks, Marquette. Probably frequent throughout. EUPHORBIACEAE Euphorbia maculata L,. (Chamaesyce maculata L., Small). Milk Purs- lane.—Waste places in Marquette. Probably frequent throughout. Euphorbia Cyparissias 1, (Tithymalus cyparissias L., Hill). Cypress Spurge.—More or less in and about cities and villages throughout. Usually abundant in cemeteries. CALLITRICHACEAE Callitriche palustris |. Vernal Water Starwort.—In shallow still water on margin of Independence Lake at Big Bay. Probably frequent through- out. ANACARDIACEAE Rhus glabra 1, Smooth Sumach.—Rocky hills near Huron Mountain Club. Plentiful throughout. Rhus Toxicodendron 1. (Toxicodendron radicans L., Kuntze). Poison Ivy.—Noted more or less throughout in sandy open ground, on or near lake shores, and rocky hills. AQUIFOLIACEAE Ilex verticillata (1,.) Gray. Winterberry, Black Alder—tIn low ground throughout. Frequent. Ilex verticillata tenuifolia (Torr.) Wats. (J. bronsxensis Britton). Northern Winterberry—On border of Rush Lake and rocky sides of Can- yon Lake near Huron Mountain Club. Probably frequent throughout. Similar to the preceding but usually the berries are larger, not so numerous and an orange red. Nemopanthus mucronata (1,.) Trel. Mountain Holly—Damp open or partially shaded ground near Huron Mountain Club. Probably frequent throughout. ACERACEAE Acer pennsylvanicum 1, Striped Maple—In hardwoods near Huron Mountain Club. Probably frequent throughout. Often called snakewood. Acer spicatum Jam. Mountain Maple—Moist woods and places near Huron Mountain Club. Common throughout. Acer saccharum Marsh. Sugar Maple.—One of the important trees making up the hardwood forests in Marquette County. Often the dominant tree and usually mixed with yellow birch and others. 32 C.K: Deper Acer saccharinum L. Silver Maple—Not native in Marquette County, but apparently often planted in cities and villages as an ornamental tree. Acer rubrum L,. Red Maple.—Rich woods with other trees, and some- times in dry sandy ground. In the latter habitat it is a shrub or small scraggy tree. Common throughout. Acer Negundo lL. Box Elder.—-Not native but occasionally set out as an ornamental tree in cities and villages. Not noted as escaping. SAPINDACEAE Aesculus Hippocastanum 1, Horse-chestnut.—Occasionally set out as an ornamental tree in cities and villages. Not escaping. BALSAMINACEAE Impatiens biflora Walt. Spotted Touch-me-not—Damp rich open or shaded ground near Huron Mountain Club. Abundant throughout. RHAMNACEAE Rhamnus alnifolia 1/Hér.. Alder-leaved Buckthorn—Damp open or partially shaded places near Marquette. Probably frequent throughout. Ceanothus americanus L, New Jersey Tea.—Gravelly open ground near Marquette. Apparently infrequent. VITACEAE Psedera vitacea (Knerr) Greene. (Parthenocissus quinquefolia L,., Planch.). Virginia Creeper, American Ivy.—Borders of hardwoods near Huron Mountain Club. Frequent throughout. Vitis vulpina L,. Riverside Grape-—Bank of Pine River at bridge near Huron Mountain Club. Not noted elsewhere. Apparently infrequent. TILIACEAE Tilia americana 1, Basswood.—Frequent in hardwoods throughout. Trees often very large. MALVACEAE Malva rotundifolia 1,. Low Mallow.—Common as a weed in gardens and waste places of cities and villages throughout. TLYPERICACEAE Hypericum perforatum 1, Common St. John’s-wort—Dry open ground in and about Marquette. Frequent throughout. Hypericum virginicum 1,. (Triadenum virginicum L., Raf.). Marsh St. John’s-wort—Marshy or swampy ground and often in water and on lake shores near Huron Mountain Club. Frequent throughout. PLANTS OF MARQUETTE CoUNTY 33 CISTACEKAE Helianthemum canadense (.?) Michx. (Crocanthemum canadense L., Britton). Rockrose, Long-branched Frostweed.— Sandy usually open ground near Huron Mountain Club. Probably frequent throughout. Hudsonia tomentosa Nutt. False Heather, Woolly Hudsonia——On and near sandy beach of Lake Superior near Huron Mountain Club. Probably frequent throughout. Lechea stricta Leggett. Bushy Pinweed.—Rocky ground on summit of Huron Mountain near Huron Mountain Club. Plentiful. Probably to be found throughout. VIOLACEAE Viola cucullata Ait. Marsh Blue Violet—Wet open or partially shaded places near Huron Mountain Club. Common throughout. Viola nephrophylla Greene. Northern Bog Violet—Wet open or par- tially shaded places and along streams near Huron Mountain Club. Prob- ably frequent throughout. Viola papilionacea Pursh. Meadow Violet—Moist open or partially shaded ground near Huron Mountain Club. Probably frequent throughout. Viola septentrionalis Greene. Northern Blue Violet—Partially shaded alluvial ground along Salmon Trout River near Huron Mountain Club. Abundant. Probably to be found throughout. Viola Selkirkii Pursh. Selkirk’s Violet, Great-Spurred Violet—Damp shaded ground near Huron Mountain Club. No doubt frequent throughout. Viola lanceolata 1, Wance-leaved Violet.— Open wet boggy places, moist meadow-like ground and often in wet sand on shores near Huron Mountain Club. Frequent throughout. Viola pallens (Banks) Brainerd. Northern White Violet—In open or partially shaded wet, springy ground near Huron Mountain Club. Common throughout. Viola rentfolia Gray. Kidney-leaved Violet—Rich shaded ground es- pecially in hardwoods, near Huron Mountain Club. Common throughout. Viola scabriuscula Schwein. (V. eriocarpa Schwein.) Smooth Yel- low Violet.—Rich damp shaded ground especially in hardwoods near Huron Mountain Club. Frequent throughout. Viola canadensis lL, Canada Violet—Shaded ground at Big Bay. Abun- dant. Probably to be found throughout. Viola conspersa Reichenb. American Dog Violet—Common in low shaded ground near Huron Mountain Club. Noted in particular on rocky ground about the lighthouse, Marquette. Frequent throughout. Viola arvensis Murr. European Field Pansy, Wild Pansy.—Sandy open ground in and about city of Marquette. Plentiful. Miss Nellie M. Fair- banks. CACTACEAE Opuntia Rafinesquii Engelm. (O. humifusa Raf.). Western Prickly Pear.—On rocks at or near summit of Huron Mountain near Huron Moun- tain Club. Plentiful. Miss Anne Russell, Detroit, Michigan. Reported on rocky hills about city of Marquette. 34 C.K. Dopey THYMELAEACEKAE Dirca palustris L. Leatherwood.—Rich woods at Canyon Lake near Huron Mountain Club. Probably frequent throughout. ~ ELAEAGNACEAE Shepherdia canadensis (1,.) Nutt. (Lepargyraea canadensis L., Greene). Canadian Buffalo Berry—Dry or rocky open ground near Huron Mountain Club. Frequent throughout. ONAGRACEAE Ludvigia palustris (1,.) Ell. (Isnardia palustris L.). Marsh Purslane. —Wet places and in shallow water near Marquette and Big Bay. Plentiful. Epilobium angustifolium 1, (Chamaenerion angustifolium 1,., Scop.). Great Willow-herb, Fireweed.—In low open or partially shaded ground and often in clearings. Plentiful throughout. E pilobium densum Raf. (&. lineare Muhl.). Linear-leaved Willow-herb. —Open low marshy ground near Marquette and Negaunee. Frequent. Epilobium adenocaulon Haussk. Northern Willow-herb. — Rich damp and usually open ground near Big Bay. Probably common throughout. Oenothera biennis 1, Common Evening Primrose——Usually in open dry and even in sandy places, especially on and near sandy beaches of Lake Superior. Frequent throughout. Circaea alpina 1, Smaller Enchanter’s Night Shade-——Damp woods near Huron Mountain Club. Probably plentiful throughout. HALORAGIDACEAE Myriophyllum heterophyllum Michx. Various-leaved Water Milfoil— In shallow water of Third Pine Lake, by the “Styx” near Huron Mountain Club. Probably frequent throughout. ARALIACEAE Aralia racemosa 1, American Spikenard—Rich shaded ground near Huron Mountain Club. Frequent throughout. Araha hispida Vent. Bristly Sarsaparilla—Dry or sandy open or par- tially shaded ground near Marquette and Big Bay. Probably frequent throughout. Aralia nudicaulis 1,. Wild Sarsaparilla—Rich shaded ground near Hu- ron Mountain Club. Common throughout. Panax quinquefolium L, Ginseng—Probably not growing wild at pres- ent in Marquette County. Should be looked for in rich woods. Cultivated at Big Bay by T. H. Carlson who is also attempting to plant it in nearby woods. Panax trifolium 1, Dwarf Gingseng, Ground-nut.—Rich woods es- pecially rich alluvial ground along Salmon Trout River near Huron Moun- tain Club. Probably plentiful throughout. PLANTS OF MARQUETTE CouNTY 35 UMBELLIFERAE Sanicula marilandica L. Black Snake-root—Rich shaded ground near Huron Mountain Club and at Big Bay. Common throughout. Osmorhiza Claytont (Michx.) Clarke, Britton. (Washingtonia Claytoni Michx.). Woolly Sweet Cicely—Rich shaded ground near Huron Moun- tain Club. Probably frequent throughout. Cicuta bulbifera L. Bulb-bearing Water Hemlock.—Wet swampy open places near Huron Mountain Club. Carum Carvi Ll. Caraway.—About dwellings, on roadsides and in waste places of cities and villages throughout. Frequent. Stum cicutaefolium Schrank. Hemlock Water Parsnip—Wet muddy places near Big Bay. Plentiful throughout. Heracleum lanatum Michx. Cow Parsnip.— Damp open or shaded ground at Big Bay. Frequent throughout. CORNACEAE Cornus canadensis L. (Chamaepericlymenum canadense L., Asch. & Graebn.). Dwarf Cornel, Dwarf Dogwood—Damp shaded ground near Huron Mountain Club. Common throughout. Cornus circinata L?Her (C. rugosa Lam.). Round-leaved Dogwood.— Shaded rocky ground about Sugar Loaf Mountain near Marquette. Plen- tiful throughout. Cornus Amomum Mill. Silky Cornel—Damp ground on margin of woods near Big Bay. Probably plentiful throughout. Cornus Baileyi Coult. & Evans. Bailey’s Dogwood.—Open dry ground about Sugar Loaf Mountain near Marquette. Probably frequent throughout. Cornus stolonifera Michx. Red-osier Dogwood.—In wet places, often in damp sand near shores. Common throughout. Cornus paniculata L/Her. (C. femina Mill.). Panicled Dogwood. — Thickets, stream banks and borders of woods near Marquette. Common throughout. Cornus alternifolia L. f. Alternate-leaved Dogwood.—Borders of woods and banks of small streams near Huron Mountain Club. Frequent through- out. ERICACEAE Chimaphila wnbellata (L.) Nutt. Pipsissewa, Prince’s Pine.—Dry woods near Huron Mountain Club. Frequent throughout. Pyrola secuida L. One-sided Wintergreen—Rich woods near Huron Mountain Club. Plentiful throughout. Pyrola elliptica Nutt. Shin Leaf.—Dry shaded ground especially in dry woods near Huron Mountain Club. Probably frequent throughout. ™ Pyrola asarifolia Michx. Liver-leaf Wintergreen—Damp shaded ground about Sugar Loaf Mountain near Marquette. Probably frequent through- out. 36 C. K. Dopcr Pyrola asarifolia incarnata (Fisch.) Fernald. (P. uliginosa Torr.). Bog Wintergreen—Damp or boggy ground about Sugar Loaf Mountain near Marquette. Probably frequent throughout. Monotropa uniflora 1, Indian Pipe.—Rich usually shaded ground near Huron Mountain Club. No doubt common throughout. Ledum groenlandicum Oeder. Labrador Tea—In boggy places and damp thickets near Huron Mountain Club. Common throughout. Kalmia polifolia Wang. Pale Laurel—Boggy open ground near Huron Mountain Club. Frequent throughout. Andromeda glaucophylla Link. Bog Rosemary.—Boggy places and wet shores near Huron Mountain Club. Common throughout. Chamaedaphne calyculata (L.) Moench. Leather-leaf—Open boggy ground near Huron Mountain Club. Common throughout. Epigaea repens 1, Trailing Arbutus, May Flower.—Sandy ground, mostly under pines in the vicinity of the Huron Mountain Club, where it is abundant. Common throughout. Gaultheria procumbens L,. Wintergreen, Checkerberry——Usually in dry sandy shaded ground, sometimes in sphagnum swamps. Abundant near Huron Mountain Club and common throughout. Arctostaphylos Uva-ursi (U.) Spreng. (Uva-ursi Uva-ursi L., Britton). Red Bearberry, Bearberry.— Sandy shores and rocky hills near Huron Mountain Club. Plentiful throughout. Chiogenes hispidula (1,.) T. & G. Creeping Snowberry—Damp woods and boggy ground near Huron Mountain Club. Frequent throughout. Gaylussacia baccata (Wang.) C. Koch. Black Huckleberry—Usually in shaded sandy ground near Huron Mountain Club. Plentiful throughout. Vaccinium pennsylvanicum Lam. (V. angustifolium Ait.). Low Sweet Blueberry, Blueberry.—Dry sandy open or partially shaded ground near Huron Mountain Club. Common throughout. _ Vaccinium pennsylvanicum nigrum Wood. (V. nigrum Wood, Britton). Low Black Blueberry—-Usually associated with the above species in small or large patches. Frequent throughout. Vaccinium canadense Kalm. Canada Blueberry—Dry, swampy, open or partially shaded ground near Huron Mountain Club. Common throughout. Often growing with and near the preceding. Vaccinium membranaceum Doug]. Thin-leaved Bilberry—Damp shaded ground near Huron Mountain Club. Apparently infrequent. In Chippewa County known as rabbit berry. Berries large black slightly acid but very palatable. Vaccimum ovalifolium Sm. Oval-leaved Bilberry—Rich woods near Huron Mountain Club. Apparently infrequent. Known in Chippewa County near Whitefish Point, as myrtilloid berry. Berry medium sized blue, slightly acid, but very palatable. Vaccinium Oxycoccos L. (Oxjcoccos oxycoccos 1,., MacM.). Small Cranberry.—Borders of and open places in tamarack-black spruce swamps. Probably frequent throughout. PLANTS OF MarguetrE County 37 Vaccinium macrocarpon Ait. (Oxycoccus macrocarpus Ait., Pursh). American Cranberry.—In boggy and swampy places throughout. Plentiful. Often cultivated. PRIMULACEAE Lysimachia terrestris (L.) BSP. Bulb-bearing Loosestrife. — Low marshy open ground near Huron Mountain Club. Frequent throughout. Lysimachia Nummularia L. Moneywort.—A weed in lawns and waste places, Marquette. Probably to be found throughout. An escape from cul- tivation. Lysimachia thyrsiflora L,. (Naumburgia thyrsiflora 1, Duby.). Tufted Loosestrife. --- Wet swampy places, often in shallow water near Huron Mountain Club. Common throughout. Trientalis americana (Pers.) Pursh. Star Flower.—Rich shaded ground near Huron Mountain Club. Common throughout. OLEACEAE Fraxinus americana L. White Ash.—Large trees along the Salmon Trout River near Huron Mountain Club, which are at present believed to be this species. No mature fruit or leaf specimens were secured. Noted June 2 and 7, 1917. Fraxinus pennsylvanica Marsh. Red Ash.—About Rush Lake near Hu- ron Mountain Club. Apparently infrequent. Fraxinus pennsylvanica lanceolata (Borkh.) Sarg. Green Ash. —In damp ground or border of hardwoods and along streams near Huron Moun- tain Club. Frequent throughout. Fraxinus nigra Marsh. Black Ash.—Swampy places and wet low banks of streams near Huron Mountain Club. Common throughout. Syringa vulgaris L. Common Lilac—Noted as escaping from cultiva- tion and persisting in dry sandy ground in Marquette. GENTIANACEAE Menyanthes trifoliata L. Buckbean.—In bogs and shallow water near Huron Mountain Club. Common throughout. APOCYNACEAE Apocynum androsaemfolium L, Spreading Dogbane—Dry open or partially shaded ground near Marcuette. Frequent throughout. CONVOLVULACEAE Convolvulus sepium . Hedge Bindweed—Moist open places near Marquette. Apparently infrequent. Convolvulus arvensis L.. Field Bindweed.—In waste places in and about cities and villages throughout. Noted in particular at Big Bay and in Mar- quette. 38 C. K. Dopcr BORAGINACEAE Cynoglossum officinale L. Wound’s Tongue——Roadsides near Big Bay. Probably occasional throughout. Lappula echinata Gilibert. (L. lappula L., Karst.). European Stickseed. —Roadsides and waste places about cities and villages as a weed throughout. LABIATAE Scutellaria lateriflora 1, Mad-dog Skullcap—Damp shaded ground near Marquette. Frequent throughout. Scutellaria galericulata 1, Marsh Skullcap.—Wet marshy places near Huron Mountain Club. Common throughout. Nepeta Cataria L. Catnip.—As an ocasional weed in waste places of cities and villages throughout. Nepeta hederacea (1..) Trevisan. (Glecoma hederacea L.). Ground Ivy—A weed in lawns and waste places of cities and villages. Frequent throughout. Prunella vulgaris 1. Self-heal.—Roadsides and margins of woods near Marquette. Frequent throughout. Galeopsis Tetrahit L. Hemp Nettle-—Roadsides, borders of woods and in waste places in and near Marquette. Probably frequent throughout. Leonurus Cardiaca 1, Motherwort.——A weed in gardens and waste places in and about cities and villages throughout. Monarda mollis 1, Pale Wild Bergamot.—Dry open ground about Sugar Loaf Mountain near Marquette. Apparently infrequent. Hedeoma hispida Pursh. Rough Pennyroyal.—Dry open ground near Marquette. Apparently infrequent. Lycopus uniflorus Michx. Northern Bugle Weed.—Low damp open or partially shaded ground near Marquette. Probably frequent throughout. Lycopus americanus Muhl. Cut-leaved Water Hoarhound.—Damp open or shaded ground near Huron Mountain Club. Common throughout. Mentha arvensis L. Corn Mint.—As a weed in waste places of cities and villages. Frequent throughout. Mentha arvensis canadensis (L.) Briquet. (M. canadensis L.). Amer- ican Wild Mint—Damp open or shaded ground near Huron Mountain Club. Common throughout. Mentha arvensis lanata Piper. Woolly Wild Mint—Damp sandy ground at Big Bay. Plentiful. Large and vigorous. Probably frequent through- out. SOLANACEAE Solanum Dulcamara I,. Climbing Nightshade.—Occasional in and about cities and villages throughout. Known in Europe as bittersweet. Solanum nigrum L. Garden Nightshade, Deadly Nightshade.—Noted as a weed in gardens in cities and villages throughout. PLANTS OF MARQUETTE CouUNTY 39 SCROPHULARIACEAE Verbascum Thapsus L. Common Mullein—A weed along railroads, in pastures, fields, and about cities and villages. Common throughout. Linaria vulgaris Hill. (Llinaria L., Karst.). Butter and Eggs——A weed in gardens and waste places of cities and villages. Plentiful throughout. Collinsia parviflora L,. (C. tenella Pursh, Piper). Small-flowered Coll- insia— Summit of Huron Mountain in cracks and crevices of rocks. Plen- tiful. B. Gladewitz. A small plant and easily overlooked. Probably fre- duent in similar situations. Scrophularia leporella Bicknell. Hare Figwort——Rich open or partially shaded ground near Marquette. Probably frequent throughout. Chelone glabra L. Snake-head, Turtle-head—Damp usually partially shaded ground near Huron Mountain Club. Mimulus ringens L,. Square-stemmed Monkey Flower.— Wet open ground near Huron Mountain Club. Common throughout. Veronica americana Schwein. American Brooklime.—Along brooks, in ditches and other wet places near Huron Mountain Club. Common through- out. Veronica serpyllifolia L. Thyme-leaved Speedwell—Damp or dry grassy open ground. Plentiful throughout. Veronica peregrina 1, Purslane Speedwell.—Noted only as a garden or street weed in Marquette and Negaunee. Plentiful. Melampyrum lineare Lam. Narrow-leaved Cow Wheat.—Dry shaded ground near Huron Mountain Club. Frequent throughout. LENTIBULARIACEAE Utricularia vulgaris americana Gray. (U. macrorhiza Le Conte). Great- er Bladderwort—Ponds and slow streams near Huron Mountain Club. Probably frequent throughout. Utricularia intermedia Hayne. Flat-leaved Bladderwort—Usually in shallow water of marshy and swampy places near Huron Mountain Club. Frequent throughout, Utricularia cornuta Michx. (Stomoisia cornuta Michx., Raf.). Horned Bladderwort.—Sandy shores of small lakes near Huron Mountain Club. Probably frequent throughout. PHRYMACEAE Phryma Leptostachya L. Lopseed.—Moist shaded ground near Turin. Apparently*infrequent. Reported in the Michigan Flora. PLANTAGINACEAE Plantago major L,. Common Plantain. cities and villages throughout. Plantago Rugelii Dene. Rugel’s Plantain—Along railroads, on road- sides, in fields and waste places of cities and villages. Plentiful throughout. Plantago lanceolata L,. English Plantain.—Occasional as a weed in fields and waste places of cities and villages throughout. Often a vicious field weed. Lawns and waste places in 40 C. K. Dopcr RUBIACEAE Galium trifidum L. Small Bedstraw.—Wet marshy places near Huron Mountain Club. Frequent throughout. Galium Claytoni Michx. Clayton’s Bedstraw.— Damp and swampy places near Marquette. Probably frequent throughout. Galium tinctorium L,. Stiff Marsh Bedstraw.—Damp shaded ground near Huron Mountain Club. Frequent throughout. Galium asprellum Michx. Rough Bedstraw.— Damp shaded ground mostly along streams near Huron Mountain Club. Common throughout. Galium triforum Michx. Sweet-scented Bedstraw.—Rich shaded ground especially in woods near Huron Mountain Club. Common throughout. Mitchella repens L. Partridge Berry—Dry woods near Huron Moun- tain Club. Common throughout. CAPRIFOLIACEAE Diervilla Lonicera Mill. (D. diervilla L., MacM.). Bush Honeysuckle. —Dry sandy open or partially shaded ground about Marquette. Common throughout. Lomcera tatarica L. ‘Tartarian Honeysuckle——Rocky ground about the lighthouse at Marquette. Apparently infrequent. An escape from cultiva- tion. Lonicera canadensis Marsh. American Fly Honeysuckle——In woods throughout the county. Common. Lonicera oblongifolic (Goldie) Hook. Swamp Fly Honeysuckle ——Wet Swampy ground near Huron Mountain Club. Probably frequent through- out. Lonicera Caprifolium LL. Perfoliate Honeysuckle——Sandy or rocky ground in and about Marquette. An escape from cultivation. Miss Nellie M. Fairbanks. Lonicera hirsuta Eat. Hairy Honeysuckle-—Damp mostly shaded ground and in rocky places near Marquette. Frequent throughout. Lonicera dioica l, Smooth-leaved Honeysuckle-—Dry open or partially shaded ground and in rocky places near Marquette. Probably frequent throughout. Symphoricarpos racemosus Michx. Snowberry.—Dry rocky places in and about Marquette. Probably frequent throughout. Symphoricarpos racemosus laevigatus Fernald. Garden Snowberry.— Sandy open ground in Marquette. An escape from cultivation. Linnaea borealis americana (Forbes) Rehder. ‘Twin Flower.—Borders of damp woods or in rocky places near Huron Mountain Club. Common throughout. Viburnum Opulus americanum Ait. (V. opulus L.). High Bush Cran- berry, Cranberry Tree—Borders of damp woods and along streams near Marquette and Big Bay. Frequent throughout. Viburnum acertfolium 1, Maple-leaved Arrow-wood.—Rocky woods about Marquette. Frequent throughout. PLANTS OF MARQUETTE CouNTY 41 Viburnum Lentago L. Nannyberry.—Rich woods and banks of streams near Huron Mountain Club. Frequent throughout. Sambucus canadensis 1, Sweet Elder, Black-berried Elder.—Noted near railroad at Birch in rich open ground. Probably frequent throughout but not so plentiful as the following. Sambucus racemosa 1,. Red-berried Elder—Rich open or shaded ground near Huron Mountain Club. Common throughout. CUCURBITACEAE Echinocystis lobata (Michx.) T. & G. (Micrampelis lobata Michx., Greene). Wild Balsam-apple—Rocky bluffs along Lake Superior in Mar- quette. Apparently an escape from cultivation. CAMPANULACEAE Campanula rotundifolia L. Harebell, Bluebell—Sandy shores and open rocky places near Huron Mountain Club. Frequent throughout. Campanula aparinoides Pursh. Marsh Bellflower——Wet, grassy, open ground near Huron Mountain Club. Frequent throughout. LOBELIACEAE Lobelia Dortmanna L. Water Lobelia——Shallow water of Rush Lake near Huron Mountain Club. Abundant. Perhaps to be found in other lakes of the county. COMPOSITAE Eupatorium purpureum L. Joe Pye Weed.—Low ground mostly on bor- ders of woods and thickets near Marquette and Big Bay. Frequent through- out. Eupatorium purpureum maculatum (1,.) Darl. (E. maculatum 1,.). Spotted Joe Pye Weed.—Wet marshy places near Marquette and Negau- nee. Frequent throughout. Eupatorium perfoliatum 1, Common Thoroughwort, Boneset—Low usually open ground near Huron Mountain Club. Plentiful throughout. Solidago hispida Muhl. Hairy Goldenrod.—Dry sandy open or partially shaded ground near Marquette. Common throughout. Solidago Randii (Porter) Britton. Rand’s Goldenrod.—QOn and near sandy beach of Lake Superior and rocky summit of Huron Mountain near Huron Mountain Club. Probably frequent throughout. Solidago uliginosa Nutt. Bog Goldenrod.—Wet, open ground near Hu- ron Mountain Club. Probably plentiful throughout. Solidago canadensis 1, Canada Goldenrod.—Rich open ground and in thickets near Marquette. Frequent throughout. Solidago altissima 1, Tall Goldenrod—Rich open ground near Big Bay. Probably frequent throughout. Solidago serotina Ait. Late Goldenrod.—Rich open ground or in thick- ets along Salmon Trout River near Huron Mountain Club. Frequent throughout. 42 C.K, Dopex Solidago graminifolia (L.) Salisb. (Authamia graminifolia L., Nutt.). Flat-topped Goldenrod.—Moist open ground near Huron Mountain Club. Common throughout. Aster macrophyllus L. Large-leaved Aster—Rich shaded ground near Huron Mountain Club. Frequent, often abundant, throughout. Aster laevis 1. Smooth Aster.—Dry open or partially shaded ground near Marquette. Probably plentiful throughout. Aster lateriflorus (L,.) Britton. Calico Aster—Border of woods and in thickets near Marquette. Common throughout. Aster Tradescanti ,. ‘Tradescant’s Aster.—Low open or partially shad- ed ground near Marquette. Common throughout. Aster puniceus L. Red-stalk Aster-—Damp open or partially shaded ground near Marquette. Frequent throughout. Erigeron philadelphicus l,. Philadelphia Fleabane, Field Daisy.—Occa- sional in and about cities and villages as a weed. Often a field weed. Erigeron annuus (1,.) Pers. Sweet Scabious.—As an occasional weed only in and about cities and villages throughout. Erigeron ramosus (Walt.) BSP. Daisy Fleabane—Roadsides, fields, pastures and waste places throughout. Frequent. Erigeron canadensis 1, (Leptilon canadense L,., Britton). Horseweed. —A weed throughout in fields, gardens, pastures and waste places. Often abundant. Antennaria canadensis Greene. »Canadian Cat’s Foot—Dry mostly open ground near Marquette. Probably frequent throughout. Antennaria fallax Greene. Large-leaved Cat’s Foot.—Summit of Huron Mountain, on rocky ground, near Huron Mountain Club. Frequent through- out. Autennaria neodioica Greene. Smaller Cat’s Foot——Rich open or par- tially shaded ground near Huron Mountain Club. Frequent throughout. Anaphalis margaritacea (1,.) B. & H. Pearly Everlasting-—Dry open or partially shaded ground near Huron Mountain Club. Abundant through- out. Gnaphalium decurrens Ives. Clammy Everlasting——Fields, pastures, roadsides, usually in dry open ground throughout the county. Gnaphalium uliginosum 1, Tow Cudweed.—Damp or dry open ground in fields and roadsides about Marquette and Big Bay. Frequent throughout. Adenocaulon bicolor Hook. Adenocaulon.—Rich open woods at or near foot of Huron Mountain near Huron Mountain Club. Apparently infrequent. Ambrosia trifida L,. Great Ragweed.—Noted as a weed in Marquette and at Big Bay. Probably frequent throughout in cities and villages. Ambrosia artemisiifolia L. (A. elatior L.). Common Ragweed.—A com- mon weed of cultivated grounds, and in and about cities and villages throughout. Ambrosia psilostachya DC. Western Ragweed—A weed noted in waste places of Marquette, Negaunee and Ishpeming. Becoming frequent. Rudbeckia hirta l,. Yellow Daisy, Black-eyed Susan.—Dry open ground near Marquette. Probably frequent throughout. PLANTS OF MARQUETTE CouNTY 43 Helianthus annuus Ll. Common Sunflower.—Rubbish heaps in Mar- quette. Perhaps not a permanent escape. Helianthus iuberosus 1, Jerusalem Artichoke—Occasional in waste places about Marquette, Negaunee, and Ishpeming. An escape from culti- vation. Bidens frondosa 1,. Beggar-ticks——Damp open ground about Marquette. Probably common throughout. Bidens cernua L. Nodding Bur-marigold.—Wet open ground near Mar- quette. Probably frequent throughout. Achillea Millefolinm L. Common Yarrow-—Common as a weed in fields, pastures, on roadsides and in waste places, throughout the county. Anthemis Cotula lL. May Weed.—About farm dwellings, on roadsides, and in waste places of cities and villages throughout. Chrysanthemum Leucanthemum pinnatifidum Lecoq & Lamotte. Ox- eye Daisy.—As a weed in fields, meadows, along railroads and on roadsides throughout. As a field weed becoming too frequent. Tanacetum vulgare 1, Common 'Tansy—On roadsides and in waste places near cities and villages throughout. Tanacetum huronense Nutt: Lake Huron Tansy.—Sandy beach of Lake Superior above Marquette. Apparently infrequent in this county. Artenusia caudata Michx. ‘Tall Wormwood.—Sandy beach of Lake Su- perior throughout. Often plentiful. Artemisia biennis Willd. Biennial Wormwood.—Waste places in Mar- quette. Apparently infrequent. Introduced from the west. Petasites palmatus (Ait.) Gray. Palmate-leaf Sweet Coltsfoot—Damp shaded ground near Marquette. Probably frequent throughout. Erechtites hieracifolia (L.) Raf. Fireweed—Moist open or shaded ground near Big Bay. Probably frequent throughout. Senecio aureus L. Golden Ragwort.—Wet open or partially shaded ground near Marquette. Probably frequent throughout. Arctium minus Bernh. Common Burdock—A weed well distributed and in various situations throughout the county. Cirsium lanceolatum (1,.) Hill. Common Bull Thistle—In pastures, waste places and on roadsides. Frequent throughout. Cirsium Pitcheri (Torr.) T. & G. Pitcher’s Thistle—Sandy beach of Lake Superior. Apparently infrequent in this county. Cirsium discolor (Muhl.) Spreng. Field Thistle—Rich open or par- tially shaded ground near Marquette. Probably frequent throughout. Cirsium muticum Michx. Swamp Thistle——Wet open or partially shaded ground near Marquette. Frequent throughout. Cirsium arvense (1...) Scop. Canada Thistle—Cultivated grounds road- sides, waste places, and along railroads. Becoming too common throughout. Tragopogon porrifolius L. Oyster Plant—Along railroads and in waste places of cities and villages. An escape from cultivation. Tragopogon pratensis I, Yellow Goat’s-beard.—In and about depot grounds and freight yards of cities and villages and along railroads. Ap- parently frequent throughout. 44 C. K. DopcE Taraxacum officinale Weber. (Leontodon taraxacum L,.). Common Dan- delion.—Very common in pastures and fields throughout. Sonchus arvensis L. Field Sow Thistle—About depot grounds and freight yards in cities and villages and along railroads. Frequent through- out. Sonchus oleraceus 1, Common Sow Thistle—In gardens, fields and waste places as a weed. Plentiful throughout. Sonchus asper (.) Hill. Spiny-leaved Sow Thistle—A weed in culti- vated grounds and waste places of cities and villages. Plentiful throughout. Lactuca scariola integrata Gren. & Godr. Entire-leaved Prickly Lettuce. —Becoming frequent in cultivated grounds and waste places of cities and villages throughout. Lactuca canadensis Ll. Wild Lettuce.— Damp rich open or partially shaded ground near Huron Mountain Club. Plentiful throughout. Lactuca spicata (Lam.) Hitche. Tall Blue Lettuce-—Rich open or par- tially shaded ground near Huron Mountain Club. Common throughout. Hieracium canadense ‘Michx. Canada Hawkweed.—Dry open or par- tially shaded ground near Marquette. Probably frequent throughout. Hieracium umbellatum L,. (H. scabriusculum Schwein). Narrow-leaved Hawkweed.—Dry open or partially shaded ground near Marquette. Noted by the late Prof. C. F. Wheeler and reported in the Michigan Flora. MOUNTAIN CLUB HURON va Nee BA \ | \\ Ez eer, LELEND - ale DAY oy oe sh Shee “F | aS I lane ene ~1 R.285W. R.27W.- al is i 1 alii ) pall we ; . f om | Wu eS pie 1 1 1 Lb | 1 i 1 ao isa PPE ar oO ee | 1 — aes 1 = es EA) MAS (L DROMUS DRO UNIVERSITY OF MICHIGAN MUSEUM OF ZOOLOGY Miscellaneous Publications No. 6 A Synopsis of the Classification of the Fresh- Water Mollusca of North America, North of Mexico, AND A Catalogue of the More Recently Described Species, With Notes BY BRYANT WALKER PART I—SYNOPSIS ANN ARBOR, MICHIGAN PUBLISHED BY THE UNIVERSITY DECEMBER 30, 1918 ADVERTISEMENT The publications of the Museum of Zoology, University of Michigan, consist of two series—the Occasional Papers and the Miscellaneous Publica- tions. Both series were founded and are being supported by Dr. Bryant Walker, Mr. Bradshaw H. Swales and Dr. W. W. Newcomb, except that the cost of the illustrations and of distribution are borne by the University. The Occasional Papers, publication of which was begun in 1913, serve as a medium for the publication of brief original papers based principally upon the collections in the Museum. ‘The papers are issued separately to libraries and specialists, and, when a sufficient number of pages have been printed to make a volume, a title page, table of contents and index are sup- plied to libraries and individuals on the mailing list for the entire series. The Miscellaneous Publications include papers on field and museum technique, monographic studies and other papers not within the scope of the Occasional Papers. The papers are published separately, and, as it is not intended that they shall be grouped into volumes, each number has a title page and when necessary a table of contents and index. ALEXANDER G. RUTHVEN, Director of the Museum of Zoology, University of Michigan. PREFACE It has been many years since the students of the North American fresh- water mollusca have had at their service a concise and complete synopsis of the classification in general use. The several monographs published by the Smithsonian Institution of W. G. Binney and Prime in 1865 and Tryon in 1873, together with Lea’s last Synopsis in 1870, gave a very complete representation of the systematic arrangement in use at that time. Tryon’s “Monograph of the Fresh-water Mollusca of the United States,” a continuation of Haldeman’s Monograph of 1842, did not include the Pleuroceridz and did not add substantially to systematic classification. Since that time, although our knowledge on the subject has greatly in- creased, there have been no general monographs published covering the entire field and the recorded advances in classification are only to be found in many scattered publications not always accessible to the ordinary ~ collector. Certain groups have, indeed, been studied monographically. Simpson’s monumental “Synopsis of the Naiades’’ (1900) followed by his ‘“Descrip- tive Catalogue” of 1914, revolutionized the classification of that great group and pointed the way to a natural classification that has since been greatly elaborated by Ortmann in many scattered papers. Baker’s elaborate “Lymnzidze of North and Middle America” (1911) marked another great advance in systematic classification. Sterki’s recent “Preliminary Catalog of North American Spheeriide” (1916) embodies the results of many years of careful study of that family and is, it is to be hoped, but the fore-runner of a complete, illustrated mono- graph of that most interesting, but difficult, group. With these exceptions, the many changes in systematic nomenclature resulting from the investigations of many different workers have been published piece-meal, as it were, and in many different publications. The need of a concise synopsis, showing the state of the science at the present time as adopted by those who are most familiar with the subject, has been met by the Museum of Zoology of the University of Michigan in its endeavors to further the study of Natural History in the schools of the state and Part I of this paper is an attempt to place before the amateur student a summary of the classification of the fresh-water mollusca which is in current use. It is to be understood, of course, that with our constantly increasing knowledge, many changes will, undoubtedly, be necessary in the future and are to be expected and hoped for. The arrangement here pre- sented is what the compiler understands to be the state of the science at the present time. Part II is an out-growth of a card catalogue that the writer has main- tained for his own convenience for many years. In the last half century the knowledge of our fauna has enormously increased, but the many new species 4 BryANt WALKER have been described in many scattered publications not always accessible to the student and not to be found except by laborious and time-taking search. The convenience of a catalogue giving references to all of the new species described since the Smithsonian monographs were published is obvious. In addition to such references, it has been deemed of service to add under many of the species, old and new, the published opinions of many writers as to their specific validity and relationships. It will be understood, of course, that such quoted opinions are not necessarily those of the compiler and are not endorsed by him unless expressly so stated. They are given solely for what they are worth as representing the views of the author at the time of their publication. The whole subject is brought down to May 1, 1918. Bryant WALKER. Dated July 1, 1918. A SYNOPSIS OF THE CLASSIFICATION OF THE FRESH-WATER MOLLUSCA OF NORTH AMERICA, NORTH OF MEXICO. (CUEISS) ais so a a a GASTROPODA. Smnclags es ees seein cee ee BUTHYNEURA. (DNRC rep Nik Sia 2 aia S sam Me eRe SC PULMONATA. “SyLils (00 24 | ch aan ear eee ect gue BASOMMATOPHORA. Superfamily LIMNOPHILA. The North American Limnophile are in the main inhabitants of fresh water, but are occasionally found in slightly brackish water along the sea coast. ‘The epidermis is smooth; the contractile tentacles are flattened or cylindrical; the eyes are placed on the inner bases of the tentacles. The genital orifices are separated, the male orifice is near the tentacle, the female at the base of the neck, near the respiratory orifice. ‘The jaw is simple or composed of three pieces. The radula has numerous rows of small teeth. Key to the families of Limnophila. I. Shell spiral, dextral, spire more or less elongated...... Lymneide, II. *Shell discoidal, with the columellar margin simple, iC ASMalMl aia hela oon Seven Sarat ea rE Se Planorbide. 17S IIS SunRRA ea ie on On ae Physide. IV. Shell patelliform or spiral, dextral, neritoid or planorboid with the columellar margin broadly dilated.........4 Ancylide. Family LYMNAAIDAE. Shell spiral, dextral, spire usually elongated and acute, but in some groups small and depressed. Animal dextral. Head with a broad, short muzzle dilated at the end. Foot rounded behind. ‘Tentacles flattened. Jaw composed of three plates, a large one in the center, with two small, narrow laterals. Radula broad; central tooth small, simple or bicuspid, the laterals bi- or tricuspid. The marginals bi-, tri-, or multicuspid or serriform. Genus LYMN/A“A Lamarck, 1799. Shell spiral, dextral, thin, unicolor or occasionally with longitudinal stripes ; spire usually acute ; aperture large, oval, rounded below, extremities united by a thin parietal callus; columella more or less twisted or plicate ; lip thin, sometimes with an internal thickening or varix. Animal dextral; head large; tentacles flattened, triangular; eyes sessile on the inner bases of the tentacles. ‘ BryANt WALKER Key to the subgenera of Lymnea. Spme Glevaredticng a ccaes ace ay Sie teeters ce eee 2 Spire SHOrt ).°7 Acti. ves ceee tes ate ctr ene Sa eR eee 6, Body-whorloreatiy iitlated (et pie utter ere ee veins Lymnea s. Ss. Body=swhorl sligitly «titla ted goes vm abe cateiesecs sees ese ech eee 2: Shell very slender; spire longer than the aperture; lip con- PITIMOUMG saa ets ev kite he oe eee ees re RIOR. er nce et te erea ae ere Acella. Shell more or less inflated; spire shorter than the aperture; lip. not COmtIMUOUS \. ees hay he steeee sree oe isa e ee nee ie ere 4. Oo Columella smooth, reflected over the umbilicus.............. Galba. Suctace witiemmpressed> Spiral litesin: =. clue, semen tree Stagnicola. Surtace longiidmally costate: -c. ce eee ee Polyrhytis. Body-whorl large, not inflated, elongate............ Pseudosuccinea. BRedy-whorldlarme, imtlatedsom, sot. ag eanconie eoeet ae een eet sp Shell large, solid, bulimiform; lip not expanded.......... Bulimnea. Shell moderate, spire very short, usually thin, lip more or less: expand ctlt Roose Aes: ahueieese eee oe ee een ee see 8. Shell thin, lip widely expanded, surface polished; spire Sctibe: whorls, rovnd Cd wep s,s cuss enact as eee ere ee Radix. Shell thicker; lip occasionally expanded; lines of growth distinct; spire broad, whorls shouldered....... Stagnicola (pars). L | le@olumella “twasred#om plicdtes.).. 2 asc. ss eect ae ee G3 | [ 4 [ ( 1 [ Subgenus LYMNZMA s. s. Shell large, thin, with an acute, slender spire and expanded body-whorl; axis gyrate, forming a (usually) pervious spiral coil without a true umbilicus; the callus on the body-whorl closely appressed; the outer lip flar- ing more or less, simple, sharp, normally without any thickening. Penis-sac very large; penis very short (about one-quarter the length of the penis-sac) ; penis retractors normally two, very large; anterior termination of the prostate bulb-shaped. Radula with unicuspid cen- Fig. 1 tral and bicuspid lateral teeth, marginals serrate. Synopsis FrESH-WATER MOoLtusca 7 Type: L. stagnalis L,. Example: J,. stagnalis appressa Say, fig. 1. Radula, fig. 2. Subgenus PSEUDOSUCCINEA Baker, 1908. Shell thin, succineiform; spire short; body-whorl large, elongated, not inflated; surface sculptured with spiral, incised lines ; axis gyrate. “Prostate long, narrowly cylindrical with a slight bulbous termination; penis thick, about half as long ff as penis-sac; lateral teeth bicuspid.” Fig. 4 Subgenus RADIX Montfort, 1810. Shell thin, globose-oval; spire short, acute; body- whorl large, inflated; aperture very large; lip ex- panded. “Prostate short, pear-shaped when viewed lateral- ly; penis very slender, longer than penis-sac; first lat- eral tooth tricuspid, balance bicuspid.” Type: L. auricularia L,., fig. 6. Radula, fig. 7. 8 BRYANT WALKER Subgenus BULIMNEA Haldeman, 1841. Shell large and solid, bulimiform, with an imper- vious axis, a twisted or subplicate pillar, the callus on the body-whorl and pillar closely appressed and the outer lip not thickened or expanded. “Prostate very large, irregularly elongate-ovate ; penis very large, one-fourth longer than penis-sac, gradually enlarging in diameter toward the distal end; lateral teeth tricuspid.” i Type: L. megasoma Say, fig. 8. Radula, fig. 9. Fig. 8 Subgenus ACELLA Haldeman, 1841. Shell thin, smooth, acute, extremely slender; aperture expanded at the margin; the inner lip not appressed, a moderate chink behind it; axis gyrate, pervious, not plicate; outer lip simple, sharp. “Prostate large, flatly cylindrical; penis thick, about four-sev- enths the length of penis-sac; lateral teeth bicuspid; the mesocone with a distinct entoconic swelling.” Type: L. haldemani “Desh.” W. G. Binn., fig. 10. Radula, fig. 11. Big saa Subgenus GALBA Schrank, 1803. Shell small, turreted; spiral sculpture wanting or subobsolete ; columella smooth; inner lip flatly reflected over the umbilicus. “Prostate long-ovate; penis a trifle shorter than penis-sac, of narrow diameter; lateral teeth bicuspid.” Type: L. truncatula Mull. Fig. 12 Example: L. caperata Say, fig. 12. Radula, fig. 13. SyNopsis FRESH-WATER MOLLUSCA 9 Section PSEUDOGALBA Baker, 1913. Shell as in Galba, but with the inner lip less expanded. Gen- italia in Galba. Lateral teeth tricuspid. Fig. 15 Type: L. humilis Say, fig. 14. Radula (L. obrussa Say), fig. 15. Animal (ibid.), fig. 16. Subgenus STAGNICOLA Leach, 1830. Shell varying from elongate to short-ovate; outer lip (usually ) somewhat thickered within; columella distinctly plicate; inner lip appressed; axis slightly or not all perforate; surface with strong, spirally impressed lines. “Prostate elongate-pyriform; penis from three-fourths to four- fifths the length of penis-sac, very thick; lateral teeth bicuspid.” Type: L. palustris Muller, fig. 16. Radula, fig. 18. Section POLYRHYTIS Meek, 1876. Shell longitudinally costate. Soft anatomy unknown. Type: L. kingtt Meek. Example: L. utahensis Call, fig. 19. Fig. 19 Family PLANORBID. Key to the subfamilies of Planorbide. DEIN CIS OCCT Raa Nek en a ee Planorbine. SILL WSU URELIO S98 SERS SE aR Pompholigine. IO Bryant WALKER Subfamily PLANORBIN H. and A. Adams, 1858. Shell discoidal, ultra-dextral or sinistral. Animal sinistral, having the pulmonary, genital and excretory orifices on the left side. Tentacles long, slender and cylindrical. Jaw in three segments. Radula with the numerous teeth arranged in nearly horizontal rows, central small and bi- cuspid, marginals tricuspid, laterals multicuspid. In the formation of the keys and descriptions of the various subdivisions of the family, the shells are treated with reference to their apparent mode of spiral growth. Key to the genera of Planorbine. Te peniure dentate, (wit Lilia eee on, one on cae eee Segmentina. 2) re NPectuce: Witnolts tmvermealyteeti omc me. eeu ee ee meee Planorbis. Genus PLANORBIS Miller, 1774. Shell discoidal, dextral or sinistral, spire very much depressed, not usually rising above the margin of the body whorl; no real columella ; aper- ture oblique; outer lip simple and sharp or thickened. Animal smooth; head short; tentacles long, slender and cylindrical; eyes sessile on the inner bases of the tentacles; foot short, narrow, obtuse at both ends; jaw and radula as in the subfamily. Key to the subgenera of Planorbts. (Aperture simple slip thickened: |. .q.t2na: eee eee Helisoma. i Aperture sigiple nee anes ere Gyraulits. a i Body-whorl acutely carinated or lenticular............... Hippeutis. Subgenus PLANORBIS s. s. Shell sinistral, large, with a moderate number of gradually increasing whorls, rounded above and below; aperture slightly and gradually expanded, with its margin simple and sharp. Represented in our fauna by a single section. Synopsis FrRESH-WATER MOortusca Il Section PLANORBINA Haldeman, 1842. Shell like Planorbis, s. s., but vertically compressed, with smaller and more numerous whorls and a very oblique aperture. Type: P. olivaceus Spix. Example: P. glabratus Say, fig. 20. Subgenus HELISOMA Swainson, 1840. Shell dextral or sinistral, few whorled, the whorls carinate above and rapidly enlarging; base funicular; aperture suddenly ex- panding and thickened. Key to the sections of Helisoma. 1. Shell dextral, carinated above and below, spire and base IRI SURE UN 21 eS page a ee aa ae Helisoma s. Ss. 2. Shell sinistral, early whorls flattened and carinate above, ase MMIC UMA ee teehee OL Pe are ea coche cee pers, « w akem he & Pierosoma. Section of HELISOMA s. s. Shell dextral, moderate size, few whorled, the whorls carinated above and below and rapidly enlarging; spire and base funicular, aperture suddenly expanded, with a thick- ened peritreme. Fig. 22 Fig. 23 Type: P. bicarinatus Say, fig. 21. Radula, fig. 22. Animal, fig. 23. Section PIEROSOMA Dall, 1905. Shell sinistral, large, high, with few transversely sculptured whorls; the early whorls carinate and flattened above, funicular below; in the adult the flattened apex is usually depressed below the upper level of the body whorl; the aperture is suddenly expanded and somewhat thickened. Fig. 24 Fig. 26 Type: P. trivolvis Say, fig. 24. Radula, fig. 25. Animal, fig. 26. 12 BryANt WALKER Subgenus PLANORBELLA Haldeman, 1842. Shell sinistral, depressed, whorls more numerous than in /Telisoma,; apex scarcely, if any, depressed below the level of the body-whorl; base funicu- lar; body-whorl constricted behind the widely expanded, campanulate aperture. Type: P. campanulatus Say, fig. 27. Radula, fig. 28. Fig. 28 Subgenus TROPIDISCUS Stein, 1850. Shell dextral, moderately large, depressed, upper surface convex, base flattened, adult periphery angular or carinate, the aperture oblique, slightly expanded, simple. Type: P. umbilicatus Miller. a epee Example: P. cultratus d’Orb., fig. 20. Fig. 29 Subgenus HIPPEUTIS Agassiz, 1837. Shell dextral, small, lenticular, with a small number of rapidly increas- ing whorls, the last enveloping a large part of the preceding whorl; apex slightly depressed, base with a narrow umbilicus, aperture oblique with a thin sharp margin. Type: Helix fontanus Lightfoot. Section MENETUS H. and A. Adams, 1855. Like Hippeutis, but the last whorl not enveloping the (©) preceding whorls to so large an extent. Type: P. opercularis Cld., fig. 30. = yp p Soe) ee Subgenus GYRAULUS Agassiz, 1837. ee g Shell dextral, small, with few, rapidly increasing whorls, <0 ) fully exposed above and below, with a nearly median peri- phery, rounded or obtusely angulated, but not accurately car- inated. Fig. 30 SyNopsis FRESH-WATER MOLLUSCA 13 Key to the sections of Gyraulus. Poe oinidce spirally striate ang MiSpid 26 ee cs ewes Gyraulus s. s. Peo iitace stiooth, or fiiely Strate: . hal pc ook poe oe ee eles, 3s Torquis. PE SUIGEACE COSEHEG: Ct Oooo thie Paths ALS APL usirme Vie Se ee Armiger. Section GYRAULUS, s. s. Periphery rounded, or angulated, surface spirally striate and hispid; aperture simple, sharp-edged, oblique. Type: P. albus Miiller. Example: P. hirsutus Gld., fig. 31. Section TORQUIS Dall, 1905. Like Gyraulus s. s., but with more rounded, less rapidly increasing whorls, not hispid nor spirally striate; aperture expanded and _ slightly Fig. 33 Type: P. parvus Say, fig. 32. Radula, fig. 33. Section ARMIGER Hartmann, 1840. Shell dextral, very small, with few rapidly increasing, costate whorls, the coste projecting at the periphery; the form in a general way like Gyraulus. iype:. P. crista L,., fig. 34. Genus SEGMENTINA Fleming, 1817. Shell discoidal, dextral, whorls few, rounded or carinated above and below, with one or more sets of laminz or teeth in the interior of the shell. Animal as in Planorbis, tentacles filiform; foot narrow anteriorly, wider behind, obtusely rounded at both ends. Radula, central tooth bicuspid; laterals tricuspid ; marginals 4-5 cuspid. Type: Nautilus lacustris Lightfoot. 14 Bryant WALKER Subgenus PLANORBULA Haldeman, 1842. Shell rather small, whorls few, slowly and regularly increasing, rounded or carinated above and below; aperture somewhat expanded, lip more or less thickened within; a single persistent set of 6 dentiform lamellz at short distance within the aperture. Wk A f fl Fig. 36 i =. Type: Planorbis armigerus Say, fig. 35. Aperture, fig. 36. Radula, fig. 37. Subfamily POMPHOLIGINA® Dall, 1866. Shell spiral, dextral, flattened above; the body whorl very large. Animal sinistral. ‘Tentacles moderate, stout, cylindrical, slightly globose at the tips; eyes sessile, near the inner base of the tentacles; foot short, bluntly rounded behind; jaw single, subcordiform; radula with the teeth in nearly horizontal rows, central tooth small, bicuspid, laterals wide. Key to the genera of Pompholigine. Shell arimpemi@nrake = ..2tte oes cores Weim er heater Pompholy.x. shell deelyertiambtlicates* +. certs se nat ster ete ereget iene eee Carinifex. i No Genus POMPHOLYX Lea, 1856. Shell spiral, dextral, globosely depressed, imperforate; whorls few, spire short, obtuse; last whorl very wide, ventri- cose; aperture very large, wide, subcircular, expanded; lip thin: columella thickened. Animal as in the subfamily. Type: P. effusa Lea, fig. 38. Radula, fig. 39. Synopsis FRESH-WATER MOLLUSCA 15 Genus CARINIFEX W. G. Binney, 1863. Shell spiral, dextral, inflated, angular; spire terraced; whorls numerous, visible above, last whorl very large, broad above, rapidly attenuated below, visible above, last whorl very large, broad above, rapidly atten- uated below, umbilicus funnel shaped; aperture triangular, broad above, narrow below; inner lip slightly thickened ; outer lip thin, acute, angular above, flexuose. Animal sinistral, resembling Planorbis, but with much shorter tentacles; jaw single; radula simliar to that of Pom- pholyx. \\ \ lio il) \ At eg Fig. 41 Type: Planorbis newberryi Lea, fig. 40. Radula and jaw, fig. 41. Family PHYSID/%. Shell spiral, sinistral, thin, smooth or with microscopic transverse striae, shining, spire more or less produced. Animal sinistral, having the pulmonary, genital and excretory orifices on the left side; tentacles slender, cylindrical; foot narrow, pointed behind; jaw single, arcuate, with a vertical fibrous accessory process on the superior margin; radula with the teeth arranged in oblique rows. Central tooth wide, base with projecting processes before and behind, multicuspid; laterals obliquely bent, comb-like, multicuspid, with a peculiar process at their ex- ternal angle. 1. Shell elongate, slender, smooth, polished, inner edge of the mantle simple, not digitate, not extending beyond CMS UIs Mapes eect tia Ment Sern RAUL da cob), .42 Dl Aas tee A plexa. 2. Shell less elongated, body whorl usually inflated, smooth or with microscopic revolving striz ; inner edge of the mantle digitate or lobed, extending partly over the shell. .Physa. Genus. PHYSA Draparnaud, 1801. Shell sinistral, oblong, thin, translucent, shining; spire acute, usually short; whorls convex; aperture ovate, rounded below; columella twisted, lip thin, acute, sometimes thickened within. Animal as in the family, but with the inner margin of the mantle digitate or lobed, and extending over the shell. 16 BryANt WALKER Section PHYSA s. s. Shell smooth. Fig. 42 Type: Bulla fontinalis L. Example: P. gyrina Say, fig. 42. Radula: P. humerosa Gld., fig. 43. Animal: P. heterostropha Say, fig. 44. Section COSTATELLA Dall, 1870. Shell longitudinally costate. Type: P. costata Newc., fig. 45. Genus APLEXA Fleming, 1822. Shell sinistral, elongated, slender, smooth, shining; spire acute; lip simple, sharp, columella but slightly twisted. Animal similar to that of Physa, but with the inner edge of the mantle simple, not digitate nor reflected over on the body whorl. " MA are antl \ | Fig. 46 Fig. 47 Fig. 48 Type: Bulla hypnorum L,, fig. 46. Radula, fig. 47. Animal, fig. 48. Family ANCYLID/. Shell patelliform or dextrally spiral, neritiform or planorbiform. Animal (fig. 49) sinistral or dextral, with a large oval foot; tentacles short, blunt, cylindrical; eyes sessile on their inner bases ; jaw in three parts or the whole segmented in plates; radula with the teeth arranged in rows nearly horizontal or slightly curved, central tooth small, unicuspid or bicuspid, laterals bi- eae cuspid or comb-like, marginals comb-like or subobsolete. Fig. 49 Synopsis FRESH-WaATER MoLtusca 17 Key to the subfamilies of Ancylide. © SUDEP STG E av 8 ses U0 a see ena on Ae aa 2. 4 | phellplanorbitormy or netititormm; oi. 2... ed... Neoplanorbine. Peel ain pen timiemenred sa 2a. tora’ alten ede s bas + ¥s geese 3 a elestial With pitt apes: 14 5% salsa 2 6 San Se ass Rhodacmeme. Snag re apexes Dinediale ec. cles | og ose ee eaten Lancine. 5 Shell small, apex more or less posterior and excentric...Ferrissiine. Subfamily LANCINA® Hannibal, 1914. Shell large for the family, ovate-elliptical, depressed or roundly arched ; apex submedial, not prominent, smooth or concentrically striate. Animal with the jaw as in Lymncea with two accessory plates. Radula also lymnzid in character. Central tooth unicuspid or tricuspid; laterals bicuspid with large quadrate bases; marginals comb-like, the cusps extending beyond the base. Key to the genera of Lancine. Pee Seer lew SUING CIPC Ea an) 4> Sibes Sen soe 2 ae aan serene Feeelaewe Lanx. Petecnm lcenapex subterminal ... ¢ . c's oe rows ss te vce Fisherola. Genus LANX Clessin, 1880. Shell large, broadly ovate, roundly arched, rather solid, apex not prominent, smooth or concentrically striate, subcentral. Radula as in the subfamily. Type: Ancylus newberryi Lea, fig. 50. Radula: Lanz pattelloides (Lea), fig. 51. 18 BryANt WALKER Subgenus WALKEROLA Hannibal. Shell as in Zanr, but thinner and more depressed. Fig. 53 Type: Lanx klamathensis Hann., fig. 52. Radula, fig. 53. Genus FISHEROLA Hannibal, 1912. Shell rounded-ovate, somewhat broader anteriorly, depressed-conic, finely concentrically striate, apex small, indistinct, subterminal, but not inclined. Anat- omy unknown. Type: /isherola lancides Yann., fig. 54. Subfamily FERRISSIINA, Walker, 1917. Shell small, thin, broadly ovate to oblong; apex more or less posterior and eccentric, apex (in North American species) smooth or radially striate, Animal having the jaw segmented in plates. Radula with a bicuspid central, laterals obliquely reflected with from two to five small cusps, arranged like the teeth of a comb, marginals also comb-like, cusps not (usually) extending to the basal line. Key to the genera of Ferrissiine. Shell never septate: ii. os acter age cieedee ere ee meee Ferrissia. Shell iaamaturity septate. et oe ee ee ee Gundlachia. Genus FERRISSIA Walker, 1903. Shell ovate to oblong, conic, more or less elevated, apex excentric and posterior, radially striate or smooth. Animal as in the subfamily. Synopsis FRESH-WATER MOoLLusca 19 Key to the subgenera of Ferrissia. Sen -elevated. apex radially striate 2... ..3 5.0.5.0 1 wes ok Ferrissia s. s. Bm meleprecseder pew SIMOOLI 2. 5 ss). cates a4 auld ees erate awe se Levapex. Subgenus FERRISSIA s. s. Shell conic, elevated, apex radially striate. Type: Ancylus rivularis Say, fig. 55. Radula, fig. 56. Fig. 55 et eae AEP HY PN NN pees | ci fo tg 57 i a (ey a Fig. 56 Shell more or less depressed, apex smooth. Type: Ancylus fuscus C. B. Ads., fig. 57. Radula: Ferrissia diaphana (Hald.), fig. 58. Fig. 57 CA DARPA ARN Fig. 58 Genus GUNDLACHIA Pfeiffer, 18409. Shell small, thin, ancyliform, obliquely conical; apex posterior, excentric, smooth or radially striate, inclined to the right ; basal side more or less closed by a flat, horizontal septum; aperture broad-oval; margin continuous, simple, entire. Animal as in Ferrissia. Radula with a bicuspid central and tricus- pid laterals and marginals, the cusps short and broad. Key to the subgenera of Gundlachia. PMc smooth or concentrically wrinkled... ..).0..0.6...'- Gundlachia s. s. ODE ECG SS sits rata ee en oe ae Kincaidella. 20 BRYANT WALKER Subgenus GUNDLACHIA sg, s. Apex smooth. Fig. 59 _ PTT PADUA SPS less Lee Fig. 60 Type: Gundlachia ancyliformis Pfr. Example: Gundlachia hinkleyi \Walk., fig. 59. Radula, fig. 60. Subgenus KINCAIDELLA Hannibal, 1912. Apex radially striate. Type: Ancylus fragilis Try. = G. californica Row. Example: G. meekiana Stimp., fig. 61. Radula, fig. 62. no Fig. 62 Subfamily RHODACMEIN A Walker, 1917. Shell patelliform, conical, elevated or depressed, apex tinged with pink. Animal having the jaw composed of numerous segmented plates. Radula with a long, slender central, unicuspid or faintly bicuspid, and-with the base widely expanded in some species; the first lateral very large, with an enormous mesocone, the blade-like cusp extending beyond the base, the ectocone is back of the mesocone, entirely separated from it and has several small cusps; there is no endocone; the next four laterals are similar in shape, but diminish rapidly in size towards the margin; these are succeeded by two or three transition teeth, smaller and with more or less imperfect cusps. The marginals are very small, rapidly decreasing in size towards the outer edge, with large quadrate bases wider than high, vestigial, the cusps being nearly, if not quite obsolete. Synopsis FresH-WaTER MoLLusca 21 Genus RHODACMEA Walker, 1917. Shell and animal as in the subfamily. Key to the sections of Rhodacmea. - Shell elevated. Radula having the base of the central tooth ex- panded and not overlapped by the mesocone of the first WeMe te Uiee a eNeneen A ra es cancel Sots omen eRe t a ae oe Rhodacmea s. s. Shell depressed. Base of central tooth of radula not expanded, overlapped by the mesocone of the first lateral....Rhodocephala. Section RHODACMEA s. s. Shell elevated. Radula with a unicuspid central, which has the base triangularly expanded; laterals with the cusp of the mesocone extending but little beyond the base and not overlapping the base of the central tooth. Type: Ancylus filosus Con., fig. 63. Radula, fig. 64. Section RHODOCEPHALA Walker, 1917. Shell depressed. Radula with a faintly bicuspid central, which has the sides of the base straight and not expanded; laterals with the cusp of the mesocone extending far beyond the base and overlapping the base of the central tooth. Fig. 66 Type: Rhodacmea rhodacme Walk., fig. 65. Radula, fig. 66. 22 BrYANT WALKER Subfamily NEKOPLANORBIN AZ‘ Hannibal, 1912. Shell small, planorbiform or neritiform. Key to the genera of Neoplanorbine. slrellplanonbit omit to 8.eprteeeet cae ee Ao eakrs cri eene ee Neoplanorbis. hell MmenitihORMin’ 5°. clircs cnecke are tice at tots ee eee aie ee Amphigyra. Genus NEOPLANORBIS Pilsbry, 1906. Shell very minute, planorboid, dextral, subdiscoidal, nearly flat above, convex below, usually carinate at the periphery; whorls two, rapidly enlarging; aperture very : oblique, wider than high, a little dilated at the base; lip Fig. 67 thin, not continuous; columellar margin straight and broadly dilated, somewhat thickened within. Dentition and anatomy so far as known similar to Amphigyra. Type: N. tantillus Pils., fig. 67. Genus AMPHIGYRA Pilsbry, 1906. Shell minute, dextral, neritoid or crepiduliform, imperforate, with a small, depressed, lateral spire; whorls about 14, very rapidly enlarging, the last very convex dorsally; apex smooth; body- whorl spirally striate; aperture very large, transversely oval; lip continuous and full, thin; cavity of the spire very small, a thin, broad, concave, columellar plate projecting across the end next the spire. Fig. 69 Fig. 68 Animal sinistral, externally lymnezid; tentacles short, blunt, cylindric; eyes near their inner bases; a short, false gill in the pallial cavity. Radula ar- ranged as in Lymnea, central tooth unicuspid, laterals bicuspid, marginals low, wide, with four or five cusps. Type: A. alabamensis Pils., fig. 68. Radula, fig. 69. Synopsis FRESH-WaTER MOLLUSCA ag Subclass STREPTONEURA. Order PECTINIBRANCHIA. Suborder TA*NIOGLOSSA. Superfamily PLATYPODA. Key to the families of Platypoda. Be re Gerilintiy (CONGEMETIE hs cei eras oa ce bs a Me ot a vals atea ave a's a J iy | Wo SEC SUELTTO EIT (00 | RR ee ere eae ee 3: ( Shell very large; animal with both gill and lung....... Ampullariide. 2. | Selb smaller: ammal-witholbonly: 2.2... ik cae. Viriparide, MOperctlsm-crcwlar nmillispital 2 o.7.. +e dcd ie se ee ap Valvatide. 1 Mpeccimiunn paweispiiatocve 6 78) sas elas aoe oye Sate Se ont ele oe * oe 4. ( Animal with external verge, central tooth with basal denti- ee lee Pe ae PR Were ciaiedd Memes es SS ee Amunicolide. | Animal without verge; no basal denticles on central tooth Pleuroceride. Family AMPULLARIID. Shell large, spiral, globosely turbinate; aperture entire; operculum (in the North American species) corneus, concentric, with a sub-central nucleus. Animal with snout divided into two long, tentacular lobes; tentacles long and filiform; eyes on peduncles on the outer bases of the tentacles; mantle with two cervical lobes, that on the left forming a more or less elongated siphon; genital orifices on the right side in the pallial cavity; the respiratory chamber divided into two parts, the one being a lung and the other contain- ing a large gill; foot large, simple; jaws two; radula with seven rows of teeth, central large, subtrapezoidal, multicuspid, no basal denticles; the laterals and marginal narrow, uni- or bicuspid. Oviparous. Genus AMPULLARIA Lamarck, 1799. Shell dextral, globose, with a green or brown epidermis; spire short, last whorl rounded, inflated ; umbilicate ; aperture entire, angular above, rounded below, lip sirmple. 24 Bryant WALKER Type: Nerita urceus Mull. Example: A. paludosa Say, fig. 70. Radula, fig. 71. Animal, fig. 72. Family VIVIPARID/. Shell moderately large, turbinate, imperforate, or subperforate; whorls convex; aperture entire, subcircular or somewhat angled above; lip simple; operculum convex, concentric, nucleus subcentral, sometimes subspiral. Animal with a long snout, not divided into tentacular lobes; tentacles long and slender, in the male the right one is shorter, truncated and forms a sheath for the verge; eyes on peduncles on the exterior base of the tentacles; mantle with two cervical lobes, of which the right is the larger, forming with the mantle distinct tubular conduits for the ingress and egress of water for respiration; jaws two; radula with the teeth simple or denti- culate, central tooth large, broad, without basal denticles, laterals large, subtrigonal, marginals narrow, elongated. Ovoviviparous. Key to the Genera of Viviparide. | Operculumawholly concentrics: 40-5... curio .0s 2 ae ee 2 its | @perculuna with subcpical mucletis9.) 0-2 eee eee Lioplax. { Reflected apices of the lateralitecth simple22 a. -55-e ee Campeloma. 2) 2.4 . \, Retlectedapices;of the lateral teeth denttcilate =): eee nena 3 { Inner anarein of the operculumesimple-.-ca5 4. ne eel oe Viviparus. 3., Inner margin of the operculum reflected, forming an ele- | wared smareial, TOldee +. 20/5 peas wore 2 cs ane ee Tulotoma. Genus VIVIPARUS Montfort, 1810. Shell dextral, spiral, subconoidal; rather thin, smooth, imperforate or slightly umbilicate; light green or olivaceous, unicolored or banded with brown or tinged with purple; whorls convex, aperture entire, subcircular ; lip simple, acute; columellar and parietal margin not usually thickened ; operculum concentric, inner margin simple, not reflected. Fig. 74 Animal with foot of moderate size, not produced beyond the snout. Teeth of the radula multicuspid. Type: Helix vivipara L. Example: V. intertextus (Say), fig. 73. Radula, fig. 74. Synopsis FrEsH-WATER MoLLusca 25 Genus CAMPELOMA Rafinesque, 1819. Shell dextral, spiral, imperforate, thick and solid; olivaceous green, uni- colored, spire produced; whorls smooth, rounded or shouldered; aperture oval; lip simple, columella and parietal wall usually callously thickened. Operculum concentric, inner margin simple. Fig 78 Animal with the foot large, rather thin, much produced beyond the snout; snout small; teeth of the radula simple or only very minutely crenu- lated. a.—Anus. b-c.— Uterus. br.— Gill. m.—Mantle. Type: C. crassula Raf. Example: C. decisum (Say), fig. 75. Animal, fig. 76. Operculum, fig. 77. Radula: C. integrum (Say), fig. 78. Animal: C. subsolidum (Anth.), fig. 78a. 26 BRYANT WALKER Genus LIOPLAX Troschel, 1856. Shell dextral, spiral, thin, ovate, turreted, imperforate, spire produced; whorls rounded, or carinated; olivaceous green or dark brown; aperture oval subcircular ; lip thin, continuous ; operculum concentric, with a subspira! nucleus. Fig. 79 Fig. 80 Fig. 82 Animal with the foot very large, greatly produced beyond the snout; snout very short. Lingual teeth smooth at their apices. Type: Limnea subcarinata Say, fig. 79. Animal, fig. 80. Operculum, fig. 81. Radula, fig. 82. Genus TULOTOMA Haldeman, 1840. Shell (typically) large, solid, thick, imperforate, obtusely conic, spire elevated; whorls flattened, nodulous, carinated; peristome thin, continuous ; operculum concentric, subtriangular, with the inner margin reflected form- ing an elevated marginal fold. Animal with a moderate foot, not produced beyond the snout; snout small; lingual teeth multicuspid. Type: Paludina magnifica Con., fig. 83. Radula, fig. 84. Operculum, fig. 85. Fig. 84 Family VALVATIDA. Shell small, spiral, dextral, turbinate, or subdiscoidal; whorls rounded or carinated ; aperture entire, circular; lip simple, sharp; operculum orbicular, multispiral, whorls with a thin elevated edge. SyNopsis FrESH-WATER MOLt,USCA 27 Animal dicecious ; tentacles long, slender, cylindrical; eyes sessile on the internal bases of the tentacles; snout long; foot large, bilobed in front; gill external, plumose, protected by a long, slender pallial appendage; verge ex- terior, placed on the right side, at the base of and below the tentacle; jaws two; lingual teeth multicuspid, no basal denticles on the central tooth. Genus VALVATA Miller, 1774. The characters of the genus are those of family. Type: V. cristata Mull. Example: I’. tricarinata (Say), fig. 86. Animal, fig. 87. Radula, fig. 88. Family AMNICOLID/. Shell small, spiral, dextral, conical, imperforate or umbilicated; unt- colored; aperture entire, lip simple, acute; operculum concentric, spiral or subspiral. Animal with a long snout; tentacles long, cylindrical, with the eyes at their outer bases; foot oblong, truncate before, rounded behind; giils inter- nal; verge exserted, placed on the back, some distance behind the right tentacles ; jaws two; central tooth of the radula multicuspid and with one or more basal denticles; laterals hatchet-shaped, multicuspid; marginals slender, multicuspid. Key to subfamilies of Ammnicolide. Wipereclitinr mints piralMen. Sway ee os out RE cae Seep na» Lyogyrine. SB) ese UE Tan Ae OTICEMM RI Crs ete < ee ae fete when tetadee An Bc, eteho aon Bythinine. IB@penculamepamersminal © 0 eta. he vac en Ss 1 ae Re heed a Pbeericinideds oy ta pies mah iy SA aaa aide sss Pomatiopsine. a * LEGS? Sin cl (2 eee nega hh aR RAT RON eR aOR Er 2 eg SEI 2. ( Shell thin, subglobose to elongate; columella not thickened Amnicoline. Shleltel orielfs) eds) .e; ag 6) eicelaliel 9) ©) 02 (6 Ael)\e%\s) 9" e)\ ‘ny! ‘e), e) (0 @) Jef ey,'6) (els) ie\ ol ko).6 soe) 6','b' si is one 4 Shell thick, short, body-whorl very large; columella thick- Weitere TE Pre arte ee res aces DNR cms ees Lithogly phine. Subfamily BYTHININA£ Stimson, 1865. Shell small, spiral, dextral, turbinate, spire produced; operculum cal- careous, concentric. Foot simple; central tooth of the radula with several basal denticles. 28 BRYANT WALKER Genus BYTHINIA Leach, 1818. Shell large for the family, spiral, elevated, subperforate; aperture ova! ; peristome thin, continuous; lip simple, sharp; operculum calcereous, con- centric. Type: Helix tentaculata L,., fig. 89. Animal, fig. go. Radula, fig. 91. Subfamily AMNICOLIN® Gill, 1871. Shell small, spiral, dextral, subglobose to elongate, thin; imperforate or umbilicate; columella and parietal wall not callously thickened; operculum corneous, paucispiral. Foot simple; central tooth of the radula with several denticles. Key to genera of Amnicoline. Shell ‘smooth, pempiery comndedssy. cas. cy) -,<.ctvemese tara ee 2 Shell smooth, periphery aneilate: - ese. tne eee ae Littoridina. i) Shell stronelyecarinated ae). seen) ee ee Pyrgulopsis. shell ;spinase’ . ea camoten aise ete yeas. eats Pemercere Potamopyrgus. Shell lonertudinally rilined' 2 a: ote ope eyes ee ase eer ee Tryonia. Shellsstender spine longo. 9.5.92. 3. a nae Oe ee Paludestrina. | Shell ventricose: spirenistiaily Short ..:.. 2.0 .eseee + oa eee Amnicola. Genus AMNICOLA Gould and Haldeman, 1841. Shell small, oval-conic, rather short, spire subacute; whorls 4-6, convex; aperture oval; peritreme continuous; lip simple, sharp; columella not thick- ened. Operculum thin, corneous, paucispiral. Fig. 92 Animal oviparous; central tooth of the radula multicuspid, with a tooth- shaped process from the middle of the anterior surface, reaching beyond the Synopsis FrESH-WaATER MOLLUSCA 29 base, and with several basal denticles; laterals and marginals multicuspid. Verge short, bifid, with a globular base. Type: Paludina limosa Say, fig. 92. Animal, fig. 93. Operculum, fig. 94. Radula, fig. 95. Fig. 95 Subgenus CINCINNATIA Pilsbry, 18or. ubgenus ilsbry, 189 ve Radula more minute and the denticulations of the cusps finer and sharper. | Type: Paludina cincinnatiensis Anth., fig. 96. Radula, fig. 97. Fig. 96 Fig. 97 Genus PALUDESTRINA d’Orbigny, 1840. Shell similar to Amnicola, but more slender and elongated. Central tooth with but one basal denticle on each side, and with- out the tongue shaped process of Ammicola. Verge bifid. Fig. 98 Fig. 99 Type: Cyclostoma acutum Drap. Example: P. nickliniana (Lea), fig. 98. Radula, fig. 99. Genus TRYONIA Stimpson, 1865. Shell perforate, elongated, turreted, subulate; apex acute; surface longitudinally ribbed or plicated; whorls numerous, shouldered; aperture small, oblique, rhombo-ovate, lip sharp, thin and effuse at the base; peritreme continuous. Type: T. clathrata Stimp., fig. 100. Fig. 100 Genus PYRGULOPSIS Call and Pilsbry, 1886. Shell ovate-conical or turreted, imperforate, whorls having a single, strong carina at the periphery, which may or may not be concealed on the spire; apex acute; whorls 41% to 6; aperture ovate, peristome continuous ; 30 Bryant WALKER central tooth of the radula with but one basal denticle on each side; denticles of the lateral teeth large and angular, those of the laterals small and slender. Fig. 101 Type: Pyrgula nevadensis Stearns, fig. 101. Radula, fig. 102. Genus POTAMOPYRGUS Stimpson, 1865. Shell ovate-conic, imperforate; apex acute; whorls angulated and usually coronated with spines; body-whorl two-thirds the length of the shell; aperture ovate, lip acute. Animal with rostrum of moderate length; tentacles very long, slender, tapering and pointed; eyes on prominent tubercles ; Fig. 103 foot rather short, strongly auriculated in front; central tooth of radula trapezoidal, inferior margin nearly straight; faintly enlobate; basal teeth minute and close to the lateral margin; denticles of the intermediate tooth numerous and of equal size. Type: Melania corolla Gld. Example: P. coronatus (Pfr.), fig. 103. Genus LITTORIDINA Souleyet, 1852. Shell narrowly perforate, subpyramidal, solid, opaque, body-whorl sub- angulate at the periphery; aperture pyriform, acutely angulated above; col- umella thickened, white; peritreme not continuous, lip sharp. Verge very large, with five or six small, digitate appendices. Radula as in Amnicola. Fig. 104 Fig. 105 Type: L. gaudichaudii Soul. Example: L. monroensis (Frfld.), fig. 104. Radula, (L. hatcheri Pils.), fig. 105. Subfamily LITHOGLYPHINA® Fischer, 1885. Shell small, spiral, dextral, spire short, body-whorl large, forming most of the shell; columella usually callously thickened; operculum corneous, subspiral. Foot simple; central tooth of the radula with several basal denticles. SyNopsis FrEsH-WATER MorLusca 31 Key to the genera of Lithoglyphina. fSonell impertorate or narrowly perforate...............0ece8 «6. 2: 4 SUSI AA IICESIRAN7 0] ay SC ca Phere Rie Pa ee en na Se ne ae a ( Peritreme sinuous, lip effuse below, verge winged...... Fluminicola. 2.4 Peritreme continuous in same plane, verge simple............ Gillia. | Peritreme very oblique, lip effuse above, verge bifid. . . .Somatogyrus. { Shell spirally striate, depressed, turbinate, widely umbilicate Ce lapel NSM cede Nee TRO Er Ae AR Ad hea A SE ae ae Cochliopa. Shell smooth, globose-turbinate, umbilicus narrower, but deep. Clappia. Genus COCHLIOPA Stimpson, 1865. Shell depressed-conic; base concave, umbilicus large and deep; aperture oblique; operculum corneous, subspiral. Rostrum of moderate size; tentacles rather long and tap- ering. Teeth of the radula multicuspid, basal denticles on central tooth 2 or 3 on each side. Verge rather elongated, compressed, geniculated and bifid. 2 Fig. 106 Type: Amnicola rowellit ‘Tryon. Example: C. riograndensis P. and F., fig. 106. Genus CLAPPIA Walker, 1909. Shell minute, spiral, dextral, globose-tur- binate, narrowly but deeply umbilicate. Spire short ; body whorl large; whorls round; aper- ture large; lip simple; columellar lip thin, ap- pressed to the body-whorl only at the upper end; operculum paucispiral, nuclear whorls large, slowly and regularly increasing. Fig. 107 Rachidian tooth as in Somatogyrus, in- termediate tooth with a long peduncle and a strong tooth projecting from the infero-anterior angle, laterals multicuspid. Fig. 109 Type: C. clappii Walker, fig. 107. Operculum, fig. 108. Radula, fig. 1009. 32 Bryant WALKER Genus FLUMINICOLA Stimpson, 1865. Shell spiral, dextral, obliquely ovate, thick, solid, smooth, imperfor- ate; spire moderate, ob- Lise; “aperture ovate; columella flattened, cal- Fig. 111 loused; lip effuse and projecting anteriorly so that the peritreme is not continuously in the same plane ; operculum corneous, subspiral. Rostrum rather large; tentacles tapering, foot broad; central tooth of the radula with several basal denticles on each side; outer lateral teeth with a smaller number of denticles than the inner. Verge large, compressed, with a broad semicircular wing on the left side. Type: Paludina nuttalliana Lea, fig. 110. Radula, fig. 111. Fig. 110 Genus SOMATOGYRUS Gill, 1863. Shell dextral, spiral, usually rather thick and solid, smooth, imperforate or narrowly perforate; spire usually short; apical whorl spirally punctate or lirate ; body-whorl large, more or less inflated; aperture very oblique; lip sharp, projecting above; columella callously thickened. Operculum corne- ous, subspiral, nuclear whorls small, rapidly increasing. Fig. 112 Hige 113 Fig. 114 Rostrum rather broad, flat and square-cut; tentacles rather short and flattened ; teeth of the radula multicuspid, basal denticles on the central tooth 3-4 on each side. Verge broad, compressed and bifid. Type: Ammnicola depressa Tryon, fig. 112. Radula, fig. 113. Operculum, fig. 114. Genus GILLIA Stimpson, 1865. Shell spiral, dextral, not very thick, smooth, imperfor- ate; spire short, obtuse, body whorl large, inflated ; aperture large, oblique; peritreme con- tinuous on the same plane, lip thin, sharp; columella very slightly thickened ; operculum corneous, subspiral. Rostrum broad, subtruncate, foot oblong, rounded behind and auriculate Pig. 115 Synopsis FrEsH-WaATER MOoLLusca 33 in front; tentacles long, slender and pointed; teeth of the radula multicus- pid; central tooth with two basal denticles on each side. Verge small, simple, lunate. Type: Melania altilis Lea, fig. 115. Radula, fig. 116. Subfamily LYOGYRINZ® Pilsbry, 1916. Shell minute, conical or subdepressed. Operculum circular, multispiral. Key to the genera of Lyogyrine. Bre mamnieolionie mace ho eee lee oe Os Mena oe os oe ees Lyogyrus. STR STCATAU EV Ee olig 10s aeeae ee eee (s. g.) Hauffenia. Genus LYOGYRUS Gill, 1863. ‘ Shell very small, spiral, dextral, smooth, umbilicate ; globose- & turbinate or elongate-ovate; aperture nearly circular; peritreme Z continuous, frequently quite separated from the body-whorl. Operculum corneous, circular, multispiral. Rostrum bilobed in front, half as long as the tentacles, which are rather stout; foot auriculated in front; gill external. Dentition as in Paludestrina, basal denticles on rachidian tooth two on each side. Fig. 117 Type: Valvata pupoidea Gld., fig. 117. Genus HORATIA Bourguignat, 1887. Shell very small, amnicoliform, thick-shelled, umbilicate, almost smooth ; whorls only 3-4, convex, rapidly increasing, the last large, rounded; suture impressed; aperture very oblique, rounded; lip adherent, straight, sharp; columella thickened, incurved. Operculum corneous, transparent, purple-red, with 3-4 slowly increasing spi- rals and an almost central nucleus. Type: (first species) Horatia klecakiana Bet. Subgenus HAUFFENIA Pollonera, 1808. Shell minute, rather thin, valvateeform, widely umbilicate. Type: (first species) Horatia tellini Poll. eas we Example: Horatia micra (P. and F.), fig. 118. Fig. 118 34 Bryant WALKER Subfamily POMATIOPSINA‘ Stimpson, 1865. Foot divided by a transverse sulcus at about its anterior third. Verge simple. Rachidian tooth with one basal denticle on each side, denticles of the lateral and marginal teeth fewer and proportionately larger than in the other subfamilies. Only one genus. Genus POMATIOPSIS Tryon, 1862. Shell dextral, spiral, thin, smooth, long, turreted, umbilicated; aperture somewhat expanded; lip simple or slightly reflected; operculum corneous, subspiral. A OA Fig. 119 Fig. 121 Animal not as long as the shell, rostrum large, longer than the tentacles, which are short and subulate. Verge very large, simple, convoluted, outer margin rounded and smooth, inner margin sharp and wrinkled. Type: Cyclostoma lapidaria Say, fig. 119. Animal, fig. 120. Radula, fig. 121. Family PLEUROCERID. Animal oviparous, edge of mantle smooth; eyes on the external bases of the tentacles; no verge. ‘Rachidian tooth large, broader than long, rounded below, multicuspid; laterals subrhomboidal, multicuspid; marginals narrow, multicuspid. Shell dextral, spiral, thick and solid, globose or elongated ; aperture entire or more or less canaliculated below, operculum corneous, subspiral. Key to the genera of Pleuroceride. Apenbure Cana liculate We low 27... 9... «cc ue See cee ee 2 i, Qupenture angulated below, not canaliculate <.2....05. a0 oeeeeeee A. eo Mpervineventize, rounded belows. ....0- 1-2 ee 5 ree eee dnculosa. | shell tusitorm-canal long t2.'s. ia. 3 vee o ee ee To. 2: | slrelicconical, ‘canal short.¢ ¢ 4325 2. oath ene eee eee z Spire short; parietal wall callously thickened above and be- | Spire elongated ; no callus thickening on parietal wall... .Plewrocera. | ONE tre ac WOs oe Foe on ek aDt Deca silat aa S215 - Lithasia. | Spire short; parietal wall callously thickened above..... Eurycelon. | Men Ure ‘Entire AbOVE ~.. Tn aatyter ee eee eee Goniobasis. Aperture withvasutural shite 2262 aha eet Gyrotoma. Synopsis FresH-WatER Mors,usca 35 Genus IO: Lea, 1831. Shell large, spiral, dextral, imperforate, fusiform, smooth, tuberculate or spinose; spire elevated, base of aperture prolonged in a long canal; columella round, smooth and concave; lip thin; operculum corneous, sub- \\ WY \“ ) hi spiral. Animal as in the family. Wy Type: Fusus fuvialis Say. Example: J. spinosa Lea, fig. 122. Radula, fig. 123. Genus LITHASIA Haldeman, 18,0. Shell of medium size, spire dextral, imperforate, glo- bose conic, smooth or tuberculate ; thick and solid, spire ele- vated, obtusely conic; aperture large, rhomboidal; columella smooth, callously thickened above and below; base of the aperture shortly channelled below.. Animal as in the fam- ily. Type: L. geniculata Hald., fig. 124. Operculum, L. obovata (Say), fig. 125. Section ANGITREMA Haldeman, 1841. Shell similar to Lithasia but with the basal canal more produced. Type: Melania armigera Say, fig. 120. 36 BryANtT WALKER Genus EURYCAELON Lea, 1864. Shell large, obovate, thick, solid; spire short; body-whorl | large; columella callously thickened above, incurved below and subtruncate. Type: Anculosa anthonyi Budd, fig. 127. Genus PLEUROCERA Rafinesque, 1818. Shell usually lengthened, conic or org cerithiform, spiral, dextral, imperfor- z ate; smooth, tuberculate, spirally stri- ate or carinate; aperture moderate, 1 a qa subrhomboidal, prolonged into a short wy canal below; columella smooth, twist- Fig. 128 Fig. 129 ed, not callously thickened; lip sim- ple, sharp, sinuous, somewhat expanded. Type: P. acuta Raf., fig. 128. Radula, fig. 129. Section STREPHOBASIS Lea, 1861. Shell smooth, spire rather short, obtusely conical, body-whorl subcylindrical; aperture subquadrate ; columella thickened below, twisted and drawn back, ‘base subcanaliculate; lip acute, very sinuous. | Type: Melania plena Anth., fig. 130. Genus GONIOBASIS Lea, 1862. Shell medium size, dextral, spiral, imperforate; smooth, lon- gitudinally plicate, transversely striate or tuberculate; thick, solid, ovate-conic to elongate turreted; aperture subrhomboidal, subangular at the base but not canaliculate; columella smooth, not twisted; lip simple, acute. Fig. 132 Type: G. osculata Lea, fig. 131. Radula, G. depygis (Say), fig. 132. Synopsis FrRESH-WaATER MOLLUSCA 37 Genus GYROTOMA Shuttleworth, 1845. Shell of moderate size, spiral, dextral, imperforate, smooth or spirally striate, rather thick and solid, conical or globosely ovate; aperture oval or elliptical, with a sutural slit or fissure above, entire below; lip thin and sharp. Fig. 133 Type: G. ovoidea Shutt., fig. 133. Radula, fig. 134. Fig. 134 Genus ANCULOSA Say, 1821. Shell of moderate size, dextral, imperforate (except in one species), smooth, tuberculate, spirally striate, sulcate or carinate, thick, solid, sub- globose with a very short spire or thinner and conical; aperture oval or sub- circular, entire, rounded below; columella callously thickened; lip simple, acute. Type: Melania prerosa Say, fig. 135. Radula, fig. 136. aK s AeA Fig. 136 Order ASPIDOBRANCHIA. Suborder RHIPIDOGLOSSA. The Rhipidoglosse differ radically from the Tznioglosse in the charac- ter of the lingual dentition. The radula has many rows of teeth, consisting of a central, 2-5 laterals and numerous marginals arranged like the sticks of a fan. Family NERITID. Head large, rostrum divided and lobed in front; tentacles long and slender ; eyes carried on peduncles placed at the external base of the tenta- cles; foot large, truncate before and obtuse behind, sides simple; gill large, triangular, pointed, free at its extremity; branchial and excretory orifices 38 BRYANT WALKER on the right side. Radula with a very small central tooth, 2 to 5 lateral teeth of varying size and shape and numerous spatulate marginals. Shell imperforate, globose, spire short; internal divisions of the shell absorbed; aperture semi-ovate, entire, columellar region expanded and flat- tened, usually thickened; lip acute; operculum subspiral. Key to the genera of Neritide. 1. Operculum calcareous, edge with projecting processes (apophyses) articulating with the columella..........Neritina. 2. ‘Operculum! cornmeous, without apophyseste .5.-0 55-45 - oe Lepyrium. Genus NERITINA Lamarck, 1809. Shell dextral, spiral, thick and solid, subglobose; spire short; surface smooth (in American species) ; aperture semi-circular, columella flattened, straight, smooth or finely denticulate ; lip acute, inner surface smooth; oper- culum calcareous, semi-circular, paucispiral, nucleus excentric; with two apophyses, the upper shorter, sometimes dilated and crested; the lateral in the form of an arched rib. ; o\ Jp Fig. 138 Animal as in the family. Radula with the central tooth small, subquad- rangular, cusp smooth; first lateral large, second and third small, fourth very large, prolonged below and with the cusp semicircular and denticulate, laterals numerous, spatulate. Type: N. perversa Gmel. Example: N. reclivata Say, fig. 137. Radula, fig. 138. Genus LEPYRIUM Dall, 1806. Shell small, dextral, spiral, corneous, thin, semi-transpar- ent; spire very small and depressed, body-whorl large; aper- ture large, semi-circular; columella concavely flattened, cal- loused, straight, smooth; lip thin, acute operculum thin, corne- ous, paucispiral without apophyses. Radula with a wide rachidian tooth with a finely denticu- lated cusp; laterals two, the inner small and oblique, the other large with the cusp finely denticulate; laterals spatulate, numerous. Animal otherwise unknown. Fig. 139 Type: Neritina showalteri Lea, fig. 139. Synopsis I*resu-WatTerR MOoLLusca 39 Class LAMELLIBRANCHIA. Order FULAMELLIBRANCHIA. Suborder SUBMYTILACEA. Key to the families of Submytilacea. ee canmenere teil (ens kek es sre odie ie vars howe pinta ine, ost Pa J i ME RNC Ti AMECt Ta ent etn NS ass iy Gina bea WO aR aE A ole Mig Bhd ein cath 5 ( Hinge with cardinal, anterior and posterior lateral teeth......... 4. eninge withucardinal teeth only a. .) seo. a teen es Cyrenellide. | Hinge with lateral teeth only (no true cardinals) or edentate..... By ( Gills with distinct, interlamellar septa, parallel with the gill PRE lei niis eer rene ea hai) 2 oc x Geen a eae nS Nope Unionide. | Gills either without distinct, interlamellar septa or, when | present, oblique to the gill-filaments .............Margaritanide. erallireu nines sttanplets = he ninth tels eee Vera Wt cdsans tae Spheriide. 55 4 . . . . pe ligalsline. sintiate’: . 4.05 peanisiee po ease a + ie ace Cyrenide. ieimieewith cardinal: and lateral teeth. ts: fe. s.64 4. 2+ Rangide. Ba ttinee without distinet teeth 0.2.50. 0. saloon eee. Dreissensiide. Family MARGARITANID. “Diaphragm incomplete, formed by the gills; posteriorly the outer lam- ina of the outer gills not connected with the mantle for a considerable dis- tance; anterior end of the inner gills separated from the palpi by a gap; branchial and anal openings ill-defined, and the latter not closed above; no super-anal developed; gills without water-tubes and with scattered inter- lamellar connections, which in certain places form irregular rows or with continuous septa which run obliquely forwards; marsupium formed by all four gills; larva a small semicircular glochidium, without distinct hooks ; shell elongated; sculpture of the beak concentric; hinge-teeth imperfect; epidermis blackish.” (Ortmann. ) Genus MARGARITANA Schumacher, 1817. Shell elongated, usually arcuate, rounded in front, almost lacking a pos- terior ridge; beaks rather low, sculpture consisting of a few coarse, par- allel ridges which follow the growth lines; epidermis concentrically &triate, brownish or blackish; hinge-teeth generally imperfect or not fully devel- oped. two more or less perfect pseudocardinals in the left valve and one in the right, often reduced to mere tubercles; laterals short, usually imperfect or wholly wanting: cavity of the beaks rather shallow. 40 BryANt WALKER Key to the subgenera of Margaritana. Gills with scattered interlamellar connections forming irregular rows running obliquely forwards............../ Margaritana s. s. Gills with continuous septa running obliquely forwards. .Cumberlandia. Subgenus MARGARITANA s. s. Shell as in the genus. Th Hy!) f ZZ z Vij AZ, } Animal as in the family, but having the gills without water-tubes and with scattered interlamellar connections which in certain places form irreg- ular rows, running obliquely forwards. Fig. 141* Type: Mya margaritifera L., fig. 140. Animal, fig. 141. * The following lettering applies to all the figures of the animals of the Margar- itanide and Unionide except as otherwise stated: s an.—anal opening. p.—foot. br.—branchial opening. pp.—papille on margin of mantle. f—flaps of margin of mantle. sa.—supra-anal opening. i—inner. gill. mp.—marsupium. o.—outer gill. Synopsis FrEsH-WaTER MOoLLusca 41 Subgenus CUMBERLANDIA Ortmann, 1912. Shell as in the genus. Animal as in the family, but having the gills with incomplete water- tubes and with continuous septa, which run obliquely forwards. Fig. 143 Type: Unio monodonta Say, fig. 142. Animal, fig. 143. Family UNIONID#. “Diaphragm complete, formed by the gills; posteriorly the outer lamina of the outer gill connected with the mantle to its posterior end; anterior end of the inner gills separated from the palpi by a gap; branchial and anal openings sharply separated from one another by the diaphragm; anal open- ings very rarely not closed above and without supra-anal, generally closed and with a supra-anal opening (which very rarely may be obliterated) ; gills with water-tubes and distinct, interlamellar septa, running parallel to the filaments. Marsupium in all four gills or only in the outer gills; larva a glochidium. Shell of very variable shape; sculpture of the beak more or less reduced, of various types, but originally of the concentric or zig-zag pattern; hinge teeth perfect or imperfect; epidermis plain or with color- markings.” (Ortmann. ) 42 Bryant WALKER Key to the subfamilies of Unionide. Water-tubes simple in theverayid* female: 2.) 1). 2 2 eee z Water-tubes in the gravid female divided ‘into three tubes, of which only the centre one is used as an ovisac..... Anodontine. marstupium always sharp and not distending........... Unionine. Male and female shells usually different; edge of the gravid marsupium distending and bulging out beyond the original { J [ Male and female shells usually alike; edge of the gravid 2 (Sedge otthetgill: >... .. Wibpe tem oreer teen eee Lampsiline. Subfamily UNIONINA® (Swainson, 1840) Ortmann, 1910. “Inner lamina of the inner gills generally free from the abdominal sac (sometimes, in extralimital forms, connected) ; supra-anal opening some- times not separated from the anal, normally present, the closed part rather short; branchial opening well-defined; no papillae nor flaps on the edge of mantle in front; marsupium formed by all four gills or by the outer gills only; edge of marsupium always sharp and not distending; water-tubes not divided in the gravid female; glochidium semielliptic or semicircular, with- out spines; shell generally heavy and solid, rounded to elongated, mostly with dull-colored epidermis; sculpture of the beak generally rather indis- tinct, concentric or pustulous or with indications of double loops or zig- zag bars; hinge always complete, with rather strong teeth; generally no difference of sex shown in the shell.”’ (Ortmann. ) Key to the genera of Unioninea. i Il fourweills: serving’ as niaestpia 1-55. see 2: ste |, Outercills only Senyine as manrsupia;()--..m8-. sn 6. (. Mailetand'fematershells alilgeten sc sce. eee cere ee a 2. H Malevandetemalershells differents oe 5: 02251. eee Tritogonia. { Hinge with perfect pseudocardinals and laterals................ 4. 3.4 | “Hinge teeth rudimentary ormwantine.....:.......6.4 00806 Gonidea. ("Surface plicate 5... <0... . pees oa We ee hae er 5. Ar ADULTACe PUStIIOSE |v... SERRE eee a eee eee Ouadrula. (oStiréace smacthy: 14 4... aetetaeiae acon ea ee eee ee Fusconaia. Beaks sculptured with coarse, concentric or somewhat dou- ble-looped ridges, which do not extend over the surface. Amblema. 5: Beaks sculptured with strong, zig-zag ridges extending over the supper surface... -... deren oer rare ee aera Megalonaias. Synopsis FresH-WatER Mort,usca 43 Sat eemIMeGl GUNG amet ya Os Ae es tisk. fous atl Skee Sar Gi a Uma CersMOOtI OG SPY r asc... ae eas te dn BUREN te in &. PeNcresmeen amine tom tse SSeS hee ee nad Rotundaria. 4 irs Piacheawhite Ok tinged withy pink’. 1: gee. o Se nsyie dens Plethobasus. Flingé' with perfect psewdocardinals and Jaterals...........s..4% 9. “| Ebimces teethcimpertect, vestioial. -... atectacs ss a scale Lastena. Snell-short, rounded,,.quadrate or oblique... ...<5.0).. osc Ses lee of. Sheu (ustallymcioneite and straightwes <2 00>. scick sis sada DD. ( Beak sculpture distinct, subconcentric, rounded upon the | SS SUTRAS Oem tenes Citic rhe hs. RMERROES ve th aie tw. ahe. 3,» 6 Lexingtonia. 10 Beak sculpture coarser, inclined to be more or less double- 115. DypLSCG ae ae es tee eco coe Re ee Pleurobema. { Beak sculpture running parallel with the growth-lines and TI Pied ot. Une ypOsteMOle SLOP. -... meade soe Se geist ate. Elliptio. | heakeseulpture conceutime, rounded: behind: : 2 ..¢0740 2 6. Uniomerus. Genus QUADRULA (Rafinesque, 1820) Agassiz. Shell triangular, quadrate or rhomboid; solid, inflated with rather coarse prominent beaks, which are generally sculptured with a few coarse, irreg- ular, subparallel ridges that are inflated where they cross the posterior ridge; posterior ridge ordinarily well developed; disk sculptured or smooth; epi- dermis usually dull-colored, dark and rayless or feebly rayed; hinge plate heavy, wide, flattened; pseudocardinals solid, direct, ragged; laterals double in the left and single in the right valve; cavity of the beaks deep and com- pressed. Marsupium occupying all four of the gills throughout, the whole smooth and pad-like. Key to the sections of Quadrula. Posterior slope with a radial furrow above the posterior ridge MBLC eo SE SE GEN NCES 90 aS Ouadrula s. s. No radial furrow above posterior ridge.................. Theliderma. 44 BRYANT WALKER section QUADRULA s. s. Shell quadrate or rhomboid; surface pustulous, with a high, rounded or sharp posterior ridge, above which on the posterior slope is a decided Fig. 145 radial furrow; umbonal region high; epidermis shining, usually painted with a beautiful pattern of triangular spots or chevron-shaped lines.. Type: Unio cylindricus Say, fig. 144. Animal, Q. metanevra Raf., fig. 145. Section THELIDERMA (Swainson, 1840) Simpson. Shell rounded, quadrate to rhomboid, solid, pustulous; beaks rather prominent, sculpture consisting of a few, rather coarse, subparallel ridges; anterior end rounded, base often arcuate, posterior end truncate, high and angled behind the ligament, epidermis rarely rayed, never as in Quadruia ss. Type: Unio lachrymosus Lea, fig. 146. Synopsis FresH-WaATER Moriusca 45 Genus TRITOGONIA Agassiz, 1852. Shell solid, elongate, rhomboid, having a strong, irregular posterior ridge, obliquely truncated behind in the male, in the female this region is somewhat compressed and expanded into a broad wing; base curved; whole surface, except the rounded wing of the females, covered with pustules; beaks rather low, incurved and turned forward over the well developed lunule; beak sculpture strong, consisting of irregular, subparallel ridges which are curved upwards behind and fine radiating ridges in front of and behind them; epidermis dark olive; hinge plate rather narrow; pseudo- Fig. 147 cardinals strong, ragged; laterals long and straight, near to the pseudo- cardinals; cavity of beaks rather deep and compressed; female shell more compressed than that of the male. Marsupium occupying all four gills. Type: Unio tuberculatus Bar., fig. 147. 46 BRYANT WALKER Genus MEGALONAIAS Utterback, 1915. Shell large, heavy, obovate or rhomboid, alate post-dorsally, disk oblique- ly folded; beaks sculptured with coarse, double-looped corrugations, which extend over the upper surface of the disk as nodulous plications; epidermis Fig. 148 dark-brown or blackish; beak cavities narrow and deep; anterior muscle scars deep and filled with a nacreous deposit, posterior scars large and in- distinct. Fig. 149* Type: Unio heros Say, fig. 148. Animal, fig. 149. * The following lJettering applies to figs. 149, 165 and 2o1: A.—Anal opening. F.—Foot. P.—Palp. Avy.—Antero-ventral margin. J.—Inner gill. S.—Supra-anal opening. B.—Branchial opening. M.—Marsupium. Synopsis FresH-WATER MOLLusca 47 Genus AMBLEMA Rafinesque, 1819. Shell more or less alate; beaks prominent, sculptured with coarse, con- centric or somewhat double-looped ridges which do not extend over the surface of the shell; surface of the valves usually sculptured with oblique folds; posterior slope generally having small radial plications, which curve Fig. 150 upwards behind; epidermis brownish or blackish; anterior muscle scars large, distinct, very shallow, the anterior edge smooth, the rest apparently filled with roughened shelly matter; posterior scars large, shallow, indis- tinct; escutcheon large and dark. Marsupium occupying all four gills. Type: A. costata Raf. Example: A. undulata (Bar.), fig. 150. Animal, A. trapezoides (Lea), Cea eae 48 Bryant WALKER Genus FUSCONAIA Simpson, 1900. Shell round, rhomboid, triangular or short elliptical, with a moderate posterior ridge; beaks high and full, curved inward and forward, sculp- tured with a few coarse, parallel ridges, which curve upward behind; ep1- dermis dark; surface not sculptured ; hinge plate of moderate width; pseudo- cardinals strong; nacre white, salmon or purple. All four gills marsupial. Type: Unio trigonus Lea, fig. 152. Animal, (F. rubiginosa (Lea) ), fig. 153. Genus ROTUNDARIA (Rafinesque, 1820) Simpson. : =>. = SA —— —y = =z Shell rounded; slightly truncated above in front; posterior ridge low beaks prominent, curved inward and forward over a strongly marked SyNopsis FresH-WaAtTER MoLzusca 49 ule; beak sculpture consisting of numerous, fine, irregular, broken, some- what concentric corrugations ; posterior three-fifths of the shell tuberculate ; epidermis brown; nacre purple. Only the outer gills serving as marsupiumn. o=mp Fig. 155 Type: Obliquaria (Rotundaria) tuberculata Raf., fig. 154. Animal, fig. 155. Genus PLETHOBASUS Simpson, rgoo. Shell large, irregularly oval, inflated, solid, somewhat suddenly swollen at the posterior base; posterior ridge low and rounded; beaks rather high, near the anterior end, having a few strong ridges, which are curved up- Fig. 156 ~ wards behind; a row of low, irregular tubercles extends from near the beaks to post-basal part of the valves; epidermis tawny yellow to dark brown; hinge plate solid, not flattened; pseudo-cardinals triangular, rough; 50 BryANt WALKER cavity of the beaks not deep; front part of the shell very heavy, thinner behind. Outer gills only serving as marsupium. o= mp Fig. 157 Fig. 159 Shell solid, triangular to rhomboid, usually with a prominent umbonal region; beaks at or near the anterior end of the shell, incurved and pointed forward over a small, but well developed lunule; beak sculpture coarse, con- sisting of a few, often broken, ridges, which curve upward posteriorly ; pos- terior ridge present, but low and rounded; epidermis showing the rest periods plainly, tawny to olive, often. ornamented with rays which show a tendency to break into square spots; hinge rather strong, plate generally narrow ; pseudocardinals double in both valves. Cavity of the beaks shal- low. Outer gills only serving as marsupium. Type: Unio clava Lam., fig. 158. Animal, fig. 159. Synopsis FrESH-WATER MOLLUSCA 51 Genus- LEXINGTONIA Ortmann, 1914. “Shell subauadrate or subtrapezoidal, with slightly elevated beaks and well developed hinge teeth. Beaks not much anterior. Outer surface without sculpture. Epidermis lighter or dark- er brownish, with rather indistinct rays, which are narrower or wider and do not break up into blotches. Beak sculpture distinct, consisting of rather numerous (six to eight), rather crowded, subconcentric ridges, which Fig. 160 form an indistinct, rounded angle up- on the posterior ridge and are in front of this somewhat wavy and corrugated, but without showing any distinct zigzag pattern. Towards the disk, they disappear. Nacre whitish or pinkish.” Animal having only the outer gills marsupial and subcylindrical, red placente. Type: Unio subplanus Con., fig. 160. Genus ELLIPTIO Rafinesque, 1819. Shell ineguilateral, ovate to elongated, rounded in front and pointed or biangulate behind, with a more or less developed posterior ridge, often be- coming slightly arcuate when old; beaks only moderately full, generally sculptured with coarse ridges, which run parallel with the growth lines or are somewhat doubly looped, sometimes broken and showing fine radiating lines behind; surface smooth, slightly concentrically ridged or pustulous: epidermis generally rather dull colored, rayless or fully rayed; hinge-plate narrow, two pseudocardinals and two laterals in the left valve and one pseudocardinal and one lateral in the right, with rarely a vestige of a sec- ond lateral; cavity of the beaks not deep or compressed. Marsupium occu- pying the whole length of the outer gills only, forming a thick, smooth pad when filled with young. Key to the sections of Elliptio. SSS offs Cae A a Oe ta roy oc ee Canthyria. Bnei smootinor feeply corrugated... ..... 008. 002 Ne. Elliptio s. s. 52 BryANt WALKER Section ELLIPTIO’s. s. « Shell elongated, rhomboid or oval, usually more or less biangulate be- hind; beak sculpture consisting of a few rather strong ridges, which are nearly parallel to the growth lines or slightly doubly looped; the surface smooth or feebly corrugated. Type: Unio crassidens Lam., fig. 161. Animal, fig. 162. TD | ai Section CANTHYRIA Swainson, 1840. He. Shell inflated, suboval, spinose, . with a high, rather sharp posterior ridge, above which it is somewhat truncated ; beaks rather compress- ed; epidermis smooth and shining, -)\ variegated with angular blotches; / tre; pseudocardinals rather com- pressed; laterals short, remote; beak cavities rather deep. Fig. 163 Type: Unio spinosus Lea, fig. 163. SyNopsis FrESH-WATER MOoLLusca 53 Genus UNIOMERUS Conrad, 1853. Shell trapezoidal, with a rounded posterior ridge and pointed or feebly biangulate behind; beaks not prominent, sculptured with curved, rather strong, concentric ridges; epidermis generally rayless; pseudocardinals us- ually compressed, laterals delicate, slightly curved. Outer gills only marsupial. Fig. 165 Type: Unio tetralasmus Say, fig. 164. Animal, fig. 165. Genus LASTENA Rafinesque, 1820. Shell elongated, subsolid, inequilateral, generally wider in front, rounded and truncate at the anterior base, pointed at the post-basal region, and hav- ing a low posterior ridge, with one or more secondary ridges above it; beaks low, sculptured with a few coarse, irregular, longitudinal folds; epi- 54 BryANt WALKER dermis shining, often rayed; a single imperfect tooth in each valve and sometimes vestiges of laterals; nacre purplish shading to blue at the edge; pallial line radially ridged. Animal having the foot very large, as long as the shell when extended, of a subcylindrical, compressed shape, with a distal swelling. The middle portion of the outer gills only marsupial. Glochidia semicircular, slightly oblique, inequivalve, without points or hooks. Type: Anodonta (Lastena) lata Raf., fig. 166. Genus GONIDEA Conrad, 1857. Shell elongated, subtriangular, much narrowed in front, wide behind, inflated, subsolid, usually with a high, sharp posterior ridge; beaks rather sharp but not high, the sculpture consisting of a few, strong, concentric bars: epidermis rayless; hinge with a rudimentary pseudocardirtal and lat- eral in each valve, though these are sometimes wanting; pallial line with a trace of a sinus behind; nacre lurid to purplish. Female having all four gills marsupial; with well developed septa, run- ning parallel to the gill filaments and forming water tubes; septa not all con- tinuous, but are often interupted, chiefly so toward the proximal (basal) part of the gill and, towards the edge, frequently shorter septa are inter- calated. Glochidia subovate or nearly subcircular, without hooks. Type: Anodonta angulata Lea, fig. 167. Subfamily ANODONTINA£ Ortmann, 1912. “Inner lamina of inner gill free from the abdominal sac, or more or less connected with it, rarely entirely connected; supra-anal opening well sep- arated from the anal, sometimes the connection of the mantle separating it from the anal is very long and the supra-anal is quite short; branchial open- ing well defined, no papille or flaps in front of it on the edge of the mantle; marsupium formed by the outer gills in their length, distending, when Synopsis FRESH-WATER MoLLusca 55 charged. and the thickened tissue at the edge capable of stretching out in a direction transverse to the gill, but not beyond the edge (or only slightly so); water-tubes in the gravid female divided longitudinally into three tubes, with only the one in the middle used as an ovisac, and closed at the base of the gill; glochidium semicircular or triangular, with a spine (hook), in the middle of the ventral margin of each valve; shell generally very heavy, often thin, never round, but more or less elongated; color of epi- dermis generally bright and with color markings; sculpture of the beak double-looped or concentric, in the latter case often extremely heavy; hinge rarely complete and, if so, of peculiar structure; generally there is a dis- tinct tendency toward the reduction of the hinge-teeth, and often they are completely absent; sexual differences in the shell very rarely present.” (Ortmann. ) Key to the genera of Anodontina. BR ealercomip pune CONCEMINIC: v 5c 59.06 2. os Gee gees @ cy ole os epoca at 2 T.4 Peieaia-scuipture double tooped 24.4.0... /yete Massel «eee A. ei colencemlpeirem tries cies ete Sieve pence hoz Anodontoides. J Pas Obes rec piieescontse sts nist, sac nen es eae ete 2 ipieseudocarumals well developed ....:.... . Wee eee ey. 32-02 Alasmidonta. [iP eserdocarcimdis. ridimentary so. . 2. cna aes tines 2 +s Strophitus. fullinge wholly edentuldtis 22.470. .\02. Bio ae wees. 3 Anodonta. 4.4 felimecrtecthy more or less developed =. .,..hoeiey a. an. 2 tens e+ 5: ( Beak sculpture tubercular; surface tubercular or folded......... 6. 5.4 Beak sculpture not tubercular; surface smooth except on INIT SIGN etn Neen sss So eee eee ee { Beak sculpture strong and continuous with tubercular sur- | CERES AR ge DR ear ge ke Aone aan ee ee Arcidens. 6. Beak sculpture poorly developed and not continuous with enetir eenec EMEC r ns cies). apa et es ns eee os Arkansia. Beak sculpture double-looped; pseudo-cardinals fully de- VLD DIE EP Ray apne ae ces aa OC 5 eo Lasmigona. Wc Beak sculpture open behind; a single pseudo-cardinal in SERIE | a ee ean, SOE Pea Simpsoniconcha. 56 Bryant WALKER Genus STROPHITUS Rafinesque, 1820. Shell elliptical to rhomboid, inflated, subsolid, pointed or biangulate be- hind, with a low posterior ridge, which is sometimes double; beaks full, sculpture consisting of a few, strong, concentric ridges, which curve sharply Fig. 168 upwards behind; epidermis rayed or rayless, shining; hinge line incurved in front of the beaks; teeth rudimentary, a vestigial, compressed tooth in each valve, and sometimes a secondary tooth; laterals rarely present. Mar- supium occupying the whole of the outer gills, consisting of short, hort- ‘zontal ovisacs, which run directly across the gills. Type: Anodonta undulata Say. Example: S. edentulus (Say), fig. 168. Animal, fig. 169. Genus ANODONTA Lamarck, 1799. Shell elliptical, thin, inflated, often slightly winged posteriorly; beak sculpture consisting of rather numerous, more or less parallel ridges, usually somewhat doubly looped and becoming slightly nodulous on the loops; sur- face generally smooth, shining; hinge edentulous, reduced to a mere line, regularly curved; nacre dull. Synopsis FrEsH-WaATER MOoLLusca 57 Marsupium occupying the whole outer gills, when filled forming a smooth, very thick, liver-colored pad. Type: Mytilus cygneus L,. Example: A. cataracta Say, fig. 170. Animal, fig. 171. \\\ \ \’ \ val al Kc \\\\ \ \ t AN \\ Fig. 171 Genus ANODONTOIDES Simpson, 18908. Shell elliptical, inflated, thin, with a faint posterior ridge, sometimes constricted at the centre of the base; beaks rather full, with a few, not very coarse, subparallel, concentric ridges, which are curved up rather suddenly, behind and back of these are fine radiating ridges; epidermis smooth and < < — = === = : sSSS== shining, often rayed; hinge line slightly incurved in front of the beaks, edentulous or bearing the merest rudiments of teeth; nacre bluish-white. Marsupium occupying the outer gills only. Type: Anodonta ferussaciana Lea, fig. 172. Genus ARCIDENS Simpson, 1900. Shell subsolid, inflated, subrhomboidal, with full high beaks; beak sculp- ture very strong, consisting of irregular corrugations, which fall into two loops, at the base of which the ridges are swollen into knobs that continue out in two radiating rows on to the disk of the shell; in front of and be- hind the beaks are many fine, radial wrinkles, the posterior ones being zig- 58 BRYANT WALKER zagged; surface of the shell covered with oblique folds and wrinkles; epi- dermis dark olive, shining; left valve with two elongated, compressed pseu- docardinals, the posterior under the beak and curved upwards, cutting off the hinge plate in the right valve, which has a single, compressed pseudo- cardinal in front; laterals numerous, short, blurred; nacre white. Mar- supium occupying the outer gills. Fig. 174 Type: Alasmodonta confragosa Say, fig. 173. Animal, fig. 174. Genus ARKANSIA Ortmann and Walker, 1912. Shell moderately thick, subrotund to subovate or subrhomboidal, in- flated, with full beaks. Disk sculptured with irregular, oblique folds, which are sometimes indistinct. Beak sculpture poorly developed. consisting of two to three double-looped bars, the loops slightly swollen or tubercular, dis- Synopsis FRESH-WATER MoLLusca 59 appearing toward the disk and not continuous with the sculpture of the latter. Hinge well developed, with strong pseudocardinals, a very strong Le ih y RAS WISE ae SS EAN TMA SS \\ N\A NN interdental projection in the left valve and well developed, strong, but rather short laterals. Outer gills only marsupial. Type: Arkansia wheeleri O. and W., fig. 175. Genus LASMIGONA Rafinesque, 1831. Shell elliptic-rhomboid, compressed; beaks low, their sculpture consist- ing of strong bars; one pseudocardinal in the right valve and two in the left, the hinder somewhat /\-shaped, cutting off the hinge-plate in the right valve; laterals generally imperfect. Marsupium thick, padlike, filling the outer gills. Key to the subgenera of Lasmigona. Shell corrueated: on posterior slope. 2.422.240. 62 se Lasmigona s. S. SHAFeT Gv har@ tls) AL et Rego enc ara ee cP OR ea 2 Mele a eas eet Le ee cere e naytaet nan Seen pn AM spalae Oak cos cay he, so a sre ee 2 2 BMI OC Leek Ver NICO iia arts tha gis sack ag aphetas Sc wate oe Pterosyna. { Lateral teeth compressed, moderately developed......... Platynaias. ac [ ateralteeth nearly or.quitetwanting 022i. ba. doco. 2 Alasnuinota. 60 BryANt WALKER Subgenus PLATYNAIAS Walker, 1917. | Shell smooth, subsolid, shining, rayed; beak sculpture sharply double- looped; teeth delicate; laterals compressed, moderately developed. VLE. CLL Jif Lea ore Sax Fig. 176 Type: Symphynota compressa Lea, fig. 176. Subgenus LASMIGONA sg. s. Shell subrhomboid, compressed, corrugated behind; beaks low, their sculpture consisting of several coarse ridges, which generally fall into two slight loops, and often with radiating ridges in front and behind; epidermis shining; laterals partly developed, consisting of blurred ridges, which slope diagonally downward and backward on the hinge-plate; cavities of the beaks shallow. 7 Type: Alasmidonta costata Raf., fig. 177. SyNopsis FRESH-WATER MOLLUSCA 61 Subgenus ALASMINOTA Ortmann, 1914. Shell elongated elliptical, rather small. Surface without sculpture. Pseudocardinals delicate; laterals nearly or quite wanting. Beak sculpture not heavy, consisting of four to six rather fine, sharp bars, the first one or $8 Re Fig. 178 two subconcentric, the following ones sharply double-looped, the posterior loop smaller, separated from the anterior by a deep, sharp, re-entering angle. Type: Margaritana holstonia Lea, fig. 178. Subgenus PTEROSYNA Rafinesque, 1831. Shell large, ovate-rhomboid, inflated in the post-basal region; beaks much compressed; their sculpture sharply and strongly doubly looped; epi- dermis dark, scarcely rayed; teeth very heavy. Vig. 179 Type: Alasmodonta complanata Bar., fig. 179. 62 BrYANT WALKER Genus ALASMIDONTA Say, 1818. Shell generally rhomboid, inflated, with a well developed posterior ridge, which ends in a point when it is single or a biangulation when double; beaks full and high, with coarse, concentric or slightly doubly looped bars ; epidermis rayed, shining; hinge with two pseudocardinals in the left valve and one in the right; laterals usually wanting or imperfect, present in Pro- lasmidonta, cavity of the beaks deep; nacre bluish. Marsupium occupying the entire outer gills. Pee | | / | | Subgenus PRESSODONTA Simpson, 1900. Shell small, decidedly rhomboid, surface gen- erally painted with unbroken rays; beak sculpture slightly corrugated; teeth compressed. aype: Key to the subgenera of Alasmidonta. Lateral teeth“ present: 2 ema erent eee ak Prolasmidonta. iateral ‘teeth absent or jobseleteme: oer. oe sete een ee 2 Shell solid,. pseudocardinalsssolidjtsttmpy =. coe etn oo eee 3. Shell thinner, pseudocardinals compressed or imperfect......... 4. Shell ovate-rhombord, inhlatedt sees a. eee Alasmidonta s. s. Shell) smaller, ‘comipressediewery solid. oe ere Pegias. Pasterior slope slightly cortusated.: oh. emis: ee ee Rugifera. Posterior slope smooth ... PE PGME TER Grek EME I oii s 22 5. Shell rhomboid, posterior ridge low, rounded.......... Presscdonta. Shell sutbtriangular, posterior ridge high, sharp........... -Bullella. Subgenus ALASMIDONTA:s. s. Shell ovate-rhomboid, solid, inflated, shining, with very strong, generally con- j centric, beak sculpture; pseudocardinals ' solid, stumpy, somewhat radiately ridg- ed; laterals short, very imperfect or wanting; beak cavities deep, compressed. Type: Monodonta undulata Say, fig. 180. Unio calceolus Lea, fig. 181. SyNopsis FRESH-WaTER MOLLUScA 63 Subgenus PROLASMIDONTA Ortmann, 1914. Lateral hinge teeth present, but their number re- versed, two in the right, one in the left valve. Beak sculpture moderately heavy, bars with an angle upon the posterior ridge and a slight sinus in front of it. Inner lamina of inner gills free. Female shell recog- nizable by a slight swelling in the region of the pos- terior ridge. Type: Unio heterodon Lea, fig. 182. Subgenus PEGIAS Simpson, 1900. Shell small, thickened in front, with a sharp posterior ridge, in front of which is a wide, radial impression, ending in a basal sinus; above this ridge is another, making the shell decidedly biangu- late and truncate behind; beak sculpture consisting of sub-conic corrugations, generally swollen on the posterior ridge; epidermis decorticated, but showing a few, dark radial rays on the base of the shell; pseudocardinals rather solid; laterals wanting. Type: Margaritana fabula Lea, fig. 183. Subgenus RUGIFERA Simpson, 1900. Shell elongated, rhomboid, inflated, surface brilliantly painted with radi- ations which often break into dappled or splashed pattern of color; pos- terior slope slightly corrugated; teeth very imperfect; laterals wanting. Type: Alasmodonta marginata Say, fig. 184. 64 Bryant WALKER ; Subgenus BULLELLA Simpson, 1900. Shell thin, greatly inflated, somewhat triangular, with a high, sharp pos- terior ridge; beaks very full, having exceedingly strong, concentric sculp- ture, extending well on to the disk; pseudocardinals reflexed, compressed. —— ——— — —— ——— Type: Margaritana arcula Lea, fig. 185. Genus SIMPSONICONCHA Frierson, 1914. Shell sntall, elongate elliptical, rounded in front and behind, often slightly incurved at the central base; beak rather sharp, but Zjj not full; sculpture consisting of fine par- Z allel ridges which are looped up in the mid- dle and open behind; epidermis brownish, rayless ; teeth imperfect, a single, irregular, compressed tooth in each valve; laterals nearly or quite wanting; anterior end of the shell much thickened; nacre dull whitish. Marsupium occupying the whole of the outer gills. Type: Alasmodonta ambigua Say, fig. 186. Subfamily LAMPSILINA® Ortmann, 1912. “Inner lamina of inner gills rarely more or less free from the abdominal sac, generally connected with it throughout; supra-anal opening separated from the anal, rarely entirely closed; branchial openings well defined ; edge of the mantle in front of the branchial opening smooth to crenulated or with peculiar papilla or a flap; marsupium rarely formed by the whole outer gill, generally only by or within the posterior part of the outer gill; edge of marsupium, when charged, distending and bulging out beyond the original edge of the gill, generally assuming a beaded appearance: water- tubes simple in the gravid female; glochidium semicircular or semi-elliptic, Synopsis FrEsH-WaTER MOLLUSCcA 65 without spine, rarely celt-shaped and with two spines; shell heavy or lighter, rounded, or oval to elongate; color of the epidermis rarely dull, mostly bright, with color markings; sculpture of the beak generally double-looped, but often obsolete. more rarely concentric; hinge generaliy complete, with well developed teeth, which only in rare cases show a tendency to become re- duced; sexual differences more or less noticeable in the shell, often very strongly expressed.” (Ortmann. ) Key to the genera of Lampsiline. Wine: aticeneniale shells valiret oy. ko ee aka es Mh eee ee le as ae ve ‘ Male and female shells different...........:.... SRS ee Re oe Be Biren clononte tianOmlar $5. sik. Sata teen werent ee ee Ptychobranchus. SHelbrounded-trianoilat-or OVal-..').5 eset tt we ee Dee weit a Shell oval, with a medial row of large tubercles......... Obliquaria. r Shell rounded-triangular, nodulously wrinkled or lachrymose.... 4. Epidermis painted with delicate mottlings on a light ground, Pee NICS SILA WEN Ve oece- 2.2/4 sce Garces othe akon ke ....Cyprogenia. 4. i Epidermis painted with radiating hair-lines, beak cavities Mea pratile COLEESSeMn. we orcia.. sc oot tye cine erage, = oes Dromus. Female shell more or less expanded in the post-basal region...... 6. 5.4 Female shell slightly swollen just behind the middle of the CRIES oye ge 2b a Eine Rae eR eh aie PRN tet 7b Medionidus. rarpcalenmiai oie win sets aii, see feos jhs ces Seteena na he ein 2S aah 7¢ 6. Mintel, maa Lot ete WwAn dy sei ee dar bars) wiediare ble 3 he eee he 8. Pseudocardinals perfect; glochidium celt-shaped, with two eh se eee esse oo ee a sO SEP SES. eile a, oe ae rae ae le PPMP AIRE IAI IE eg Seto nae ba Proptera. 7. Pseudocardinals rudimentary; glochidium semicircular, with- GELS IGUTNR Eg aS ei ee a Paraptera. Peemdocrnmuaisamell developed, complete on. 2... fs sce eee ee ee 9. 8. Pseudocardinals divided into irregular lamine............. Glebula. { Shell with a distinct posterior ridge, dorsal-slope smooth........ 10. 9.4 Shell usually without a distinct posterior ridge, or when dis- Peruines dorsal-slope.tadiately scilptired ) nies. ee ek wees EL: { Hinge heavy and strong; hinge-plate wide and flat......... Plagiola. To. 4 leldmnee delicate hinge-plate marrow... 25.0 .:.2 6 2s. 2 Amygdalonaias. 66 BRYANT WALKER Marsupial expansion of the female shell of the same texture as the rest of the Shell ie. cnet dio eee ae ee ae 12. 11.4 Marsupial expansion of the female shell of different texture | from the rest of the shell and usually radiately sculp- lpereatmunedl yale cae wrx a See ee ee eS 2 5s hse Ree Truncilla, { Inner edge of the mantle in front of the branchial opening differentiated with papilla or flaps....-.-.-+++++++seseeeees hy 12: Inner edge of mantle in front of branchial opening without papiliz orcdiaps oo soen oe OES, gation +0" Seapets ae pag 16. Shell ‘simiooth: ...4s.< -sesp ate aetna eleva Sere Se Na 14. He: Shell strongly sculptured posterionly. --22- ee. aes ote eee Lemiov. i Beak -sculpture £7 Ae to Ae - J UNIVERSITY OF MICHIGAN MUSEUM OF ZOOLOGY Miscellaneous Publications No. 6 A Synopsis of the Classification of the Fresh- Water Mollusca of North America, North of Mexico, AND A Catalogue of the More Recently Described Species, With Notes BY BRYANT WALKER PART II—CATALOGUE ANN ARBOR, MICHIGAN PUBLISHED BY THE UNIVERSITY DECEMBER 30, 1918 enge ae A CATALOGUE OF THE MORE RECENTLY DESCRIBED FRESH-WATER MOLLUSCA OF NORTH AMERICA, WITH NOTES Subclass FUTHYNEURA. Order PULMONATA. Suborder BASOMMATOPHORA. Superfamily LIMNOPHILA. Family LYMNAAIDA. Genus LYMNZEA Lamarck, 1788. Baker’s elaborate monograph of the Lymneide of North America (No. 6) brings the subject down to Igrt. Simpsonia having been used by Rochebrune in 1904 for a group of Naiades, Baker (7, p. 120) has substituted Pseudogalba for his Lymnzeid group. The synonymy will be as follows: Simpsonia Baker, Chi. Acad. Sci., Sp. Pub. 3, 1911, p. 236. Non Rochebrune, 1904. Pseudogalba Baker, Naut. XX VI, 1913, p. 120. For a proposed revision of Baker’s arrangement and key. see Colton (22, p. 116 and 23, p. 119) and reply by Baker (8, p. 20). For the author’s arrangement, see ante p. 6. LYMNZA APICINA Lea. Although Lea’s name solida for this species has page precedence as stated by Hannibal (53, p. 146), and has been preferred by Haldeman, Tryon and Binney, according to Kuster (Con. Cab., Lymneus, p. 48) solida Phil. has priority and Lea’s species must be known by the later name. LYMNA‘A AURICULARIA (L.). This species seems to have become thoroughly acclimated and is extend- ing its range in this country very rapidly. In addition to the localities cited by Baker (6, p. 182), it has been listed from Toledo, O., by Goodrch (47, p. 11) ; Lake Erie, Kingsville, Ont., by Allen (1, p. 60) ; Philadelphia, Pa., by Long (72, p. 27) ; Colorado Springs, Colo., by Henderson (55, p. 84); Charles River, Cambridge, Mass., by Clapp (19, p. 116) and Johnson (59, p. 83) and has more recently been found in the Detroit River at Belle Isle by Dr. H. B. Baker and at Gibralter, Wayne Co., and La Plaisance Bay, Monroe Co., by Goodrich and at Harbor Beach, Huron Co., Mich., by Wal- ter Koelz. : 94 BryANnt WALKER LyMN2A cooper! Hannibal. Lymnea Cooperi Hannibal, Proc. Mal. Soc. Lond., X, 1912, p. 143, pl. VI, fig. 13a — c. Type locality: Spring at Wright’s, Santa Cruz Mountains, Cal. LYMNEA MONTANENSIS Baker. Lymnea montanensis Baker, Naut., XX VI, 1913, p. 115. Type locality: Hayes’ Creek, Ward, Montana. Family PLANORBID/E. Subfamily PLANORBIN/ H. and A. Adams, 1858. Genus PLANORBIS Miller, 1774. Dr. Dall (32, p. 80) has proposed the following arrangemeut of this genus: Subgenus Planorbis s. s. Type Planorbis corneus Miller. Section Planorbina Haldeman. Type P. olivaceus Spix. Subgenus Helisoma Swainson. Type P. bicarinatus (Say) Sowerby. Section Pierosoma Dall. Type P. trivolvis Say. Section Planorbella Haldeman. Type P. campanulatus Say. Subgenus Tropidiscus Stein. Type P. umbilicatus Miller. Subgenus Hippeutis Agassiz. Type P. fontanus Lightfoot. Section Menetus H. and A. Adams. Type P. opercularis Gould. Subgenus Gyraulus Agassiz. Type P. albus Miiller. Section Torquis Dall. Type P. parvus Say. Section Armiger Hartmann. Type: P. crista_ i. For a criticism of Dall’s arrangement, see Kennard (64, p. 47) and reply by Dall (33, p. 141). Brown and Pilsbry have proposed (15, p. 212) a new section, Trop- icorbis, type P. liebmanni Dkr., but without any definition. CATALOGUE RECENTLY DESCRIBED MoLLUSCA 95 PLANORBIS ALABAMENSIS Pilsbry. Planorbis alabamensis Pilsbry, Naut., VIII. 1895, p. 114. Type locality: Woodville, Ala. Dall (32, p. 92) considers this a local race of dilatatus Gld., but it seems sufficiently distinct. PLANORBIS ALABAMENSIS AVUS Pilsbry. Planorbis alahamensis avus Pilsbry, Naut., X1X. 1905, p. 34. Type localitv: Caloosahatchee Pliocene, Fla. Also from the marl at Lake Panasoffkee, Fla. PLANORPIS ANTROSUS Conrad. Vanatta has shown (138, p. 138) that as the earliest available name this must take precedence over the familiar P. bicarinatus Say, non La- marck, 1804. Conrad’s type has disappeared and until topotypes can be obtained, the uncertainty as to whether Conrad’s species is typical bicarinatus Say or equivalent to var. angistomus Hald. prevents the proper re-arrangement of the forms. According to Vanatta (1. c.) P. bicarinatus major and bicarinatus minor «.. Beck (Index. Moll., 1837, p. 118) are synonyms of the typical form. Planorbis biangulatus Say., Con. Icon., Planorbis, 1877, Sp. 25, pl. IV, .¢. 25, said to be from Brazil is apparently this species. PLANORBIS ANTROSUS ANGISTOMUS Haldeman. flanorbis bicarinatus angistomus Haldeman, Mon., 1844, p. 7; Walker, Naut, 1909, XXIII, p. 4, pl. II, fig. 4-5. mae ‘ype locality not given. PLANORBIS ANTROSUS AROOSTOOKENSIS Pilsbry. Planorbis bicarinatus aroostookensis Pilsbry, Naut., VIII, 1895, p. 115; Walker, Naut:,,1909;, XXIIT-p: 7; pid; fig: r & 3. Type locality: Salmon Brook, Woodland, Aroostook Co., Me. PLANORBIS ANTROSUS CORRUGATUS Currier. Planorbis bicarinatus corrugatus Currier MSS, DeCamp, Kent Sci. Inst. Misc. Pub., 1868, p. 8; Walker, Naut., XXIII, 1909, p. 5, pl. I, fig. 10. Type localjty: Perch Take, Kent Co., Mich. PLANORBIS ANTROSUS PERCARINATUS Walker. Planorbis bicarinatus major Walker, Naut., VI, 1893, p. 136. Planorbis bicarinatus percarinatus Walker, Naut., XXIII, 1909, p. 6, pl. I, fig. 12. Type locality: Crystal Lake, Benzie Co., Mich. 96 BryANt WALKER PLANORBIS ANTROSUS PORTAGENSIS Baker. Planorbis bicarinatus portagensis Baker, Naut., XXII, 1908, p. 45; Walker, Naut.,-1909, XXIII, p.-3; pl Tis; 0: Type locality: Portage Lake, Aroostook Co., Me. PLANORBIS ANTROSUS ROYALENSIS Walker. Planorbis bicarinatus royalensis Walker, Naut., XXIII, 1909, p. 9, pl. I, fic; LT. Type locality: Siskowit Lake, Isle Royale, Mich. PLANORBIS ANTROSUS STRIATUS Baker. Planorbis bicarinatus striatus Baker, Naut., XV, 1902, p. 120: Tr. Acad. Sci.. St. Louis, 1906, XVippsoeplie bk fic, 1 Type locality: Coldspring Park, Milwaukee, Wis. Also recent. See Walker 151a, p. 7. PLANORBIS ANTROSUS UNICARINATUS Haldeman. Planorbis bicarinatus unicarinatus Haldeman, Mon., 1844, p. 7; Walker, Naut., XXIII, 1909, p. 3, pl. I, fig. 6-8. Type locality: Schuylkill River, Pa. PLANORBIS ARCTICUS Moller. Planorbis arcticus Moller, Index Moll. Gronl., 1842, p. 5; Morch, Am. J. of Con., 1868, IV, p. 32, pl. 4, fig. 9. Type locality: Kudsuk, Greenland. Also Fort Chimo, Ungava, Labrador. PLANORBIS ARIZONENSIS Pilsbry and Ferriss. Planorbis filocinctus Pilsbry and Ferriss, Pr. A. N. S. P., 1906, p. 165, pl. IX, figs. 1-3, not of Sandberger. Planorbis arigonensis Pilsbry and Ferriss, Pr. A. N. S. P., 1915, p. 390. Type locality: San Pedro River, Benson, Ariz. PLANORBIS BILLINGSI Lea. Planorbis billingsit Lea, Pr. A. N. S. P.,- 1864, p: 111; Jour. A. N.S. P., 1866, VI, p..15, pl. 23s. 72); Obs. ;- 1866; 1, pi 415; ples ee: Type locality: Ottawa River, Canada. See parvus. ; PLANORBIS CALLIOGLYPTUS Vanatta. Planorbis callioglyptus Vanatta, Naut., IX, 1895, p. 54. Type locality: Freeport, Wash. Is opercularis planulatus Cooper according to Dall (32, p. 92). CATALOGUE RECENTLY DrEscriBED MoLLusca 97 PLANORBIS CAMPANULATUS MINOR Dunker. Planorbis campanulatus minor Dunker, Con. Cab., 1850, Limneiden, p. 52, pl. 9, fig. Io. Type locality not given. PLANORBIS CAMPANULATUS RUDENTIS Dall. Planorbis campanulatus rudentis Dall, Rep. Harriman Exp., 1905, XIII, p. 90. Type locality: Knee Lake, Keewatin, Canada. The citations of P. multivolvis from Newfoundland by Farrar (37, p. 36), and from Michigan by Walker, prior to 1907, refer to this form and not to Case’s species. PLANORBIS CAMPANULATUS SMITHII Baker. Planorbis campanulatus smithii Baker, Naut., XXV, 1912, v. 118. Type locality: Douglas Lake, Cheboygan Co., Mich. PLANORBIS CARIB US d’Orbigny. This name has priority for the species commonly known as tumidus Pfr. PrLANorBis Carus Pilsbry and Ferriss. Planorbis carus Pilsbry and-Ferriss, Pr. A. N. S. P., 1906, p. 164, pl. IX, figs. 4-5. Type locality: Canyon of the Pecos River, above High Bridge, Val Verde Gorm Nexas. PLANORBIS CENTERVILLENSIS Tryon. Planorbis centervillensis Tryon, Mon., 1870, p. 210, pl. 7, fig. 7-0. Type locality: Centerville, Cal. Is a form of opercularis according to Dall (32, p. 92). PLANORBIS CIRCUMSTRIATUS Tryon. Planorbis circumstnatus Yryon,Am. J. of Con.) II, 1866, p. 113, pl. to, figs. 6-8. Type locality: Weatogue, Conn. As identified by Sterki, this species is certainly distinct from P. parvits Say. It ranges from Connecticut to Colorado. PLANORBIS COARCTATUS Sowerby. Planorbis coarctatus Sowerby, Con. Icon., Planorbis, 1876, Sp. 21, pl. 3, fig. 21 a-b. Type locality :? This is apparently an immature campanulatus. 98 Bryant WALKER PLANORBIS COMMU’TATUS ‘“Dunker” Sowerby. Planorbis commutatus Sowerby, Con. Icon., Planorbis, 1876, Sp. 63, pl. 8, fig. 63. ‘lype locality: North America. PLANORBIS CORPULENTUS Say. Is a valid species. See Walker 144, p. 133. The citation of this species from Florida or South Carolina by Melvill (74, p. 167) is no doubt erroneous. Sowerby’s figures of this species in the Conchologia Iconica are binneyi Tryon. PLANORBIS CRISTA L,. Planorbis nautileus, Walker, Naut., 1897, X, p. 117; Hanham, Naut., 1897, X, 130; Taylor, Nat, 1807.5 paa30: Planorbis costatus. DeTarr and Beecher, Leaflet, Albany, 1878; Baker, Naut., 1906, XIX, p. 120. Range: Maine, west to Alberta and Illinois. PLANORBIS CULTRATUS Orbigny. Planorbis cultratus Orbigny, Hist. Cuba Moll., 1853, (Fr. ed.) I, p. 196, pl. 14, figs. 5-8; Pilsbry; Naut., 1889, ITI, p. 63, pl. I, fig. 1-3. Type locality: Cuba? Listed from Miami, Fla., by Rhoads (113, p. 48) and by Pilsbry from Hidalgo, Tex. (88, p. 63) and by Pilsbry and Ferriss from Devil’s River. Val. Verde Co., Tex: (Giebjop: 2165). PLANORBIS DECLIVIS Sowerby. Planorbis declivis Sowerby, Con. Icon., Planorbis, 1876, Sp. 28, pl. 4, fig. 20. Type locality: ? Von Martens (73, p. 397) thinks that this is probably a form of triuvol- vis. The name is pre-occupied by Tate (1870), for a species from Nica- ragua. PLANORBIS DEFLECTUS Say. Planerbis deflexus Sowerby, Con. Icon., Planorbis, 1876, Sp. 88. pl. XI, fig. 88. Dall (32, p. 94) considers this doubtfully distinct from hirsutus and apparently identical with the European form known as draparnaudi or draparnaldi Shepp. PLANORBIS DILATATUS Gould. Includes buchanensis Lea according to Tryon (132, p. 209) ; virens and elevatus according to Vanatta (136. p. 55); and lens, alabamensis and buchanensis according to Dall (32, p. 92). CATALOGUE RECENTLY DrEscriBED Mol Lusca 99 PLANORBIS DILATATUS PENNSYLVANICUS Pilsbry. Planorbis dilataius pennsylvanicus Pilsbry, Naut., XXX, 1916, p. 96. Type locality: Glenolden, Delaware Co., Pa. PLANORBis DURYI Wetherby. Planorbis duryi Wetherby, Jour. Cin. Soc. Nat. Hist., 1879, p. 7, fig. 4. Type locality: Everglades, Fla. ‘Types No. 9712 Coll. Walker. PLANORBIS DURYI INTERCALARIS Pilsbry. Planorbis duryi intercalaris Pilsbry, Amer. Nat., 1887, XXI, p. 287. Planorbis intercalaris Rhoads, Naut., 1899, XIII, p. 47. Type locality: Florida. PLANORBIS EUCOSMIUS Bartsch. Planorbis eucosmius Bartsch, Pr. U. S. Nat. Mus., 1908, XX XIII, p. 699, pl. 57, figs. 1-3. Type locality: Greenfield Pond, Wilmington, N. C. PLANORBIS FUCOSMIUS VAUGHANI Bartsch. Planorbis eucosmius vaughant Bartsch, Pr. U. $. Nat. Mus., XX XIII, 1908, p. 699, pl. 57, figs. 4-6. Type locality; Burkes Place, La. PLANORBIS EXACUOUS Say. Commonly known as exacutus Say, but Dall (32, p. 91) restores the name as originally used by Say. Henderson and Daniels (56, p. 56), after careful consideration, have done likewise. It includes buchanensis Lea ac- cording to Vanatta (fide Simpson) (136, p. 54). PLANORBIS EXACUOUS MEGAS Dall. Planorbis exacuous megas Dall, Rep. Harriman Exp., XIII, 1905, p. 91. Type locality: Birtle, Manitoba. PLANORBIS GLABRATUS Say. Haldeman’s and Binney’s figures do not represent this species, which is entirely distinct from trivolvis. It is not uncommon in Florida and in the United States does not range outside of that state, (Pilsbry, 91, p. 321). Dall (32, p. 84) includes lentus, which was described from New Orleans, but this does not accord with Fischer and Crosse’s (38, p. 65) identification cf Say’s type. The true glabratus belongs to section Planorbina and not to Pierosoma. 100 BrYANT WALKER PLANORBIS GRACILENTUS Gould. Tryon (132, p. 192) and Fischer and Crosse (38, p. 72) consider this distinct from liebmanni, to which it is referred by Binney. PLANORBIS HAVANENSIS Pfeiffer. Listed from New Orleans, La., as a Segmentina by Pilsbry (85, p. 20) and Hinkley (58, p. 36). PLANORBIS HiIRSUTUS Gould. Is considered distinct from albus Mull. by Vanatta (136, p..55). Dall (32, p. 94) refers it to the so-called albus Mull., but in view of the uncer- tainty as to the proper name to be used for that species, retains Gould’s name. Kennard (64, p. 49) considers the American and European species as distinct. PLANORBIS IHIRSUTUS BOREALIS Westerlund. Planorbis borealis Westerlund, Mal. Bl., XXII, 1875, p. 77. Type locality: Port Clarence, Alaska; Northern Sweden. Stated by Dall (32, p: 94) to be “‘merely a somewhat delicately scu!p- tured mutation.” PLANORBIS HORNI Tryon. Planorbis horni Tryon, Am. J. of Con., I, 1865, p. 231, pl. 22, fig. 16. Type locality: Fort Simpson, British America. Listed as a var. of trivolvis by Pilsbry (95, p. 65). Dall (32, p. 89) states that there is a doubt as to whether the types came from Fort Simp- son on the Mackenzie River, or Fort Simpson, British Columbia, but that the figure looks more like the Pacific variety (subcrenatus), which he re- fers to trivolvis. PLANORBIS INTERTEXTUS Sowerby. Planorbis intertextus Sowerby, Con. Icon., Planorbis, 1876, Sp. 123, pl. 14, fig. 123a-b. Type locality: Florida. Through the courtesy of Mr. FE. A. Smith of the British Museum I have been able to examine one of the cotypes of this species. Both the description and the figure given by Sowerby are quite erroneous and very misleading. It has no resemblance whatever to antrosus, but is undoubt- edly a young shell of one of the many southern mutations of trivolvts. PLANORBIS JENKSII Carpenter. Planorbis jenksiu Carpenter, Central Falls QR. I.) Visitor, Mar. 2, 1871; Conca x. tke7,.pe2, Type locality: Pawtucket, R. I. CATALOGUE RECENTLY DESCRIBED MoLLusca IOI PLANORBIS LENTUS Say. Fischer and Crosse (38, p. 65) restrict this species to the New Orleans ferm described by Say and do not consider that the figures given by Gould, Haldeman, Dunker and others represent the species. Dall (32, p. 86) refers it to glabratus. PLANORBIS LIEBMANNI Dunker. Is referred to orbiculus by Fischer and Crosse (38, p. 71), but is stated by Pilsbry (91, p. 322) to be distinct. ‘ It is the type ‘of section Tropicor- bis Brown and Pilsbry. PLANORBIS MAGNIFICUS Pilsbry. Planorbis magnificus Pilsbry, Naut., 1903, XVII, p. 75; Bartsch, Pr. U. S. Nat. Mus., XX XIII, 1908, p. 697, pl. 57, figs. 7-9. Type locality: Cape Fear River, Wilmington, N. C. PANORBIS MULTIVOLVIS Case. Is a valid species and has been rediscovered at Howe Lake, Marquette Co., Mich. See Walker, 149, p. 61. Earlier citations of this species from Michigan, except the original one, and Newfoundland refer to P. cam- panulatus rudentis. PLANORBIS NATHORSTI Westerlund. Planorbis nathorsti Westerlund, Vega Exp., IV, 1887; p. 168. Type locality: Aulatsivik, West Greenland. PLANORBIS OCCIDENTALIS Cooper. Planorbis occidentalis Cooper, Pr. Cal. Acad. Sci., IV, 1870, p. 99. Type locality: Not given. Range: Washington Terr. to San José, Cal. See trivolvis. Is the mature form of tuwmens according to Cooper (26, p. 89). PLANORBIS OPERCULARIS Gould. Planorbis Ienticularis Sowerby, Con. Icon., Planorbis, 1876, Sp. 110, pl. 13, fig. I10. Includes planulatus Cooper, centervillensis Tryon and multilineatus Van. (oregonensis Van.) as varieties according to Dall (32, p. 92), with callio- glyptus Van. as a synonym of planulatus. PLANORBIS OPERCULARIS MULTILINEATUS Vanatta. Planorbis opercularis oregonensis Vanatta, Naut., IX, 1895, p. 54; non Tryon, 1865. Planorbis opercularis multilineatus Vanatta, Naut., XIII, 1899, p. 48. Type locality: Salem and Portland, Oregon. 102 BrYANt WALKER PLANORBIS ORBICULUS Morelet. Planorbis orbiculus Morelet, Test. Noviss., 1849, I, p. 17. Includes haldemani Dunker (1850) non haldemani C. B. Adams (1844). Fischer and Crosse also include liebmanni, but Pilsbry (1. c.) considers it to be distinct. PLANORBIS OREGONENSIS Tryon. Planorbis oregonensis Tryon, Am. J. of Con., I, 1865, p. 231, pl. 22, fig. 17. Type locality: Pueblo Valley, Oregon. See trivolvis. PLANORBIS PARVUS Say. Includes billingsii Lea according to Vanatta (136, p. 54) and Dall (32, p- 95) and circumstriatus Tryon according to Vanatta (1. c.). PLANORBIS PARVUS WALKERI Vanatta. Planorbis parvus walkerit Vanatta, Naut., XVI, 1902, p. 58. Type locality: Hartland, Vt. Also Michigan. PLANORBIS PERFORATUS “Gould?” Sowerby. Planorbis perforatus Sowerby, Con. Icon., Planorbis, 1876, Sp. 105, pl. 12 stipe TOS: Type locality: United States. Gould never described a Planorbis under this name. Clessin (20, p. 227) suggests that the species is perhaps from Fast Asia. PLANORBIS PLANULATUS Cooper. Is doubtfully referred to P. opercularis Gld. as a variety by Cooper (25, e100). PLANORBIS PLEXATA Ingersoll. Planortis plevata Ingersoll, U. S. Geol. & Geog. Surv. Terr., 1874, p. 402, text-fig. Type locality: St. Mary’s Lake, Antelope Co., Col. Is a var. of trivolvis according to Stearns (121, p. 105) and Cooper (26,79: S85). PLANORBIS RUBELLUS Sterki. Planorbis harni Pilsbry, Naut., IV, 18@1, p. 137, sine desc. Planorbis exacutus rubellus Sterki, L. and F. W. Moll., New Phila., 1894, Pp: 7- Planorbis rubellus, Pilsbry, Naut., XIII, 1890, p. 51. Type locality: Stone Creek Valley, Odbert’s Station, O. CATAT,OGUE RECENTLY DrEscrirED Mor Lusca 103 PLANORBIS SAMPSONI Ancey. Planerbis sampsoni Ancey in Sampson, Bull. Sedalia N. H. Soc., No. 1, 1885, p. 10, text-fig. Type locality: Flat Creek, Sedalia, Mo. PLANORBIS SCALARIS (Jay). Paludina scalaris Jay, Cat., 3rd ed., 1839, p. 112, pl. 1, figs. 8-9. Physa scalaris Haldeman, Mon., 1842, p. 34, pl. IV, fig. 9; W. G. Binney, L. and F. W. Shells, Pt. II, 1865, p. 96, fig. 164. Ameria scalaris ‘Tryon, Mon., 1870, p. 168; Dall, Ann. N. Y.: Lyc. N. H., 1X;-1870, ‘p: 356;. Naut., I1I,-1880, p.. 8: Planorbis scalaris Pilsbry, Con. Ex., II, 1888, p. 113. Physa (Thomsonia) carinifera Ancey, Le Nat., 1886, p. 358. Type locality: Everglades of Florida. Pilsbry (86, p. 287) states that this species is a Planorbis. PLANORBIS SINUOSUS Bonnet. Planorbis sinuosus Bonnet, Rev. & Mag. Zool., 1864, p. 280, pl. XXII, fig. 3. Type locality: New Mexico. j Is referred to glabratus Say by Tryon (129, p. 183). Fischer and Crosse (38, p. 67) question this approximation, but as their opinion is based on Binney’s figure (11, fig. 179), which does not represent Say’s species, it is not of much value. However as glabratus is not known to occur outside of Florida, Tryon’s suggestion is wrong anyway. Dr. Pilsbry informs me that it is P. tumidus Pfr. PLANORBIS SUBCRENATUS DISJECTUS Cooper. Planorbis subcrenatus disjectus Cooper, Pr. Cal. Acad. Sci., (2) III, «890, Poo4 spk Ty ice. 20: Type locality: Tuolumne Meadows, Cal. PLANORBIS TENUIS Phil. Listed from the drift of the Santa Cruz River, Tucson, Ariz., by Pilsbry and Ferriss (109, p. 400). PLANoRBIS TRASKII Lea. Planorbis trasku. ea, Jour. A. N.S. P.,-V]1, 1866; p. 157, pl. XXIII, fig. 70; Obs., XI, 1866, p. 113, pl. X XIII, fig. 70. Type locality» Kern Lake, Cal. Dall (32, p. 88) considers this specifically distinct from P. ammon. 104 Bryant WALKER PLANORBIS TRIVOLVIS Say. Includes subcrenatus Cpr., with oregonensis Tryon, occidentalis Cooper, and tumens Cooper, non Cpr., as synonyms and probably hornii Tryon ac- cording to Dall (32, p. 89). Puilsbry also (95, p. 65) lists hornii as a variety. PLANORBIS TRIVOLVUS BINNEYI Tryon. Planorbis corpulentus Gould, U. 8. Expl. Exp., 1852, p. 114, fig. 130; Hal- deman, Mon., 1844, p. 19, pl. III, figs. 7-9; W. G. Binney, L. & F. W. Shells, pt. II, 1865, p. 114, figs. 191-2; Sowerby, Con. Icon., 1877, Sp. Aj play dre>45 pl. Xe tence Planorbis binneyi Tryon, Am. J. of Con., III, 1867, p. 197. Type locality: West Coast. PLANORBIS UMBILICATELLUS Cockerell. Planorbis wmbilicatus Taylor, J. of Con., IV, 1885, p. 351, text-fig. non Miller (1774). Planorbis umbilicatellus Cockerell, Con. Ex., 1885, II, p. 68. Type locality: Brandon and Birtle, Manitoba. Ranges from New York to South Dakota. See also Vanatta (137, p. 117). PLANORBIS VERMICULARIS Gould. Is referred to parvus by Vanatta (136, p. 55), but is considered dis- timet by, Dall. 32. %p..05)). Genus SEGMENTINA Fleming, 1817. Subgenus PLANORBULA Haldeman, 1842. SEGMENTINA ARMIGERA (Say). Dr. Pilsbry informs me that he has seen the type of Planorbis lautus H. Ads. and that it is a young specimen of this species. SEGMENTINA ARMIGERA CAMPESTRIS Dawson. Segmentina armigera campestris Dawson, Rep. Brit. N. A. Boundary Com., 1875, Pp. 349. Type locality: Red River Valley, Canada. SEGMENTINA cuRistyi Dall. Segmentina christyi Dall, Rep. Harriman Exp. XIII, 1905, p. 99, pl. 11, figs. IO-II. Type locality: High Bluff, Manitoba; Fort Smith, Mackenzie River. Re- ported from South Dakota by Walker (151, p. II). SEGMENTINA CRASSILABRIS Walker. Segmentina crassilabris Walker, Naut., XX, 1907, p. 122, pl. 7, figs. 4-6. Type locality: Hamtramck, Wayne Co., Mich. CATALOGUE RECENTLY DrscrrBpED Mort,usca 105 SEGMENTINA DECLIVIS (Tate). Planorbis declizis Tate, Am. J. of Con., V, 1869, p. 159. Type locality: San Augustin, Acoyapa, Nicaragua. Cited-by Dall (32, p. 98) from Umpqua River, Oregon. Hannibal (53, p. 158) states that it has not been found by any of the local collectors in that region and questions the authenticity of the locality of Dall’s specimens. SEGMENTINA opstRucTA (Morelet). Planorbis obstructus Morelet, Test. Noviss. I, 1849, p. 17. Planorbis berendti ‘Tryon, Am. J. of Con., II, 1866, p. 10, pl. 2, figs. 14-16. Type locality: Carmen Island, Yucatan. “Occurs abundantly in Texas as far north as Austin.” (Pilsbry 91, p. 322. See also Pilsbry and Ferris, 106, p. 166.) In the absence of a figure of this species in any American publication, I have quoted that of berendti Tryon from Mexico, which is considered a synonym by Fischer and Crosse (38, p. 78) and von Martens (73, p. 398). SECMENTINA WHEATLEYI (Lea). Planorbis wheatleyi Lea, Jour. A. N. S. P. VI, 1866, p. 158, pl. 23, fig. 71; Oia, us 1860.qe 13, pl. 23-fis. Zi. Segmentina wheatleyi Walker, Naut. XX, 1907, p. 123, pl. VII, figs. 7-9. Dall (32, p. 97) has proposed a new section, Haldemanina, for this species, based on the “complex, dentiform and ridgelike” lamelle, but these differ from those of the other species (armigera and crassilabris) only in degree. See Pilsbry and Ferriss (106, p. 166) and Walker (1. c.). Subfamily POMPHOLIGIN/: Dall, 1866. Genus POMPHOLYX Lea, 1856. PoMPHOLYX LEANA H. and A. Adams. Pompholyx leana H. and A. Adams, Pr. Zool. Soc. London, 1863, p. 434. Type locality: West Columbia. POMPHOLYX SOLIDA Dall. Pompholyx var. solida Dall, Ann. N. Y. Lyc. Nat. Hist., IX, 1870, p. 335, pl. Il, fig. 7a. Type locality: West Columbia. Dall states that his species is clearly not effusa Lea, but that in the ab- sence of typical specimens of P. leana H. and A. Adams described from West Columbia, it still remains doubtful whether it belongs to the latter species. 106 BrYANtT WALKER Genus CARINIFEX W. G. Binney, 1863. Megastropha Lea, 1866 CARINIFEX NEWBERRYI MINOR Cooper. Carnifex newberryi var. ? minor Cooper, Pr. Cal. Acad. Sci., IV, 1870, p. 98. Type locality not stated. CARiNIFEX PONSONBYI E. A. Smith. Carinifex ponsonbyi E. A. Smith, P. Z. S. Lond., 1875, p. 536, text-fig. Planerbis ponsonbyt Sowerby, Con. Icon., Planorbis 1876, Sp. 80, pl. X, figs. 80a-b. Type locality: California. Call (16, p. 140) states that the figure in the P. Z. S. is interchanged with that of Diala leithii described at the same time. Family PHYSIDZ. Genus PHYSA Draparnaud. Dall (32, p. 100) has proposed the following arrangement: Section PHYSA s. s. Type P. fontinalis L. Section costTATELLA Dall. Type P. costata Newcomb. For an excellent revision of the Eastern American species, see Crandall, Now27. Von Martens (73, p. 368) has proposed the subgenus Alampetis for the North American and Mexican species with a dull, not glossy, surface and (often) thickened lip. He gives no type, but mentions P. ancillaria as an example. PHYSA ALBOFILATA Ancey. Physa albofileta Ancey, in Sampson, Rep. Geol. Surv. Ark., IT, 1891, p. 194. Type locality: West Leatherwood Creek, Eureka Springs, Carroll Co., Ark. See gyrina. PHYSA ALTONENSIS Lea. Physa-altonensis Lea, Pr. A. N.. Se Paci864, para; JourwAs Neo see 1866, p. 164, pl. 24, fig. 82; Obs., XI, 1866, p. 120, pl. 24, fig. 82. Type locality: Alton, Ills. Is elliptica according to Tryon (132, p. 163) and an abnormal ayrine according to Crandall (27, p. 71). CATALOGUE RECENTLY DrEscRIBED Mollusca 107 PrysA AMPULLACEA Gould. Includes P. lordi Bd., propinqua Try. and coniformis Try. as varieties according to Cooper (25, p. 98). According to Henderson and Daniels (56, p. 52) it is possible that I.ea’s P. nuttallii may be this species. If so it would have priority. PHYSA AMPULLACEA COLUMBIANA Hemphill. Physa ampillacea columbiana Hemphill, Naut., 1V, 1890, p. 27. Type locality: Columbia River, Astoria, Oregon. PHYSA AMYCDALUS Sowerby. Physa amygdalus Sowerby, Con. Icon., Physa, 1873, Sp. 65, pl. 8, fig. 65. Type locality: Texas. PHYSA ANATINA J.ea. Enyce ondima, Lea;‘Pr. A. N.S. P., 1864, p. 115; Jour. A. N: S$: P:, VI, 1866, p- 171, pl. 24, fig. 94; Obs., XI, 1866, p. 127, pl. 24, fig. 94. Type locality: Northern tributary of the Arkansas River, Kans. PirySA ANCILLARIA Say. According to von Martens (73, p. 374) Physa subarata Mke. belongs to this species and not to P. heterostropha Say as supposed by Binney and is represented by fig. 1, pl. III of Haldeman’s Monograph. PHYSA ANCILLARIA CRASSA Walker. Physa ancillaria crassa Walker, Naut., XIV, 1901, p. 98. Type locality: Higgins Lake, Roscommon Co., Mich. Types No. 1471 Coll. Walker. PHYSA ANCILIARIA MAGNALACUSTRIS Walker. Physa ancilleria magnalacustris Walker, Naut., XIV, 1901, p. 97. Type locality: Frankfort, Benzie Co., Mich. Types No. 9214 Coll. Walker. Puysa APLECTOIDES Sterki. Physa aplectoides Sterki, Pr. O. St. Acad. Sci., IV, 1907, p. 381. Type locality: Portage and Tuscawaras Co’s., O. Also Isle Royale and Schoolcraft County, Michigan. 108 BRYANT WALKER PHYSA AUREA Lea. Is a synonym of elliptica and not of heterostropha according to Tryon (a32, p, 163) and’ Crandall (27; gg55): PITYSA BILLINGSII Heron. Physa billings: Heron, Tr. Ott. F. Nat. Club, I, 1880, p. 62, pl. 2, fig. 5. Type locality: Billings’ Bridge, Ottawa, Ont. Is a var. of integra according to Crandall (27, p. 15). PHYSA BINNEYANA Ancey. Physa diaphana Tryon, Am. J. of Con. I, 1865, p. 224, pl. 23, fig. 11, non Krauss (1848). Physa binneyana Ancey, Le Nat., 1886, p. 358. Type locality: Oakland, Cal. PHYSA BLANDI Lea. Phitsa bland Vea, Pa AN. S. P1864, :p.- 1103 jours. NS) Pay eeu: p. 168, pl. 24, fg.88; Obs.; XI, 1866, p. 124, pl. 245 fig. 88. Type locality: California. Includes distinguenda ‘Try. and “?” is the same as grosvernori Lea and nuttallii Lea according to Cooper (25, p. 97). Both of the latter names have priority. PHYSA BREVISPIRA Lea. Physa bremspira Lea, Pr AN: S.eP., 1864; p 116; our. ALN. SP. Ae 1866, p:°173, pik 2aj tig. Oe, Obs. 2c, 1866, puizoepl 2. fess Type locality: Ottawa River, Ont. PHYSA CARLTONII Lea. Physa-carlionu Lea, Pr. AN. Si P1860, p..1255 Jour, A. NASe 2. Vila 1874, p. 63, pl. 21, fig. 19 fObs., XITL 1874, p67; pli zr fie; ro. Type locality: Mount Diablo, Cal. PHYSA CONIFORMIS Tryon. Physa coniformis Tryon, Am. J. of Con., II, 1866, p. 6, pl. II, fig. s. Type locality: Humboldt River, Oregon. PHysaA CoopERI Tryon. Physa coopert Tryon, Am. J. of Con., I, 1865, p. 224, pl. 23, fig. 9. Type locality: Crane Lake Valley, Cal. Is a variety of P. triticea Lea according to Cooper (25, p. 97). CATALOGUE RECENTLY DrEscriBED Mortusca 109 PHYSA CRANDALLI Baker. Physa rhomboidea Crandall, Naut., XV, 1901, p. 44, pl. II, figs. 6-7. non Meek and Hayden (1856). Physa crandalli Baker, Tr. Acad. St. Louis, XVI, 1906, p. 8. Type locality: Cedar and Muddy Creeks, Sedalia, Mo. Also Dardenelles and Sulphur Springs, Ark., and Las Vegas, N. M. Types No. 40775 Coll. Walker. According to Springer (120, p. 513) is a synonym of P. humerosa. PHYSA CROCATA Lea. fenysa crocaia Lea, Pr. A. N. 5. P., 1864, p. 114; Jour. Av N. S. P., V1, 1866, p. 169, pl. 24, fig. 90; Obs., XI, 1866, p. 125, pl. 24, fig. go. Type locality: Lafayette, Walker Co., Ga. Is closely allied to microstoma Hald. according to Crandall (27, p. 70). PHYSA CUPREONITENS Cockerell. Physa cupreonitens Cockerell, J. of Con., VI, 1889, p. 63. Type locality: Hot Spring, Wellsville, Colo. Though described as a distinct species, in the text it is called a sub- species of heterostropha. PHYSA CUBENSIS Pfeiffer. Physa cubensis Pfeiffer, Wiegm. Archiv., I, 1839, p. 354. Physa heterostropha peninsule Pilsbry, Naut., XIII, 1899, p. 48; ibid, XIV, 1899, p. 70. Type locality: Cuba. Also Miami and elsewhere in Florida. See Rhoads (113, p. 48). PHysa DEFoRMIS Currier. Physa deformis Currier, Am. J. of Con., ITI, 1867, p. 112, pl. 6, fig. r. Type locality: Grand Rapids, Mich. Is elliptica Lea according to Crandall (27, p. 54). PHYSA DISTINGUENDA Tryon. Physa distinguenda Tryon, Am. J. of Con., I, 1865, p. 225, pl. 23, fig. 6. Type locality: Marysville and Stockton, Cal. PHYSA DORBIGNYANA Lea. Physa striata Lea, Pr. A. N. S. P. 1864, p. 115, non d’Orbigny (1853), nec Menke (1830). Physa dorbignyana Lea, Jour. A. N. S. P., VI, 1866, p. 166, pl. 24, fig. 85; Obs-,2<), 1866.sp. 123, pl. 24, fig. 85. Type locality: Monterey, Cal. Is a synonym of P. virgata Gld. according to Pilsbry and Ferriss (108, Pp. £05): 1IO BryYyaANt WALKER PHYSA ELLIPTICA Lea. Is a valid species according to Crandall (27, p. 54) and includes troos- _ tiana Lea and minor Crandall as varieties and aurea, febigert and nicklinu Lea and deformis Currier as synonyms. Baker’s figures (4, pl. 34, fig. 5), copied by Blatchley and Daniels (14, pl. I, fig. 118) do not represent the true elliptica. PHYSA ELLIPTICA MINOR Crandall. Physa elliptica minor Crandall, Naut., XV, 1901, p. 55. Type locality: Grand Rapids, Mich. Types No. 14469 Coll. Walker. PHYSA FEBIGERI Lea. Physa febigeri Lea, Pr. A. N. 5. P., 1864, p. 114; Jour. A. N.S. P., VI, 1866, p. 174, pl. 24, fig. 99; Obs., XI, 1866, p. 130, pl. 24, fig. go. Type locality: Logan Co., O. Is elliptica according to Tryon (132, p. 163) and Crandall (27, p. 55). PHYSA FORSHEYI Lea. Physa. forshey:, Lea, Pris N:- Sak, 18645 p, 1i4e ote AI noo tee ele 1866, p. 172, pl. 24, fie: 95 3Obs., 2X1, 1866) ps 28) pl. 24) tie. OE: Type locality: Rutersville, Texas. Includes zhitei Lea according to Crandall (27, p. 67). Puysa FRAGILIS Mighels. Is a pathologic form of ancillaria according to Morse (75, p 43). PHYSA GROSVERNORI Lea. Physa grosvernori Lea, Pr. A. N. 8. P., 1864, p. 114; Jour. AN. 9. P.; VI, 1866, p. .275,: pl::24, fig. 10o5-Obs.,. x1, 1800.49 :51 31, salen ies Woe According to Cooper (25, p. 97) includes P. traskw Lea, occidentalis Try., dorbignyana Lea and sparsestriata Try. as varieties. Type locality: Santa Rita Valley. Is a var. of forsheyi according to Crandall (27, p. 69). PHYSA GYRINA Say. Includes cylindrica Newc., altonensis, hawnit and smithsoniana Lea as synonyms and albofilata Ancey, hildrethiana Lea and oleacea Tryon as vari- eties according to Crandall (27, p. 45). CATALOGUE RECENTLY DESCRIBED MoLLusca TIE Pirysa HALEI Lea. iehysa hale: Lea, Pr. A. N. $: P., 1864, p. 114; Jour. A: N. S. P., VI, 1866, pe 105. pl 24, ne. 83; -Obs., XI, 1866; p. 121, pl. 24, fig. 83. Type locality: Alexandria, La. PHYSA HAWNII Lea. Physa hawni Lea, Pr. A. N.S. Pe 1864, p. mes our ANS. -P., V1, 1866, p. 165, pl. 24, fig. 84; Obs., XI, 1866, p. 121, pl. 24, fig. 84. Type locality: Verdigris River, Kans. Is gyrina according to Tryon (132, p. 162) and Crandall (27, p. 54). PHYSA HETEROSTROPHA Say. Includes Jata and primeana Tryon according to Crandall (27, p. 29). PHYSA HETEROSTROPHA ALBA Crandall. Physa heterostropha alba Crandall, Naut., XV, 1901, p. 29. Type locality: Cedar Lake, Capachet, N. Y. Types No. 40747 Coll. Walker. PuysA HUMEROSA Gould. Includes rhomboidea Crandall (crandalli Baker), according to Springer 20, p. 513). PHYSA INTEGRA Haldeman. Includes billingsii as a var. according to Crandall (27, p. 56). Puysa LATA Tryon. Physa lata Tryon, Am. J. of Con., I, 1865, p. 227, pl. 23, fig. 7. Type locality: Juniata River, Hallidaysburg, Pa. See heterostropha. Puysa Lorpi Baird. Physa parkeri Currier, Kent Sci. Inst., Misc. Pub., 1868, p. 7 (no desc.) ; DeCamp, Kent Sci. Inst., Misc. Pub., No. 5, 1881, p. 15, pl. 1, fig. 3. Type locality (parkeri): Houghton Lake, Mich. Types (parkeri) No. 11997 Coll. Walker. Henderson and Daniels (56, p. 75) suggest that the Michigan and Can- adian forms differ markedly from the typical western form. PrysA MALLEATA Tryon. Physa malleata Tryon, Am. J. of Con., I, 1865, p. 225, pl. 23, fig. 14. Type locality: Hell Gate River, Oregon. I12 BryANt WALKER PHYSA MARGARITA Iesson. Physa margarita Lesson, Rev. Zool., 1840, p. 356. Type locality: Newfoundland. . PHYSA MEXICANA CONOIDEA Fischer and Crosse. Physa mexicana conoidea Fischer and Crosse, Moll. Mex., II, 1886, p. IOI, pl. 39, figs. 8-8a. Type locality: Mehedin, Mexico. Also McLennan Co., Texas, see Strecker, 1a6)) p- 1a PHYSA NIAGARENSIS Lea. Physa niagarensis Lea, Pr. A. N. Sink Lobd; pn Ti4s- four A. Nees bwy 1866, p. 168, pl. 24, fig. 97; One XI, 1866, p. 124, pl. 24, fig. 97. Type locality: Niagara River, N. Y. Is referred to integra by Tryon, (132, p. 167), but Crandall (27, p. 55) considers it distinct. PHYSA NICKLINII Lea. Physa nicklina Lea, Pr. A. N. S. P., 1864, p. 114; Jour. A. N. S. P., VI, - 1866, p. 175, pl. 24, fig. 101; Obs., XI, 1866, p. 131, pl. 24, fig. ror. Type locality: Callaghan’s, Alleghany Co., Va. Is elliptica according to Tryon (132, p. 163) and Crandall (27, p. 55). PHYSA NUTTALLII Lea. Physanutialln Lea, Pee N= S. Tee e1864> pat TOs out Ne Saale 1666, p:-171, pli 24rte. OF 9Obs., Xl 1866p. 127. spe eA, tiem oes Type locality: Lewis River, Oregon. See ampullacea. PHYSA OCCIDENTALIS Tryon. Physa occidentalis Tryon, Am. J. of Con., I, 1865, p. 226, pl. 2, fig. 8. Type locality: San Francisco and numerous other localities in California and Oregon. PuysA OLEACEA Tryon. Physa oleacea Tryon, Am. J. of Con., II, 1866, p. 6, pl. II, fig. 6. Type locality: Bridgeport, Ala., and Lake Superior. Is elliptica according to Tryon (132, p. 163). Crandall states (27, p. 45) that Tryon himself admitted this obvious error and considers it to be a var. of gyrina. Baker (5, p. 492) considers it to be simply an immature stage of tvpical gyrina. CATALOGUE RECENTLY DrEscriBED MoLLusca 113 PHYSA OSCULANS Haldeman. Includes mesvicana Phil. according to Fischer and Crosse (38, p. 100), Pilsbry (91, p. 323) and von Martens (73, p. 370). “Physa osculans is readily distinguishable from the eastern forms, P. heterostropha, integra and gyrina, but several described Californian Physas present no differences from the Mexican species and must be considered synonyms.” (Pilsbry, 1. c.) PHYSA PARVA Lea. meusorporva lea, Pr, AW. N.S. P:,.1864, p. 115; jour. A. N.S. P., VI, 1866, oi77. ple 24, hie. 104 Obs, XL, 1866, p2re3; pl: 245) fig-T04. Type locality: Verdigris River and Roca Creek, Kans. Is doubtfully referred to P. malleata Try. as a variety by Cooper (25, p. 97). Probably a young grosvernori, Tryon (128, p. 169); is gyrina, Tryon (132, p. 162); probably a young anatina, Crandall (27, p. 71). PHYSA POLITISSIMA Tryon. Physa politissima Tryon, Am. J. of Con., I, 1865, p. 226, pl. 23, fig. 13. Type locality: Sacramento, Cal. Is a variety of P. binneyana Ancey (P. diaphana Try.) according to Cooper (25, p. 97). Ts probably a synonym of friticea, and both are “dwarfed and arrested aspect (s)” of gyrina according to Stearns (122, p. 51). PHYSA POMILIA Conrad. Physa pomilia Conrad, Am. J. of Sci., XXV, 1834, p. 343; Am. J. of Con., ki ere66. p27S.pl, 15, figs. 1-3. Type locality: Randon’s Creek, Claiborne, Ala. Includes showalteri Lea according to Tryon (132, p. 162) and Crandall (27, p. 90). PHYSA PRIMEANA Tryon. Physa primeana Tryon, Am. J. of Con., I, 1865, p. 227, pl. 23, fig. 12. Type locality: Long Island, N. Y. Is heterostropha according to Crandall (27, p. 29). PHYSA PROPINQUA Tryon. Physa propinqua Tryon, Am. J. of Con., I, 1865, p. 223, pl. 23, fig. 5. Type locality: Jordan Creek, Idaho. PiysA RIVALIS Sowerby. Physa rivalis Sowerby, Con. Icon., Physa, 1873, Sp. 31, pl. 4, fig. 31. Type locality: Columbia River. This is not the P. rivalis of Maton and Rackett (1807) nor of Sowerby (1821-6). Clessin (20, p. 331) considers it a synonym of hildrethiana Tea. 114 Bryant WALKER Pitysa sAFrorpit Lea. Fypysa-sofferdn Vea, Pr. A.. N2SyPgereody ip. 115 Jour, A. Noe aa 1866, p. 166, pl. 24, fig. 87; Obs., XI, 1866, p. 123, pl. 24, fig. 87. Type locality: Lebanon, Wilson Co., Tenn.; Verdigris River, Kans., and Nashville, Tenn. Is gyrina according to Tryon (132, p. 162). PHYSA SHOWALTERI Lea. | Physa showaltert lea, Pr. A> NAS2E., 1804, p. L153 Jour, x. Noe Peay 1866, p. 170, pl. 24, fig. 92; Obs., XI, 1866, p. 126, pl. 24, fig. 92. Type locality: Uniontown, Ala. PHYSA SMITHSONIANA Lea. Physa smithsoniana Lea, Pr. A. N.S. P., 1864, p. 115; Jour. A. N.S. P., VI, 1866, p. 169, pl.24, fig..97 ; Ops, 2c) 1806; p. 25, pl 24 io rar. Type locality: Loup Fork of the Platte River. Is gyrina according to Crandall (27, p. 54). PHYSA SPARSESTRIATA Tryon. Physa sparsestriata Tryon, Am. J. of Con., I, 1865, p. 224, pl. 23, fig. Io. Type locality: San Joaquin Valley, Cal. PHYSA SUBROTUNDA Sowerby. Physa subrotunda Sowerby, Con. Icon., Physa, 1873, Sp. 87, pl. 10, fig. 87. Type locality: North America. PHySsA TENUISSIMA Lea. Physa tenuissima Lea, Pr A. NESeP., 1864, p. 14s sfoursAueNe seals 1866, p. 167, pl. 24, fig. 86; Obs., XI, 1866, p. 123, pl. 24, fig. 86. Type locality: Alexandria, La. Is referred to Aplexa by Tryon (132, p. 17). See Crandall (27, p. 71). His shell now in my collection is a dead, bleached specimen of A plexra hyp- norum. PHYSA TRASKII Lea. Physa-trasku Lea, Pr. A. N.S. Pie64)p.115, Joan AGN. Saher ear p- 163, pl. 24, fig. 80; Obs., XI, 1866, p. 119, pl. 24, fig. 80. Type locality: Rio Los Angelos, Cal. Ts a synonym of P. virgata Gld. according to Pilsbry and Ferriss (108, Dp. 198). CaraLOGuUE RECENTLY DescrrsED MoLLusca 115 PHYSA TRITICEA Lea. Physa triticea Lea, Jour. A. N. S. P., VI, 1866, p. 177, pl. 24, fig. 103 ; Obs., XI, 1866, p. 132, pl. 24, fig. 103. Type locality: Shasta Co., Cal. Is a form of gyrina and probably includes politissima Tryon, according fo-Stearns (122, p. 51). PHYSA TROOSTIANA Lea. Is elliptica according to Tryon (132, p. 163) and Crandall (27, p. 55). PHYSA VENUSTA Lea. Physa venusta Lea, Pr. A. N. S. P., 1864, p. 116; Jour. A. N. S. P., VI, 1866, p. 168, pl. 24, fig. 89; Obs., XI, 1866, p. 124, pl. 24, fig. 80. Type locality: Fort Vancouver, Oregon. Very closely allied to, if not identical with, P. virginea Gld. according to Tryon (128, p. 170), who also remarks in 1870 (Mon., p. 138) that it groups with gyrina Say. Prysa vinosa Gould. Crandall (27, p. 42) considers this to be a var. of ancillaria, but it seems to be sufficiently distinct. PHYSA VIRGATA Gould. Listed from Muscatine, Ia., by Nelson (76, p. 182). In all probability an erroneous identification. Is a variety of P. humerosa Gld. according to Cooper (25, p. 98). Widely, if sparsely, distributed in Arizona and New Mexico according to Pilsbry and Ferriss (107, p. 144). PHYSA VIRGATA ALBA Cockerell. Physa virgata mut. alba Cockerell, Jour. Mal., IX, 1902, p. 138. Type locality: Salt River, Tempe, Ariz. The varietal name is preoccupied by Crandall, P. heterostropa alba, 1901. PHysA WALKERI Crandall. Physa walkeri Crandall, Naut., XV, 1901, p. 57, pl. II, fig. 5. . Type locality: Petoskey, Mich. Types No. 3483 Coll. Walker. PHYSA WARRENIANA Lea. Physa warreniana Lea, Pr. A. N. S. P., 1864, p. 115; Jour. A. N. 8. P., VI, feoo, peas, pl. 24, he. St; Obs.; XI, 1866) p..120, pl. 24, fig. Sr: Type locality: Long Fork of the Platte River; Milwaukee, Wis.; Grand Rapids, Mich. Is a var. of sayii according to Crandall (27, p. 44). 116 BryANt WALKER PHYSA WHITEI Lea. Physa whitet Lea, Pr. A. N.S. P., 1864, p. 114; Jour. A. N.S. P.,; VI, 1866; p. 172, pl. 24,:fig..06 ; Obs, XG 1866; p 128, pl. 24, fisx06. Type locality: Walker Co., Ga.; Verdigris River, Kans. Is forsheyi according to Crandall (27, p. 69). PHYSA WOLFIANA Lea. Physa wolfiana Lea, Pr. A..N. S. P., 1860, p. 125; Jour. A. Nos. Ps Vit, 1874; p63, pl. 21,'fig. 20; Obs: XITL, 1874, p. 67,/pl er me. 20: Type locality: Hot Springs, Colo. Genus APLEXA Fleming, 1822. APLEXA HORDACEA (Lea). Physa' homdacea Lea, Pr. A. N.S. P., 1864, p. 116; Jour.-A. N. 6, B:, Vip 1866, p. 176, pl. 24, fig. 102; Obs., XI, 1866, p. 132, pl. 24, fig. 102: Type locality: Vancouver Island, Oregon. Referred to Aplexa by Tryon (132, p. 170), and doubtfully by Dall (32, p. 113), but its generic position still remains to be definitely settled by an examination of the animal. Dall (1. c.) states that the types came from Vancouver, Wash. and not from Vancouver Island, B. C. Is a variety of P. venusta Lea according to Cooper (25, p. 97). APLEXA HYPNORUM L,. Clessin (20, p. 287) distinguishes the American form (P. elongata Say) on the ground that the European form has a more slender shell and never a short spire as is the case with both the American varieties recognized by him, but the concensus of opinion is against him. APLEXA HYPNORUM ARCTICA (Clessin). Physa elongata arctica Clessin, Con. Cab., Limnzeiden, 1886, p. 287, pl. 41, fig. 5. Type locality: Hudson Bay. APLEXA HYPNORUM GLABRA (DeKay). Physa glabra DeKay, N. Y. Moll., 1843, p. 80, pl. 5, fig. 83. Physa elongatina Lewis, Pr. B. S. N. H., V, 1855, pp. 122, 208. Range: Conn., N. Y., and Michigan. This form seems to be entitled to recognition as a well marked race. APLEXA HYPNORUM TRYONI (Currier). Bulinus tryoni Currier, Am. J. of Con., III, 1867, p. 112, pl. 6, fig. 2. Type locality: Grand Rapids, Mich. CATALOGUE RECENTLY DrEscrRIBED MoLLusca 117 Family ANCYLIDA. For a revision of the patelliform genera of this family, see Walker, No. 160. Subfamily LANCINA® Hannibal, 1914. Genus LANX Clessin, 1880. LANX ALtus (Tryon). Ancylus altus ‘Tryon, Am. J. of Con., I, 1865, p. 230, pl. 22, fig. 15. Type locality: Klamath River, Cal. Is probably only a var. of newberryi according to Pilsbry (95, p. 65). LANx cRASsuS (Haldeman). Ancylus crassus Haldeman, Mon., 1844, p. 14, pl. 1, fig. 8. LANX KOOTANIENSIS (Baird). Ancylus kootaniensis Baird, Pr. Zool. Soc., Lond., 1863, p. 69; W. G. Bin- ney, L. and F. W. Shells, II, 1865, p. 144, fig. 242; Tryon, Mon., 1070, p> 227,-pl. Li, figs. 11-12. Ancylus (Levapex) kootaniensis Dall, Alaska, XIII, 1905, p. 110, fig. 82. Is doubtfully referred to L. crassus Hald. as a variety by Cooper (25, P-y100)): LANX NEWBERRY! (Lea). Ancylus newberryi Wea, Jour. A. N.S. P., VI, 1866, p. 185, pl. 24, fig. 116; Obs., X1, 1866, p. 141, pl. 24, fig. 116. LANX NUTTALLIT (Haldeman). V ellettia nuttallii Haldeman, Mon., 1841, pt. 3, p. 3 of cover. Acroloxus nuttallii Binney, L. and F. W. Shells, II, 1865, p. 147. LANX PATELLOIDES (Lea). Ancylus patelloides Lea, Jour. A. N. 8. P., VI, 1866, p. HOS: Br BATS ATL 7s Obs., XI, 1866, p. 141, pl. 24, fig. 117. Is not a marine species as stated by Tryon (132, p. Baye? See Pilsbry (93, Pp. 60). Includes altus Try. and subrotundus Try. and doubtfully newberryi Lea as varieties according to Cooper (25, p. 100). 118 BRYANT WALKER LANX PRAICLARUS (Stimpson). (Mss. ?) Ancylus preclarus “Stimpson” Lea, Obs. XI, 1866, p. 141. This apparently undescribed species is*referred to and distinguished from newberryi by Lea. LANX SUBROTUNDUS (Tryon). Ancylus subrotundus Tryon, Am. J. of Con., I, 1863, p. 230, pl. 22, fig. 14. Type locality: Umpqua River, Oregon. Subgenus WALKEROLA Hannibal, 1912. LANX (WALKEROLA) KLAMATHENSIS Hannibal. Lanx (Walkerola) klamathensis Hannibal, Pr. Mal. Soc. Lond., X, 1912, poi40; pl. Villienie--2re Type locality: Upper Klamath Lake, Ore. Genus FISHEROLA Hannibal, 1912. FISHEROLA LANCIDES Hannibal. Fisherola lancides Hannibal, Pr. Mal. Soc. Lond., X, 1912, p. 152, pl. VIII, fig. 35. Type locality: Snake River, Washington. Genus ACROLOXUS Beck. Does not occur in our fauna. Of the two species referred to it by Binney, one, 4. nuttalli, is a Lanx and the other, A. filosus, is a Rhedacniea. Subfamily FERRISSIINA® Walker, 1917. Genus FERRISSIA Walker, 1903. FERRISSIA BOREALIS (Morse). Ancylus borealis Walker, Naut., XVIII, 1904, p. 80, pl. 6, figs. 14-16. FERRISSIA CAURINA (“W. Cooper,’ W. G. Binney). Ancylus caurinus, J. G. Cooper, Pr. Cal. Acad. Sci., IV, 1870, p. roo. Tryon (132, p. 229) refers this species to fragilis, but it is an error. Is doubtfully referred to Ferrissia fragilis Try. as a variety by Cooper (25, p. Too), but later (26, p. 83) he considers it distinct. Dall (32, p. 110) also doubtfully refers it to fragilis. FERRISSIA CAURINA SUBALPINA -(J. G. Cooper). Ancylus caurinus subalpinus J. G. Cooper, Pr. Cal. Acad. Sci., (2), III, 1890, p. 82, pl. 1, figs. 27-28. Type locality: Yosemite Valley and Bloody Canyon, Cal. Also Oregon. CATALOGUE RECENTLY DEscrisED MoL_Lusca 119 FERRISSIA FRAGILIS (Tryon). As suggested by J. G. Cooper (26, p. 83), and Hannibal (53, p. 148), this is probably the non-septate form of Gundlachia californica. FERRISSIA HALDEMANI (Bourguinat). . Ancylus haldemani Walker, Naut., XVIII, 1904, p. 78, pl. 6, figs. 9-13. FERRISSIA HENDERSONI (Walker). Ancylus hendersoni Walker, Naut., X XI, 1908, p. 138, pl. 9, figs. 8-10. Type locality: Lake Waccamaw, N. C. FERRISSIA NOVANGLLA (Walker). Ancylus novanglie Walker, Naut., XXI, 1898, p. 138, pl. 9, figs. 5-7. Type locality: Cambridge, Mass. FERRISSIA OVALIS (Morse). Ancylus ovalis Walker, Naut., XVIII, 1904, p. 79. FERRISSIA PARALLELA (Haldeman). Ancylus parallelus Walker, Naut., X VIII, 1914, p. 77, pl. 5, figs. 1-9. FERRISSIA PUMILA (Sterki). Ancylus pumilus Sterki, 8th Ann. Rep. O. St. Acad. Sci, 1900, p. 36; sep- arate, p. 7; Walker, Naut.; XVIII, 1904, p. 82, pl. 6, figs. 20-22: Type locality: Tuscawaras River, Tuscawaras Co., O. It is possible that this will prove to be the non-septate form of Gund- lachia meekiana. FERRISSIA RIVULARIS (Say). Ancylus rivularis Walker, Naut., XVIII, 1904, p. 25, pl. 1, figs. 1-10, 13-14. FERRISSIA SHIMEKII (Pilsbry). Anciylus obliquus Shimek, Bull. Lab. Nat. Hist., St. Univ. Ia., I, 890, p. 214, pl. III, figs. 5a-c, non Broderip and Sowerby (1832), nor C. B. Ads. (1850), nor Krauss (1853). Ancylus shimeku Pilsbry, Naut., 1V, 1890, p. 48; Walker, Naut., XVIII, 1904, p. 81, pl. 6, figs. 17-109. Type locality: Deadman’s Run, Lincoln, Neb. Pilsbry (1. c. and 54, p. 63) has suggested that this may be the non- septate form of a Gundlachia, perhaps meekiana. This was controverted by Walker (1. c.), but nevertheless may be correct. 120 BRYANT WALKER FERRISSIA TARDA (Say). Ancylus tardus Walker, Naut., XVIII, 1904, p. 27, pl. I, figs. 11-12, 16-23; pl. II, figs. 1-23. FERRISSIA WALKERI (Pilsbry and Ferriss). Ancylus walkeri Pilsbry and Ferriss, Pr. A. N. S. P., 1906, p. 564, fig. 5. Type locality: Rogers, Benton Co., Ark. Subgenus LAS VAPEX Walker, 1903. FERRISSIA DIAPHANA (Haldeman). Ancylus diaphanus Walker, Naut., XVII, 1903, p. 17, pl. II, figs. 13-18. FERRISSIA EXCENTRICA (Morelet). Ancylus excentricus Morelet, Test. Noviss., II, 1851, p. 17; Pilsbry, Naut., THI, 1880, p:..64, pl. Lies 4; Walker Naut,; XVLL 19025 pa27-plds figs. 19-21. Type locality: Lago de Ita, Peten, Guatemala. Also Comal Creek, New Braunfels and Barton Creek, Travis Co., Texas. Frrrissta Fusca (C. B. Adams). Ancylus fuscus Walker, Naut., XVII, 1903, p. 15, pl. I, fig. 1-9. FERRISSIA FUSCA EUGRAPTA (Pilsbry). Ancylus eugraptus Pilsbry, Naut., IX, 1896, p. 1309. Anciylus fuscus cugraptus Walker, Naut., XVII, 1903, p. 17, pl. I, figs. 13-18. Type locality: Illinois River, Havana, Ills. FERRISSIA HEMISPH RICA (Walker). Ancylus hemisphericus Walker, Naut., X XI, 1908, p. 140, pl. 9, figs. 14-16. Type locality: Georgia. Also Decatur, Ala. FERRISSIA KIRKLANDI (Walker). Ancylus kirklandi Walker, Naut., XVII, 1903, p. 29, pl. II, figs. 1-12. Type locality: Grand Rapids, Mich. FERRISSIA OBSCURA (Haldeman). See Walker (Naut., XVII, 1903, p. 25, pl. I, figs. 16-18) for the Floridan form doubtfully referred to this. Rediscovered in the south fork of the Powell River at Big Stone Gap, Wise Co., Va., by Goodrich (48, p. 92), and quite different from the supposed Florida examples. CATALOGUE RECENTLY DeEscRIBED Mont,usca I2I FERRISSIA PENINSULA: (Pilsbry and Johnson). Ancylus peninsule Pilsbry and Johnson, Naut., IX, p. 138; Walker, Naut., MOVE 1903). 28: pl--IT, figs: 19-21. Type locality: St. John’s River, Fla. SPECIES INCERTAY SEDIS. ANCyLUS CALCARIUS DeKay. ANCYLUS OREGONENSIS Clessin. Ancylus oregonensis Clessin, Con. Cab., Ancylinen, 1882, p. 66, pl. 8, fig. 1. Type locality: Salem, Oregon. Also listed from the Sacramento River, Reading, Shasta Co., by Pilsbry (93, Pp. 60). Genus GUNDLACHIA Pfeiffer, 1840. The validity of this genus has been a subject of considerable discussion. See Dall (31, p. 97) and Walker (148, p. 14, and 160, p. 3). Dall has also published a very interesting series of observations on the relations of An- cylus and Gundlachia (34, p. 175). Subgenus GUNDLACHIA s. s. GUNDLACHIA ANCYLIFORMIs Pfeiffer. Gundlachia ancyliformis Pfeiffer, Zeitsch. fur Mal., 1849, p. 98; Ibid, 1853, p. 180, pl. I, figs. 1-16. Type locality: Lagune Injinio, San Vincente, Cuba. Listed by Simpson (117, p. 96), from Palma Sola, Fla. GUNDLACHIA HJALMARSONI Pfeiffer. Gundlachia hjalmarsoni Pfeiffer, Mal. Blatt., V, 1858, p. 197. Type locality: Santa Rosa, Honduras. Has been recorded and figured by Clapp (18, p. 77), from the drift of the Rio Grande, at Brownsville, Texas. Subgenus KINCAIDELLA Hannibal, 1912. This group includes: G. meekiana Stimp., californica Row., and stimp- soniana S. Smith. ‘GUNDLACHIA STIMPSONIANA S. Smith. Gundlachia stimpsomana S. Smith, Ann. N. Y. Lyc. Nat. Hist., IX, 1870, p. 390, fig. 6; Walker, Naut., X XI, 1907, p. 15, pl. IV. Type locality: Greenport, Long Islend, N. Y. Also on Shelter Island, N. Y. 122 Bryant WALKER Subfamily RHODACMEIN AZ Walker, 1917. Genus RHODACMEA Walker, 1917. Subgenus RHODACMEA s. s. RHODACMEA FILOSA (Conrad). Ancylus filosus Conrad, New F. W. Shells, 1834, p. 57; Haldeman, Mon., 1844, p. 10, pl. I, fig. 9; Binney, L. and F. W. Shells, II, 1865, p. 147, fig. 248; Walker, Naut., XVIII, p. 75; pl. 6, figs. 7-8. Acroloxus filosus Tryon, Mon., 1870, p. 232. Type locality: Black Warrior River, south of Blount Springs. RHODACMEA CAHAWBENSIS Walker. Ancylus filosus Walker, Naut., XVIII, 1904, p. 76, pl. VI, figs. 1-6. Rhodacmea cahawbensis (Walker, Naut., XXXI, 1917, p. 7, pl. I, figs. 4-6. Type locality: Cahawba River, Gurnee, Shelby Co., Ala. RHODACMEA ELATIOR (Anthony). Ancylus elatior Anthony, Ann. N. Y.. Lyc. Nat.:Hist., VI, 1855, p. 158, pl. V, fig. 20; Binney, L. and-F. W. Shells,-II, 1865, p. 140, fie.2234; Walker, Naut., X VIII, 1904, p. 78, pl. V, figs. 10-12. Type locality: Green River, Ky. RHODACMEA’ HINKLEYI ‘Walker. Ancylus rhodaceus “Walker,” Hinkley, Naut., XX, 1906, p. 40, not de- scribed. ; Ancylus hinkleyi Walker, Naut., XX1, 1908, p. 139, pl. IX, figs. 11-13. Type locality: Ohio River, Golconda, IIls. Section RHODOCEPHALA Walker, 1917. RHODACMEA RHODACME Walker. Rhodacmea rhodacme Walker, Naut., XX XI, 1917, p. 8, pl. I, figs. 1, 2 and 8. Type locality: Coosa River, Williamsville, Shelby Co., Ala. RHODACMEA GWATKINIANA Walker. Rhodacmea gwatkiniana Walker, Naut., XXXI, 1917, p. 9, pl. I, figs. 3, 7 and 9. Type locality: Coosa River, Butting Ram Shoals, Coosa Co., Ala. CATALOGUE RECENTLY DescrinED MoLLusca 123 Subfamily NEOPLANORBIN4E Hannibal, 1912. Genus NEOPLANORBIS Pilsbry, 1906. NEOPLANORBIS CARINATUS Walker. Neoplanorbis carinatus Walker, Naut., XXI, 1908, p. 127, pl. 9, figs. 17-18. Type locality: Duncan’s Riffle, Coosa River, Coosa Co., Ala. NEOPLANORBIS SMITHII Walker. Neoplanorbis snithii Walker, Naut., X XI, 1908, p. 126, pl. 9, figs. 1-2. Type locality: Higgin’s Ferry, Coosa River, Chilton Co., Ala. NEOPLANORBIS TANTILLUS Pilsbry. Planorbis tantillus “Pilsbry” Hinkley, Naut., XVIII, 1904, p. 54. Nude name. Neoplanorbis tantillus Pilsbry, Naut., XX, 1906, p. 51, pl. 3, figs. 3-5. Type locality: Wetumpka, Ala. NEOPLANORBIS UMBILICATUS Walker. Neoplanorbis umbilicatus Walker, Naut., X XI, 1908, p. 126, pl. 9, figs. 3-4. .Type locality: The Bar, Coosa River, Chilton Co., Ala. Genus AMPHIGYRA Pilsbry. Amphigyra Pilsbry, Naut., XX, 1906, p. 49. Type: Amphigyra alabamensis Pils. AMPHIGYRA ALABAMENSIS Pilsbry. Amphigyra alabamensis Pilsbry, Naut., XX, 1906, p. 50, pl. ITI, figs. 1-2. Type locality: Wetumpka, Ala. Subclass STREPTONEURA. Order PECTINIBRANCHIA. Suborder TASNIOGLOSSA. Superfamily PLATYPODA. Family AMPULLARID®. Genus AMPULLARIA Lamarck, 1799. AMPULLARIA BOREALIS Valenciennes. W. G. Binney (12, p. 430), has definitely ascertained that this species was based on the well known Natica heros Say. 124 BryANt WALKER AMPULLARIA CALIGINOSA Rve. Ampullaria caliginosa Reeve, Con. Icon., Ampullaria, 1856, pl. XXV, fig 118. Type locality: Unknown. Not listed by Sowerby in his recent catalogue (119, pp. 345-362). Listed from several localities in Florida by Dall and Simpson. AMPULLARIA MIAMIENSIS Pilsbry. Ampullaria mamiensis Pilsbry, Pr. A. N. S. P., 1899, p. 365. Type locality: Miami, Dade Co., Fla. AMPULLARIA PALUDOSA Say. This name must be used for Say’s species as his first name depressa was preoccupied by Lamarck. AMPULLARIA PINE! Dall. Ampullaria pinei Dall, Naut., XII, 1898, p. 75. Type locality: Homosassa River, Fla. AMPULLARIA ROTUNDATA Say. Sowerby has recently (119, -p. 357) referred this species with doubt to paludosa, overlooking Say’s statement that the operculum was calcareous and Binney’s figure in his edition of Say, pl. 75. It is no doubt an Old World species as suggested by Binney. In a recent letter, Mr. Sowerby says that he has “not the slightest doubt that it is a small specimen of the Indian 4. globosa Sw.” Family. VIVIPARID/. Genus VIVIPARUS Montfort, 1810. VIVIPARUS CONTECTUs (Millet). . This European species has become fully acclimatized at Washington, D. C., and at Philadelphia, Pa. (Bailey, 2, p. 60). VIVIPARUS CONTECTOIDES W. G. Binney. Tryon’s contention (132, p. 17) that this species should be known as V. lineata Kuster non Val. (Con. Cab., Paludina, 1852, p. Io, pl. 2, figs. 6-9) is not well founded. Limeata is preoccupied and linearis (Ibid, p. 19) is “of course” a misprint for lineata as stated by Tryon (131, p. 197) and Binney (13, p. 295). This species has been introduced and fully acclimated in Fairmont Park, Philadelphia, Pa. (Vanatta, 139, p. 84), and in the Public Garden in Bos- ‘ton, Mass. (Johnson, 62, p. 72). CATALOGUE RECENTLY DrscriBpED MoLtusca 125 VIVIPARUS CONTECTOIDES COMPACTUS Pilsbry. Viwifarus contectoides compactus Pilsbry, Naut., XXX, 1916, p. 42. Type locality: Doherty, Ga. VIVIPARUS CONTECTOIDES IMPOLITUS Pilsbry. Viwiparus contectoides impolitus Pilsbry. Naut., XXX, 1916, p. 4T. Type locality: Paint Rock River, Jackson Co., Ala. VIVIPARUS GEORGIANUS ALTIOR Pilsbry. Vivipara georgiana altior Pilsbry, Naut., V, 1892, p. 142. Type locality: Hitchin’s Creek, Fla. VIVIPARUS GEORGIANUS FASCIATUS Tryon. Viwipara georgiana fasciata Tryon, Mon., 1870, p. 17. Type locality not specified. VIVIPARUS GEORGIANUS LIMNOTHAUMUs Pilsbry. Vivipara georgiana limnothauma Pilsbry, Naut., VIII, 1895, p. 116. Type locality: Hitchin’s Creek, Fla. Also Lake George, opposite Drayton's Island, Fla. VIVIPARUS HALDEMANIANUuS “Shuttleworth” Frauenfeld., Vivipara haldemaniana “‘Shuttleworth” Frauenfeld, Verh. k. k. zool.-bot. Gesell. Wien, 1862, p. 1162. Type locality: Black Creek, Fla. Tryon (130, p. 374) says that this is “doubtless” V’. lineata Val. (con- tectoides \V. G. Binn.), but this is not likely as that species does not range so far south. It is more probable that it is either georgianus (Lea) or waltonii Try. If the latter, it would have priority. Tryon (131, p. 197) suggests that the Florida contectoides listed by Binney are “perhaps” his qwaltont. VIVIPARUS HALEANUS (Lea). This is- apparently a valid species as stated by Tryon. It also occurs in Itchaway-Notchway Creek, Baker Co., Ga., and fossil in a peat bed at Lake Panasoffkee, Fla. VIVIPARUS INTERTEXTUS (Say). Hannibal (53, p. 193) has proposed a new subgenus, Callina, having this species as the type. The distinction seems to be based on the rounded whorls and perforate shell of this species as compared with the imperforate shell and subcarinate body-whorl of typical Viviparus. But as the embryonic 126 BryANtT WALKER young of intertextus are strongly angulated and those of Il’. viviparus are quite acutely carinated the distinction does not seem to be well taken. If, however, for any valid reason, it should be found desirable hereafter to separate the two groups, the name will be available. VIVIPARA LINEATA (Valenciennes). W. G. Binney (13, p. 295) from an examination of the type states that this is the l’. bengalensis (Lam.) from India. VIVIPARUS MALLEATUS Reeve. This Japanese species has been introduced into a number of localities on the Pacific coast and has been listed under various names :— Paludina japonica Wood, Naut., V, 1892, p. 114; Ibid, VI, 1892, p. 51. Vivipara stelmaphora Stearns, Naut., XV, 1901, p. 91. Vivipara lecythoides Hannibal, Naut., XXII, 1908, p. 33. Viviparus malleatus Hannibal, Naut., XXV, I9QII, p. 31. Hannibal (53, p. 194) has made this species the type of a new subgenus, Cipangopaludina, which he refers to Jdiopoma Pils., (98, p. 189) originally proposed as a subgenus, but which he raises to generic rank. As the valid- ity of both of these changes must be ultimately determined by a study of the Asiatic species, they may well be held in abeyance until that has been done. VIVIPARUS JAPONICUS v. Martens. This species has been introduced into British Columbia (Pilsbry and Johnson, 110, p. 144) and California (Hannibal, 52, p. 32). It has also recently appeared in the Muddy River, Brookline, Mass. (Johnson, 60, p. 35 and 61, p. 48). Hannibal (53, p. 194) refers it to Jdiopoma Pils. VIVAPARA MULTICARINATA (Haldeman). This name was proposed by Haldeman for the Paludina carinata Val., which was erroneously stated by the author to be from Mexico, carinata having already been used by Swainson for an Indian species of the same genus. W. G. Binney (12, p. 430), states that the types in the Jardin des Plantes, Paris, are labelled in Valenciennes’ handwriting “Philippines.” It is undoubtedly a form of I’. Durroughianius Lea. VivIPARUS WALKERI Pilsbry and Johnson. Viviparus walkeri Pilsbry and Johnson, Naut., XX VI, 1912, p. 48, pl. XXX, figs. 6-7. Type locality: Juniper Creek, Lake Co., Fla. CataAlLOGUE RECENTLY DeEscriBED MoLLusca 127 \VIVIPARUS WALTONII Tryon. Vivipara waltonu Tryon, Am. J. of Con., II, 1866, p. 108, pl. 10, fig. 2. Type locality: St. John’s River, Fla. VIVIPARUS WAREANUS (Shuttleworth). This species is distinct from georgianus Lea. Genus CAMPELOMA Rafinesque, 1819. Melanthro W. G. Binney non Bowditch. Pilsbry has recently (105, p. 111) proposed to substitute Ambloxis Raf. for Campeloma Raf. For the same reasons that I have urged in support of the retention of Anculosa Say, it seems to me that the preference should be given to Campeloma. CAMPELOMA DECISUM (Say). The undescribed forms of this species from Michigan listed as vars. flava Currjer MSS. and melanostoma Currier MSS. (Walker, 142, p. 138) are of doubtful validity. Binney is in error in referring the following species to decisum as syno- nyms: integrum Say, geniculum Con., milesii Lea, obesum Lewis, rufum Hald., and subsolidum Anth. Melantho fecunda mentioned, but purposely left undescribed, by Lewis in 1868 (66, p. 135) and listed as a distinct species in 1869 (67, p. 34) does not seem to be separable from decisum, judging from the author’s original specimens now in my collection. Call’s remark (17, p. 135) that this is the female of obesuin Lewis is wholly wrong. CAMPELOMA FLORIDENSE Call. “Campeloma floridense Call MSS.” (as synonym of C. limum), Call, Bull. Washb. Coll. Lab. of Nat. Hist., I, 1886, p: 159, pl. 6, fig. 7-; Pilsbry, Naut., XXX, 1917, p. 42. Type locality not specified. Apparently restricted to the St. John’s River and tributary creeks in Florida. It has very generally been considered to be the C. limum (Anth.). CAMPELOMA GENICULUM (Con.). The exact status of this species still remains to be settled. Call at one time considered it a valid species (15a, p. 157), but later (17, p. 134) treated it as a variety of decisum. Lewis remarks (71, p. 41) that all the Alabama species exhibit this peculiarity. Under this aspect of the case, the species, to which Conrad’s form should be referred, can only be determined by an examination of his original type. 128 Bryant WALKER CAMPELOMA INTEGRUM (Say). Is a valid species and quite distinct from decisum. CAMPELOMA INTEGRUM oBEsuM (‘‘Lewis” Tryon). ? Paludina obesa “Lewis” W. G. Binney, L. and F. W. Shells, III, 1865, p. 47, fig. 95. Melantho obesa Lewis, Am. J. of Con., IV, 1868, p. 134. Melantho obesus Lewis, Pr. A. N. 8. P., 1875, p. 336, pl. XXIII, figs. 4-5. Vizipara obesa ‘Lewis’ Tryon, Mon., 1870, p. 25, pl. 13, fig. 6. Type locality: Ohio Canal, Columbus, O., and Michigan. Tryon seems to have been the first to have formally described this well marked form, although Lewis had already referred to it by that name in his papers on Melantho in 1868 and 1869. Binney figured what he supposed to be it, but Lewis seems to think (1. c.) that he did not do so. Binney states that “Paludina obesa’ is preoccupied, but I have not been able to check the reference. If that is true and Binney’s figure represents the true obesa of Lewis, his remarks and figure are sufficient to fix that name on the form and consequently it would have to receive a new name. Typically very distinct, this form seems to bear the same relation to integrum that gibbum does to rufum. : Call’s statement (17, p. 135) that Lewis’ tvpe of this form is the male and the type of his undescribed fecunda the female of the same species is an error. CAMPELOMA LEWIsI1 Walker. Campeloma lewisu ‘Walker, Naut., XVIII, 1915, p. 126, pl. V, fig. 3. Type locality: Yallabusha River, Grenada, Miss. This is the Melantho coarctata of W. G. Binney. For full synonymy see Walker, 154, p. 126. CAMPELOMA LIMuUM (Anthony). According to Pilsbry (103, p. 43) Melantho decampii W. G. Binn. is a synonym of this species, which has been very generally misunderstood. ‘The Florida form usually known by this name is C. floridense Call. CAMPELOMA MILESII (Lea). Is apparently a valid species. If not, it should be referred to decisum rather than to subsolidum. See Walker, 146, p. 121. CAMPELOMA PONDEROSUM COARCTATUM (Lea). This is the Paludina coarctata and P. incrassata of Lea and the Vivipara nolant of Tryon. For full synonymy see Walker, 154, p. 125. CAMPELOMA RUFUM (Haldeman). Ts a valid species. ; CATALOGUE RECENTLY DrEscriBED MoLLusca 129 CAMPELOMA RUFUM GIBBUM (Currier). Melantho gibba Currier, Am. J. of Con., III, 1867, p. 112, pl. 6, fig. . Type locality: Grattan, Mich. CAMPELOMA RUFUM GENICULIFORME Pilsbry. Campeloma rufum geniculiforme Pilsbry, Naut., XXX, 1916, p. 42. Type locality: Dooley Co., Ga. CAMPELOMA RUFUM MERIDIONALF Pilsbry. Campeloma rufum meridionale Pilsbry, Naut., XXX, 1916, p. 42. Type locality: Crozier’s Branch, Cabarrus Co., N. C. Also Little Sugar Creek, N. C. and Georgia. CAMPELOMA SPILLMANIT (Lea). Paludina spillmanu Lea, Pr. A. N.S. P., 1867, p. 81; Jour. A. N. S. P., VI, 1868, p. 343, pl. 44, fig. 29; Obs., XII, 1868, p. 103, pl. 44, fig. 29. Lioplax spillmanu Tryon, Mon., 1870, p. 35, pl. 14, fig. 7; pl. 15, fig. 8. Type locality: Jackson Co., Ala. Tryon (1. c.) gives the type locality as Jackson Co., Miss. Numerous specimens from several streams near Mooresville, Limestone Co., Ala., col- lected by Rev. H. E. Wheeler agree with the descriptions and figures given by Lea and Tryon and are Campelome. The embryonic young are strongly and acutely bicarinated, differing in this respect from all the other species of the genus. The operculum is wholly concentric. These shells agree very exactly with the cotypes of C. decampit W. G. Binn. in the DeCamp collec- tion. If this identification and approximation are correct, spillinanii Lea will follow decampii into the synonymy of C. limum (Anth.). CAMPELOMA suBSOLIDUM (Anthony ). Is a valid species. Whether the Paludina evxilis of Anthony is a sexual form as believed by Lewis and others or an individual or local mutation is unsettled. The fact that it has not been found in southwestern Michigan, where the species is a common one would seem to cast a doubt on its being a sexual variation. Genus LIOPLAX Troschel, 1856. LtiorLAxX ELLIoTTIt (Lea). Ts a valid species. LIOPLAX PILSBRYI Walker. Lioplax pilsbryt Walker, Naut., XVIII, 1905, p. 133, pl. IX, figs. 1-3. Type locality: Chipola River, Fla. Also Econfine River and Mud Creek, Fla. 130 Bryant WALKER Genus TULOTOMA Haldeman, 1840. TTULOTOMA ANGULATA (Lea). The opinion of Lewis (71, p. 24) and Wetherby (164, p. 207) that this is specifically distinct from magnifica Con. is no doubt correct. TULOTOMA COOSAENSIS (Lea). This species described as a Paludina and referred to Vivipara by Binney and to Lioplax by Tryon (132, p. 36) is a Tulotoma as stated by ‘Wetherby (i64;p; 212). Family VALVATID/. Genus VALVATA O. F. Miiller, 1774. VALVATA BICARINATA Lea. Is a valid species. See Walker, 146, p. 124 and 147, p. 209. VALVATA BICARINATA CONNECTANS Walker. Valvata bicarinata connectans ‘Walker, Naut., XX, 1906, p. 30. Type locality: Lake Michigan, New Buffalo, Mich. VALVATA BICARINATA NORMALIS Walker. Valvata bicarinata normalis Walker, Naut., XV, 1902, p. 125, fig. 5. Type locality: Not specified. Habitat: Muscatine, Ia. and Utica, Ills. VALVATA BICARINATA PERDEPRESSA Walker. Valvata bicarinata perdepressa Walker, Naut., XX, 1906, p. 30, pl. I, figs. 15-16. Type locality: Lake Michigan, Michigan City, Ind. VALVATA CALLI Hannibal, Naut., XXIII, 1910, p. 107. Type locality: Marl-deposit, Upper Lahontan Quaternary, Summer Lake, Or. VALVATA HUMERALIS CALIFORNICA Pils. Valvata humeralis californica Pilsbry, Naut., XXII, 1908, p. 82. Type locality: Bear Lake, San Bernardino Co., Cal. CATALOGUE RECENTLY DEscrisED Mor tusca 131 VALVATA LEWISI Currier. Valvata striata Lewis, Pr. A. N. S. P., 1856, p. 260; non striata Philippi, 1836-1844; Binney, L. and F. W. Shells, Pt. III, 1865, Pts, hers. Valvata lewisi Currier, Kent Sci. Inst. Misc. Pub., 1868, p. 9. Type locality: Little Lakes, N. Y. VALVATA LEWIS! HELICOIDEA Dall. Valvata lewisi helicoidea Dall, Rep. Harriman Exp., XIII, 1905, p. 123, pl. ie diese I-2.~ Type locality not specified. Range: “With the type form, to some extent everywhere, but especially toward the Northwest”. VALVATA MERGELLA West. Valvata mergella Westerlund, Vega Exped. Vetens. Iakt., [V, 1885, p. 209, pl. V, figs. 22 a-d. Type locality: Port Clarence, near Bering Strait, Alaska. VALVATA OBTUSA Drap. This European species has been listed from the mouth of the Genessee fiver, N.Y., by Baker’(3, p. 71). VALVATA PISCINALIS Miller. This European species has recently been found by Latchford (65, p. 10) at Honisher Bay, Toronto, Ont. VALVATA SINCERA DANIELSI !Walker. Valvata sincera danielsi Walker, Naut., XX, 1906, p. 28, pl. I, figs. 10-11. Type locality: Cannon Lake, Rice Co., Minn. VALVATA SINCERA NYLANDERI Dall. Valvata (sincera var.?) nylanderi Dall, Rep. Harriman Exp., XIII, 1905, p. E22 Type locality: Aroostook Co., Me. VALVATA TERRA-NOV-E Ferussac. Type locality: ? Specimens under this name are in the Museum of Paris according to Binney (12, p. 430), but it does not appear to have ever been described. 132 Bryant WALKER VALVATA TRICARINATA Say. This species is the type (by designation) of the subgenus Tropidina H. and A. Adams, 1858, but as it 1s based upon the carinated whorls of the typical form and the species varies from ecarinate to tricarinate, it does not seem worthy of recognition. VALVATA TRICARINATA BASALIS Vanatta. Valvata tricarinata basalis Vanatta, Naut., XXVIII, 1915, p. 105, fig. Type locality: Hudson River, N. Y. VALVATA TRICARINATA INFRACARINATA Vanatta. Valvata tricarinata infracarinata Vanatta, Naut., XXVIII, 1915, p. 104, fig. Type locality: White Pond, N. J. VALVATA TRICARINATA PERCONFUSA Walker. Valvata tricarinata confusa Walker, Naut., XV, 1902, p. 124, fig. 2, non V. confusa West. (1897). Valvata tricarinata perconfusa Walker, Naut., XX XI, 1917, p. 26. Type locality not specified. VALVATA UTAHENSIS Call. Valvata sincera utahensis Call, Bull. U. S. Geol. Surv., No. 11, 1884, p. 44, pl.- V1, figs. 1-3: Valvata utahensis Call, Pr. Davenport A. N.S., V, 1886, p. 4, pl. I, figs. 1-3. Type locality: Utah Lake, Utah. Family AMNICOLID. Subfamily BYTHININA® Stimpson, 1865. Genus BYTHINIA Leach, 1818. BYTHINIA PERFECTA Frauenfeld. Bythinia perfecta Frauenfeld, Verh. der k. k. zool-bot. Ges. Wien, 1862, p. £1545 Ubi; 1865,-p..527, pl soe Type locality: Columbia, North America. Frauenfeld states that as the types are without the opercula, he could not tell whether the species was a Bythinia or an Amnicola. If the locality is correct, it is surely not a Bythinia. It may be a Fluminicola. BYTHINIA TENTACULATA (L,.). This well known European species has been introduced by commerce and has spread from the Hudson west to Lake Michigan. CATALOGUE RECENTLY DrEscriBED MoLLusca 133 Subfamily AMNICOLINE Gill, 1871. Genus AMNICOLA Gould and Haldeman, 1840. AMNICOLA AUGUSTINA Pilsbry. Amunicola augustina Pilsbry, Naut., XVII, 1904, p. 113; Walker, Naut., XIX, 1906, p. 117, pl. V, figs. 13-14. Type locality: St. Augustine, Fla. Also at Tuscumbia, Ala., and fossil in a peat deposit at Lake Panasoffkee, Fla. AMNICOLA BAKERIANA Pilsbry, Naut., XXXI, 1917, p. 44. Type locality: Oneida Lake, N. Y. AMNICOLA PAKERIANA NIMIA Pilsbry. Amunicola bakeriana nimia Pilsbry, Naut. XX XI, 1917, p. 45. Type locality: Oneida Lake, N. Y. AMNICOLA CLARKIEI Pilsbry. Amnicola clarkei Pilsbry, Naut., XX XI, 1917, p. 45. Type locality: Oneida Lake, N. Y. AMNICOLA COMALENSIS Pilsbry and Ferriss. - Amnicola comalensis Pilsbry and Ferriss, Pr. A. N. S. P., 1906, p. 171, fig. 37; Pilsbry, Naut., XIII, 1910, p: 98. Type locality: Comal Creek, New Braunfels, Texas. Also Guadalupe River at the same place. AMNICOLA DESERTA Pilsbry. Amnicola deserta Pilsbry, Naut., XXIX, 1916, p. III. Type locality: Washington Co., Utah. AMNICOLA FERRUGINEA Calkins. Amnicola ferruginea Calkins, Valley Nat., II, 1880, p. 6, text fig. Type locality: Calumet River, Il. Baker (4, p. 331) refers this to 4. limosa Say. AMNICOLA FLORIDANA Frauenfeld. Amunicola floridana Frauenfeld, Verh. der k. k. zool.-bot. Ges. Wein., 1863, p. 1028; Ibid, 1865, p. 529, pl. X. Type locality: East Florida. 134 BryANt WALKER AMNICOLA FLORIDANA CONVEXA Pilsbry. Amnicola floridana convexa Pilsbry, Trans. Wag. Free Inst. Sci., III, Pt. IT, 1892, p. 338. Type locality: Pliocene marl of the Caloosahatchie and Shell Creek, Fla.; also living in the fresh-water of Florida at the present time. AMNICOLA HARPERI Dall. Amnicola harperi Dall, Naut., XXIV, 1913, p. 2. Type locality: Marl deposit, Lake Panasoffkee, Fla. AMNICOLA JOHNSONT Pilsbry. Aimnicola johnsoni Pilsbry, Naut., XIII, 1899, p. 21. ; Type locality: St. Augustine. Fla. Also fossil at Lake Panasoffkee, Fla. AMNICOLA LIMOSA (Say). Includes 4. ferruginea Calkins according to Baker. The figure given for this species by Dall (32, p. 117, fig. 84) is incor- rect, being a copy of Binney’s figure (No. 165) of A. pallida Hald. ~AMNICOLA LIMGSA PORATA (‘Say ). Includes A. orbiculata Lea as a synonym according to Pilsbry (92, p. 44). AMNICOLA LUSTRICA Pilsbry. Amunicola lustrica Pilsbry, Naut., IV, 1890, p. 53. Type locality not specified. Range: “New York to Illinois and Minnesota’. AMNICOLA Micrococcus Pilsbry. Ammnicola micrococcus Pilsbry, N. Am. Fauna, No. 7, Pt. II, 1893, p. 277, fig. 1; U. S. Nat. Mus., XXIV, 1901, p. 286, fig. 4. Type locality: Oasis Valley, Nev. Also Death Valley, Inyo Co., Cal. AMNICOLA MILIARIA Parreys. Frauenfeld (40, p. 1027) states that in the Cuming Collection there is a set of this European species labelled “Spring Garden Lake, East Florida”. He considers the shells to belong to this species without doubt, so that as he suggests there has probably been a mixing of labels. AMNICOLA MISSOURIENSIS Pilsbry. Amnicola missouriensis Pilsbry, Naut., XII, 1898, p. 43. Type locality: Carter Co., Mo. CATALOGUE RECENTLY DescRIBED MoLLusca 135 AMNICOLA NEOMEXICANA Pilsbry. Amnicola neomexicana Pilsbry, Naut. XXIX, 1916, p. III. Type locality: Socorro, New Mexico. AMNICOLA NUTTALLIANA “Lea” Frauenfeld. Amnicola nuttalliana Frauenfeld, Verh. der k. k. zool.-bot. Ges. Wien., 1863, p. 1029. The form thus listed by Frauenfeld from Silver Spring, Fort King, Fla., is probably Gillia wetherbyi Dall, as suggested by Dall (28, p. 258). AMNICOLA OLIVACEA Pilsbry. Amnicola olivacea Pilsbry, Naut., VIII, 1895, p. 115. Type locality: Huntsville, Ala. AMNICOLA ONEIDA Pilsbry. Amnicola oneida Pilsbry, Naut., XX XI, 1917, p. 46. Type locality: Oneida Lake, N. Y. __— AMNICOLA PALLIDA Haldeman. The figure given by Dall for this species (32, p. 117. fig. 85) is a copy of Binney’s figure (No. 168) of A. cincinnatiensis. AMNICOLA PARVA Lea. Is a valid species. See Pilsbry, 92, p. 44. AMNICOLA PILSBRYI Walker. Amunicola parva Marsh, Con. Ex., II, 1888, p. 91. Amnicola pilsbryi Walker, Naut., XIX, 1906, p. 116, pl. V, figs. 11 and 16. Type locality: Rockford, Ils. AMNICOLA SANCTIJOHANNIS Pilsbry. Amnicola sanctijohannis Pilsbry, Naut., XIII, 1899, p. 20. Type locality: St. John’s River, Astor, Fla. Also Silver Spring Run, Ma- rion Co., and Wekiva River, Fla. _+ AMNICOLA SCHROKINGERI Frauenfeld. Amunicola schrokingeri Frauenfeld, Verh. der k. k. zool.-bot. Ges. Wien, 1863, p. 1030; Ibid, 1865, p. 528, pl. X. Tvpe locality: Massachusetts. If the specimens from several localities in Maine are correctly identified, this species seems to be distinct from limosa, to which it is referred by Tryon (132, p. 52). Closely related to, but very much smaller than A. winkleyi according to Pilsbry (102; p.-1)- 136 Bryant WALKER AMNICOLA WALKERI Pilsbry. Amnicola walkeri Pilsbry, Naut., XII, 1898, p. 43; Walker, Naut., XIX, 1900,°p, 007, (pl. V shies Type locality: High Island Harbor, Beaver Ids., Lake Michigan. Range: Upper St. Lawrence drainage from Ottawa, Ont., to Lake Mich- igan. -AMNICOLA WINKLEYI Pilsbry. Amunicola winkieyi Pilsbry, Naut., XX VI, 1912, p. 1, pl. I, figs. 9-10. Type locality: Saco, Me. Section CINCINNATIA Pilsbry, 1891. AMNICOLA CINCINNATIENSIS (Anth.). Anmicola scarbcroughi Tryon MSS. is a synonym according to Tryon (132, p. 54). Baker’s remark (4, p. 336), that Binney’s fig. 162 is*an error is incor- rect. That figure does not represent this species, but is the radula of A. sayana Anth. (Pomatiopsis cincinnatiensis Lea). ‘The same author (loc. cit., pp. 335 and 343) has reversed the svnonymy of the two species, which explains his remark that Haldeman’s figures of the two species are inter- - changed. The figure given for this species by Dall (32, p. 118, fig. 87) is also incorrect, being Binney’s fig. 166 of A. limosa. AMNICOLA EMARGINATA (Kuster). Paludina obtusa lea, Pr. Am. Phil. Soc., II, p. 34, (1841), non P. obtusa Phil. (1837). Pauldina emarginata Kuster, Con. Cab., Paludina, 1852, p. 50, pl. 10. figs. 3-4- Cincinnatia binneyana Hann., Pr. Mal. Soc. Lond., X, 1912, p. 190. There seems to be no valid ground for the new name proposed by Han- nibal. AMNICOLA PERACUTA Pilsbry and Walker. Amnicola peracuta Pilsbry and Walker, Pr. A. N. S. P., 1889, p. 88, pl. IH, fig. 20. Type locality: Spivey’s Lake, Navarro Co., Texas. Genus PALUDESTRINA d’Orbigny, 1841. Bythinella Mogq.-Tand, 1851. Stimpsonia Clessin, Mal. Blatt., XXV, 1878, p. 151. Except as noted, all the species referred to Bythinella by Binney belong to this genus. CaTaLocuE RECENTLY DrscrrsEp Moriusca 137 PALUDESTRINA ACUTISSIMA “Whit.” (Frauenfeld). ? Amnicola acutissima ““Whit.”, Frauenfeld, Verh. der k. k. Zool.-bot. Ges. Wien, 1863, p. 207. Hydrobia acutissima “‘Whit.”, Frauenfeld, Ibid, 1863, p. 1021. Type locality: ? Frauenfeld queries as to who “Whit.” is. The fact that Pal. emarginata and other North American species are in the Cuming Collection named by “Whit.” makes it a possibility that this is also an American species. It is possible that ‘‘Whit.” stands for T. J. Whittemore, who was a well known collector in Massachusetts in 1840. PALUDESTRINA -EQUICOSTATA (Pilsbry). Bythinella equicostata Pilsbry, Pr. A. N. S. P., 1889, p. 86, pl. III, fig. 16. Type locality: Sumpter Co. and Haulover Canal, Fla. PALUDESTRINA ALDRICH (Call and Beecher). Bythineila aldrichi Call and Beecher, Bull. Wash. Coll., I, 1886, p. 190, pl. VII, figs. 11-14. P Type locality: Tributary of Black River, Reynolds Co., Mo. PALUDESTRINA BREVISSIMA (Pilsbry). Bythinella brevissima Pilsbry, Naut., IV, 1890, p. 64. Type locality: Haulover Canal, Indian River, Fla. “LYDROBIA” CALIFORNICA Tryon. Hydrobia californica Tryon, Am. J. of Con., I, 1865, p. 221, pl. 22, fig. 11. Is an 4lssimenia according to Pilsbry (96, p. 123). PALUDESTRINA CORRIGATA (Frauenfeld). Hydrobia corrigata Frauenfeld, Verh. der k. k. Zool.-bot. Ges. Wien, 1863, p. 1021; Ibid, 1865, p. 525, pl. VIII. Type locality: Boston, Mass. PALUDESTRINA DIABOLI Pilsbry and Ferriss. Paludestrina diabeli Pilsbry and Ferriss, Pr. A. N. S. P., 1906, p. 170, fig. 36. Type locality: Devil’s River, Val Verde, Texas. Also Rio San Filipe in the same county. PALUDESTRINA HEMPHILLI (Pilsbry). Bythinella hemphilli Pilsbry, Naut., IV, 1890, p. 63. Type locality: Kentucky Ferry, Snake River, Idaho. 138 BryANtT WALKER PALUDESTRINA IMITATOR Pilsbry. Paludestrina imitator Pilsbry, Naut., XII, 1899, p. 124. Type locality: Santa Cruz, Cal. PALUDESTRINA LONGINQUA (Gould). Amunicola longinqua Gould, Pr. B. $. N. H., V, 1855, p. 130. Pomatiopsis intermedia Tryon, Am. J. of C., I, 1865, p. 220, pl. 22, fig. 8. Bythinella intermedia Tryon, Mon., 1870, p. 46. See Pilsbry, 96, p. 122. »PALUDESTRINA MINUTA (Totten). Turbo minutus Totten, Am. Jour. Sci., O. S., XX VI, 1834, p. 369, fig. 6. Cingula minuta Gould, Rep. Invert. Mass., 1841, p. 265, fig. 171. Type locality: Mass. and Rhode Island. This species is referred to Paludestrina by Pilsbry (99, p. 90). PALUDESTRINA MONAS Pilsbry. Paludestrina monas Pilsbry, Naut., SLED, 1899, p. 21. Type locality: Wekiva River, Fla. PALUDESTRINA MONROENSIS (Dall). Bythinella monroensis Dall, Pr. U. S. Nat. Mus., VIII, 1885, p. 256, pl. 17, fig. 99. Type locality: Brook from Benson’s mineral spring into Lake Monroe, Enterprize, Fla. PALUDESTRINA NICKLINIANA (Lea). This species is the type of Clessin’s genus Stimpsonia. PALUDESTRINA NICKLINIANA ATTENUATA (Haldeman). Amunicola attenuata Hald., Mon., pt. 4, 1842, p. 3 of wrapper; Mon., 1844, p. BOS lel ehioeste: Bythinella attenuata Binney, L. and F. W. Shells, pt. IIT, 1865, p. 68, fig. 132. PALUDESTRINA PROTEA (Gld.). Amnicola protea Gould, Pr. Bost. S$. N. H., V, March, 1855, p. 129. Melania exigua Conrad, Pr. A. N. S. P., April, 1855, p. 269. Tryonia protea Binney, L. and F. W. Shells, III, 1865, p. 72, fig. 140. Bythinella protea Stearns, N. Am. Fauna, No. 7, Pt. II, 1893, p. 278. Paludestrina protea Stearns, Pr. U. S. Nat. Mus., XXIV, 1901, p. 277, pl. XIX-XXI. CATALOGUE RECENTLY DEscriBED MoLLusca 139 Hydrobia seemant Frauenfeld, Verh. der k. k. zool.-bot. Gesell. Wien, 1863, p. 1025; Ibid, 1865, p. 525, pl. VIII. Bythinella seemani Pilsbry, N. Am. Fauna, No. 7, Pt. II, 1893, p. 278. For an elaborate and fully illustrated account of the variation of this protean species, see Stearns’ paper cited above. PALUDESTRINA SALSA Pilsbry. Paludestrina salsa Pilsbry, Naut., XTX, 1905, p. go, pl. III, fig. ro. Type locality: Cohasset, Mass., in brackish water. PALUDESTRINA STEARNSIANA Pilsbry. Paludesirina stearnsiana Pilsbry, Naut., XII, 1899, p. 124. Type locality: Oakland, Cal. Also Marin, Tuolumne, Contra Costa and Santa Cruz Counties, Cal. Also Ash Canyon and Tanner Canyon, Huachuca Mts., Ariz., Pilsbry and Ferriss (106a, p. 516). Genus TRYONIA Stimpson, 1865. Pilsbry (96, p. 122), states that Tryonia is probably only a subgenus of Paludestrina. TRYONIA CLATHRATA Stimpson. This species, described from fossil specimens, has been found living in the Pahranagat Valley, Nev. (Stearns, 123, p. 281). Genus PYRGULOPSIS Call and Pilsbry, 1886. Pyrgulopsis Call and Pilsbry. Pr. Davenport A. N. S., V, 1886, p. 9. Type: Pyrgula nevadensis Stearns. PyYRGULOPSIS LETSONI (Walker). Anwnicola letsont Walker, Naut., XIV, p. 113 (1901) ; Letson, Bull. Buffalo poc. Nat. Sci.,; Vil, root, p. 241; fig. 165. Type locality: Post-Glacial deposit, Goat Island, Niagara River, N. Y. Also fossil at Bowmanville, Ills. and living at La Plaisance Bay, Lake Erie, Monroe Co., Mich. Goodrich (in lit.) has suggested that this species should be referred to Pyrgulopsis and I fully agree with him. PYRGULOPSIS NEVADENSIS (Stearns). Pyrgula nevadensis Stearns, Pr. A. N. S. P., 1883, p. 173, text fig.; Call and Beecher, Am. Nat., XVIII, 1884, pp. 851-855; Call and Pilsbry, Pr. Davenport Acad. Nat. Sci., V, 1886, p. 10, pl. II, figs. 1-10. Type locality: Walker and Pyramid Lakes, Nev. 140 Bryant WALKER PyYRGULOPSIS SCALARIFORMIS (Wolf). Pyrgula scalariformis Wolf, Am. J. of Con., V, 1869, p. 198, pl. 17, fig. 3. Pyrgulopsis scalariformis Shimek, Bull. Lab. Nat. Hist. St. Univ. Ia., LU, 1892, p. 168, pl. xii, figs. 3a-d. Pyrgula scalariformis mississippiensis Pilsbry, Am. Nat., 1886, p. 5. No description. Pyrgulopsis mississippiensis Call and Pilsbry, Pr. Davenport A. N. S., V, 1886, p. 13, pl. II, figs. 14-16; Walker, Naut., XIX, 1906, p. 116, pl. 5, lige DS; Type locality: Illinois River, Tazwell Co., Ills. Holocene. According to Shimek (1. c.) mississippiensis is a synonym of scalari- formus. PyRGULOPSIS OZARKENSIS Hinkley, Pyrgulopsis ozarkensis Hinkley, Pr. U. S. Nat. Mus., 49, 1915, p. 588, pl. Vege 2: Type locality: North Fork of White River, above Norfolk, Ark. PyYRGULOPSIS SHELDONI ( Pilsbry). Amnicola sheldoni Pilsbry, Naut., IV, 1890, p. 52.° Type locality: Lake Michigan, Racine, Wis. This species seems to be a Pyrgulopsis rather than an Ammnicola. PyRGULOPSIS WABASHENSIS Flinkley. Pyrgulopsis wabashensis Hinkley, Naut., X XI, 1908, p. 117. Type locality: Wabash River, The Chains, Posey Co., Ind. Genus POTAMOPYRGUS Stimpson, 1865. PorAMOPYRGUS CORONATUS (Pfeiffer). Paludina coronata Pteitter, Wiegm. Archiv., I, 1840, p. 253- Type locality: Cuba. Listed from Miami River, Fla., by Rhoads (113, p. 47). POTAMOPYRGUS SPINOSUs (Call and Pilsbry). Pyrgulopsis spinosa Call and Pilsbry, Pr. Davenport A. N. S., V., 1886, p. 14, pl. II, fig. 17-19. Hydrobia texana Pilsbry, Ibid, V, 1886, p. 33, pl. III, fig. 1-6. Type locality: spinosus, Comal Creek, Texas; terana, Guadelupe River and Comal Creek, Tex. According to Pilsbry (91, p. 327) this species is a Potamopyrgus and is doubtfully distinct from coronatus. Te.xana is the ecarinate form. CATALOGUE RECENTLY DESCRIBED MoLLusca 14! Genus LITTORIDINA Souleyet, 1852 a LITTORIDINA MONROENSIS (Frauenfeld). Hydrobia monroensis Frauenfeld, ee der k. k. Zool.-bot. Gesell. Wien, 1863, p. 1023; Pilsbry, Pr. A. N.S. P., 1889, p. 88, pl. ITI, figs. 17-19. Bythinella monroensis ‘Tryon, Mon. 1870, p. 48. Type locality: Lake Monroe, Fla. The generic position of this species, which was doubtfully referred to Littoridina by Pilsbry (94, p. 22) has since been confirmed by him. It is not the Bythinella monroensis of Dall (28, p. 256). Subfamily LYTHOGLYPHINA: Fischer, 1885. Genus COCHLIOPA Stimpson, 1865. CocHLIOPA ROWELLI Tryon. 5 The occurrence of this species in California is considered doubtful by Pilsbry (100, p. 91) and by Pilsbry and Ferriss (106, p. 172). Rowell how- ever insists that the types were collected by him “near Baulinas Bay (not Clear’ Lake), Marin Co., Cal.” (114, p. 10). It is known to inhabit Nicaragua. CocHLIOPA RIOGRANDENSIS Pilsbry and Ferriss. Cochliopa riograndensis Pilsbry and Ferriss, Pr. A. N. S. P., 1906, p. 171 pl. IX, figs. 10-13. Type locality: Rio San Filipe, near the Rio Grande, Val Verde Co., ‘Texas. Also Devil's River in the same county. by Genus FLUMINICOLA Stimpson, 1865. Hannibal (53, p. 186) has proposed a new subgenus Heathella, “readily distinguished by its globose form,” having F. seminalis Hds. as the type and including F. fusca, merriami, erythropoma, columbiana and minutissima. FLUMINICOLA COLUMBIANA Hemphill. Fluminicola columbiana “Hemphill,” Pilsbry, Naut., XII, 1899, p. 125; Pr., Wes. Nat; Mus:, XXIV, 1901; p. 285. fig. 3. Type locality: Columbia River, Washington, near Wallula and near mouth of Snake River; Snake River, near Weiser, Idaho. FLUMINICOLA ERYTHROPOMA Pilsbry. Fluminicola fusca minor Stearns, N. Amer. Fauna, No. 75 PEEL, 18035 p: 282. No description. Fluminicola erythropoma Pilsbry, Naut., XII, 1899, p. 125. Type locality: Ash Meadows, Nye Co., Nev. 142 Bryant WALKER FLUMINICOLA MERRIAMI Pilsbry and Beecher. Fluminicola merriami Pilsbry and Beecher, Naut., V, 1892, p. 143; Stearns, N. Am. Fauna, No:7; pt: 2i.1893."p. 262, fig. 2. Type locality: Pahranagat Valley, Nev. FLUMINICOLA MINU‘ISSIMA Pilsbry. Fluminicola minutissima Pilsbry, Naut., XXI, 1907, p. 76, pl. IX, fig. 1. Type locality: Price Valley, Weiser Canyon, Washington Co., Idaho. FLUMINICOLA MODOCcL Hannibal. Fluminicola modoci Hannibal, Pr. Mal. Soc. Lond., X, 1912, p. 187, pl. VII, fig. 30. Type locality: Fletcher’s Spring, south end of Goose Lake, Cal. FLUMINICOLA NEVADENSIS ‘Walker. Fluminicola nev'adensis Walker, Occ. Pap. Mus. Zool., Univ. Mich., No. 29, 1916, p. 6, text-fig. Type locality: Cortez foot-hills, Humboldt Valley, Elko Co., Nev. FLUMINICOLA SEMINALIS (Hinds). Paludina senunalis Hinds, Voy. Sulphur, 1844, p. 59, pl. 16, fig. 22. Lithoglyphus cumingiu Frauenfeld, Verh. der k. k. Zool.-bot. Ges. Wien., 1863, p. 195; Ibid, 1865, p. 530, pl. XI. ; Amnicola tubiniformis Tryon, A. J. of Con., I, 1865, p. 219, pl. 22, fig. 5. Type locality: seminalis, Sacramento River, Cal. cumingu, California. turbiniformis, Crane Lake Valley and Surprise Valley, Cal. This synonymy is according to Pilsbry (96, p. 123). FLUMINICOLA SEMINALIS DALLI (Call). Amnicola dallt Call, Bull. U.S. Geol. Surv., No. XI, 1884, p. 45, pl. VII, figs. 4-6. Fluminicola seminalis dalli Pilsbry, Naut., XIII, 1899, p. 123. Type locality: Mountain streams near Pyramid Lake, Nev. Genus SOMATOGYRUS Gill, 1863. For a description of the peculiar apical sculpture of this genus, see Walker, 156. SOMATOGYRUS ALDRICHI W alker. Somatogyrus aldricht Walker, Naut., XIX, 1901, p. 114, ol V,, fig.<9: Type locality: Coosa River, Chilton Co., Ala. CATALOGUE RECENTLY DrEscrIBED Mollusca 143 SoOMATOGYRUS AMNICOLOIWES Walker. Somatogyrus amnicoloides Walker, Naut., XXIX, 1915, p. 52, fig. 3. Type locality: Ouachita River, Arkadelphia, Ark. SOMATOGYRUS AUREUS Tryon. Somatogyrus aureus Tryon, A. J. of Con., I, 1865, p. 220, pl. 22, fig. 9. Type locality: Tennessee River. SOMATOGYRUS BIANGULATUS Walker. Somatogyrus biangulatus Walker, Naut., XIX, 1906, p. 99, pl. V, fig. 6. Type locality: Tennessee River, Florence, Ala. SOMATOGYRUS CONSTRICTUS Walker. Somatogyrus constrictus Walker, Naut., XVII, 1904, p. 135, pl. V, fig. 3. Type locality: Coosa River, Wetumpka, Ala. SOMATOGYRUS COOSAENSIS Walker. Somatogyrus coosaensis Walker, Naut., XVII, 1904, p. 137, pl. V, figs. 6-8. ‘Type locality: Coosa River, Wetumpka, Ala. SOMATOGYRUS CRASSILAPRIS Walker. Somatogyrus crassilabris Walker, Naut., XXIX, 1915, p. 53, fig. 4; Hink- ley, Pr. U. S. Nat. Mus., XLIX, 1915, p. 580, pl. 78, fig. 1. Type locality: North Fork of White River, Norfolk, Ark. SOMATOGYRUS CRASSUS Walker. Somatogyrus crassus Walker, Naut., XVII, 1904, p. 138, pl. V, figs. 11-12. Type locality: Coosa River, Wetumpka, Ala. SOMATOGYRUS CURRIERIANUS (Lea). Ammola currieriana Lea, Pr. A. N. S? P., 1863, p. 118; Jour. A. N. S. P., VI, 1866, p. 186, pl. XXII, fig. 118; Lea, Obs., XI, 1866, p. 142, pl. XXII, fig. 118. Somatogyrus currierianus Walker, Naut., XVII, 1904, p. 137, pl. II, figs. 8-9. Type locality: Huntsville, Ala. SOMATOGYRUS DECIPIENS Walker. Somatogyrus decipiens Walker, Naut., XXII, 1909, p. 80, pl. I, figs. 10-11. Type locality: Coosa River, The Bar, Chilton Co., Ala. 144 Bryant WALKER SoMATOGYRUS EXCAVATUS Walker. Somatogyrus excavatus Walker, Naut., XTX, 1906, p. 100, pi. V, fig. 7. Type locality: Shoal Creek, Florence, Ala. SOMATOGYRUS GEORGIANUS Walker. Somatogyrus georgianus Walker, Naut., XVII, 1904, p. 139, pl. V, fig. 13. Type locality: Chattooga River, Chattooga Co., Ga. SOMATOGYRUS HENDERSONI Walker. Somatogyrus hendersoni Walker, Naut., XXII, 1900, p. 87, pl. VI, fig. 2. Type locality: Coosa River, Duncan’s Riffle, Chilton Co., Ala. SomMATOGYRUS HINKLEYI Walker. Sematogyrus hinkleyit Walker, Naut., XVII, 1904, p. 135, pl. V, figs. 1-2; Naut., XXII, 1909, p. 87, pl. VI, fig. 8-9. Type locality: Coosa River, Wetumpka, Ala. SomAtTOGYRUS HUMEROsUS Walker. Somatogyrus humerosus Walker, Naut., XIX, 1906, p. 98, pl. V, fig. 2. Type locality: ‘Tennessee River, Florence, Ala. SoMATOCYRUS INTEGER (Say). Includes Paludina fontinalis Phil. erroneously referred to subglobosus by Binney. Both integer and fontinalis are referred to Lithoglyphus by Frauenfeld (39, pp. 194 and 179). SomArocyrus NANus Walker. Somatogyrus nanus Walker, Naut., XVII, 1904, p. 136, pl. V, fig. 4. Type locality: Coosa River, Wetumpka, Ala. SOMATOGYRUS oBTrustUsS Walker. Somatoayrus obtusus Walker, Naut., XVII, 1904, p. 138, pl. V, fig. ro. Type locality: Coosa River, Farmer, Ala. SOMATOGYRUS PARVULUS ‘Tryon. Somatogyrus parvulus Tryon, A. J. of Con., I, 1865, p. 221, pl. 22, fig. 10. Type locality: Powell’s River, Tenn. Binney’s figure of “Gillia sp?” (1. and F. 'W. Shells, IIT, p. 115, fig. 220) is this species according to Tryon (131, p. 198). CATALOGUE RECENTLY DrEscRIBED Mollusca 145 SoMATOGYRUS PENNSYLVANICUS Walker. Somatogyrus peunsylvanicus Walker, Naut., XVII, 1904, p. 140, pl. V. figs. 15-16; Naut., XIX, 1906, p. 116, pl. 5, figs. 17-18. Type locality: Columbia, Pa. Also Potomac River, Harper’s Ferry, Va. SoMATOGYRUS PILSBRYANUS Walker. Somatogyrus pilsbryanus Walker, Naut., XVII, 1904, p. 142, pl. V, figs. 20-21. Type locality: Tallapoosa River, Tallassee, Ala. SoMATOGYRUS PUMILUS (Conrad). Anculotus pumilus Conrad, New F. W. Shells, 1834, p. 62; Binney, L. and FW shells’ LE o1865."p. So. Anculosa pumila Conrad, A. J. of Con., II, 1866, p. 278, pl. XV, fig. 5. Somatogyrus pumilus Walker, Naut., XTX, 1906, p. 115, pl. V, fig. ro. Type locality: Black Warrior River, Ala. Also Cahatchee Creek, Shelby Co; Ala. In his original description, Conrad quotes his species from Bayou Teche, La., but for some reason did not in his subsequent one in 1866. SoMATOGYRUS PYCMaA:US ‘Walker. Somatogyrus pygmeus Walker, Naut., X XII, 1909, p. 88, pl. VI, fig. 3. Type locality: Coosa River, The Bar, Chilton Co., Ala. SOMATOGYRUS OUADRATUS Walker. Somatogyrus quadratus Walker, Naut., XTX, 1906, p. 98, pl. V, figs. 3-4. Type locality: Tennessee River, Florence, Ala. SOMATOGYRUS SARGENT! Pilsbry. Somatogyrus sargenti Pilsbry, Naut., VIII, 1895, p. 102; Walker, Naut., MeVAL 1 904,"p: 139, pl.: V; fir.: 14. Type locality: Mud Creek, a tributary of the Tennessee River, Ala, SOMATOGYRUS STRENGI Pilsbry and Walker. Somatogyrus strengi Pilsbry and Walker, Naut., XIX, 1906, p. 99, pl. V, fig. 5. ‘ Type locality: Tennessee River, Florence, Ala. Also Wabash River, Posey Co: Ind: SOMATOGYRUS SUBGLOBOSUS (Say). Subglobosus (1825) has priority over isogonus (1829). Paludina fon- tinalis Phil. doubtfully referred to this species by Binney is a synonym of mteger Say. ; 146 BryANt WALKER SOMATOGYRUS SUBSTRIATUS Walker. Somatogyrus substriatus Walker, Naut., XIX, 1906, p. 97, pl. V, fig. 5. Type locality: Tennessee River, Florence, Ala. SOMATOGYRUS TENNESSEENSIS Walker. Somatogyrus tennesseensis Walker, Naut., XIX, 1906, p. 114, pl. V, fig. 8. Type locality: Shoal Creek, Florence, Ala. Somarocyrus TROTHIS Doherty. Somatogyrus trothis Doherty, Quar. J. of Con., I, 1878, p. 341, pl. IV, fig. 1. Type locality: Ohio River, Campbell Co., Ky. SoMaAToGyRUS UMBILICATUS Walker. Somatogyrus umbilicatus Walker, Naut., XVII, 1904, p. 137, pl. V, ng. : Type locality: Coosa River, Wetumpka, Ala. a SoMATOGYRUS VIRGINICUS Walker. Somatogyrus virginicus Walker, Naut., XVII, 1904, p. 141, pl. V, figs. 18-19. Type locality: Barnard’s Ford, Rapidan River, Va. SoMATOGYRUS WALKERIANUS Aldrich. Somatogyrus walkerianus Aldrich, Naut., XVIII, 1905, p. 140, text-fig. Type locality: Conecut (Conecuh?) River, Escambia Co., Ala. SoMATOGYRUS WHEELERI Walker. Somatogyrus wheeleri Walker, Naut., XXIX, 1915, p. 51, figs. 1-2. Type locality: Ouachita River, Arkadelphia, Ark. Genus GILLIA Stimpson, 1865. GILLIA WETHERBY (Dall). Hydrobia ? wetherbyi Dall, Pr. U. S. Nat. Mus., 1885, p. 258, pl. XVII. fig. 10. ? Ammicola nuttalliana Frauenfeld, Verh. der k. k. Zool.-bot. Ges. Wien., 1863, p. 1029. Type locality: Lake Eustis, Fla. Genus CLAPPIA Walker, 1909. Clappia Walker, Naut., XXII, 1909, p. 89. Type: Clappia clappi Walker. CLAPPIA CLAPPI Walker. Clappia clappi Walker, Naut., XXII, 1909, p. 89, pl. VI, figs. 1, 4 and 7. Type locality: Coosa River, Duncan’s Riffle, Chilton Co., Ala. CATALOGUE RECENTLY DeEscriIBED MoLLusca 147 Subfamily LYOGYRINAE Pilsbry, 1916. Genus LYOGYRUS Gill, 1863. LyocyRuS BROWNI (Carpenter). Amnicola brownii Carpenter, Central Falls (R. I.) Weekly Visitor, April, 1872. Valvata Seine brownii Carpenter, Naut., III, 1889, p. 69. Type locality: Cunliff’s Pond, Elmville, R. I. The citation of this species from Minnesota by Sargent (115, p. 126) is no doubt erroneous. Lyocyrus DALLI Pilsbry and Beecher. Lyogyrus dalli Pilsbry and Beecher, Naut., VI, 1892, p. 62. Type locality: 'Wekiva River, Fla. LyOGYRUS GRANUM (Say). Paludina grana Say, J. A. N.S. P., II, 1822, p. 378. Amnicola granum W. G. Binney, L. and F. W. Shells, III, 1865, p. 86, fig. 170. This species is known only from the Atlantic drainage in southeastern Pennsylvania and New Jersey. Western records for it and L. brownii are in all probability based upon some of the smaller species of Ammnicola. LyoGyRUS LEHNERTI Ancey. Liogyrus lehnerti Ancey, Con. Ex., II, 1887, p. 79. Type locality: Potomac River, Washington, D. C. According to Pilsbry (87, p. 113) and part of the original lot in my col- lection, this is a reversed Amnicola limosa Say. Genus HORATIA Bourguignat, 1887. Horatia Bourguignat, Etude sur les noms gen. des petites Paludinidees &c., 1887, p. 47. Westerlund (163, 4th Supp., p. 23) remarks that these small shells have the form of the smallest Pseudamnicolas, the structure of Lithoglyphus, the color of many Bythinellas and are allied to the Valvatas in their oper- culum. Typical Horatia is not represented in the North American fauna. Subgenus HAUFFENIA Pollonera, 1808. Hauffenia Pollonera, Boll. Mus. Zool. ed. Anat. Comp. Univ. Torino, XIII, 1808, p. 3. HoratiaA (HAUFFENIA) micrA (Pilsbry and Ferriss). Valvata micra Pilsbry and Ferriss, Pr. A. N.S. P., 1906, p. 172, pl. IX, figs. 7-9. Horatia (Hauffenia) micra Pilsbry, Naut., XXX, 1916, p. 83. Type locality: Guadalupe River, New Braunfels, Texas. 148 BRYANT WALKER HorATIA MICRA NUGAX (Pilsbry and Ferriss). Valvata micra nugax Pilsbry and Ferriss, Pr. A. N. S. P., 1906, p. 173, pl. IX, fig. 6. Horatia (Hauffenia) micra nugax Pilsbry, Naut., XXX, 1916, p. 83. Type locality: Guadalupe River, New Braunfels, Texas. Subfamily POMATIOPSINA® Stimpson, 1865. Genus POMATIOPSIS Tryon, 1862. POMATIOPSIS BINNEYI Tryon. Pomatiopsis binneyt Tryon, Pr. A. N. S. P., 1863, p. 148, pl. I, fig. ro. Bythinella binneyi W. G. Binney, L. and F. W. Shells, III, 1865, p. 60, figs. 130-137. ; This species is a true Pomatiopsis according to Pilsbry (96, p. 123). POMATIOPSIS CALIFORNICA Pilsbry. Pomatiopsis californica Pilsbry, Naut., XII, 1899, p. 126. Type locality: San Francisco, Cal. POMATIOPSIS CINCINNATIENSIS (Lea). Cyclostoma cincinnatiensis Lea, Pr. Am. Phil. Soc., I, 1840, p. 280. Amnicola sayana “Anthony” Haldeman, Mon., 1844, p. 19, pl. I, fig. 11. This species having proved to be a Pomatiopsis, Lea’s name takes pre- cedence over that of Anthony which was proposed on the supposition that the species was an Amnicola. Baker (4, p. 343) has erroneously attributed the species to Anthony and his description, figures and synonymy are those of Amnicola cincmnatiensis Anth. POMATIOPSIS HINKLEYI Pilsbry, Naut., X, 1896, p. 37. Type locality: Black Falls, Florence, Ala. POMATIOPSIS LAPIDARIA (Say). Paludina lustrica Say, quoted as Ammnicola lustrica by Haldeman and authors generally and as Pomatiopsis lustrica by Binney according to Pils- bry (80, p. 53) is the young of this species. PCOMATIOPSIS ROBUSTA Walker. Pomatiopsis robusta Walker, Naut., XXI, 1908, p. 97, text-fig. Type locality: Jackson Lake, Wyo. CATALOGUE RECENTLY DrEscrinrp MoLLusca 149 PLEUROCERIDA. It has been suggested that the family name should be properly Pleuro- ceratide. But Stejneger (Herpetology of Japan, Bull. 58, U. S. Nat. Mus., p. 24) in a similar case has decided that the change is not necessary. The errata given by Tryon on p. 427 of his “Strepomatid” are not in- cluded in his index and have, therefore, been noted under the several species as they are likely to be overlooked. Pilsbry (Pilsbry and Rhoads, 111, p. 496) has proposed the following re- arrangement of this family: Genus IO Lea. Type Fusus fluvialis Say. Genus LITHASIA Haldeman. Type Anculosa (Lithasia) geniculata Hald. Section ANGITREMA Haldeman. Type Melanta armigera Say. Genus PLEUROCERA Rafinesque. Type ? Section STREPHOBASIS Lea. Types S\. spillmani, cornea and clarkii Lea (all= plena Anth.) Genus ELIMIA H. and A. Adams. Type Melania acutocarinata Lea. Genus GYROTOMA Shuttleworth. Genus ANCULOSA Say. Dr. Pilsbry has more recently decided that Goniobasis should be restored to its former position as a generic term, on the ground that Elimia was a composite group. It will be noticed that no mention is made of Eurycelon in this arrange- ment. While, as Tryon remarks (134, p. 341), the genus as aggregated by him is made up of incongruous elements and upon a revision of the family will no doubt be dismembered, the typical group, of which anthonyi and crassa are leading terms form a very distinct group, which seems entitled to recognition. Genus LITHASIA Haldeman, 1840. — LITHASIA CURTA Lea. Ponasuneuria Lea, Pr: A. N.S. P:, 1868, p. 153; Lea, Jour. A. N.S. P., VI, 1868, p. 340, pl. 54, fig. 24; Obs., XII, 1868, p. 100, pl. 54, fig. 24. Type locality: Northern Alabama and Tuscumbia. 150 BryANT WALKER LITHASIA CYLINDRICA Lea. Lithasia cylindrica Lea, Pr. A. N.S. P:, 1866, p.-1333 Jour A; - NSF) awe 1868, p. 341, pl. 54, fig. 26; Obs., XII, 1868, p. ror, pl. 54, fig. 26. Type locality: Coosa River, Ala. LITHASIA OBOVATA (Say). Pilsbry (101, p. 47) has figured the operculum of this species. For an account of the early stages of growth in this species, see Walker, No. 143. LITHASIA OBOVATA BICONICA Pilsbry. Lithasia obovata biconica Pilsbry, 27th Ann. Rep. Dep’t. Geol. & Nat. Resc. Ind., 1908, p. 604, fig. 23a. Type locality: Wabash River, Gibson Co., Ind. LITHASIA PLICATA Wetherby. Lithasia plicata Wetherby, Jour. Soc. N. H. Cin., 1876, p. 9, pl. I, fig. 1. Type locality: Green River, Jackson Co., Ky. LITHASIA PURPUREA Lea. Lithasta purpurea Lea, Pr. A. N.S. P., 1868, p.153; Jour. A. N.S. P., VI, 1868, p. 340, pl. 54, fig. 23; Obs., XII, 1868, p. 100, pl. 54, fig. 23. Type locality: Cahawha River, Centreville, Bibb Co., Ala. LIrH ASIA WHEATLEYI Lea. Lithasm wheailey: Vea, Pr. vA. IN. be P., 1806, p..133; Jour As No Sok ve 1868, p. 341, pl. 54, fig. 25; Obs., XII, 1868, p. Io1, pl: 54, fig. 25- Type locality: Cahawba_ River, Ala. Section ANGITREMA Haldeman, 1841. LiIrHASIA ANGULATA (Wetherby). Angitrema angulata Wetherby, Jour. Cin. Soc. N. H., 1876, p. 11. pl. I, fig. 5. Type locality: Stone’s River, Rutherford Co., Tenn. LirHastA PARVA (Wetherby). Angitrema parva Wetherby, Jour. Cin. Soc. N .H., 1876, p. 9, pl. I, fig. 2. Type locality: Stone’s River, Rutherford Co., Tenn. CATALOGUE RECENTLY DEscriBED MOoLtusca 151 PLEUROCERA Rafinesque, 1818. Pilsbry (105, p. 114) from a consideration of the literature concludes that the type of Pleurocera Raf. is verrucosa Raf. and that it consequently takes the place of Angitrema Hald. and that for this group Ceriphasia Sw. should be used. ‘This has been controverted by Walker, No. 161, who argues that the type of Pleurocera had never been properly designated and desig- nates P. acuta Raf. as the type, thus retaining the name for the group with which it has commonly been known. In a similar case, Dall (20, p. 1141) had already taken the same position as advocated by Walker. PLEUROCERA ACUTA Rafinesque. Pleurocera acuta Rafinesque, Enumeration and Account, 1831, p. 3; Walker, Occ. Pap. Mus. Zool., U. of M., No. 38, p. 8. Ts identical with and has precedence over P. subulare lea and is the type of Pleurocera Raf. It includes according to Goodrich (409, p. 122) tractum Anth., neglectuim Anth., intensum Rve., pallidum Lea and Jabiatuim Vea. PLEUROCERA AFFINE (Lea). Trypanostoma affine Lea, Obs., XI, 1866, p. tor, pl. 23, fig. 57. PLEUROCERA ALTIPETOM (Anthony). Trypanostoma corneum Lea, Obs., XI, 1866, p. 104, pl. 23, fig. 63. PLEUROCERA ALVEARE (Conrad). Includes P. plicatum Tryon. PLEUROCERA ARATUM (Lea). Trypanostoma cinctum Lea, Obs., XI, 1866, p. 103, pl. 23, fig. 60. PLEUROCERA BICINCTUM Tryon. Pleurocera bicinctum Tryon, Am. J. of Con., II, 1866, p. 4, pl. I, fig. 2. Type locality: Bridgeport, Ala. PLEUROCERA CARINATUM (Lea). Trypanostoma carinatum Lea, Obs., XI, 1866, p. 104, pl. 23, fig. 62. This species as suggested by Tryon is probably the young of some other species. If, however, it should prove to be a valid one, it will have to be renamed as Pleurocera (Strephobasis) carinatum Lea has priority. PLEUROCERA CASTANEUM (Lea). Trypanostoma castaneum Tea, Pr. A. N. S. P., 1868, pers2;jour As Ns See Mil e605, p- 338, pl. 64, fig. 205:Obs.4, X11, 1868, p. 98, pl. 54, fig. 20. Type locality: Coosa River, Ala. 152 BryANt WALKER PLEUROCERA CURRIERIANUM (Lea). Trypanostoma currierianum Lea, Obs., XI, 1866, p. 103, pl. 23, fig. 61. PLEUROCERA CYLINDRACEUM (Lea). Trypanostoma cylindraceum Lea, Obs., XI, 1866, p. 98, pl. 23, fig. 51. Includes P. roanense Lea, according to Pilsbry (111, p. 498). PLEUROCERA EXIMIUM (Anthony). Has priority over gradatum (Anth.). PLEUROCERA GLANDULUM (Anthony). Is undoubtedly the young of one of the earlier described species of Strephobasis included in the synonymy of plena. PLEUROCERA GRADATUM (Anthony). Trypanostoma curtatum Lea, Jour. A. N. S. P., V1, 1866, p. 143, pl. 23, fig.53. Obs: X1, 1860, p: 09, pl, 2ennien 53: PLEUROCERA LESLEYI (Lea). Trypanostoma lesleyi Lea, Obs., XI, 1866, p. 102, pl. 23, fig. 59. PLEUROCERA LEWIsII (Lea). As suspected by Tryon this form is only a striate variety of elevatum. PLEUROCERA LYONII (Lea). Trypanostoma lyonii Lea, Jour. A. N.S. P., VI, 1866, p. 144, pl. 23, fig. 55; Obs., XI, 1866, p. Ioo, pl. 23, fig. 55. PLEUROCERA NAPOIDEUM (Tea). Trypanostoma napoideum Lea, Obs., XI, 1866, p. 99, pl. 23, fig. 54. PLEUROCERA NUCIFORME (Lea). Trypanostoma nuciforme Lea, Pr. A. N. S. P., 1868, p. 152; Jour. A. N. ». Ee VI) 1868, -p. 337, pli54. fig) 19 >"Obs,, XU, 1668, sp. @7aplewsas fig. 19. Type locality: Connesauga Creek, Whitfield Co., Ga. PLEUROCERA PUMILUM (Lea). Is probably the young of some other species. If valid however, it will have to be renamed as Pleirocera (Stephobasis) pumilum Lea has priority, unless that too should prove to be a synonym. CATALOGUE RECENTLY DrEscrIBED Mollusca 153 PLEUROCERA ROANENSE (Lea). Trypanostoma roanense Lea, Obs., XI, 1866, p. 98, pl. 23, fig. 52. PLEUROCERA SUBROBUSTUM (Lea). Trypanostoma subrobustum Lea, Obs., XI, 1866, p. 97, pl. 23, fig. 50. PLEUROCERA TEREBRALE (Lea). Trypanostoma terebrale Lea, Pr. A. N.S. P., 1868, p. 153; Jour. ANS. PS NES 330; pl. 54, figs 22>Obs., XII, 1868, p90, pl..54, fig. 22. Type locality: Jackson Co., Ala. PLEUROCERA UNDULATUM (Say). Includes MW. excurata Con. and rorata Rve., T. moniliferum Lea and spillmani Lea, and Io nobilis Lea, nodosa Lea, robusta Lea and variabilis Lea according to Pilsbry (111, p. 497). PLEUROCERA UNIVITTATUM (Lea). Trypanostoma univittatum Lea, Obs., XI, 1866, p. rot, pl. 23, fig. 58. PLEUROCERA'VENUSTUM (Lea). Trypanostoma venustum Lea, Pr. A. N.S. P., 1864, p. 12; Jour. A. N.S. P., VI, 1866, p. 149, pl. 23, fig. 66; Obs., XI, 1866, p. 107, pl. 23, fig. 66. Type locality: Big Prairie Creek, Ala. PLEUROCERA WHEATLEYI (Lea). Irypanosioma wheatleyi Lea, Pr. A. N. S. P., 1868, p. 153; Jour. A. N. ee) Vie668, (p:7336, pl $4, fie. 21; Obs. X1lk 1868, ps 90, pl. 54, biper2 I Type locality: Coosa River, Ala. Section STREPHOBASIS Lea, 1861. PLEUROCERA CURTUM (Haldeman). Dr. James Lewis (68, p. 224) suggested a long synonymy for this species, “not as being conclusive, but as being in many particulars deserving of in- quiry”. ‘Tryon (133, p. 88) in a review of Lewis’ paper, states that upon the invitation of Dr. Lea, he had re-examined the types of the species included in the proposed synonymy, which resulted in a renewed assurance that his original determinations respecting them were correct. Later, (Tryon, 134, p. 424) Lewis states that on reviewing the matter, he had ascertained “that one of Say’s species (hitherto treated as superfluous) was really entitled to take precedence of curtum’”. 154 BRYANT WALKER PLEUROCERA BITANIATUM (Conrad). According to Pilsbry (111, p. 499) Strephobasis clarkii Lea is not a synonym of this species, but of plena Anth. PLEUROCERA LYONII (Lea). Strephobasis lyonu Lea, Jour. A. N. S. P., VI, 1866, p. 151, pl. 23, fig. 65; Obs.,. XI, 1866, p. 107, pl. 23, fig. 65. The figure given by Tryon (134, p. 46, fig. 93) for this species is erron: eous and is that of Eurycwlon wmbonatum (loc. cit. fig. 658). If Pleurocera (Trypanostoma) lyonii Lea is a valid species, it has prior- ity over this one, which, if also a valid species, will have to have anew name. But in the present chaotic state of our knowledge of the family, it does not seem advisable to propose one until the double question of synonymy is definitely settled. PEUROCERA PLENUM (Anthony). Includes clurkii (Lea), corneum (Lea) and spillmani (Lea) according to Pilsbry (111, p. 499) and P. glandulum (Anth.). “STREPSIORASIS” HARTMANIANA Lea MSS. Ts listed by Lewis (1. c.) among the probable synonyms of P. curtuni (Hald.). It does not appear to have ever been described. Genus GONIOBASIS Lea, 1862. Hannibal (53, p. 179) has designated G. osculata Lea as the generic type. GONIOBASIS ACUTIFILOSA (Stearns). Melania (? Goniobasis) acutifilosa Stearns, Pr. U. S. Nat. Mus., XIE, 1890, Peep, ple Vij chic. .G: Type locality: Eagle Lake, Cal. GONIOBASIS ACUTIFILOSA SISKIYOUENSIS Pilsbry. Goniobasis acutifilosa siskiyouensis Pilsbry, Naut., XIII, 1899, p. 65. Type locality: Fall River, Siskiyou Co., Cal. GONIOBASIS ALBANYENSIS Lea. Goniobasis albanyensis Lea, Pr. A. N.S. Bs, 1864, p. 43 spout aA. eens VI, 1866, p. 140, pl. 23, fig” 49; Obs., XI, 1866, p. 97, pl. 23, fig. 49. Type locality: Albany and Blue Spring, Baker Co., Ga. CATALOGUE RECENTLY DeEscriBED MoLLUSCA 155 GONIOBASIS ARACHNOIDEA (Anthony). Includes baculum Anth. according to Lewis ( 69, p. 114). GONIOBASIS ARATA Lea. Goniobasis arata Lea, Pr. A. N. S. P., 1868, p. 151; Jour. A. N. S. P., VI, 1868, p. 329, pl. 54, fig. 4; Obs., XII, 1868, p. 80, pl. 54, fig. 4. Type locality: Connesauga Creek, Whitfield Co., Ga. GONIOBASIS ATERINA Lea. Goniobasis aterina Lea, Jour. A. N.S. P., VI, 1866, p. 136, pl. 23, fig. 42; Obs., XI, 1866, p. 92, pl. 23, fig. 42. GGNIOBASIS BACULOIDES Lea. eantopasts baculoides Lea, Pr. A. .N:.S. P., 1869, p. 125; Jour. A: N.S. P., WINE 1374, p: 62: pk 21, fig. 16 ; Obs.,, XIII, 1874; p. 66, pl. 21, fig: 18: Type locality: Coosa River, Ala. GONIOBASIS BIFASCIATA Lea. ieoniovasis bifasciata_Lea, Pr. A. N: S. P., 1868, p. 151; Jour. A. N. S..P., NO ees, ppssinpl-54,.fe..72; Obs... XI] 1668, pe O1.. pl. §4; figs 7. Type locality: Jackson Co., Ala. GONIOBASIS BOYKINIANA (Lea). Includes hallenbeckii Lea according to Pilsbry (90, p. 124). See also catenaria. GONTOPASIS BUI,BOSA (Gould). Includes nezberryi Lea according to Pilsbry (95, p. 66). GONIOBASIS CALIFORNICA (Clessin). Mclama californica Clessin, Mal. Blett., V, 1882, p. 180, pl. 4, figs. 9-9. Type locality: California. See nigrina. GONIORBASIS CATENARIA (Say). Includes G. sublirata (Con.), foridensis (Rve.), etowahensis (Lea), pap- illosa (Anth.), and downieiana Lea and probably also boykiniana (Lea), hallenbeckii Lea, bentonensis Lea and couperi Lea, according to Pilsbry (90, p. 124). GONTOBASIS CINGENDA Anthony. Goniobasis cingenda Anthony, Am. J. Con., IT, 1866, p. 146, pl. 7, fig. 3. Type locality: North Carolina. 156 Bryant WALKER GONIOBASIS CIRCUMLINEATA Tryon. Goniobasis circumlineata Tryon, Am. J. of Con., II, 1865, p. 244, pl. 24, figs. 14-15. Type localities: Mission San Antonio, Shasta Co.; Pit River and Feather River, Cal. Probably only a variety of G. nigrina according to Pilsbry (95, p. 66). GONIOBASIS CLATHRATA Lea. Gomobasts.-clathrota: Lea; Pr. Az N.S) 25 3868.4p. 151; jour Ak. Beak VI, 1868, p- 331, pl. 54; fig. 8:2Obs., XT, 1868; "p.-o1, pl. 54, fies: Type locality: Jackson Co., Ala. GONIOBASIS CLAVULA Lea. Goniobasis clavula Lea, Pr. A. N: S. P., 1868, p: 152; Jour. A. N. S$, P., V1, 1868, -p. 335, pl 54, fig. 15; Obs., X11, 1868, p. 95, pl. 54: eas: Type locality: Jackson Co., Ala. GONIOBASIS COCHLIARIS Lea. Goniobasis cochliarts ea, Pr. A. N.-S. P., 1868, p. 152; Journ. A. N. S. P. VI, 1868, p: 336, pl. 54, fig. 16; Obs:, XII, 1868, p. 96, pl. 54, fig. 16. Type locality: Shelby Co., Ala. GONTOBASIS COILUMBIENSIS Whiteaves. Goniobasis columbiensis Whiteaves, Natt., XIX, 1005, p:61, pl: 21, tags TI-12. Type locality: Upper Columbia River, B: C. This is probably only a form of livescens. GONIOBASIS COMALENSIS Pilsbry. Goniobasis pleuristriatus comalensis Pilsbry, Naut., 1V, 1890, p. 49. Goniobasis comalensis Pilsbry and Ferriss, Pr. A. N. S. P., 1906, p. 167, fig. 24-31. Type locality: Comal Creek, New Braunfels, Texas. Melania pluristriata Say is a Pachycheilus and is not found in Texas. All the Texan records of that species refer to comalensis. The var. marmocki mentioned, but not described by Wetherby (Am. Nat. XII, 1868, p. 254) seems to have been abandoned by him. His shells came from Helotes, Bexar Co., Texas and the set of “pleuristriata” in his collection, now in my possession, includes both the striate and smooth forms. There are none labelled var. marmockt. CATALOGUE RECENTLY DEscRIBED MoLtusca 157 GONIOBASIS COMALENSIS FONTINALIS Pilsbry and Ferriss. Goniobasis comalensis fontinalis Pilsbry and Ferriss, Pr. A. N. S. P., 1906, p. 198, figs. 32-35. Type locality: Spring, New Braunfels, Texas. GONIOBASIS CONNESAUGAENSIS Lea. Goniobasis connesaugaensis Lea, Pr. A. N.S. P., 1868, p. 152; Jour. A. N. See VV lelovenp.-233;, pl. 54, fie. 193 Obs: X11 1808, pe 93; pls; fig. II. Type locality: Connesauga Creek, Whitfield Co., Ga. GONTOBASIS CONTIGUA Lea. Gomiobvasis contigua ea,.Pr. A. N.S. P., 1868, p.-152; Jour. A. N.S. P., VI, 1868, p. 334, pl. 54, fig. 12; Obs., XII, 1868, p. 94, pl. 54, fig. 12. Type locality: Connesauga Creek, Whitfield Co., Ga. GONIOBASIS CRANDALLI Pilsbry. Goniobasis crandalli Pilsbry, Pr. A. N. S. P., 1890, p. 301, pl. V, fig. 4-5. Type locality: Mammoth Spring, Ark. GONIOPASIS CUMBERLANDENSIs Lea. Goniobasis cumberlandensis Lea, Pr. A. N. S. P., 1863, p. 155; Jour. A. N. poets V1; Peoo,, p..132, pl..23; fie (35 ;'Obsi xh 1866; pr88"pl. 39, fig. 35. Tryon (134, p. 272) considers this a synonym of G. adusta (Anth.), and erroneously quotes it as G. cumberlandienstis GONIOBASIS DECAMPII Lea. Goniobasis decampi Lea, Jour. A. N. S. P., VI, 1866, p. 138, pl. 23, fig. 45; Obs., XI, 1866, p. 94, pl. 23, fig. 45. GONIOBASIS DEPYGIS (Say). Is considered a variety of livescens by Baker (4, p. 327) and Sterki (124A, p. 385), but is distinct. CONIOBASIS EMERYENSIS Lea. Gontobasis emeryensis Lea, Pr. A. N: S. P., 1864, p. 3; Jour. A. N. S. P., Wi, 1606, p. 1127, pl. 23, fir..43; Obs., XI, 1866, p. 93, pl. 23, fiz. 43. Type locality: Rocky Creek, Head Branch of Emery River, Tenn. 158 Bryant WALKER GONIOBASIS FRATERNA Lea. Goniobasis fraterna Lea, Pr. A. N.S. P., 1864, p. 111; Jour. A. N. S$. P., VI, 1866, p. 139, pl. 23, fig46; Obs., XI, 1806, pis95, ph. 23, fie: 46: Type locality: Bibb Co. and Cahawba River, Ala. GONIOBASIS GESNERI Lea. Gomiobasis gesnert Lea, Pr. A. N.S. P., 1868, p. 151; Jour. A. N. 8. P., VI, 1868 p. 330, pl. 54, fig. 5; Obs., XII, 1868, p. go, pl. 54, fig. 5. Type locality: Uchee River, Ala. GONIOBASIS GOULDIANA Lea. Goniobasis pulchella Lea, Pr. A. N.S. P., 1868, p. 151, non pulchella Anth. (1850). Goniobasis gouldiana Lea, Jour. A. N.S. P., VI, 1868, p. 332, pl. 54, fig. 9; Obs., XII, 1868, p. 92; pl.-54, fig. o. Type locality: North Alabama. GONIOBASIS GRANATOIDES Lea. Goniobasis granatoides Lea, Pr. A. N. 5. P., 1868, p. 152; Jour. A. N.S. P., VI, 1868,:p: 335; pl. 54, fig. 14; Obs, XII, 1868, p. o5) pl. 54, ie. 4 Type locality: Connesauga Creek, Whitfield Co., Ga. GONIOBASIS INDIANENSIs Pilsbry. Goniobasis indianensis Pilsbry, 27th Ann. Rep. Dep’t. Geol. and Nat. Res. Ind., 1903, p. 606, fig. 28. Type locality: Blue River, Wyandotte, Crawford Co., Ind. GONIOBASIS LASVIGATA (Lea). This species, if a valid one, will have to be called leaii Brot as Melania levigata is preoccupied by Lamarck. GONIOBASIS LAQUEATA (Say). Includes: deshayesiana (Lea) and “probably” also cerea, corrugata, cos- tulata, cinerella, circincta, lyonii, plicatula, rugosa and sparus, all of Lea and athleta and glauca of Anthony according to Pilsbry (111, p. 499). GonropAsis LAWRENCE! Lea. Goniobasis lawrence Lea, Pr, A. N. S. P., 1869; p. 125; Jour. A. N.S. P., VIII, 1874, p. 62, pl: 21, fig. 17; Obs., XH, 1874;:p66, pli 21, ne. 17: Type locality: Washita River, Hot Springs, Ark. - CATALOGUE RECENTLY DeEscrisED MoLLusca 159 GONIOBASIS LUTEOCELLA Lea. Gontobasis luteocella Lea, Pr. A. N.S. P., 1868, p. 151; Jour. A. N. S. P., VI, 1868, p. 332, pl. 54, fig. 10; Obs., XII, 1868, p. 92, pl. 54, fig. 10. Type locality: Connesauga Creek, Whitfield Co., Ga. and Oconee River. GONIOBASIS MILESII Lea. Gomiobasis miles Lea, Jour. A. N.S. P., VI, 1866, p. 135, pl. 23, fig. 40; Obs., XI, 1866, p. 90, pl. 23, fig. 40. GONIOBASIS MURRAYENSIS Lea, Pr. A. N. S. P., 1868, p. 152; Jour. A. N. 8. P., VI, 1868, p. 334, pl. 54, fig. 13; Obs., XII, 1868, p. 92, pl. 54, fig. 13. ‘Type locality: Swamp Creek, Whitfield Co., Ga. GONIOBASIS NICRINA Lea. Includes californica (Cless.) and draytonit Lea and perhaps should be included in bulbosa (Gld.) according to Pilsbry (95, p. 66). GONIOBASIS ORNATA Lea. Goniobasis ornaia-Lea, Pr. A: N. S..P., 1868, p. 152; Jour. A. N.S. P., VI, T8668, p.. 337, pl. 54; fig: 18; Obs., X11, 1868, p..97, pl: 54, fig. 18. Type locality: Connesauga Creek, Whitfield Co., Ga. GONIOBASIS OSCULATA Lea. Erroneously printed as “inosculata’” by Tryon (134, p. 302) but corrected in the errata. Is the type by designation of Goniobasis Lea. GONIOBASIS OZARKENSIS Call. Goniobasis ozarkensis Call, Bull. Washburne Coll., I, 1886, p. 189, pl. 7, fig. I-10. Type locality: Blue Spring, Shannon Co., Mo. GONIOBASIS PENNSYLVANICA Pilsbry. Goniobasis pennsylvanica Pilsbry, Naut., XXX, 1916, p. 4. Type locality: Ohio River, Coraopolis, Pa. GONIOBASIS PLEBEIUS Anthony. Is a valid species and includes cubicoides (Anth.) according to Pilsbry (97, p- 458). 160 BryaANt WALKER GONTOBASIS PLICATA-STRIATA Wetherby. Goniobasis plicata-striata Wetherby, Jour. Cin. Soc. Nat. Hist., 1876, p. 10, pial, te 3; Type locality: Stone River and Mill Creek, Rutherford Co. and Sinking Creek, Shelbyville, Tenn. GONIOBASIS PLICIFERA Lea. Includes silicula (Gld.), rudens (Rve.), bairdiana Lea and shastensis (Lea) according to Pilsbry (95, p. 66). GONIOBASIS PLICIFERA BULIMOIDES Tryon. Goniobasis plicifera bulimoides Tryon, Am. J. of Con., I., 1865, p. 238, pl. 24, figs. 5-6. Type locality: Wahlamat River, Ore. GONIOBASIS PLICIFERA OREGONENSIS Tryon. Goniobasis plicifera oregonensis Tryon, Am. J. of Con., I., 1865, p. 238, pl. 24, fig. 4. Type locality not specified. GONIOBASIS PORRECTA Lea. Goniobasis porrecta Lea, Jour. A. N.S. P., VI, 1866, p. 139, pl. 23, fig. 47; Obs. 20 1866, p. 25. pli a3, ie. 47. GONIOBASIS PROXIMA (Say). Includes symmetrica as a variety according to Pilsbry (111, p. 499). Reeve in his errata states that the shell figured by him (Fig. 275) as this species does not represent it according to Anthony. The figure is not cited by Tryon and has not, apparently, been identified. GONIOBASIS PULLA Lea. Gomobasis pulla Lea, Pr. A. ‘N.S. P.,.1864, p. 112; Jour. A:.N.1S..P., VIL 1866; p. 130, pl..23, fie. 325-Obs., Xi5 1860: pGb, playsets sae: Type locality: Cumberland Gap, Tenn. GONIOBASIS PUPAFORMIS Lea. Goniobasis pupeformis Lea, Pr. A. N.S. P., 1864, p. 112; Jour. A. N.S. P., VI, 1866;.p:.130, pl. 23. fig..31;.Obs.;. 261, 1866, -p.86,, ply 23. fies sre Type locality: Coosa River, Ala. CaTALOGUL RECENTLY DEscRIBED MoLLusca 161 GONIOBASIS ROM Lea. Goniobasis rome Lea, Pr. A. N. S. P., 1864, p. 111; Jour. A. N. S. P., VI, 1866, p. 129, pl. 23, fig. 30; Obs., XI, 1866, p. 85, pl. 23, fig. 30. Type locality: Rome, Ga. GONIOBASIS RUFESCENS (Lea). Reeve in his errata states that he had been informed by Anthony that the shell figured by him as this species (Fig. 279) does not represent the species. This figure is not cited by Tryon and, apparently, has not been identified. GONIOBASIS SIMILIS Lea. Gomiobasis similis Lea, Pr. A. N. S. P5, 1868, p. 151; Jour. A. N. S. P., VI 1868, p. 328, pl. 54, fig. 2; Obs., XII, 1868, p. 88, pl. 54, fig. 2. Type locality: Connesauga Creek, Whitfield Co., Ga. GONIOBASIS SMITHSONIANA Lea. Goniobasis smithsomana Lea, Pr. A. N. S. P., 1864, p. 112; Jour.,, A. N. Pei, 1OO0 137. pl..22, figs 44:3-Obs,, xt 1800; pz 93,. pl. 23, fig. 44. Type locality: North Georgia and East Tennessee. GONIOBASIS STEARNSIANA Call. Goniobasis stearnsiana Call, Pr. Davenport Acad. Nat. Sci., V, 1886, p. 6, fig. 43. Type locality: Dyke’s Creek, Floyd Co., Ga. GONIOBASIS SUBRHOMBICA Lea. Gomobasis subrhombica Lea, Pr. A. N. S. P., 1864, p. 111; Jour. A. N. Pak. Vi, 1600p. 132).ply 23, fig.-34;-Obs. Xt, 1866; p: 88, pl..23; fig. 34. Type locality: Hog Creek, North Georgia. GONIOPASIS SULCATA Lea. iSamovasis suicata lea, Pr. A. N: S: P., 1868, p. 151; Jour. A. N. S. P., ie 1868, p. 320, pl. 54, fig. 3; Obs., XII, 1868, p. 89, pl. 54, fig. 3. Type locality: Cahawba River, Ala. GONIOBASIS UNDULATA Tryon. Goniobasis undulata ‘Tryon, Am. J. of Con., II, 1866, pl. II, fig. 4. Type locality: Georgia. 162 Bryant WALKER GONIOBASIS VENUSTA Lea. Gontobasis venusta Lea, Pr. A. N.S. P., 1868, p. 152; Jour. A. N. S. P., VI, 1868, p. 336, pl..54, fig. 17; Obs., XII, 1868, p. 96, pl. 54, fig. 17. Type locality: Coosa River, Ala. GONIOBASIS VERSA Lea. Melania blanda Wea, Pr. A. N. 8. P., 1861, p. 122, non blanda Lea (1841). Goniobasis blanda Lea, Jour. A. N. S. P., V, 1863, p. 242, pl. 35, fig. 44; Obs., IX, 1863, p. 64, pl. 35, fig. 44. Goniobasis versa Lea, Jour. A. N. S. P., VI, 1866, p. 127; Obs., XI, 1867, Pp. 93. Goniobasis leai Tryon, L. and F. W. Shells, Pt. IV, 1873, p. 163, non Brot (1862-8). Type locality: Yellowleaf Creek, Ala. GONIOBASIS VIRGINICA Gmelin. Add to the synonymy, Gon. virginica bilirata De Gregorio (35, p. 39). GONIOBASIS VIRIDISTRIATA Lea. Goniobasis viridistriata Lea, Pr. A. N. S. P., 1864, p. 4; Jour. A. N.S. P., VI, 1866,-p. 140, pl. 23; fig 48; Obs., XI, 1866, p. O06; pl) 235 fig 48. Type locality: Flint River, Ga. — GONIOBASIS WHEATLEYI Lea. Gontobasis wheatley: Lea, Pr. A. N. S. P., 1868, p. 151; Jour. A. N. S. P., Wii-1 868, p-.328, pl. 54, fig. 13,Obs., X11 1868.0 p. Senpl 545, tesa Type locality: Coosa River, Ala. GONIOBASIS WHITFIELDENSIS Lea. Goniobasis tenebrcsa Lea, Pr. A. N. S. P., 1868, p. 151, non Mel. tenebrosa Lea (1841). Goniobasis whitfieldensis Lea, Jour. A. N. S. P., VI, 1868, p. 330, pl. 54, fig? 6;"Obs., WL, T1868, p: 00, ply 54 ne 0. Type locality: Connesauga Creek, Whitfield Co., Ga. Genus GYROTOMA Shuttleworth, 1845. Schizostoma Lea (1842), non Brown (1835). GyROTOMA LEWISII (Lea). Schizotoma lewisu Lea, Pr..A. N. S. P.; 1869, p. 125; Jour. A. N. S. P., VIN, 1874, p..61, pl. 25, fig. 16; Obs:, XIII, 18745 p.65, plo er, figs 1G. Type locality: Coosa River, Ala. CATALOGUE RECENTLY DEscRIBED Mollusca 163 GyROTOMA SHOWALTERI (Lea). Schizostoma showalteri Lea, Pr. A. N.S. P., 1864, p. 112; Jour. A. N.S. P., VI, 1866, p. 149, pl. 23, fig. 56; Obs., XI, 1866, p. 105, pl. 23, 56. This is a different species from that described by Lea under the same name in 1860, which proved to be a synonym of G. cariniferum (Anth). The name being preoccupied, this species will have to be re-named, if it should prove to be a valid one on a revision of the genus. GYROTOMA WHEATLEYI (Lea). Schizostoma wheaileyw: Lea, Pr. A. N.S: P.; 1868, p. 153; Jour. A. N.S. P., Pil Looe. py 842, pl 5A; ies 275 Obs. X11 1868, ps 102; pl. 54, fig. 27. Type locality: Coosa River, Ala. Genus ANCULOSA Say, 1821. Pilsbry has recently come to the conclusion (105, p. 109) that Leptoxis Raf. should be preferred for this genus. In my paper on Pleurocera (161, p. I) I reluctantly acquiesced in his conclusion on the ground that the ques- tion was zoological rather than one of Code construction. ‘Since that time I have again gone carefully over the ground and have become satisfied that, if all incompetent evidence is eliminated, there is not sufficient grounds to justify the change. I have therefore retained Say’s well known name for the genus. ANCULOSA ARKANSENSIS Hinkley. Anculosa arkansensis Hinkley, Pr. U. $. Nat. Mus., 49, 1915, p. 587, pl. 78, fie 3: Type locality: North Fork of White River, above Norfolk, Ark. ANCULOSA DOWNIEI Lea. Anculosa downie: Lea, Pr. A. N.S. P., 1868, p. 153; Jour. A. N. S. P., VI, 1868, p. 342, pl. 54, fig. 28; Obs., XII, 1868, p. 102, pl. 54, fig. 28. Type locality: Connesauga Creek, Whitfield Co., Ga. and Coosa River, Ala. ANCULOSA HARPETHENSIS Pilsbry. Anculosa harpethensis Pilsbry, Pr. A. N. S. P., 1896, p. 499. Type locality: Big Harpeth River, Bellevue, Tenn. ANCULOSA MINOR Hinkley. Anculosa minor Hinkley, Naut., XX VI, 1912, p. 47, pl. I, figs. 7-8. Type locality: Tennessee River, Florence, Ala. 164 Bryant WALKER ANCULOSA SUBGLOBOSA Say. Does not include A. tintinnabulum Lea and virgata Lea as stated by Tryon (134, p. 404). See Walker (150, p. 110). ANCULOSA TINTINNABULUM Lea. Is a valid species and does not include virgata Lea. See Walker (1. c.). ANCULOSA TRYONI Lewis. ' Anculosa tryoni Lewis, Am. J. of Con., VI, 1870, p. 221, pl. 12, fig. 8. Type locality: Holston River, Tenn. Tryon (133, p. 87) considers this to be the teniata Con., but the approx- imation is at the best very doubtful. ANGULOSA UMBILICATA Wetherby. Anculosa umbilicata Wetherby, Jour. Cin. Soc. Nat. Hist., 1876, p. 11, pl. I, fig. 4. Type locality: Stone River, Rutherford Co., Tenn. ANCULOSA VIRGATA Lea. Is not the young of tintinnabulum as stated by Tryon (134, p. 404). but is a valid species. See Walker (150, p. 110). Genus MESECHIZA Lea. Has no standing. The type species is the young of Angitrema cerru- cosa. See Hinkley (57, p. 56). Family NERITID/L. Genus NERITINA Lamarck, 1809. NERITINA RECLIVATA Say. v. Martens (73, p. 472) considers that this is only a variety of N. lineo- lata Lam. NERITINA RECLIVATA PALM Dall. Nertiina reclivata palme Dall, Pr. U. S. Nat. Mus., 1885, p. 250. Type locality: Palma Sola, Fla. Genus LEPYRIUM Dall, 1896. Lepyrium Dall, Naut., X, 1896, p. 15. Type: Neritina showalteri Lea. LEPYRIUM SHOWALTERI CAHAWBENSIS Pilsbry. Lepyrium showalteri cahawbensis Pilsbry, Naut., XX, 1906, p. 51. Type locality: Cahawba River, Ala. CaTALOCUE RECENTLY DEscRIBED Mollusca 165 Class LAMELLIBRANCHIA. Order EULAMELLIBRANCHIA. Suborder SUBMYTILACEA. Family MARGARITANIDE. Ortmann (79, p. 223) has raised the genus Margaritana to the rank of a family and (80, p. 13) has proposed a new genus, Cumberlandia, for M. monodonta (Say), both based upon anatomical peculiarities For the distribution of the genus in this country, see Walker, Nos. 152 and 153, Ortmann (8o, p. 14) and Utterback (135, p. 99). MARGARITANA MARGARITIFERA (L.). Unio ocmulgeensis dominus De Gregorio, Moll. di aq. dul. di Amer., 1914, p. 13, pl. 7, fig. a-c. Family UNIONID.. Simpson's “Descriptive Catalogue of the Naiades” brings the subject down to January I, 1913. For the revised classification so far as it has progressed, see Part I. In view of the relatively small number of species that have been exam- ined anatomically and the consequent element of uncertainty as to the sys- tematic position of the remainder that must continue until the animals can be critically examined, it has seemed better, for convenience of reference, in this porton of the work to retain the generic names given by Simpson, noting, however, under such species as have been examined anatomically their proper place in the revised classification. Recent attempts to revive Rafinesque’s names for many of the species have created considerable confusion as to the proper nomenclature to be followed. Vanatta’s valuable paper on “Rafinesque’s Types of Unio” (140, p. 549), reviewed by Walker (158, p. 43), has given definite information as to what Rafinesque in 1831 understood or claimed to be the species that he had described in 1820. It has been too hastily assumed by some that these determinations have definitely settled the validity of all of the Rafinesqueian species involved. This is far from correct. It is not claimed, except in one instance, that the so-called types in the Poulson collection are the original types of Rafinesque. And, even if they were, reference to them for the purpose of determining an otherwise unidentifiable description is prohibited by the International Code (Op. Int. Co., 1). The requisites for a sufficient description are defi- nitely specified by the Code (Art. 25) and these provisions as defined by the decisions of the International Committee must be applied to each indi- vidual case. 166 BrYANT WALKER Dr. Pilsbry in Vanatta’s paper has very aptly stated the situation as fol- lows: “The use of a Rafinesquian name depends upon whether it could be identified by descriptions published prior to any other recognizable name for the same species. That it can be recognized from the types or other specimens from Rafinesque does not entitle his names to acceptance unless the published descriptions are adequate. This question of the adequacy of published diagnoses must be considered for each species separately.” In the same connection, see Walker, (157, p. 74). Subfamily UNIONINA® (Swainson, 1840), Ortmann, 1910. Genus QUADRULA Rafinesque, 1820. QUADPRULA ASKEWI (Marsh). Frierson (41, p. 136) refers this to beadleiana Lea. But Ortmann (81, p. 21) states that it does not group with that species, but is a Fusconaia of the widata group. QUADRULA BEADLEIANA (Lea). Includes Q. chickasawhensis (Lea) and askewi (Marsh) according to Frierson (41, p. 136). But Ortmann (79, p. 268) says that it is an Elliptio. QUADRULA BURSAPASTORIS (B. H. Wright). Is a Fusconaia according to Ortmann (81, p. 90). QUADRULA COCCINEA (Conrad). Is a Pleurobema according to Ortmann (77, p. 101) and a variety of QO. obliqua (Lam.) (78, p. 117 and 79, p. 263). Utterback (135, p. 190) quotes catillus Con., which Simpson has con- sidered a synonym of Q. coccinea, as a variety of Q. obliqua (Lam.) and on p. 193 of the same paper considers it identical with Q. solida (Lea), having priority and gives it specific rank as such. QUADRULA COOPERIANA (Lea). At first referred to Pleurobema by Ortmann (78, p. 117), this species is now included in Plethobasus by him (79, p. 261). QuUADRULA CYLINDRICA (Say). Unio cylindricus propetypicus De Gregorio, Moll. di aq. dul. di Amer., 1914, p. 11, pl. 4, fig. 1. Unio cylindricus acrispatus De Gregorio, Ibid, p. 11, pl. 4, fig. 2. Vanatta (140, p. 556) states that the Unio solenoides Raf. of the Poul- son collection is this species. CATALOGUE RECENTLY DeEscriBED MoLLusca 167 QUADRULA EBENUS (Lea). Is Obovaria obovalis Raf. of the Poulson collection according to Van- atta (140, p. 558). If identifiable from the origina! description, obovalis would have precedence. QUADRULA FRIERSONI (B. H. Wright). Is a Pleurobema according to Ortmann (81, p. 30). QUADRULA HEROS (Say). This species is the type of Megalonaias Utterback. Frierson (45, p. 61) has identified Barnes’ Unio giganteus as this species and gives it priority. QUADRULA INTERMEDIA (Conrad). Unio tuberosus perlobatus De Gregorio, Moll. di aq. dul. di Amer., 1914, p. 9, pl. I, fig. 3. QUADRULA KIRTLANDIANA (Lea). Is a variety of Q. subrotunda (Lea) according to Ortmann (78 p. 116). QUADRULA LACHRYMOSA (Lea). This species has been identified as the Obliquaria quadrula Raf. by Say, Conrad and others and Vanatta (140, p. 556) states that the shell so labelled in the Poulson collection is Lea’s asperrimus. If identifiable from the orig- inal description, it would have priority. QUADRULA LACHRYMOSA CONTRARYENSIS Utterback. Ouadrula lachrymosa contraryensis Utterback, Amer. Mid. Nat., IV, 1915, p. 138, pl. XVIII, figs. 47a-b. Type locality: Lake Contrary, St. Joseph, Mo. PLEUROBEMA MISSOURIENSIS (Marsh). The type of this species has been figured by Walker (155, p. 140, pl. V, figs. 1-2) and it appears to be a Ouadrula allied to Q. subrotunda (Lea). QUADRULA OBLIQUA (Lam.). Includes pyramidata Lea, cocinea Con., and plena Lea according to Ort- mann (78, p. 117 and 79, p. 264) and isa Pleurobema. Vanatta states (140, p- 557) that the Obliquaria lateralis Raf. of the Poulson collection is this species. 168 Bryant WALKER QUADRULA PERUVIANA (Lamarck). Lamarck in his original description refers to the figure in the Encyc. Meth., pl. 248, fig. 7, but the reference was overlooked by Simpson in his Synopsis, but was supplied in the Desc. Catalogue. The species is the form commonly called plicata Say by collectors and is characterized by its prom- inent beaks. It is quite different from plicata Say from Lake Erie. QUADRULA RARIPLICATA (Lamarck). This species, which has been referred to plicata Say by Simpson and authors generally, is neither typical peruviana (Lam.) nor typical plicata (Say). The type which 1s still preserved in the Museum at Geneva, Switzer- land, is the Ohio River form, which has commonly passed as plicata (Say), and is sufficiently distinct to have varietal rank at least. ‘There is some reason to believe that Say’s plicata is an off-shoot from this race rather than of undulata Bar. as has been suggested by Ortmann (79, p. 246). It is also possible that it rather than wndulata should be considered the costata Raf. QUADRULA PLICATA (Say). As stated by Ortmann (70, p. 246) the type of this species came from Lake Erie and is undoubtedly the form described by Lea as Unio hippopeus. It has been referred to undulata Bar. by Ortmann (1. c.), but there is appar- ently some ground for considering it as more closely allied to rariplicata (Lam.).. Pending the settlement of this question, it would seem better to keep it separate from either. QUADRULA PLENA (Lea). According to Ortmann (78, p. 117) this is probably only a form of obliqua (Lam.). Vanatta states (140, p. 558) that the Obovaria cordata Raf. of the Poulson collection is this species. If identifiable from the orig- inal description, Rafinesque’s name would have priority. QUADRULA PUSTULATA (Lea). According to Vanatta (140, p. 557) the Obliquaria nodulata Raf. of the Poulson collection is this species. If identifiable from the original descrip- tion, nodulata would have precedence. QUADRULA PUSTULOSA (Lea). According to Vanatta (140, p. 5560) the Obliquaria retusa Raf. of the Poulson collection is “probably” this species. The specific name is not pre- occupied by Unio retusa Lam. and, if identifiable from the original descrip- tion, Rafinesque’s name would have priority. Utterback (135, p. 131) has suggested that the species should be known as bullata Raf., but see next note. CATALOGUE RECENTLY DEscriBED MoLLusca 169 QUADRULA PUSTULOSA PERNODOSA (Lea). According to Vanatta (140, p. 557) the Obliquaria bullata Raf. of the Poulson collection is this form, but the name is preoccupied by Obliquaria fleruosa bullata Raf. and Lea’s name will stand. QUADRULA PYRAMIDATA (Lea). Unio plenus interduos De Gregorio, Moll. di aq. dul. di Amer., 1914, p. 18. According to Ortmann this is probably only a form of obliqua (Lam.). Vanatta states (140, p. 557) that the Obliquaria rubra Raf. of the Poulson collection is this species. If identifiable from the original description, rubra would have priority. QUADRULA REFULGENS (Lea). Includes spherica (Lea) according to Frierson (41 p. 136). QUADRUL,A RUBIGINOSA (Lea). Unio validus continuus De Gregorio, Moll. di aq. di Amer., 1914, p. 21. This species has been identified as the Obliquaria flava Raf. by Say, Conrad and others. According to Vanatta (140, p. 557) the O. flava Kaf. of the Poulson collection is this species. If identifiable from the original de- scription, Rafinesque’s name would have priority. According to Ortmann (78, p. 116) rubiginosa is a variety of undata (Bar.). FuscoNaAtaA SELECTA Wheeler. Fusconaia selecta Wheeler, Naut., XXVIII, 1914, p. 76, pl. IV. Type locality: Cache River, Nemo, Craighead Co., Ark. QuUADRULA SPHRICA (Lea). Is a synonym of refulgens (Lea) according to Frierson (41, p. 136). QUADRULA SUBROTUNDA (Lea.) According to Vanatta (140, p. 558) the Obliquaria sintovia Raf. of the Poulson collection is this species. If identifiable from the original descrip- tion, sinto.ia would have precedence. The species is a Fusconaia according to Ortmann (79, p. 244). FUSCONAIA SUBROTUNDA LEUCOGONE Ortmann. Fusconaia subrotunda leucogone Ortmann, Naut., XXVII, p. 89. Type locality: Elk River, Gassaway, Braxton Co., W. Va. 170 Bryant WALKER QUADRULA ‘TRAPEZOIDES (Lea). Bariosta ponderosus Raf. is a synonym and Bariosta Raf. is a synonym of Amblema Raf., unless the species should prove to be generically distinct according to Frierson (42, p. 7). FuscoONAIA UNDATA TRIGONOIDES ‘‘Frierson”’ Utterback. Fusconaia undata trigonoides “Frierson” Utterback, Amer. Mid. Nat., IV, IQI5, p. 107, pl. XV, figs. 30A-D. Type locality: Platte River, Agency Ford, Mo. QUADRULA UNDULATA (Barnes). This species has been identified as the Amblema costata Raf. by Conrad. Frierson and others. Vanatta states (140, p. 556) that the Amblema costata Raf. of the Poulsen collection is also this species. If identifiable from the original description, Rafinesque’s name would have priority. In consider- ing this question attention should be given to the possibility that costata Raf. may be the rartplicata Lam. Costata has been designated by Frierson (42, p. 7) as the type of Amblema Raf. QUADRULA UNDULATA PILSBRYI (Marsh). According to Utterback (135, p. 119) this is a synonym of Q. perplicata quintardu (Cragin). Genus TRITOGONIA Agassiz, 1852. The recent discovery of Sterki (’07, p. 48) that in the type species, T. tuberculata, all four of the gills are utilized for marsupia, removes the genus from the Digene of Simpson to the Tetragene. But in view of the re- markable dimorphism of the species, which is apparently a sexual and not a senile charcter as has been suggested by Ortmann, the subordination of the genus to Quadrula as proposed by him would seem to be inexpedient. TRITOGONIA TUBERCULATA (Barnes). Ouadrula tritogonia Ortmann, Naut., XXII, 1909, p. 101. Quadrula parkeri Geiser, The Academician I, 1911, p. 15. The new names proposed by Ortmann and Geiser can not be used, even if the species is referred to Quadrula. If Rotundaria, with tuberculata Raf. as its type, be given generic rank, Barnes’ name can still be used in Quad- rula (Ortmann, 78, p. 116); but if not, the species would take the name of obesa Simp. (Vanatta, Naut., XXIII, 1910, p. 102). Obliquaria verrucosa Raf. is identified as this species by Conrad and the shells so labelled in the Poulson collection are also that species according to Vanatta (140, p. 554). If identifiable from the original description Rafinesque’s name has priority. CATALOGUE RECENTLY DrEscrirnED MoLLusca 171 Genus AMBLEMA Rafinesque, 1820. Amblema Rafinesque, Monographie, 1820, p. 314. Crenodcnta Schluter, Verz. meiner Conch., 1836, p. 33; Simpson, Syn., 1900; p. 766; Desc: Cat., 1914, p. 813. Type: Amblema costata Raf. Genus MEGALONATIAS Utterback, 1915 Meg-lonaias Utterback, Amer. Mid. Nat., IV, 1915, p. 123. Type: Unio heros Say. Genus ROTUNDARIA Rafinesque, 1820. Rotundaria Rafinesque, Monographie, 1820, p. 308; Simpson, Syn., 1900, p. 794; Desc. Cat., 1914, p. 903. Type: Obliquaria tuberculata Raf. Genus FUSCONATA Simpson, rgoo. Fusconaia Simpson, Syn., 1900, p. 784; Desc. Cat., 1914, p. 865. Type: Unio trigonus Lea. Genus PLETHOBASUS Simpson, 1900. Plethobasus Simpson, Syn., 1900, p. 764; Desc. Cat., 1914, p. 895. Type: Unio esopus Green. Genus PLEUROBEMA Rafinesque. PLEUROBEMA #SOPUS (Green). This species has been referred to Obliquaria cyphya Raf. by Conrad, Call, Ortmann and others. Vanatta (140, p. 556) states that the shell in the Poulson collection so labelled is this species. If identifiable from the original description, cyphya would have priority. This species is the type of Simpson’s section Plethobasus, which Ort- mann (70, p. 259) has raised to generic rank. PLEUROBEMA ARGENTEUM PANNOSUM Simpson. This is a Fusconaia and a synonym of F. ozarkensis (Call) according to Ortmann (84, p. 63). ~ PLEUROBEMA BARNESIANUM (Lea). Issa Fusconaia and includes meredithii Lea, pudicum Lea, lyoni Lea, tellicoensis Lea and lenticulare Lea according to Ortmann (84, p. 59). 172 Bryant WALKER PLEUROBEMA BIGBYENSE (Lea). Is a Fusconaia and a variety of F. barnesiana (1,ea) and includes esta- brookianum Lea, fassinans Lea and fassinans rhomboidea Simp. according to Ortmann (84, p. 59). PLEUROBEMA BREVE SUBELLIPTICUM Simpson. Is a Fusconaia and a synonym of F. ozarkensis (Call) according to Ort- mann (84, p. 63). PLEUROBEMA CLAVUS (Lamarck). Unio consanquineus De Gregorio, Moll. di aq. dul. di Amer., 1914, p. 16. Unio anaticulus ohiensis De Gregorio, Moll. di aq. dul. di Amer., 1914, p. Binal. TO hea Vanatta (140, p. 555) states that Unio elliptica Raf., Obliquaria scal- enia Raf. and Pleurobema cuneata Raf. as labelled in the Poulson collection are this species and expresses the opinion that Pleurobema mytiloides Raf. is also. PLEUROBEMA CONRAD? Vanatta. This species was originally described by Conrad as Unio maculatus, but, as shown by Vanatta (140, p. 559), that name had already been used by Rafinesque for a variety of his Unio nigra and he has proposed conradi as a specific name for the Pl. maculatum (Con.) of Simpson’s Desc. Catalogue. PLEUROBEMA cor (Conrad). The types of this species came from the Flint and Elk rivers in northern Alabama and as Frierson (44, p. 102) has shown is closely related to, if ~ not identical with, edgarianum Lea or some other species of that group. PLEUROBEMA CRUDUM (Lea). Is a synonym of Fusconaia barnesiana tumescens (Lea) according to Ortmann (84, p. 59). PLEUROBEMA ESTABROOKIANUM (Lea). Is a Fusconaia and a synonym of F’. barnesiana bigbyensis (Lea) accord- _ ing to Ortmann (84, p. 59). PILEUROBEMA DOLLABELLOIDES (Lea). Unio tornhatoni duckensis De Gregorio, Moll. di aq. dul. di Amer., 1914, p. 20, pl; Ta hies.-2aze. CaTALOGuT RECENTLY DeEscriBED MoLLusca 173 PLEUROBEMA FASSINANS (Lea). This species and its variety rhomboides Simp. are Fusconaias and syn- onyms of F. barnesiana bigbyensis (Lea) according to Ortmann (84, p. 59). PLEUROBEMA LENTICULARIS (Lea). Is a Fusconaia and a synonym of F. barnesiana (Lea) according to Ort- mann (84, p. 59). PLEUROBEMA LEWISII (Lea). That this species is distinct from P. cor (Con.), to which it was referred by Simpson, has been shown by Walker (160, p. 114) and U. crapulus Lea, if not distinct, is made a synonym. PLEUROBEMA MEREDITHIL (Lea). Is a Fusconaia and a synonym of F. barnesiana (Lea) according to Ort- mann (84, p. 59). LAMPSILIS OZARKENSIS (Call). Is a Fusconaia and includes Pleurobema argenteum pannosum Simp., breve subellipticum Simp. and utterbacki Fr. according to Ortmann (84, p. 63). PLEUROBEMA PUDICUM (Lea). Is a Fusconaia and a synonym of F. barnesiana (Lea) according to Ort- mann (84, p. 59). PLEUROBEMA SIMPSONI Vanatta. Pleurobema simpsoni Vanatta, Pr. A. N. S. P., 1915, p. 559. Originally described as Unio striatus Lea. Lea’s name is not preoccu- pied by Rafinesque, but as suggested to me by Frierson it seems to have been by Goldfuss for a fossil species. I have not been able to examine, or to have examined, Goldfuss’ original description. Lea refers to Goldfuss’ species in his Synopsis, but as usual gives no exact citation. A paleonto- logical friend has supplied the following references, which seem sufficient to settle the question. Goldfuss’ description of his Unio striatus is to be found in his “Pertrefakten Deutschlands,” II, 1839, p. 182, pl. 132, fig. 3. Bronn in his “Index Paleontologicus,’ II, p. 1345, includes the species among his “omnia dubii generis’. D’Orbigny in his “Prodrome de Paleon- tologie” includes it in his genus Hesione (1847). These facts were probably known to Lea and explain why he did not rename his species, as in other 174 Bryant WALKER instances he had claimed that the reference of the prior species in such cases to another genus “liberated” his own subsequent name. This under the Code is quite erroneous. PLEUROBEMA UTTERBACKI Frierson. Pleurobema utterbacki Frierson, Amer. Mid. Nat., IV, 1915, p. 197, pl. V, figs. I2a-b and pl. XX, figs. 63a-d. Type locality: ‘White River, Hollister, Mo. This species is a Fusconaia and a synonym of F. ozarkensis (Call) ac- cording to Ortmann (84, p. 63). Genus LEXINGTONIA Ortmann, 1914. Lexingtonia Ortmann, Naut., XXVIII, 1914, p. 28. Type: Unio subplanus Conrad. “This genus stands near Pleurobema and Elliptio and differs from either chiefly by the subcylindrical, red placente, and by the beak sculpture.” Genus UNIO Retzius, 1788. UNIO coMPLANATUS (Dill). Unio pullatus majusculus De Gregorio, Moll. di aq. dul. di Amer., 1914, p. 27, pl. 8, figs. a-d. Haas has recently (50, p. 54) figured the original type of Spengler’s Unio violaceus, which appears to be an abnormal specimen of this species and proposes to give precedence to Spengler’s name. I have shown else- where (162, p. 3) that this is not competent under the Code and that Dill- win’s name should be retained. UNIO CRASSIDENS Lam. According to Ortmann (79, p. 266) this is the Unio nigra Raf. Vanatta (140, p. 555) states that the shell so labelled in the Poulson collection is this species. If identifiable from the original description, Rafinesque’s name would have priority. Utterback (135, p. 199) has quoted me as authority for the statement that Quadrula trapezoides (Lea) should be considered as the Unio crassidens of Lamarck. I have expressed that opinion in corre- spondence and if the process of elimination could be strictly applied that would be the result. But Dr. Pilsbry has suggested that the provisions of the Code in regard to designating generic types apply equally well to the cases of composite species, of which this is an example, and that, if this rule does apply, Lea’s statement (Obs., I, p. 199) of his examination of Lamarck’s types in 1832 amounted to such a designation and can not now CATALOGUE RECENTLY DrEscRIBED Mollusca 175 be changed. Pending further consideration and final decision of the ques- tions involved, it would be better to allow the accepted identification of crassidens to stand. Unto cisposus Barnes. Unio propeverutus De Gregorio, Moll. di aq. dul. di Amer., 1914, p. 8, pl. 3, figs. Ia-c. This species has been referred to Unio dilatata Raf. by Conrad and oth- ers. Vanatta states (140, p. 355) that both U. dilatata Raf. and Obliquarta sinuata Raf. as represented in the Poulson collection are this species. If identifiable from the original descriptions, both of Rafinesque*s names have priority. Diulatata has page priority in his Monograph. UNIO PUSILLUS Lea. Lea’s name is not preoccupied in Unio by Obliquaria pusilla Raf. (1820) as stated by Vanatta (140, p. 555) and will stand. UNIO RAFINESOUEI Vanatta. Unio fuscatus Lea, Simpson, Desc. Cat., 1914, p. 643. Vanatta has shown (140, p. 559) that fuscata was twice used by Raf- inesque as varietal names for species of Unio and has proposed the name given above for Lea’s species. UNIO TUOMEYI Lea. Unio arctior fisheropsis De Gregorio, Moll. di aq. dul. di Amer., 1914, p. 15, pl. 5, figs. 3a-c. Genus LASTENA Rafinesque, 1820. From an examination of the soft anatomy, Ortmann has recently (81, p. 106) shown that this group belongs in the Unionine and not in the Ano- dontine. Hemistena Raf. is a synonym according to Frierson, (42, p. 7). LASTENA LATA Raf. Unio dzhiscens oriensopsis De Gregorio, Moll. di aq. dul. di Amer., 1914, p- 39, pl. 7, figs. 2a-b. This variety (?) is made the type of a new genus or subgenus, Sayuzio, the author does not seem to know which it should be considered. Genus GONIDEA Conrad, 1857. Ortmann has recently found from an examination of the soft anatomy (83, p. 50) that this genus belongs to the Unionine. 176 Bryant WALKER Subfamily ANODONTINA£ Ortmann, 1912. Genus STROPHITUS Rafinesque, 1820. STROPHITUS EDENTULUS (Say). Anodonta foliopsis De Gregorio Moll. di aq. dul. di et {OLA 1p. 295. pr XI, figs. 4a-b. There seems to be some uncertainty as to the proper name to be used for this species. Say described his edentulus in 1829. Swainson had al- ready in 1822 described his Anodon rugosus from the “United States.” Lea (Obs., I, p. 39) says that “it is well known” that Swainson’s rugosus is the adult of Say’s widulata, which has priority. Simpson, who considers un- dulatus and edentulus distinct, for some unexplained reason includes Swain- son’s species under edentulus as a synonym. Dall (32, p. 127) “on the face of the returns” gives the species to Swainson. I have not been able to ex- amine Swainson’s description and figure myself. Ortmann (78, p. 118) unites both species under the prior name of undulatus. If this is correct, the exact identity of Swainson’s species becomes immaterial. Otherwise his description and figure should be critically examined again to determine, if possible, to which species it belongs. STROPHITUS EDENTULUS SITEFFERIANUS (Lea). This seems to be a well marked race characteristic of the Tennessee drainage and as such entitled to varietal rank. Genus ANODONTA Lamarck, 1799. ANODONTA CATARACTA Say. Anodonta ? subcylindracea propexilis De Gregorio, Moll. di aq. dul. di Amer, TOL4-p. 33, pl. XT fies. ta-e. ANODONTA GRANDIS Say. Anodonta venusta De Gregorio, Moll. di aq. dul. di Amer., 1914, ps 25pph DO igs This species founded on a single deformed valve is made the LV DeLOr sa new subgenus, Nayadina. ANODONTA IMBECILIS Say. Anodonta phalena De Gregorio, Moll. di aq. dul. di Amer., 1914, “p» 34s ple XI, figs. 3a-e. ais This species is Lastena ohioensis Raf. and a Lastena according to Utter- back (135, p. 260). ANODONTA SUBORBICULATA Say. According to Utterback (135, p. 256) this species is a Lastena. CATALOGUE RECENTLY DescrrBED MoLLusca 177 Genus ANODONTOIDES Simpson, 1808. ANODONTOIDES FEUSSACIANUS (Lea). Anodonta ferussaciana incertopsis De Gregorio, Moll. di aq. dul. di Amer., 1914, p. 34, pl. XI, fig. 5. Genus LASMIGONA Rafinesque, 1831. Symphynota Simpson (non Lea), Syn., 1890, p. 662; Desc. Cat., 1914, p. 480. Frierson (Naut., XXVIII, 1914, p. 40) has shown that the original type of Lea’s Symphynota was Unio alatus Say and it is therefore a synonym of Proptera Raf. and that consequently Lasmigona Raf. as the earliest avail- able name becomes the generic type. Subgenus PLATYNAIAS Walker, 1918. Platynaias Walker, Occ. Pap., Mus. Zool., U. of M., No. 49, 1918, p. 1. Type: Symphynota compressa Lea. As the result of the disappearance of Symphynota Lea from this genus, the group typified by S. compressa Lea was left without any name and Platynaias has been proposed for it. LASMIGONA COMPRESSA (Lea). Unio compressa ? lindus De Gregorio, Moll. di aq. dul. di Amer., r9r4, p. 14, pl. 6, figs. ta-d. Frierson (43, p. 57) has argued that Rafinesque’s Unio viridis is this species and consequently has priority. This has been contested by Walker (157, p. 74). Vanatta (140, p. 554) states that Poulson’s shell labelled Unio viridis fuscata from the Kentucky River, is the Sym. viridis Con. of Simpson’s Synopsis. ‘This I have verified from a personal inspection of the shell. For the reason stated by Walker, (1. c., p. 78) Lea’s name is not superceded by alasmedontina Stimp. and will stand as the specific name. LASMIGONA virtpIs Rafinesque. For the conflicting opinions in regard to this species see the preceeding note. Under all of the evidence that has been adduced I think that Raf- -inesque’s name should be given precedence, with subviridis Con. (24, App. p. 4), viridis “Con.” Simp. and tappanianus Lea as synonyms. Subgenus ALASMINOTA Ortmann, 1914. Alasminota Ortmann, Naut., XXVIII, 1914, p. 41. Type: Margaritana holstonia Lea. Frierson (42, p. 7) has identified Rafinesque’s Alasmodon badium as this species and has designated it as the type of Sulcularia Raf. If the species is identifiable from the original description and is Lea’s holstonia, Sulcularia has precedence over Alasminota. 178 Bryant WALKER Genus ALASMIDONTA Say, 1818. Subgenus PRESSODONTA Simpson, 1900. I have recently (162, p. 2) proposed to supercede this name with that of Calceola Sw., 1840, on the ground of priority. Dr. Dall has since called iny attention to the fact that Calceola had already been used by Lamarck in 1799 for a coral. Simpson’s name will therefore stand. Subgenus PROLASMIDONTA Ortmann, 1914. Prolasmidonta Ortmann, Naut., XXVIII, 1914, p. 44. Type: Unio heterodon Lea. a Subgenus PEGIAS Simpson, 1900. According to Ortmann (81, p. 45) this group is a subgenus of Alusnii- donta. Subgenus RUGIFERA Simpson, 1900. ALASMIDONTA MARGINATA Say. Unio calceolus sciotincola De Gregorio, Moll. di aq. dul. di Amer., 1914, p. 24, pl. UX, figs 3: Frierson (42, p. 7) has identified Alasmodon scriptum Raf. with this species, but Say’s name has priority. ALASMIDONTA RAVENELIANA (Lea). Frierson (42, p. 7) has identified Alasmodon atropurpureum Raf. as be- ing this species. If identifiable from the original description, it has priority. On the basis of these identifications, he would substitute Decurambis Raf., 1831. for Rugifera Simp. as the subgeneric name. Genus SIMPSONICONCHA Frierson, 1914. Hemulastena Simpson, non Agassiz, Syn., 1900, p. 673; Desc. Cat., 1914, p. 223: Simpsonaias Frierson, Naut., XXVIII, 1914, p. 7. (Preoccupied.) Simpsoniconcha Frierson, Naut., XXVIII, 1914, p. 4o. Type: Alasmodonta ambigua Say. For note on this name, see Walker, 162, p. 4. Subfamily LAMPSILINA® Ortmann, 1912. Genus PTYCHOBRANCHUS Simpson. Frierson, having identified (42, p. 7) Obliquaria fasciolaris Raf. with P. phaseolus (Hild.), has designated it as the type of Ellipsaria Raf. and gives the latter priority as the generic name. CATALOGUE RECENTLY DrEscrreED Mor,usca 179 PryCHOBRANCHUS PHASEOLUS (Hild.). Unio compressissimus performosus De Gregorio, Moll. di aq. dul. di Amer., ROE Op. 23.spl. “V , fre 2) Unio lanceolatus blandus De Gregorio, Ibid, p. 22, pl. VIII, fig. 2. Unio imperitus De Gregorio, Ibid, p. 15, pl. IX, fig. 1. Say, Conrad and Frierson (1914, p. 7) have identified the Obliquaria fasciolaris Raf. as this species and Vanatta (140, p. 554) states that the shell so labelled in the Poulson collection is also that species. If identifiable from the original description, Rafinesque’s name would have priority. PTYCHOBRANCHUS SUBTENTUS (Say). Unio subteritus purcheornatus De Gregorio, Moll. di aq. dul. di Amer., 1914, Pasi. Px fig. 2: Genus CYPROGENIA Agassiz, 1852. CyPROGENIA IRRORATA (Lea). Is Obovaria stegaria Raf. according to Conrad and Vanatta (140, p. 554) states that the shell so labelled in the Poulson collection is this species. If identifiable from the original description stegaria has priority. Genus PLAGIOLA (Rafinesque, 1819) Agassiz. PLAGIOLA DONACIFORMIS (Lea). Unio zig-zag illius De Gregorio, Moll. di aq. dul. di Amer., 1914, p. 11, pl. PY: tis. 3: Is an Amygdalonaias according to Ortmann (81, p. 67). PLAGIOLA ELEGANS (Lea). Unio elegans elegantopsis De Gregorio, Moll. di aq. dul. di Amer., 1914, pe iy ple. LV ie. /O: Unio elegans magnelegans De Gregorio, Ibid, p. 11, pl. V, figs. 1a-c. This species is an Amygdalonaias according to Ortmann, (79, p. 328). Say and Conrad have identified the T7runcilla truncata as this species. Vanatta states (140, p. 553) that both the T. truncata Raf. and Umo meta- plata Raf. as represented in the Poulson collection are this species. If iden- tifiable from the original description, truncata would have precedence. Meta- plata is subsequent to both. PLACIOLA SECURIS (Lea). Both Obliquaria depressa Raf. and O. lineolata Raf. are identified by Say and Conrad as this species and both authors give the preference to lineolata as the specific name. According to Vanatta (140, p. 553) these two species and also O. ellipsaria Raf. as represented in the Poulson col- lection are securis Lea. If identifiable from the original description, any one of Rafinesque’s names would have priority. 180 BRYANT WALKER Genus PARAPTERA Ortmann, 1911. The type of this genus is U. gracilis Bar. If, as claimed by Frierson and others, this is the Unio fragilis Raf., which the latter in 1831 made the type of his genus asmonos, this name would have priority over Paraptera. Frierson (42, p. 6) has suggested that if leptodon Raf. should prove to belong to this genus, Leptodea Raf., of which leptodon is the type, would have to be used. Genus OBOVARIA Rafinesque, 1810. OpovaRIaA circULUS (Lea). Obliquaria subrotunda Raf. has been identified as this species by Say and Conrad and the latter adds Obovaria striata Raf. as a synonym. ‘The specimens under both of these names in the Poulson collection, according to Vanatta (140, p. 552), are this species. If identifiable from the original descriptions, either of Rafinesque’s names would have priority. OpovaRIA ELLIpsis (Lea). Conrad has identified Amblema olivaria Raf. as this species and accord- ing to Vanatta (140, p. 553) so also is the shell in the Poulson collection under that name. If identifiable from the original description Rafinesque’s name has priority. OpovaRIA LENS (Lea). According to Ortmann (79, p. 323) this species is not specifically dis- tinct from O. circulus (Lea). Vanatta (140, p. 552)-states that the shell labelled Unio levigata Raf. in the Poulson collection is this species. If iden- tifiable from the original description, Jevigata would have precedence OpoVARIA RETUSA (Lam.). According to Vanatta (140, p. 552) this is the Obowaria torsa Raf. of the Poulson collection. Genus CARUNCULINA Simpson, 1808. This group is clearly entitled to generic rank as stated by Ortmann (81, p. 68), who has shown that the type is Unio parvus Bar. and not terasensis ea. Frierson (42, p. 7) has identified C. glans (Lea) as the Unio (To.ro- lasma) lividus Raf. and consequently substitutes To.rolasma Raf. for Car- unculina Simp. Genus LAMPSILIS Rafinesque, 1820. LAMPSILIS ALATA (Say). Vanatta (140, p. 552) states that the shell labelled Metaptera megapiera Raf. in the Poulson collection is this species. Alata is the type of Proptera Raf. CATALOGUE RECENTLY DrEscrisED MoLLusca 181 LAMPSILIS ANODONTOIDES (Lea). This species belongs to Lampsilis s. s. according to Ortmann (79, p. 346). It has been identified with U. teres Raf. by Say, Conrad and others. LAMPSILIS ARKANSENSIS (Lea). This species is a Micromya according to Ortmann (81, p. 54). LAMPSILIS AMGNA (Lea). Is a synonym of L. nebulosa (Con.) according to Ortmann (81, p. 64). LAMPSILIS BOREALIS (Gray). The citation of this species from Oneida Lake by Baker, (9, p. 257) has proved to be erorneous. See Baker, (10, p. 75). LAMPSILIS BREVICULA (Call). Is a Micromya according to Utterback (135, p. 434). LAMPSILIS CAPAX (Green). Is a Proptera according to Coker and Surber, (21, p. 179) and Ortmann (81, p. 67). I,AMPSILIS CARIOSA (Say). Unio pallescens Lea var. De Gregorio, Moll. di aq. dul. di Amer., p. 9, non Lea, 1845. LAMPSILIS CONSTRICTA (Con). Is a Micromya according to Ortmann (81, p. 66). LAMPSILIS ELLIPSIFORMIS (Con.). Is a Nephronaias according to Utterback (135, p. 341). LAMPSILIS FALLACIOSA Smith. Is a Lampsilis s. s. and doubtfully distinct from L. anodontoides ac- cording to Ortmann (79, p. 347). But Surber (127, p. 5) states that the glochidia of the two species differ both in size and shape. CARUNCULINA GLANS (Lea). Unio castus mirus De Gregorio, Moll. di aq. dul. di Amer., 1914, p. 27. Frierson (42, p. 7) has identified Unio (Toroclasma) lividus Raf. as this species. If identifiable, and it is not at all certain that it might not be the pullus of Lea, it would have priority. If it can not be definitely deter- mined what species it is, it should be rejected for indefiniteness. 182 BRYANT WALKER LAMPSILIS CRACILIS Barnes. This species has been identified by Frierson (42, p. 7) and others as the Unio fragilis Raf. (1820) and Lasmonos fragilis Raf. (1831) and either of these names, if identifiable from the original description, would have prior- ity. The example under this name in the Poulson collection according to Vanatta (140, p. 552) 1s gracilis Bar. Fragilis Raf. (1831) is the monotype of his genus Lasmonos. LAMPSILIS IRIS (Lea). . Is a Micromya according to Ortmann (79, p. 341). I,AMPSILIS LEPTODON (Raf.). Unio shepardianus f. duttonianus De Gregorio, Moll. di aq. dul. di Amer., TOMY. p-225- This species is the tvpe of Leptodea Raf. by designation (Frierson, 42, psiO}: LAMPSILIS LIENOSA (Con.). Is a Micromya according to Ortmann (79, p. 340). LAMPSILIS LIGAMENTINA (Lam.). Unio tecomensis De Gregorio, Moll. di aq. dul. di Amer., 1914, p. 26. This species was at first referred to Obovaria by Ortmann (78, p. 119) and later to Nephronaias (79, p. 325). It should rather be referred to Ac- tinonatas F. and C. According to the specimens in the Poulson collection (Vanatta, I40, p. 551) U. crassa Raf., fasciata Raf. and pallens Raf. are this species. LAMPSILIS LUTEOLA (Lam.). Vanatta (140, p. 551) states that L. fasciola Raf. is this species. LAMPSILIS NEBULOSA (Con.). Is a Micromya according to Ortmann (81, p. 64). I,AMPSILIS NIGERRIMA (Lea). Is a variety of 1. concestator Lea according to Frierson (41, p. 135). LAMPSILIS OCCIDENTALIS (Con.). Is Ptychobranchus clintonensis Simp. and has priority according to Ut- terback (135, p. 317). LAMPSILIS ORBICULATA (Haild.). Belongs to Lampsilis s. s. and does not group with L. liyamentina Lam. according to Ortmann (79, p. 353)- CATALOGUE RECENTLY DrscrrBED MoLtusca 183 I,AMPSILIS OZARKENSIS (Call). Is a Nephronaias according to Utterback (135, p. 344). Ortmann (84, | p. 62) has more recently determined it to be a Fusconaia. LAMPSILIS PARVA (Bar.). Unio pertenuis De Gregorio, Moll. di aq. dul. di Amer., ror4, p. 31, pl. VI, figs. 4a-f, non Lea, 1863. This species is the true type of Carunculina according to Ortmann (8r, p. 68). LAMPSILIS PERDIX (Lea). Is a Nephronaias according to Ortmann (79, p. 326). It is rather an Actinonaias. L,AMPSILIS PERPURPUREA (Lea). Is a Micromya according to Ortmann (81, p. 63). LAMPSILIS PICTA (I,ea). Is a Micromya according to Ortmann (79, p. 342). LAMPSILIS PLEASIT (Marsh). Is a Nephronaias according to Utterback (135, p. 343). It is rather an Actinonaias. LAMPSILIS RADIATA (Gmel.). Unio muhlfeldianus plurimaffinis De Gregorio, Moll. di aq. dul. di Amer., 1914, p. 8, pl. III, figs. 2a-d. LAMPSILIS RADIATA ONEIDENSIS Baker. Lampsilis borealis Baker, Tech. Pub. N. Y. Coll. For., 4, 1916, p. 257, fig. 44, nos. I-4. Lampsilis radiata oneidensis Baker, Naut., XXX, 1916, p. 74, pl. IT. Type locality: Oneida Lake, N. Y. LAMPSILIS RECTA (Lam.). Unio sageri Conrad, Mon., VI, 1836, p. 53, pl. xxix, fig. 1. Lampsilis recta sageri Simpson, Desc. Cat., 1914, p. 96. Lamarck’s type came from Lake Erie and is the small form character- istic of the Great Lakes that Conrad described as Unio sageri. The large, normal form from the Ohio and elsewhere may be distinguished under Rafinesque’s name. 184 Bryant WALKER LAMPSILIS RECTA LATISSIMA (Raf.). _ Unio latissima Rafinesque, Mon., 1820, p. 297, pl. Ixxx, figs. 14-15. Unio angustatus cuniculus De Gregorio, Moll. di aq. dul. di Amer., 1914, p22) pl. ox, fice. Type locality: Ohio River. This is the form represented in the Poulson collection (Vanatta 140 P5517): , LAMPSILIS SIMPSONI Ferriss. Belongs to Lasmonos Raf. according to Utterback (135, p. 307). LAMPSILIS SUBROSTRATA (Say). Belongs to Eurynia s. s. according to Ortmann (81, p. 55). LAMPSILIS TENERA (Ravenel, Mss.) Mazyck. Lampsilis tenerus (Rav. Mss.) Mazyck, Cat. Moll. S. C., 1913, p. 23. Type locality: Santee Canal, St. John’s, Berkeley, S. C. The specific name is preoccupied by Lea, 1840. LAMPSILIS TRABALIS (Con.). Is a Micromya according to Ortmann (79, p. 340). LAMPSILIS VANUXEMENSIS (Lea). Is a Micromya according to Ortmann (79, p. 342; 81, p. 65). LAMPSILIS VENTRICOSA (Bar.). According to Say and Conrad this is the Z. cardium Raf., 1820, and if so ,the latter name has priority. Vanatta (140, p. 551) states that the shell so labelled in the Poulson collection is the ventricosa Bar. According to Ortmann (79, p. 351) it “is probably only a variety of L. ovata (Say). LAMPSILIS VENTRICOSA COHONGORONTA Ortmann. Lampsilis ventricosa cohongoronta Ortmann, Naut., XX VI, 1912, p. 53. Type locality not specified. Found in the Potomac River, Hancock, Wash- ington Co., Md., and in the South Branch of the Potomac at South- branch and Romney, W. Va., and in the Shenandoah River, Harper’s Ferry, W. Va. LAMPSILIS VENTRICOSA SATURA (Lea). Frierson (41, p. 136) says that this is not a variety of L. ventricosa, but is the same as L. evcavata Lea and has priority as the proper specific name. I do not agree with this. Ortmann (81, p. 56) deals with it as a form of ventricosa and intimates a “suspicion” that it may prove to be a distinct species. CATALOGUE RECENTLY DEscrIBED MOLLUSCA 185 LAMPSILIS VENUSTA (Lea). Is a variety of L. ellipsiformis (Lea) according to Utterback (135, p. 343). T,AMPSILIS VIBEX (Con.). Is a Wicromya according to Ortmann (79, p. 340). Genio MICROMYA (Agassiz, 1852) Simpson. Micromya c&Lara Conrad. Unio propecelatus De Gregorio, Moll. di aq. dul. di Amer., 1914, p. 30, pl. VIII, figs. ta-d. This species has been identified by Frierson (42, p. 7) and Ortmann (82, p. 39) as Unio (Lemiox) rimosus Raf. and the latter (1. c.) has raised Lemiox to generic rank. The identification seems too doubtful (Walker, 162, p. 4) to ‘be relied upon, the species should therefore retain Conrad’s name and a new generic name proposed. MiIcROMYA FABALIS (Say). Unio donacopsis De Gregorio, Moll. di dul. di Amer., 1914, p. 30, pl. X, figs. 5a-b. Genus TRUNCILLA Rafinesque, 1819. TRUNCILLA CURYTISIL Frierson and Utterback. . Truncilla curtis Frierson and Utterback, Am. Mid. Nat., IV, 1916, p. 453, pl. VI, figs. 14a-d, pl. XXVIII, figs. rogA-D. Type locality: White River, Hollister, Mo. This is the form that was collected by Ferriss in 1900 and distributed by him as 7. deviata (Anth.). It has also been found in Bear Creek, Frank- lin Co., Ala., a tributary of the Tennessee. It does not seem to be more than a light colored form of that species. TRUNCILLA LEFEVREI Utterback. Truncilla lefevrei Utterback, Am. Mid. Nat., IV, 1916, p. 455, pl. VI, figs. 13 -d, pl. XXVIII, figs. ro8A-D. ‘Type locality: Black River, Williamsville, Mo. TRUNCILLA FOLIATA (Hild.). This species has been identified as Rafinesque’s Obliquaria fleruosa by Conrad and the specimen in the Poulson collection under that name is stated by Vanatta (140, p. 550) to be this species. If identifiable from the orig- inal description, Rafinesque’s name would have priority. It has also been identified by Frierson (42, p. 7) as Rafinesque’s Epioblasma biloba. 186 BrYANT WALKER TRUNCILLA BREVIDENS (Lea). Vanatta (140, p. 550) states that the shell in the Poulson collection la- belled Obliquaria interrupta Raf. is this species. If identifiable from the original description, Rafinesque’s name has precedence. TRUNCILLA PERPLEXA (Lea). This species has been identified by Conrad and others as Amblema toru- losa Raf. and Amblema gibbosa Raf. According to Vanatta (140, p. 550) the shells so labelled in the Poulson collection are this species. Either name, if identifiable from the original description, would have priority. TRUNCILLA SULCATA (Lea). Unie stewardsoni stevensoni De Gregorio, Moll. di aq. dul. di Amer., 1914. per5, spl. Vie hos Unio propesulcatus De Gregorio, Ibid, p. 30, pl. X, fig. 2. Vanatta (140, p. 550) states that the shell in the Poulson Collection under the name of Obliquaria obliquata Raf. is this species. If identifiable from the original description, it would take precedence. TRUNCILLA TRIQUETRA Rafinesque. Unio triangularis pergitbosa De Gregorio, Moll. di aq. dul. di Amer., 1914, p: 10,spl. 1], fig. 4. Unio triangularis longiusculus De Gregorio, Ibid, p. 10, pl. II, fig. 5. TRUNCILLA TRIQUETRA TRIANGULARIS (Barnes). Barnes’ types of his Unio triangularis came from Bois Blanc Island in the Detroit River and the formsis fairly entitled to varietal rank, being like nearly all of the Unionide of the Great Lakes a characteristically depauper- ate race. TRUNCILLA WALKERI ‘Wilson and Clark. Truncilla walkeri Wilson and Clark, Bur. of Fish. Doc. No. 781, 1914, p- Ao, ple es 1. Type locality: East Fork of Stone’s River, Walterville, Tenn. Subgenus DYSNOMIA Agassiz, 1852. Frierson (42, p. 7) identifies Rafinesque’s Unio or Epioblasma biloba with 7°. foliata (Hild.) and consequently replaces Agassiz’ subgeneric name by Epioblasma Raf. CATALOGUE RECENTLY DeEscripeD MoLLusca 187 Genus incerte sedis. COKERIA Marshall, 1916. Cokeria Marshall, Naut., XXIX, 1916, p. 133. Type: Cokeria southalli Marshall. COKERIA SOUTHALLI Marshall. Cokeria southalli Marshall, Naut., XXIX, 1916, p. 133, pl. IV. Type locality: James River, Huron, S. D. This genus and ‘species are founded on an unique specimen collected by the U. S. Bureau of Fisheries. It seems to combine the characters of both Quadrula and Lampsilis, although the preponderance of the characters ap- pear to be Quadruline. Indeed, but for the hinge teeth, which are quite like those of Lampsilis, it would be considered a Quadrula. It is quite possibly an abnormality of Quadrula undulata (Bar.). Unfortunately nothing is known of the soft parts. Additional material is greatly to be desired. Family CYRENID/. Genus CYRENA Lamarck, 1818: Section POLYMESODA Rafinesque, 1820. CyYRENA ALABAMENSIS Clessin. Cyrena alabamensis Clessin, Con. Cab., Cycladeen, 1869, p. 114, pl. XVIII, figs. 3-4. Type locality: Alabama. CyRENA DONACIFORMIS Sowerby. Cyrena donaciformis Sowerby, Con. Icon., Cyrena, 1878, p. 108, pl. XIX, fig. TOs. Type locality: Florida. Is floridana Con. according to von Martens (Zool. Rec., Moll., 1877, p. 81) and Dall (29, p. 1447). CyRENA PROTEXTA Conrad. Cyrena protexta Conrad, Am. J. of Con., V, 1869, p. 107, pl. XII, fig. 3. Type locality: Tampa Bay, Fla. Is floridana Con. according to Tryon, (A. J. of C., V, 1870, p. 227) and Dall (29, p. 1447). 188 Bryant WALKER Family SPHASRIIDAE. Sterki’s “Preliminary Catalog” (125, p. 429) has brought the subject down to July 1, 1916. For valuable papers on the anatomy, reproduction and growth of Sphe- rium and Musculium, see Drew (36, p. 173) and Gilmore (46, p. 16). Genus SPHARIUM Scopoli, 1777. SPHARIUM LENTICULARE Sowerby. Spherium lenticularis Sowerby, Con. Icon., Spherium, Sp. 6, pl. I, fig. 6. Type locality: ? SPH.ERIUM MAGNUM Sterki Mss. Spherim magnum “Sterki’, Springer, Pr. A. N. 8. P., 1902, p. 513. | This undescribed species is quoted from Arroyo Pecos, Las Vegas, N. M. (pleistocene) by Springer. SPHARIUM MEDIUM (Sowerby). Cyclas medium “Sowerby”, Richardson, Fauna Bor. Amer. III, 1836, p. 316. Type locality: Methy Lake, Athabaska. Probably never described, see Dall (32, p. 140). SPH.ERTUM RUGOSUM “Whitmore” Sowerby. Spherium rugosum “Whitmore” Sowerby, Con. Icon., Sphzrium, Sp. 16, plil tig 10: Type locality: ? SPH-ERIUM STAGNICOLUM (Sowerby). Cyclas stagnicola “Sowerby” Richardson, Fauna Bor. Amer. III, 1836, p. 310. Type locality: Methy Lake, Athabaska. Probably never described, see Dall (32, p. 140). Genus MUSCULIUM Link, 1807. Musculium Link, Beschr. Rostock Samm I, 1807, p. 152. Calyculina Clessin, Mal. Blatt., XIX, 1871, p. 150. Frimella Cooper, Pr. Cal. Acad. Sci., (2), III, 1891, p. 82. MvSscCULIUM COOPFRIANUM (Prime) Mss. Spherium cooperianum Prime, Cat. Corbic., Am. J. of Con., V, 1869, p. 152. Type locality: Johnson’s Pass, Eldorado Co., Cal. Listed as a new species, but never described. Probably the young of M. raymondi according to Cooper (26, p. 81). CATALOGUE RECENTLY -DESCRIBED MoLLusca 189 MUSCUIIUM ERRANS (Lewis). Cyclas errans Lewis, Ubi ? Lewis (Pr. A. N.'S. P., 1872, p. 105) says that he “proposed” this spe- cies a number of years before, but that Prime considered it a synonym of rosaceum. ‘That from further study, he is “induced” to reclaim his species. I have been unable to find any other reference to the species. Genus PISIDIUM C. Pfeiffer, 1821. The use of Corneocyclas Fer. for this genus proposed by Dall (30, p. 7 and 29, p. 1459) is disputed by Woodward (165, p. 367 and 166, p. 1) and has not been followed by Sterki (125, p. 473) and until the question has been definitely settled, it seems preferable to use the name that has been in common use for so many years. PISIDIUM BOREALE Westerlund. Pisidium boreale Westerlund, Kongl. Sv. Vet. Ak. Forh., 1877, p. 70, fig. 23. Type locality: Lusino, Siberia. ? Port Clarence, Alaska. PISIDIUM CALIFORNICUM (Newcomb ?). Listed by Berry (Naut., XXIII, 1909, p. 79) from Bluff Lake, San Rer- nardino Co., Cal. I have not been able to find any other reference to it. PISIDIUM COMPRESSUM LIMNICOLUM Sterki. Pisidium compressum limnicolum Sterki, Naut., XTX, 1905, p. 8r. Type locality: Fox River, Wis. PISIDIUM COMPRESSUM SMITHII Sterki. Pisidium compressum smithi Sterki, Naut., XTX, 1905, p. 83. Type locality: Shoal Creek, Ala. PISIDIUM NOVEBORACENSE PROCLIVE-Sterki. Pisidium noveboracense proclive Sterki, Naut. XIX, 1G06, p. 110. Type locality: New Philadelphia, O. PISIDIUM OBTUSALE C. Pfeiffer. This European species has been listed from near Lake James, Steuben Co., Ind. by Sterki (Naut., XVII, 1903, p. 43). Pistptum sipiricuM Westerlund. Pisidium sibiricum Westerlund Nachr. Blatt. Gess. 1876, p. 103; Clessin, Con. Cab., Cycladeen, 1877, p. 66,.pl. VII, figs: 15-17. Type locality: Yenesei River, Siberia. ? Port Clarence, Alaska according to Dall (32, p. 144). Dall’s reference for this species in K. Svenska Vet. Ak. Forh. is erroneous. It should be Deo, fig, 21, not p. Fo; fig, 23: 190 Bryant WALKER Family CYRENELLID-. Genus CYRENELLA Deshayes, 1835. CyRENELLA FLORIDANA (Dall). Cyrenoida floridana Dall, Naut. X, 1896, p. 52; Pr. U. S. N. M. XXIII, EQOT, ip, 020; ps 2c Tess 7. Type locality not specified. Habitat: Brunswick, Ga., south to the Everglades on the east, and, on the west, north to Charlotte Harbor and vicinity. SI oa BIBLIOGRAPHY OF THE REFERENCES IN THE NOTES EN PART II. ALLEN, J. A., Lymneza auricularia Linné in Canada. Nautilus, XXV, FOIL, p: 60. Baixky, J. T. Jr., Viviparus in Philadelphia. Nautilus, XXIII, 1909, p. 60. Baker, F. C., Shell collecting near Rochester, N. Y. Nautilus, XIV, 1900, pp. 69-71. Baker, F. C., Mollusca of the Chicago Area. Chi. Acad. Sci. Bull. Pe Pe 1e9c Pt Ly 1902. Baker, F. C., The Ecology of the Skokie Marsh Area, with special merenemce torte Mollusca, “Bull. Is. St;-Wabs Nat. sHist., VILL, IQ10, Pp. 441-499. Baker, F. C., The Lymnezide of North and Middle America, Recent and Fossil.-Chi. Acad. Sci., Sp. Pub., 3, 1911. Baker, F. C., Pseudogalba, new name for Simpsonia. Nautilus DXVE AGL3)p: 120, Baker, F. C., On the Classification of the Lymnezide. Nautilus X XIX, IQ15, pp. 20-24. Baker, F. C., The Relation of Mollusks to Fish. Tech. Pub., N. Y. St. Coll. of Forestry, 4, 1916, pp. 1-366. 3\KER, F. C., Description of a new variety of Lampsilis from Oneida Lake with notes on the L. lubeola group. Nautilus XXX, 1916, pp. 74-77, pl. I. Binney, W. G., Land and F. W. Shells of North America. -Smithson- any Mise, Cont.,.143, Part 11,1865. Brnney, W. G., Notes sur quelques especes de mollusques fluviatiles de l’Amerique du Nord. Jr. de Con., XV, 1867, pp. 427-432. Binney, W. G., Note on Vivipara lineata Val. Ann. N. Y.Lyc. N. H., IX, 1870, pp. 295-297. BLaAtcHLEy, W. S. and Daniets, L. E., On some mollusca known to occur in Indiana. 27th Ann. Rep. Dept. of Geol. & Nat. Re- sources, Indiana, 1903, pp. 579-628. Brown, A. B., and Pirssry, H. A., Fresh Water Mollusks of the Oli- gocene of Antiqua. Pr. A. N.S. P., 1914, pp. 209-213. . Caut, R. Ef., On the Genus Campeloma Rafinesque with a Revision of the Species. Washburne Coll. Bull., I, 1886, pp. 149-168. Cau, R. F., On a new post-pleiocene Lymnzid. Am. Geol. 1888, pp. 147-149. Cane, R. E., On the Geographic and Hypsometric Distribution of North American’ Viviparide. Am. Jr. Sci., XLVIII, 1894, pp. 132-140. Crapp, G. H., Gundlachia hjalmarsoni Pfr. in the Rio Grande, Texas. Nautilus XX VII, 1913, 77-8. 40. BiBLIOGRAPHY CLapp, W. F., Radix auricularia at Cambridge, Mass. Nautilus, XXXVI, 1912, p. 116. CLEsSIN, S., Die Familie der Limnziden. Con Cab., 1880. Coxer, R. E. and Surzser, T., A Note on the Metamorphosis of the mussel Lampsilis levissimus. Biol. Bull., XX, 1911, pp. 179-182. Coron, H. $., On Classification in general and the Genus Lymnza in particular. Nautilus, XXVIII, 1915, pp. 116-1109. Corron, H. S., A Provisional Key to the Sub-genera and Species of Lymnea. Nautilus, XXVIII, 1915, pp. 119-120. Conrap, T. A., New Fresh-water Shells of the United States, 1834; Appendix 1835. Cooper, J. G., The West Coast Fresh-water Univalves, No. 1. Pr. Cal. Acad; Sci., TV, 1870, pp. 92-100: Cooper, J. G., Notes on Raymond’s Subalpine Mollusca (No. 112). Pr.-Cal: Acad: Sci., 2nd: Ser., JL, 1890; pp. 70-di: Crandall, O. A., The American Physe. Naut., XV, 1901, pp. 25-30; 42-45; 54-58; 69-71. DatL, W. H., Notes on some Floridian Land and F. W. Shells, etc Proc. U. S. N. M., VIII, 1885, pp. 255-280. Datu, W. H., Contributions to the Tertiary Fauna of Florida. rans. Was. Free: Inst: Sci, 104, Pt. V; To00!. pp. 0940-12165. Ei aa, 1903, pp. 1219-1654. Dati,’ W. H., Review of the Classification of the Cyrenacea.. Pr. Biol. Soc. Wash., XVI, 1903, pp. 5-8. Dau, W. H., Gundlachia and Ancylus. Nautilus, XVII, 1994, pp. 97-98. Dati, W. H., Harriman Alaska Expedition, XIII, Land and F. W. Mollusks, 1905, pp. I-157. Dati, W. H., Notes on Planorbis and its Subdivisions. Proc. Mal. Soc. Lond., VIE, 1908, p. 141. Dati, W. H., Notes on Gundlachia and Aucylus’ Am. Nat., XLV, IOIT, pp. 175-189. De GreEcorIo, ANTonto, Su Taluni Molluschi di Acqua Dolce di America, 1014, pp. I-- Drew, G. A.. The Anatomy of Spherium sulcatum Lam. Pr. Ia. Acad. of Sci.. III, 1806, pp.’173-182. Pl. VITI-X. Farrer, W. J., Note on Planorbis multivolvis. Nautilus, VI, 1892, pi 26: Fiscner, P. and Crosst, H., Etudes sur les Mollusanes Terrestres et Fluviatiles du Mexique et du Guatemala. 1870-1802. FRAUENFELD, G. R.. Die Arten der Gattung Lithoglvphus Mhlf.. Pal- ndinella Pfr.. Assiminea Gray, in der Kaiserlichen und Cumings Samlung. Verh. der k. k. zool-bot. Gesellschaft in Wien, 186532, pp. 193-212. FRAUENTELD, G. R., Vorlaufige Aufzohlung der Arten der Gattungen Hvydrobia Htm. und Amnicola Gld.-Hald. in der Kaiserlichen und in Cumings Samlung. Verh. der k. Ik. zool.-bot. Gesellschaft in Wien, 1863, pp. 1017-1032. 59. 60. 62. BIBLIOGRAPHY 193 Frierson, L. S., A Comparison of the Unionide of the Pearl and Sabine Rivers. Nautilus, XXIV, 1911, pp. 134-136. Frierson, L. S., Remarks on Classification of the Unionide. Naut- ilus, XXVIII, 1914, pp. 6-8. Frierson, I,. S., Lasmigona subviridis Con. Redivivus. Nautilus, XXIX, 1915, pp. 57-59- Frierson, L. $., Observations on Unio cor. of Conrad. Naut., XXIX, 1916, pp. 102-104, Pl. III, figs. 1-2-3. Frierson, L. $., Observations on Unio giganteus Bar. Nautilus, XXX, 1916, pp. 61-64. GitmorE, R. J., Notes en Reproduction and Growth in certain vivi- parous mussels of the family Sphzride. Nautilus, XXXI, 1917, pp. 16-30, pl. IV-VI. GoopricH, CALvin, Lymneea auricularia L. in Ohio. Nautilus, XXV, EQUI, Pp: Ii. Goopricu, CALVIN, Spring Collecting in Southwest Virginia. Naut- ilus, XX VII, 1913, pp. 81-82 and gI-95. GoopricH, Calvin, Pleurocera subulare Lea. Nautilus, XXX, 1g17, pp. 122-124. : Haas, F., Bemerkungen tiber Spengler’s Unionen. Kobenhaven Nath. Medd., 65, 1913, pp. 51-66. Hawnnipat, H., Exotic Vivipara in California. Nautilus, XXII, 1908, P. 33- HANNIBAL, Harorp, Further Notes on Asiatic Viviparus in Califor- nia. Nautilus, XXV, I9QII, pp. 31-32. Hannipar, Harorp, A Synopsis of the Recent and Tertiary Mollusca of the Californian Province. Proc. Mal. Soc. Lond., X, 1912, pp. {12-211. Hept&y, C. and Pirssry, H. A., On the Australian Gundlachia with note on American forms. Nautilus, IX, 1895, pp. 61-68. Henprrson, J., Lymnea (Radix) auricularia Linn. in Colorado. Nautilus, XX VI, 1912, p. 84. Henprrson, J. and Danrets, L. ™., scunting Mollusca in Utah and Idaho in 1916. Pr. A. N.S. P., 1917, pp. 48-81. Hinxtey, A. A., Meseschiza grosvernori Lea. Nautilus, XXII, 1908, p. 56. Hinx.&ty, A. A., New Orleans Mollusca. Nautilus, XX VI. 1912, p. 36. Jounson, C. W., Lymnea (Radix) auricularia in Charles River, Boston. Nautilus, XX VII, 1913, p. 83. Jounson, C. W., Viviparus malleatus Rve. Nautilus, XXIX, 1915, Pp. 35. Ibid, Nautilus. XXX, 1916, p. 48. Jonnson, C. W., Viviparus contectoides Binn. in Boston, Mass. Naut- ilus, XXX, 1916, p. 72. _ Jonnson, C. W., Vivipartts malleatus and contectoides in Massachus- etts. Nautilus, XX XI, 1918, p. 107. 194 BIBLIOGRAPHY KrnNARD, A. S., Notes on Planorbis vorticellus Troschel, ete. Pr. Mal. Soc. Lond., VIII; 1908; p. 47: LATCHForD, F. R., Valvata piscinalis in Canada. Nautilus, XXVIII, 1914, p. 10. Lewis, JAMES, Observations on Melantho. Am. J. Con., I\ pp. 133-130. Lewis, ]AMEs, Observations on Melantho. Am. J. Con., V, 1869, pp. 33-30. Lewis, JAMES, On the Shells of the Holston River. Am. J. Con., VI, 1871, pp. 216-220. LEwIs, JAMEs, Shells of Tennessee, (No. 2). Pr. A. N.S. P., 1872, pp. 108-115. Lewis, JAMES, Appendix to Tryon’s Mon. of Strepomatidz. Smith- sonian Misc. Coll., No. 253, 1873, pp. 423-426. Lewis, JAMES, Fauna of Alabama, I, F. W. and Land Shells, 1876. Lons, Bayarp, Lymnza auricularia near Philadelphia. Nautilus, XXVI, 1912, pp. 27-29. v. Martens, E., Biologia Centrali- Americana, Land and F. W. Mol- lusca, 1890-1801. MELvILL, J. C., List of Mollusca obtained in South Carolina, Florida, ete. I Corn, IIT, 1881, pp. 155-173. Morse, E. &., @peenemen on the Terrestrial Pulmonifera of Maine, ete. Jour. Portland Soc. of Nat. Hist., I, 1864, pp. 1-63. NeEtson, Wm., Note on Physa virgata Gld. Jr. Con., IL). 1870, p: 182: OrtTMANN, A. E., The Breeding Season of Unionidz in Pennsylvania. Nautilus, XXII, 1909, pp. 91-95 ; 99-103. OrtMANN, A. E., A New System of the Unionidz. Nautilus, XXIII, IQIO, pp. 114-120. OrtMANN, A. E., Notes upon the families and genera of Najades. Ann. ‘Car: Mus.) VILE 1012, pp. 222-305; OrTMANN, A. E., Cumberlandia, a new genus of Najades. Nautilus, XXVI, 1912, pp. 13-14. OrTMANN, A. E., Studies in Najades. Nautilus, XX VII, 1913, pp. 88-91. Nautilus, XXVIII, 1914, pp. 20-22; 28-34; 43-47; 65-60. Nautilus, XXVIII, 1915, pp. 106-108; 129-131; 141-143. Nautilus, XXIX, 1915, pp. 93-67. Nautilus, XXX, 1916, pp. 54-57. OrtTMANN, A. E., The Anatomy of Lemiox rimosus. Nautilus, XXX, 1916, p. 39. OrtTMANN, A. E., The Anatomical Structure of Gonidea mene Lea. Nautilus, XXX, 1915, pp. 50-53. OrtMANN, A. E., A New Type of the Nayad-genus, Fusconaia, Group of F. barnesiana Lea. Nautilus, XXXI, 1917, pp. 58-64. Pirspry, H. A., Notes on some New Orleans F. W. Shells. Con. Ex., T1686; p. 20. 1868, I 107. BIBLIOGRAPHY 195 Piissry, H. A., Notes on the Larger Florida Planorbes. Am. Nat., XXI, 1887, pp. 286-7. Piussry, H. A., Lyogyrus Gill and other American Shells. Con. Ex., fr P1S8e7 pHi13. Piispry, H. A., Recent Additions to the U. S. Snail Fauna. Nautilus, III, 1889, pp. 61-64, pl. I, figs. 1-2-3. Pirspry, H. A., Preliminary Notices of New Amnicolidz. Nautilus, IV, 1890, pp. 52-53. Pitspry, H. A., Note on Goniobasis catenaria Say. Nautilus, IV, 1891, p. 124. Piysspry, H. A., Land and F. W. Mollusca Collected in Yucatan and Mexico. (Pr Av N.S: P.; 1891, pp: 310-334: Piispry, H. A., Notes on New and Little-known Amnicolide. Naut- ilus, XIII, 1898, pp. 42-44. Pinspry, H. A., Shells of Redding, Shasta Co., Cal. Nautilus, XII, 1898, p. 59. Piyspry, H. A., New Amnicolide from Florida. Nautilus, XIII, 1899, pp. 20-22. Pinspry, H. A., Mollusks collected by R. C. McGregor in Northern California. Nautilus, XIII, 1899, pp. 64-67. Pirspry, H. A., Catalogue of the Amnicolidz of the Western United States. Nautilus, XII, 1899, pp. 121-127. Piispry, H. A., Notes on certain mollusca of Southwestern Arkansas. Pr. A. N. S. P., 1900, pp. 449-459. Pinspry, H. A., New Species of Mollusca from South Africa and Burma. Pr. A. N.S. P., 1901, pp. 188-190. Piuspry, H. A., A New Brackish-water Snail from New England. Nautilus, XIX, 1905, p. go. Pitspry, H. A., Is Cochliopa rowelli a Californian Shell? Nautilus, EIS OOS. pi Or Pitspry, H. A., A New Mexican Genus of Pleuroceratide. Pr. Mal. Soc. Lond., IX, 1910, pp. 47-50. Pitspry, H. A., A New Species of Amnicola. Nautilus, XX VI, 1Ol2; pr: Pirspry, H. A., New Sub-species of Viviparus and Campeloma. Nautilus, XXX, 1916, pp. 41-43. Piissry, H. A., Note on Valvata micra P. and F. Nautilus, XXX, 1916, p. 82. Piispry, H. A., Rafinesque’s Genera of Fresh-water Shells. Naut- ilus, XXX. 1917, pp., 109-114. Pirspry, H. A.. and Ferris, J. H., Mollusca of the Southwestern Stames, UL Prac Ne Ss: P., 1906, pp: 123-175. . Prrspry, H. A. and Ferriss, J. H., Mollusca of the Southwestern States, III. Pr. A. N.S. P., 1900, pp. 495-516. Pirspry, H. A., and Ferriss, J. H., Mollusca of the Southwestern Siatess web A WN. 9. 2b. 1O10, Pp: 44=147- 196 108. 109. 110. LEE I12. ge BIBLIOGRAPHY Piyspry, H. A., and Ferriss, J. H., Mollusca of the Southwestern states, Vo (Pro A. ON; >: Por: pp 4a1oo. Pirspry, H. A. and Ferriss, J. H., Mollusca of the Southwestern States; Vz. Pr. A. N.S. "P2915, pp:-363-416; Pirspry, H. A. and Jounson, C. W., Note on Paludina japonica. Nautilus, VII, 1894, p. 144. Pirspry, H. A. and Ruwoaps, $8. N., Contributions to the Zoology of Tennessee, No. 4, Mollusca. Pr. A. N. S. P., 1896, pp. 487-506. Raymonp, W. J., Notes on the Subalpine Mollusca of the Sierra Nevada, near’ lat. 38°. Pr. Cal, Acad Nat. Sct, ende ser, ue 1890, pp. 61-69. Ruoaps, 8. N., Annotated List of L. and F. W. Shells recently col- lected in the vicinity of Miami, Fla. Nautilus, XIII, 1899, pp. 43-48. RowE LL, J., Note on Cochliopa rowelli in California. Nautilus, XX, 1900, p. 10. SARGENT, H. E., Annotated List of the Mollusca found in the vicinity of Clear Water, Wright Co., Minnesota, Part II. Nautilus, IX, 1896, pp. 125-128. Say, TuHos., The Complete Writings of Thomas Say on the Conch- ology of the United States, edited by W. G. Binney, 1858. Simpson, C. T., Gundlachia ancyliformis Pfr. in Florida. Con. Exc., IT, 1885, p. 96. Sowerby, G. B., Monograph of Planorbis, Con. Icon., 1877. Sowersy, G. B., Notes on the family Ampullariide, etc. Pr. Mal. Soc. Lond., VIII, 1909, pp. 345-362; Ibid, IX, 1910, pp. 56-64. SPRINGER, ADA, On Some Living and Fossil Snails of the Genus Physa found at Las Vegas, New Mexico.- Pr. A. N.S. P., 1902, Pp- 513-516. STEARNS, R. FE. C., Observations on Planorbis. Pr. A. N.S. P., 1881, pp. 92-110. STEARNS, R. FE. C., Notes on Physa triticea of Lea, etc. Nautilus, IIT, 1889, pp. 49-51. SreARNS, R. E. C., The Death Valley Expedition, Pt. IT, 1893, Re- port on Mollusca. U.S. Dept. of Ag., N. Amer. Fauna No. 7, pp. 269-283. STEARNS, R. E. C., Japanese Vivipara in California. Nautilus, XV. TQOT, p. OT. _Sterkt, V., A Preliminary Catalogue of the Land and Fresh-water Mollusca of Ohio. Pr. O. St. Acad. Sci., IV, 1907, pp. 367-402. SterKr, V., A Preliminary Catalog of the North American Sphz- riide. Ann. Car. Mus., X, 1916, pp. 429, 474. StrEcCKER, J. K., Jr., The Mollusca of McLennan Co., Texas. Nauti- lus, XXII, 1908, pp. 63-67. Surner, T., Identification of the Glochia of Fresh-water Mussels. Bureau of Fisheries, Doc. 771, 1912, pp. I-10. 128. 130. BIBLIOGRAPHY 197 Tryon, G. W. Jr., Catalogue of the Species of Physa inhabiting the United States. Am. J. Con., I, 1865, pp. 165-173. Tryon, G. W. Jr., Review of Bonnet’s Coquilles Nouvelles. Am. J. Conk (re65. p83 Tryon, G. W. Jr., Review of Frauenfeld’s. paper in Verh. der k. k. Zool:-bot. Ges. Wien, 1862.. Am- J.:Con.,; 1;°1865, p. 374. Tryon, G. W. Jr., Review of Binney’s L. and F. W. Shells, Pt. III. Am. «J. Gon: 111, 1867; p. 197. Tryon, G. W. Jr., A Monograph of the F. W. Mollusca of the U. S., 1870. Tryon, G. W. Jr., Notes on Dr. James Lewis’ paper “On the Shells of the Holston River.” Am. J. Con., VII, 1871, pp. 86-88. Tryon, G. W. Jr., The Strepomatide. Smithsonian Misc. Coll., No. 253; L. and F. W. Shells of North America, Pt. IV, 1873. Utrersack, W. I., The Naiades of Missouri. Amer. Mid. Nat., IV, 1915, Pp. 41-53; 97-152; 182-204; 244-273; 1916, pp. 311-327; 339-354; 387-400 ; 432-464; Separate, 1916, pp. I-200. Vanarra, EF. C., Notes on the Smaller American Planorbes. Naut- ilus, IX, 1895, pp. 52-55. VanaTta, E. G., The Geographic Distribution of Planorbis umbili- catellus. Nautilus, IX, 1896, p. 117. Vanatta, E. G., Planorbis Bicarinatus and Pleurodonte Angulata. Nautilus, XXIV, 1911, pp. 136-138. VanatTra, E. G., Viviparus contectoides in Philadelphia. Nautilus, DVI, 1912,........- 157 Equicostata (Bythinella) Pils. .......... 137 /Equicostata (Paludestrina) Pils. ........ 137 #esopus’ (Pleurobema) Green ........:.. TAG PESOS | (CEIO) GEGEN ccs jere sjheie otr since = SOn sr 71 mine: (elemrocera)) Wea: s+ smclsee ec «<0 Sr Affine (Trypanostoma) Lea .....,...... 151 Alabamensis (Amphigyra) Pils. ...... 22,0123 Alabamensis (Cyrena) Cless. .......... 187 Alabamensis (Planorbis) Pils. ........ 95, 98 Alabamensis avus (Planorbis) Pils. ...... 95 PAM ID ELIS ave VASES: slelets: «feretstola iso efelasonn olalie 106 PM ASMIGONta SAY sswics bse ei vime « 56, 62, 178 Lp jigieitaitiavoehs i OR io a ereaaeeieitrcis aieeciecae 59, 61, 177 Alasmodontina (Complanaria) Stimp. ... 177 Play Ca TPSUIS)) SAY os cqle oleeieieiereaie «7 ath = 180 LACE (QORaION ISR Reanmeicdoicicn Comecnc Vee i Albanyensis (Goniobasis) Lea .......... 154 Albofilata (Physa) Ancey ........... 106, I10 Albus «celanorbis), Mall... i... << 13, 94, 100 Aldrichi (Bythinella) C. and B. ........ rae Aldrichi (Paludestrina) C. and B. ...... 137 Aldrichi (Somatogyrus) Walk. ......... 142 EMIS MOE atta) 1SC ar veesiate. alle thence llores ies 33 Altipetum (Pleurocera) Anth. .......... rats fAilfonensis. (Physa)) Wea) oc: wie sicie sie s 106, 110 AS CAMeyLUS)) “DryOn ..sc > 06 sere 2 0 117 PMTs OMG AaTIK)) TV ONW \ scan vehs rm (ay seate, fete. 'speretss e 117 miveire GPlewrocera)!. Cotte. aces ce ai 0 50 151 Ambigua (Alasmodonta) Say ......... 64, 178 PATILeI a: ALE, Mohs suo; oie\cus forste,010 9 42. A Le One TT MATES EA AL es Mec catiaseire iaiieltreils Toe cous eiah Sceaci eye's Tar7 mArmmorn) CPlarorbis)) “Glds 3.6 0% secs sens 103 Amnica.(Leliuna) Mull oo. ec. eis wet 88, 200 PNG Olae Gre and) EMSS ss:6 cence e miale-e.s, ca 28, 133 PNENIETE OLLI CEs teicies clio sit se) bs shone tysrie cue teenies 23-27. 032 PIRI LE GIL aire a syne areas “osetia feos 27, 28,133 Amnicoloides (Somatogyrus) Walk. ..... 143 Amnicnum’ (Pisidium), Mull. ..... 0.05.00. 200 Amcena (Lampsilis) Thea... 2.020650. 5085 181 PNURLPEELOUsa eM TI See < sess nus, svslisis shasyates sate naioieys 22 123 Amputlacea: (Physga) \Gld: <. 5.05.0 oes 107 Ampullacea columbiana (Physa) Hemp. . 107 PIR TATU Ae MATT Ls sche wicieiel ecsteronecehe. s(e'iace 23, .123 PATENTAMT 22 ae njs:sfeyricis 5 e)tis Sieiers «8 ods, ote 235 429 Py ecalOnalas, C. -aNG iE, (sss ce sete tes 65, 69 mimyeaaiis CPhysa)s SDy i ‘ej eleiecia sos as 107 Anaticulus ohiensis (Unio) DeG. ........ 172 PAiatian Ce SA), Leeda. 9.) iss ss cis 0 stele ane 107,°113 Ancillaria (Physa) Say ....106, 107, 110, 115 Ancillaria crassa (Physa) Walk. ........ 107 Ancillaria magnalacustris (Physa) Walk. . 107 AN CUlOSAM SAY waicalersiontelscre SAEs 7s (1275) 0A0,eLOS EMA ICE co cod oMpOoo pCO amon ome Sy TOs 207 Ancyliformis (Gundlachia) Pfr. ...... 205 X21 ANP trem pep eda Gear elle, atten a lekescv closets 35, 149, 150 Angitremoides (Io) ‘C. C. Ads. ......... 109 Angulata (Angitrema) Weth. ........... 150 Angulatas¢Anodonta)) Kea) 2 eeeisc- se a0 sere 54 Angulata (Lithasia)) Weth............... 150 Angulatay (bulotoma)) seats are sis slice onic alee 130 Angustatus cuniculus (Unio) DeG. ...... 184 INMNelori LENE, GSyoaoc ei DONO oOAU Hi son auc) Anod omtirtas es Onitoet aiel« clieiatet-lalcle iets lle Ao. S40 D716 Anodontoides simpy fe. 5 ser <1 clay eee Sig melien Anodontoides (Lampsilis) Say .......... 181 Anthonyi (Anculosa) Budd ............. 36 Anthonyi (Eurycelon) Budd ........... 149 Antrosus (Planorbis) Con. -.. 2.5...) -- 95, 100 Antrosus angistomus (Planorbis) Hald. .. 95 Antrosus arostookensis (Planorbis) Pils. 95 Antrosus corrugatus (Planorbis) Curr. .. 95 Antrosus percarinatus (Planorbis) Walk.. 95 Antrosus portagensis (Planorbis) Baker . 96 Antrosus royalensis (Planorbis) Walk. .. 96 Antrosus striatus (Planorbis) Baker ..... 96 Antrosus unicarinatus (Planorbis) Hald. . 96 Apicina, Quymmea) Wear ce ck. ci oe nio6 ont 93 Aplectoides (Physa) Sterki ............. 107 Aplexcmbillctuiget siienveneclsteieliolewsictel olererelele icity KOs NL Arachnoidea (Goniobasis) Anth. ......... 155 Aratam@GotioDasis) aed am airiere kale) oiterena sie 155 AratumbaC@blemrocera) Wea pactere cc oclass onte 1sI Mesa Sales y Senecwanscc pocaaoo OUT 5G coe Arcticus (EL lanorbis) elo ects ««leeie' eyelets) 2 96 Arctior fisheropsis (Unio) DeG. ........ 175 Arcula e@WMarganittana)s Wea ec. 6crc:0.s «.e 01 ue 64 Argenteum pannosum (Pleurobema) Simp ie Freier ope etete uetatons/eteivafar sie reiar kere L7D Ly Arizonensis (Planorbis) P. and F. ...... 96 Arkansensis (Anculosa) Hink. .......... 163 Arkansensis (Lampsilis) Lea ........... 181 iNsdehalsjis, (0), Evel Wgeaoonanomooogedese 55, 58 ipauegor Islay, “Gbohoadpaoecun boo oeoo 13, 94 Armigera (Melania) Say ............ 35, 149 Armigera (Segmentina) Say ............ 104 Armigera campestris (Segmentina) Dawson 104 Armigerus (Planorbis) Say ...........-. 14 Askewi (Quadrula) Marsh ............. 166 As perninamis sG@Wia10) uean . sfavelc) sla cietele| «> ltiars 167 PSS Mekol seve)” Goma mocpooedudodesoGec 37 AterinapCGOntopasis)s LsGal wes is) -Anthewicyscuterere ele ieee > 155 Badium: CAlasmodon) = Raiouem-cle reece ts 177 Baindianay (Goniobasis)) weal te sic e ie ele 160 Bakeriana (Amnicola) Pils. :........... 133 Bakeriana nimia (Amnicola) Pils. ....... 133 Bariostac Rafi is iniecre sve steiteistotaciats errs + oyee 170 Barnesiana (Fusconaia) Lea ....171, 172, 173 Barnesiana bigbyensis (Fusconaia) Tea istarshetateiatetscatenetel skatorops > ovsns'(o\ vca' 084 rier, 187} Barnesiana tumescens (Fusconaia) Lea .. 172 Barnesianum (Pleurobema) Lea ........ 171 BAsOMmMaLOp Mousey: sine clelarclereien crete GenOs Beadleanas(@Quadrula), Weal. mace eric cons 166 Bengalensis (Cyrena) Lam. ......:..... 85 Bengalensis (Viviparus) Lam. .......... 126 Bentonensis (Goniobasis) Lea .......... 155 BerendtuGelanorbis)» Lryom 2.2 eset 105 Biangulatus) (elanorbis)) Sbysneassee ess: 95 Biangulatus (Somatogyrus) Walk. ...... 143 BicaniiatayGVialvata) wueaanientecmeeiie eb 130 Bicarinata connectans (Valvata) Walk. .. 130 Bicarinata normalis (Valvata) Walk. .... 130 Bicarinata perdepressa (Valvata) Walk. . 130 Bicarinatus (Planorbis) Say ....... TOA OG Bicarinatus angistomus (Planorbis) Hald. 95 Bicarinatus arostookensis (Planorbis) Pal S85 ite cuckoo MeO EE OR 95 Bicarinatus corrugatus (Planorbis) Curr.. 95 Bicarinatus major (Planorbis) Beck. .... 95 Bicarinatus major (Planorbis) Walk. .... 95 Bicarinatus minor (Planorbis) Beck. .... 95 Bicarinatus percarinatus (Planorbis) Walk. 95 Bicarinatus portagensis (Planorbis) Baker 96 Bicarinatus royalensis (Planorbis) Walk.. 96 Bicarinatus striatus (Planorbis) Baker ... 96 Bicarinatus unicarinatus (Planorbis) Hald. 96 Bicinctum (Pleurocera) Tryon ......... I51 Bifasciatay (Gonitobasis)) Weal aescaeee sce 155 Bigbyense (Pleurobema) Lea ........... 172 Billings: \(ebysa)) berons.-se. esses: 108, III Billings WCelanorbis) Weaweseeeceenee: 06, 102 Biloba ((Epioblasma)i Rata. seem nee 185, 186 Binneyana (Cincinnatia) Hann. ......... 136 Binneyana (Physa) Ancey .......... 108, 113 Binneyi (Bythinella) W. G. Binn. ...... 148 Binneyi (Planorbis) Tryon ..... Bieayes 98, 104 Binneyi (Pomatiopsis) Tryon ........... 148 Biteniatum (Pleurocera) Con. .......... 154 Biandae (Gontobasis)mlueas see cee e nee 162 Blandae (Melania)iwuea 121 GCallceolay (Swi. acieuerercvehsteletiehers/ore caek eens 178 Calceolus sciotincola (Unio) DeG. ....... 178 Californica (Goniobasis) Cless. ...... 155, 159 Californica (Gundlachia) Row. ...20, 119, 121 Californica (Hydrobia) Tryon .......... m7, Galifornieas @Melania)? Cless) oa. ace eee 155 Galifornica (Pomatiopsis) , Pils’ 2.020. a2. 148 Californicum (Pisidium) Newe. ......... 189 Caliginosa (Ampullaria) Rve. ........... 124 Galli "(CWalvata) Se Elantise eet crater 130 Callioglyptus (Planorbis) Van. ....... 96, 101 CallinaysHannibaly orem. sce Lneoee ee ooo 125 Calyeulina “Cléssi: eae sissies ee clotssischen eres 188 Campanulatus (Planorbis) Say ..... 12, 94, 97 Campanulatus minor (Planorbis) Dkr. ... 97 Campanulatus rudentis (Planorbis) Dall Sens tvet pus seater tapers TNE staat St eee Toreietebeer 97, 101 Campanulatus smithii (Planorbis) Baker . 97 Campeloma Rate cree series emcee 24, (25, 127, Ganthyria, “Sway ane ite celeit dietrese eae Sits Gps Capax (@Vampsilis) (Green) ~. mene sean ees 181 Gaperata) (fymnea) Say erie nee ae 8 Ganditimes Qeamipsilis) kates eae nie te 184 Cazibeus (Planonbis)) dt@zbas seer 97 Carinatay GE alicia) mic leececnteretoeeiereieteniete 126 Garinatum (eleurocera)s ears sc. seen cles TS. Carinatum (Strephobasis) Lea .......... Isr Carinatum (Trypanostoma) Lea ......... I5I Carinatus (Neoplanorbis) Walk. ........ 123 Carimiferad(Physa)rAgic eyes ic eeiel eee 103 Cariniferum (Gyrotoma) Anth. ......... 163 Garimifex WG: Binney eerie es sieie 14, 15, 106 Garltoniu: (@ehysa) eiueasiesenieen eee 108 Gatiosa i(lampsilis) Savin. isle erate so atte 181 @arolintanaGCy.clas) ip BOSGes seteiisle eieieraeiatels 85 (Gehabhaeuilbhoey Shhayoy Facog doote. 66, 76, 180, 183 Garus*=(Planorbis) ies and hemi clelsietelaan 97 Gastaneum (@Pleuxocera)) eae ceisler ISI Castaneum (Trypanostoma) Lea ........ I51 (Griinky sopbabey (Usa) IDECe Soe 554 snedoac 181 ‘Gataracta sCAnodonta) eo avin iseisem etic 57 ei Catenaria (Goniobasis) Say .:.....+-0 0. 155 Gatillus (GUinto) § Cons pete) teres aeiebreiciots 166 Gaurina (Peprissia) si Ge COOp rile ete 118 Caurina subalpina (Ferrissia) J. G. Coop. Caurinus (Ancylus) J. G. Coop. ......... Caurinus subalpinus (Ancylus) J. G. Coop. Centrevillensis (Planorbis) Tryon ....97, @erea (Gontobasis)* leay oo Sac ae eye ore = avec PENT ASTAM OW. Uaaalelcie aieie ci siaitisie,«) sieverels velar’ Chickasawhensis (Quadrula) Lea ........ Christy: (Segmentina) Dall.............> Cigenda (Goniobasis) Anth. ............ MBNIGITI NAIA TE IUS, §1 cctejaie'n)e|e ats /cloleleiteie.e ote 29, Cincinnatiensis (Amnicola) Anth. 135, 136, Cincinnatiensis (Cyclostoma) Lea ....... Cincinnatiensis (Paludina) Anth. ........ Cincinnatiensis (Pomatiopsis) Lea ...136, Cinctum (Trypanostoma) Lea .......... Gmerella (CGontobasis) Lea oo. ccs sees Sipangopaluditia oElarins)s 3 %<.,-1aie.s snelc seis cree Gircincta (Goniobasis): Leas. . Jtalsie ous Grrculus /(Obovaria), Weal fisasi.e crite ois cic Circumlineata (Goniobasis) Tryon ...... Circumstriatus (Planorbis) Tryon ....97, Glappie(@lappia)” Walkie ove ome.c « srais ie 31, Clappia Walk. GinnkerscAmmicola), Pils. jaiccgc ce cic ele. Clarkii (Strephobasis) Lea .......... 149, Glathratay CGontobasis) Wea) oy... ccc cee C@lathrata, CLryonia)) Stumps.» «<6.» «1 29, Rel ayama (CO TITO)) GIGAMIN. | s: Raf.” ois .i6 « seic.e 6 0,0 01s Glinchensis’ (lo) Co Cx Ads: occas oles > Clintonensis (Ptychobranchus) Simp. .... Coarctata (Melantho) W. G. Binn. ...... Goatctata, (Paludina) Wears. i ti..<0 «es (oarctattise Chlanorbis) SDYe a <\0,0% see ns Coccinea (Quadrula) Con. .......... 166, Goehinaris CGoniobasis)! Lew, cai. va cee MORMON A SUID, cc sicticts o's: s/c cisielclsteiets,'e © Bie (Sores aarasts OG Opetio) ee) Eyer eae acto nico SIGIGIND cleiiea @okeriay Marshall M .o).sksteiacs etslels © a'sie.e's 83, Columbiana (Fluminicola) Hemp. ....... Columbiensis (Goniobasis) Whiteaves .... Golumella (Eymnza) Say <2). stots oie Comalensis (Amnicola) P. and F. ....... Comalensis (Goniobasis) Pils. ........... Comalensis fontinalis (Goniobasis) era TIL Bette oc satararaiseise ates Secreta e iste sie sieychele Commutatus (Planorbis) Sby. ........... Complanata (Alasmodonta) Bar. ........ Gomplanatus, (Unio) Dill. an 3 ices Gomptessa (Lasmigona): Lea ....:....0.2.. Compressa (Symphynota) Lea ........ 60, Compressa lindus (Unio) DeG. ........ Compressissimus performosus (Unio) DeG. Compressum (Pisidium) Pme. .......... Compressum limnicolum (Pisidium) Sterki Compressum smithii (Pisidium) Sterki .. Concestator| (Lampsilis) Lea ............ Confragosa (Alasmodonta) Say ......... Sa sert ey D8 eyriCtd aes COC GDIGIECE: Hootie caine Contrormis: (Physa) Tryon, 2 .:..¢ «.\ 107, Connesaugaensis (Goniobasis) Lea ...... Conradt (Pleurobema) Vans)... 20s ee os Cauradiens. CUMI0) Mea s smizsiale » efe)e.5 s sas Consanguineus (Unio) DeG. ........... INDEX 118 118 118 IOI 158 151 166 104 155 136 203 Constricta (Lampsilis) Con, ............ 181 Censtrictus (Somatogyrus) Walk. ....... 143 Contectoides (Viviparus) W. G. Binn.124, 125 Contectoides compactus (Viviparus) Pils. 125 Contectoides impolitus (Viviparus) Pils. 125 Contectus (Viviparus) Millet ........... 124 Gontieual(Goniobasis) aed) sor els oie oclere 157 CoGpert, (aymnwea) Earnie: cis se 0 corer > le 94 Cooper Gey Sa), WESyOn! fits sic leis s'sofelete © 108 Cooperiana (Quadrula) Lea ............ 166 Cooperianum (Musculium) Pme. ........ 188 Cooperianum (Spherium) Pme. ......... 188 Coosaensis (Lioplax) Tryon ............. 130 Coosaeusis mCP aludina)) Wea fain. > + «0's ole 130 Coosaensis (Somatogyrus) Walk. ........ 143 Coosaensis @fulotoma) Wea) Fo... acs aes 130 Coosaensis (Vivipara) W. G. Binn. ..... 130 Cor) (Blemrobema) Gort 23s)< sca oe os L72> 173 GordatamcOboyaria)) Ratseren otclccrcleie clei ete, = 168 Cornea: (Strephobasis)) Wearsia-. sects 8 sree 149 Gornewe (ell inal srevehereiniste'slelatels) s' siete 86, 200 (OGrmtseiey It GapodeacoopooUmoDe 189, 200 (Glorgotaaiey (ChiskheudnaodocccoN cob uUadorOND Oo 199 Corneum (Pleurocera) Lea ............. 154 Copnenme CSpleeevtys) eee -reyetevats =) s/sleyote 01s)l= 199 Corneum (Trypanostoma) Lea .......... 151 Gorneusy Cedamocbisy) aes crc c.cvc er eversletalet abet 94 Corollas@Vielania Gide wie csarctc's © ol clolere olor 30 Gardnatis: Ge aluidinia)) weshtesarcla(cleveletelele « ole 140 Coronatus (Potamopyrgus) Pfr. ...... 30, 140 Corpulentus (Planorbis) Gould .......... 104 Corpulentus (Planorbis) Say ............ 98 Corrigata CEbydrobia) “ErildS 2s... 6. 137 Corrigata (Paludestrina) Frfld. ......... 137 Corrugata(Gonitobasis) ied) verte cle re er 158 Costata (Amblema) Raf. ....47, 168, 170, 171 Costata .¢Alasmidonta)) (Rats, 01. «(ictiereele «= 60 Goastatae Ge lysa) ee Newt telerefeld siteialn lei rs 16, 106 Costata (Symphynota) ‘Raf, ...6....5.... 200 @ostatelianap Dally cr rerecster-veicioveiereleve) of oevel siace 16, 106 Costatus (Planorbis)) DD? and B: .....-..- 98 Gostulataes( Gomtobasts pelea eee srerele aiclstel ese 158 Couperi (Goniobasis) Lea ........:..... 155 Grandaliv = (Goniobasts) (Suse fjeeisie 1eyelele aie 157 GrandallaGehysa)) Baker tera ejeis core c10 109, III (Grpyoosiess (QUinwiey): ILE 6 nooo nooo canoe 173 Crassag(Baunycoclan) etal de) <)or- nas) alelatciay alate 149 @rassa a GUinit0))melcateeecisters cre clsiereleretaretelaveleyet« 182 Crassidens@Uimio)” Wal: velere ais civic sins 0 52, ld: Crassilabris (Segmentina) Walk. ........ 104 Crassilabris (Somatogyrus) Walk. ...... 143 Crassula “(Campeloma) Raf. ...........- 25 Crassusm, CAncylus)) Eval fi... 6 «ioe s/ee!s cere 117 (Gira, (besa yal Gea ee oor ano comonon 117 Crassus (Somatogyrus) Walk. .......... 143 Crenodontar SCHUH E. fc. ceyersipalolelnrg Sel sie's (ai 171 Crista MGR amon pisi elas mies ce) eeene tarelss = 13, 94, 98 Gristatay Ovialvatay Vitaly weiss cicialsterwi)alsiel eles 27 GrocatamGeby say eqlscamme cre: < 171 @yprogenia PA ges cry cloister elel= 65, 68, 179 G@yrenay Wane rs crsteers cele cere et oneretatenan 85, 187 KGymenastr it sO Pty wate ole ocletaieletenetoreiecs -12e 199 Gyrenellam Deshi seiacineetemieniricrsrsr 88, 190 CNG STEIMGESS coo oon nob nome nao aoc 39, 88, 190 Gy remidee Posies ays wie aiets ocsshe report 39, 85, 187 WallisiGCAmmnicola) ye Galle gsr. cece crersislse iene 142 Dall Geyogyrus) Pe and Bo 2 ose eee rere orale 147 Decampii (Goniobasis) Lea .............. 157 Decampii (Melantho) W. G. Binn. ..128, 129 Decipiens (Somatogyrus) Walk. ......... 143 Decisa flava (Campeloma) Curr. ........ 127 Decisa melanostoma (Campeloma) Curr. . 127 Decisum (Campeloma) Say ...... 255, L276 026 Wechivis @Planorbis)ie Sby.: cise emilee 98 Declivis =(Planorbis)) late = cele sree 165 Declivis (Segmentina) Tate .........-... IO Decuitambisw katate aoe reyauereit 178 Deflectus: (GPlanorbis)) Say.- eerie. te 98 DEH exds |Celanoxrbis)) ISDy.. curse ieteuiel seat 98 iD Gidacive Gehysey) Cite Baoboonororc 109, 110 Dehiscens orionopsis (Unio) DeG. ...... 175 Depressa, (Amnicola)" Tryon 30.0.6... .- «6 32 Depressa’ (CAmpullaria) (Say ~~... 5... =) 124 Depressa (Obliquaria) Raf. ............. 17 Depygis (Goniobasis) Say ............ 36, 157 MWesentarn¢Ammnicola)) (ils) wee cic ele eit slete 133 Deshayesiana (Goniobasis) Lea ......... 158 DeviataoCEruneilla) cAmth~ 2205. ae cee 185 Diaboli (Paludestrina) P. and F. ........ 137 Diaphana (Ferrissia) Hald...:....... 19, 120 Diaphanay(Physa)i ry or 2 cele) irae nse 108 Diaphanus (Ancylus) Hald. ............ 120 La yregspolak aS) 00) n deole anictLaO GO De Ooo UN ONC 170 DilatatanGU nia) Ratha metic oe) steunetotar (oi aie 175 Dilatatus (Planorbis)* Glas... a. = he 98 Dilatatus pennsylvanicus (Planorbis) Pils. 99 Distinguenda (Physa) Tryon ........ 108, 109 Dollabelloides (Pleurobema) Lea ....... 173 INDEX Donaetformis (Cyrena)° Sby.. .. 0... 20 0 187 Donaciformis: (Plagiola) Lea 2.02. 00.080 179 Donaeopsis: (Unio)! DeG.. to. .hiee et. os es 185 Dorbignyana (Physa)) Lea 22... ci 109, I10 Downieana (Goniobasis) Lea ........... 155 Downer CAnculosa)) Suea Tryon 5 ..2..is...s/.s Fragilis (Ferrissia) Tryon ........... 118, Fragilis (Lasmonos) Raf. Fragilis (Physa) Migh. Fragilis (Unio) Raf. eee eee eee eee ee were ee eens satel a (aim, mis'islisce os INDEX 120 205 Fraterna, /¢Goniobasis) ,Lea .......25 0 158 Friersoni (Quadrula) B. H. Wr. ........ 167 HuscasG@ernicsia) s@.4 Be (AdSanem «nosis cas 120 Fusca eugrapta (Ferrissia) Pils. ......... 120 Busca (@Pluminicola)’ Halds: 5.0 .cck. oe. <2 141 Fusca minor (Fluminicola) Stearns ...... 141 HMuscatusm unto) luea vas aelataers misrsleve oe vie aie 175 Dy ebsferoy ele, (Srhiths wis AS Cee ee 42, 48, 171 FuscussCAmeyilus)- C2, Be Ads: -.... 2s. 19, 120 Fuscus eugraptus (Ancylus) Walk. ...... 120 Galba, GSchienlesercrectacsticia koalas a oie tiean 6, 8 Gastropadapent ve, foteieatete tos eG cctecclals mcs ae 5 Gaudichaudii (Littoridina) Soul. ........ 30 Geniculata (Anculosa) Hald. ...:....... 149 Geniculata (Withasia) Hald. ............. 35 Geniculum (Campeloma) Con. .......... 127 Georgiana altior (Vivipara) Pils. ........ 125 Georgiana fasciata (Vivipara) Tryon .... 125 Georgiana limnothauma (Vivipara) Pils. . 125 Georgianus (Somatogyrus) Walk. ....... 44 Georgianus (Viviparus) Lea ............. 125 Georgianus altior (Viviparus) Pils. ...... 125 Georgianus fasciatus (Viviparus) Tryon .. 125 Georgianus limnothaumus (Viviparus) Pils. 125 Gesneri (Gontobasis) Lea ......-...7a--. 158 Gibba (@Mielantho). Gurteeen on 22s 129 Gibbosay CAmblema) Rafts ...25. 0.6 sees 186 Gibbosusm@Wanto) eebars sats ts eee 175 Giganteusm@Uinio) ge Bataan on circ neee 167 Gulia: (Stimpresee eter ne ccc ae Bi sou THO Glabra W@ehy say ieDelmava mts aarseaide acer. 116 Glabratus (Planorbis) Say ....... 99, IOI, 103 Glandulum (Pleurocera) Anth. ...... 152, 154 Glans GG@arunculina)lueamee cease ee 180, 181 Glaucay(Goniobasis)mAnthiawssseee eae 158 GlebwlagiGonhkyr coos ween tne ore ee 65.5 70 GlobosamcAmpullaria)miswarsseees cere: 124 GonideayiGoniiiaccrttcnie as aan eee 42, 54; 175 GontobastsmMea seria ee 34, 36, 149, 154 Gouldiana (Gontobasis) Lea .)........2-.. 158 Gracilentis@Planorbis)mGldeeeseneeaeeee. 100 Gracilism@eampsilis ebatardee ect 182 Gracilisg@Unio) me Bane rst sas See 73, 180 Gradatum (Pleurocera) Anth. .......... 152 Grana “(CBaltidina) eis aysererie: ciao tele 147 Granatoides (Goniobasis) Lea ........... 158 Grandisi@Anodonta)e Say: crisis ss cloe oe nelee 176 Granum (Amnicola) W. G. Binn. ........ 147 Granume@eyocyrus) Save dessin cceen oes 147 Grosvernori (Physa) Lea ....... TOS, LOM ves Gundlachawibineassscaie esc 18, 19, 20, 121 Gwatkiniana (Rhodacmea) Walk. ........ 122 GyraulusipAwassiziyan che cece cee TOs D2, OH. Gyrina (@Bhwsa Say sac ectisencsoscee deine ¢ Spc ee te HOveHOO, LIO;0iT, Lil, Tr 3. Ted, Ts Gy Totomeasnsbittin ects oles. 345 345 \T40; 102 Haldemanra(Aneylus) wets i sss. scls olee t I19 Maldemami@Berrissts) Betel 2. -/< ees. ne 119 Haldemani (Lymnza) W..G. Binn. ...... 8 Haldemani (Planorbis) C. B. Ads. .... 102 Haldemani (Planorbis) Dkr. ............ 102 Haldemaniana (Vivipara) ‘Shutt.’ Frfld. 125 Haldemanianus (Viviparus) “Shutt.” Frfld. 125 206 INDEX HMaldemanina Dall! cccrin. aeuienmten corte ie 105 Eraleanuss CViviparus)aloea aarti tele. scr 125 HaleisGBhysa)) Lea. aigistete sietatortcc clea gare a crc Ill Hallenbeckii (Goniobasis) Lea .......... 155 Harn (elanorbis)); e1lsae. aie. te ete 102 Harpery «CAmmicola)! Wall) atreciassie sce 134 Harpethensis (Anculosa) Pils. .......... 163 Hartmaniana (Strephobasis) Lea ........ 154 Hatcher (oittoridina) = eilsee seepaeide ee cee 30 Hanutieniay oll Foycie.ce siete lari eats te lols 33, 147 Havanensis (Planorbis)) (Pir. .....-+.52 02 100 Havanensis (Segmentina) Pils. .......... 100 Hawn @bhvsa) mises jesiecuwts seteeiiek LiGyeaLt Heathella, Gilanmiball .c resi sreers ete, oie ais Shoals 141 Flelisoma owas erica erareteietersietie 10, II, 94 Pliemilastenam orp: urie store fe Gieieiei i sereeke ore 178 Hemispherica (Ferrissia) Walk. ........ 120 Hemisphericus (Ancylus) Walk. ........ 120 lemistena Rai. cejatere ore cress cvsieeereranets sce 175 Hemphilli .(Bythinella) Pils. .........<% 137 Hemphilli (Paludestrina) Pils. ....... 0: eS 7. Hendersoni (Ancylus) Walk. ........... 119 Hendersoni (Ferrissia) Walk. .......... 119 Hendersoni (Somatogyrus) Walk. ....... 144 Henslowianum (Pisidium) Shepp. ....... 200 ELE OS GN attGay | SAV si «jcc cieseterseheeisterarsistae 123 isles (CObexsbaoeO) SEN? Senmodsce qeone aoc 167 HienosuG@WU mon Say.2 = «:scuclc)sstesslateeterers a 46, 171 Elesvome rd Oxbs, «cred sepele co ceita's cicaeerare-n ate 174 rHeterodom GUinto)) eal a. s snreaeeie el Ose LTS Eeterostropha aCbhy sam soayn cise ai eis aiet sifelsle nuele ea eye ie aleys 16; 107, 10S; 109; Ili, 123 Heterostropha alba (Physa) Crand. ..111, 115 Heterostropha peninsulie (Physa) Pils. .. 109 Haldrethitarnias (eb ysa) sean settee TOs LS HinkleyriCAneylus) Walk sei. eee orntene 122 Hinkleyi (Gundlachia) Walk. ........... 20 Hinkley: (Bomatiopsis). Pils clete 158 Masmigonar Raters nicieeieieret 555595, O05 airy, WasmGnos Ha lorpencteheroiceieierctcteteieie tye mteiars 180, 182 Mastenay Rats < smcle ve tslstewsiohets srereeeyeis 43, 53, 175 juata’ (Anodonta): Rais eos heen 54 INDEX 267 REAL CEASLCN A) URAL. Yor so, ctetacoatsic eect sires 175 ata Ge wysa)) UT y Oty stile see's wherein oo eee III Bateralis, (Obliquaria)® Raf. o:5-0s-.-<- 167 PACISSt aay CU MLO wr Reale bs acc «.6, cle evetcus’s wre oe 184 Rautus CPlanorbis), H. and A: Ads...°.-.. 104 Vawrencei (Goniobasis) Lea ..........+. 138 Leana (Pompholyx) H. and A. Ads. .... 105 Beam (Gomobasis) Tryon. $5..9.5. 05s 007 162 eatin CGoniobasis) Brot) 3... 6.6 6.45 sis eee 158 merevrersGerunciHla)! Utt.» sc ens ce cms oes 185 Lecythoides (Vivipara) Hann. .........-. 126 mehnert: (Liogyrus)..Ancey ....:....23-- 147 Kehnerti (Lyogyrus) Ancey ..........s: 1A7 (LaIOS ReT e eeEe 22663, S0,, 185 EIS CODOVATIA) 012A) rake wraieilhays om a ayers ecties ane 180 PeerasuEanOrDIS): 6a fsjs\<:< so ors lelsve.< vais e ote 98 Benticulare (Spherium) Sby. ..........- 188 Benticularis (Planorbis) “‘Sby.°.¢.....5...: IOI Lenticularis (Pleurobema) Lea ....... 171, 173 Lenticularis (Spherium) Sby. .......... 188 Wentts —GPlanorbis) “Say Scie. .e ns oess 99, IOI WE HtOce ae Rate rr 5 Scie.» alieislavlstaccoes 180, 182 Leptodon (Lampsilis) Raf. ......0.%. 180, 182 WRG DEGMISCO RAL isc cvars ss sale sidleisrclanetee te e013, 163 PA UPMEIRILU Dal we atcyata Steves ace eeneke nara 38, 164 Wesley, CPleurocera) Wea 2.050.000 sess 152 mesievraCerypanostoma) Gear... +0 «+ 152 Letsoni (Amnicola) Walker ............ 139 Letsoni (Pyrgulopsis) Walker .......... 139 Leucophzatus (Mytilus) Con. .......... 84 PCAC ATH ECRLTHIO ie EMAL orc > syajslare vise eaietorexers © 180 Lewisii (Campeloma) Walker ........... 128 BewisiGGayrotoma) Wea. stcrtsc ess << «i's 162 ews (CPleurobema)) Wea ~.% .005.-. 220% 171 memisweGelenurocera) Liea eo scccteyslee sea s 152 eawiswnGschizostoma), ean so.>ca0 ales ccs s 162 MiewmAStIuG@W dlvata)) Curres s.ccaisieiaee ee 2% 131 Lewisii helicoidea (Valvata) Dall ....... 131 Wextmrtonia Ortmann ...6 63..61s it 43,7) Sip 174 Liebmanni (Planorbis) Dkr. ..... 94, 101, 102 Wienosa, Cluampsilis) “Cons, .. oc oe <2 ss 182 Ligamentina (Lampsilis) Lam. .......... 182 Ligamentinus (Actinonaias) Lam. ....... 75 MOTTTOD HEA «avons ciciqyc.cucis seis els.o-0,4 ote's 8x tyes 55.93 Limosa (Amnicola) Say ....133, 134, 135, 136 uN OSA NCE AlIGINA)ASAVe ye cries ics wn srocie ° 29 Limosa porata (Amnicola) Say .......... 134 Limum (Campeloma) Anth. ....127, 128, 129 femiearis sGoalugitia) INGSten. cus +6 ve diem 12 Lineata (Vivipara) Kuyst. ........... 124. 125 maneata, GVivipara) “Wali onic sccnec es 124, 126 ieineolata (Neritina)) Lam.) oc neces sce 164 inaneolata (Obliquaria) Raf. ...<.......<. 179 Peery bake et ctl cris ects iictuiaefere:or4 «0s ms'9 © 24, 26, 12 Berehiisia: Pal, 4e\svcc wise) a2 ¢.0.01s 54.0317 34, 35, 149 Mithoglyphine: Fischer ...:....-...- 27, 30, 141 MMEEEOTICIIA™ SOUL oy «. claisia eiece coe c's 0 ane 205 0305, 142 Moudonensis (To) ©. CG. Ads, <5 005 550055 199 Livescens (Goniobasis) Mke. ........ 156, 157 mrteolas (luampsilts)) am occ catia ce 0 182 Wervidus, (Loxolasma)) Raf: oss. sh.'--«- 180, 181 Longinqua (Amnicola) Gld. ............ 138 Longinqua (Paludestrina) Gld. ......... 138 Loge (EM GED 1 316 Begg erctneiceaicis Stas 107, III lenstrica cCAtmnicola)’ Haldts toi. -.sicens ss 148 Eustricacammicola) Pils.) s,s s.a.a0coeae 134 Enstrica Graltidina): Sayin «ccs sls a ateneets 148 Lustrica (Pomatiopsis) W. G. Binn. .... 148 Luteocella, (Gontobasis)’ Lea... 20.5... 159 LYM Neidcaewr lemtclad tet ccmcire esate eiee see 5, 93 Myrna a here) reves a citi esis! t atzvster sores 56,05 Ly ogyaimre eis. pM ni sie iste cree ss Stiale 6 Sets 275933 E-yOgymtsteGillorasrclaratolerne > aicim aide coisa cele s oe 33 Lyonme@Gontobasis)® ea 25.4. 0. 2 5h 158 Lyon (Pleurobema) Lea ............... 171 Lyonii «(Pleurocera) Lea .:.. 02.2... 152, 154 Lyon €Strephohasis)). Leac.o 4. oe cc wie 154 Lyonii (Trypanostoma) Lea ......... 152, 154 Lyttonenusts: Glo) @.1G. *Ads:> \3.c55% acces 199 Maculatum (Pleurobema) Con. .......... 172 Maculatus= @Uimip yn Cons 26, 3 occ crete soe 172 Marniticas@Paludina) Gon: %\.. «. «0s 128 Obesa (Paludina) W. G. Binn. .......... 128 Obesa. (Tritogonia) “Sumips 2 aces ps se aie oe 170 Obesa “GVavipara) is Piyone. a-ha ai ere 128 Obesus (Melantho) Lewis .............. 128 Obliqua (Quadrula) Lam. ..166, 167, 168, 169 Obhtattacia a Rat. orrite @ lelel-ieie peice ree 65, 67 Obliquata “Obliquaria) Rate sic a ata 186 Obliquus (Ancylus) Shimek ............ 119 Obovalis. (Obovaria) Raf. <....---.-.<-- 167 Obovatta Ratal céet); .aie tyes acess 20 36 Plcman(@iadtiia): Wea). sa cscs ae sae 167, 168 Plena (Strephobasis) Anth. ......... 149, 152 Pletiume@bleurocera)’ PArth” ce. wchoas acu os 154 Plenus interduos (Unio) DeG. ...;...... 169 Plethopasts Sim tite. aia clstesie sicre eile 43, 49, 171 Pleuristriata marmocki (Goniobasis) Weth. 156 Pleuristriatus comalensis (Goniobasis) Pils. 156 Pleurobema) Rafi Pirin. act. sss 43, 50, I7I Pleuroceramivad i tuanielesetis 10s. 0 sje 34; 36; “149; 1st Plewurocemde’ an iciis cles o0 ates v/s 92356 355. LAS Pleurocenatidas ate tite tr srs on iale. seisic.e cls 149 BlexatamGblanorpis) eit oc. iei< ea nie eae ache 102 Plicata, /@erthasiame Weths os, .0 ouciccacee < Ae 150 Plicatay €Ouadrala)) (Saya once sicko scorers ebdncne 168 Plicata-striata (Goniobasis) Weth. ...... 160 Plicatulas GGontohasis)iolea A on cea octets 6 15 Plicatam’ GPletrocera)? Tryon *.2.5..%%... 151 Plicifera, (GGontobasis) isea, Ges... cece 160 Plicifera bulimoides (Goniobasis) Tryon . 160 Plicifera oregonensis (Goniobasis) Tryon 160 Pluristriata “GWelania)) Say: 2 /cicc .. cc'e o 156 Politissimay GP hysae lsyone iat ot sn itse 113, II5 Polymesoda Rater a srereiaiere scepnieteiei overs nero 85, 187 Poly rly tis celepcrie dcctetani cae eters ocaeiaia ate 6, 9 Pomatiopsinz: Simp.) 5 cc screu oles ¢ 27,034, 014o Pomattopsis® Uisyareme pos, ntesiciae 2 aie sts 34, 148 Pomiliay CBhysa)i Gomera went ye ctere cto 113 Pompholy gunz Dalle sive. < sit.cicteteta este 14, 105 Pompholyse > cause stele arearee roee 14, 105 Ponderosum coarctatum (Campeloma) Lea 128 Ponderosus=-@Bhanosta)) Rat.) acon. 2. case 170 Ponsonby1 (Carinitex) eB. Al Sms «oer 106 Ponsonby1 .(Planorbis) Sbye jase a vet 106 Porreeta (Gonmtobasis) elsea- +cat ass eto 160 Potamopy ees StmMip. ies cisig> «elie viele 28, 30, 140 Powellensis .@lo), C4 Ga Adsi0's tes seis icles 199 Preclarus (Ancylus) ‘“‘Stimp.” Lea ..... 118 Preclarus i(Ganx ss stimpi lea atest 118 Prerosa, CMetanta)! Sayan itr ciniele isis ale 37 Pressodonta. (Simp st cae sist eet lee ie ae 62, 178 Primeanay (Ehysa) is Eig omnes etaneteterees ELL EES Primellan Coopas a) & gets spoiele orale starctatstasiens 188 Prolasmidontay Ort oeneec tens tara 62, 63, 178 Propecelatus) @unto) (DEG? S525. eae 185 Propesulcatus (Unio) DeG. ..........%. 186 Propeverutus (Unio) DeG. ............. 175 Propingua (@Ebysa)aslryon oo. «cee ome 107, 113 Propteray Rateama resents rie: sve-- 65, 71; 177, 180 Protea -CAmnicola)INGlde sce alse ale wavelets © 138 Protea, (Bythinellayn Stearns: <)< 6s. miset 138 Protea, (Paludestrina)) Gldu 2. .cctean ssc 138 Protea (Dryonta) WeiGs Brtiti,) StGiatae(Valvata), GewiS ca. ec+tes «ve +s Desiatis (Unto), Goldfiiss oc... sels oes rrctditise CUmiO). Wea: o's..aiis 4 dhessrctereccateserace perOphitus Rat). een. sce SS, 50s BRO pvatatas-(Phiysa)>» “Mikes” oo. cers cvierelaarsie'e INDEX 211 140 Subcarinata, (Limnea)y Say cc. ces - 00 26 Subcrenatus (Planorbis) Cpr. ....... 100, 104 Subcrenatus disjectus (Planorbis) Coop. . 103 Subcylindracea propexilis (Anodonta) DeG. 176 Subglobosa (Anculosa) Say ............ 164 Subglobosa (Congeria) Partsch .......... 84 Subglobosus (Somatogyrus) Say ....144, 145 Sublirata (Goniobasis) Con. ............ 155 Submypitlaceaeesaewseparsteretsin,« ocreleeiateraye ei 39, 165 Suborbiculata (Anodonta) Say .......... 176 Subplamusm(Uin1a) Cone tc) i etter 51, 174 Subrhombica (Goniobasis) Lea .......... 161 Subrobustum (Pleurocera) Lea ......... 153 Subrobustum (Trypanostoma) Lea ...... 153 Subrostrata Goampsilis)i Sayin. ieee riots «0 184 Subrotunda (Obliquaria) Raf. .......... 180 Subrotundar (Physa)MSDyer were cin 6.) ae 0 114 Subrotunda (Quadrula) Lea ........ 167, 169 Subrotunda leucogone (Fusconaia) Ort. .. 169 Subrotundus (Ancylus) Tryon .......... 118 Subrottindus ‘Qsanx) oiry.onl aes cir. 117, 118 Subsolidum (Campeloma) Anth. SAAT Cn bry 3.000 OClOgU Aas 25, 1275 120, 229 Substriatus (Somatogyrus) Walk. ....... 146 Subtentus (Ptychobranchus) Say ..... 67, 179 Subteritus pulchreornatus (Unio) DeG. .. 179 Subulare (Pleurocera) Lea ....'....:...- I51 Sioinisvacahts SUOpaven) (Coin, 4 ae Bedgriads co ace 177 Suleatan (Gontopasis)misean setae atts iit steie 161 Stilcatay Chartier ame hea fey erer svete ieevensieneleneti 186 Sulcatam) (Spheerium)!Wamies ss mee. sets c= 86 Sulcatism (Uintonealseamnr cicincitacterontencie cine 81 Sulcwlasia GRatiptricmimeiceis tiajcteers sates sje oie 177 Supinum (Pisidium) A. Schmidt ........ 200 Symmetrica (Goniobasis) Hald. ......... 160 Symphynota, eae. aosocn 200 LO POGtSCUS TOtelue ri actete-ereaneienieraele 10, 12, 94 Mrothisw GSOMmatozy mus)! WOM sve) .tererstds etter 146 iirtincatay iC bntinenl la) sa few sts so cones erate 179 Truncatula (Lymnea) Mill. ........... 8 Truncatum (Musculium) Lind. ......... 87 Plertinenl lake = ee tere lelel sie levee stents 66, 80, 81, 185 iryoni \CAnculosa) sGewiS: a ciatse «= -lelsielers 164 ADrawopany ((szeueaeed) (Chars Gocoodeoocnaone 116 TLR CKEY (Shahi Goacacccoedpoua0es 28, 29, 139 Tuberculata (Obliquaria) Raf. ...... 49, I71 Tuberculata (Rotundaria) Raf. ......... 170 Tuberculata (Tritogonia) Bar. .......... 170 Muabecculatas q@U mio) bat aareteretotvs i -tecaterstet= 45 Tuberosus perlobatus (Unio) DeG. ..... 167 Murlotomarellalds: «.cieetelaietaitaisielsiikisi 24520, mens O Tumens (Planorbis) Coop. ......... IOI, 104 iamens, @elanorbis)) ite ees) veee seis 104 TNerayokeey (Gem ordye)) (bea) Gooon sdbaos 345 103 dibayereseh (Oke) WEEN Rodogodgcnooeoos acc 175 Turbiniformis (Amnicola) Tryon ......... 142 airway (GIG), ANadNa Hoo boodacdpoaso35 55 199 Umbilicata (Anculosa) Weth. .......... 164 Umbilicatellus (Planorbis) Ckll. ........ 104 Umbilicatus (Neoplanorbis) Walk. ...... 123 Umbilicatus (Planorbis) Mill. ........ 12, 94 Umbilicatus (Planorbis) Taylor ......... 104 Umbilicatus (Somatogyrus) Walk. ...... 146 Umbonatum (Eurycelon) Lea .......... 154 Unakestsis':(To) [C.sG. PAdS! ji chicee sacks 199 Undata (OQuadrula) Bar. see oes ae ee ee 169 Undata trigonoides (Fusconaia) “Fr.’? Utt. 170 Undulata ‘(Amblenta)’ Bar... «sen ste 47 Undulata (Quadrula) Bar. ...... 168, 170, 187 Undulata pilsbryi (Quadrula) Marsh ..... 170 Undulata: (Anodonta) Say ...:....... 56, 176 Undulata (Goniobasis) Tryon ........... 161 Undulata (Monodonta) Say .......-eceees 62 Undulatum® (Pleurocera) say .-c-eemeee: 153 Unio WRetEzZ “s.). 1 wis cibions esters eistvie clepkaeiepetare 174 Wnromerus Cons sraatetacascatsierseiaatels cucversas 43, 53 [Gkeiterertlge Hea. aunnibocecn nemo Aornah 39, 41, 165 Unioninee | Swi va: fates olathe eta eet esate othe 42, 166 Univittatum (Pleurocera) Lea .......... 153 Univittatum (Trypanostoma) Lea ,...... 153 Usrceus, CNerita) Malls sic ac civics cicierctele cles 23 Utahensis (Llymnza)? Gallo. se sc. se neces 9 Utahensis(Valvata)) (Call ree ..cctevelsre ole soe 132 Utterbacki (Pleurobema) Fr. ,...... 173, 174 Validus continuus (Unio) DeG. ........ 169 Valvata) Mull. 4% occ cavavsisce ewe’) » pistes ee 27, 130 Walvatidles © aries ten yo vst ove etye ors eres tore 23 5620,0530 Vanuxemensis (Lampsilis) Lea .......... 184 Variabilis’ (Clio) Wea Gre arse + aaeiencsre eaters 153 WVientricosa: Gaampsilis)= bate ss creer 184 Ventricosa cohongoronta (Lampsilis) Ort. 184 Ventricosa satura (Lampsilis) Lea ....... 184 Venusta (Anodonta) DeG: .c.-e eee ee 176 Venusta (GGontohasis) = ease ene tere 162 Vienusta Geanpsilis)mlheay -ssretseeialtietalotniens 185 Menusta me bysa))laGareterticiieeiainic ite Ts, 116 Venustium> (Pletirocera) Dea <.jam <1 iele 153 Venustum (Trypanostoma) Lea ......... 153 Vermicularis (Planorbis) Gld. .......... 104 Verrucosa (Angitrema) Raf. ........... 164 Werrucosa 6(lo)) IRiveaa. -sieeneiaie ie ots 199 WVetrrucosa (€Obliquaria) (Rak. fn nt aie 170 Verrucosa (Pleurocera) Raf. .......-... I51 Wiersaa(Gontobasis)) lscay cies teil terior seieeee 162 Mibex: (lampsiis)) (Con, cancer sions LOA Winosa, .CPhysa)? \Gldit ceva = ets reetcnerneints IIs WVaolaceus: (Unio) Spenig. f.< aactece cls ele = 174 Vairens(Planorbis) (©; BaAdS wae cenl 98 Virgata’ ((Anculosa))) Wealvcinctsen = siren: 164 Mircatas (Physa) Glday sae an eien 109, 114, II5 Wirginea (@Physa)) iGldaem ce ctelrel-etetateteatane 115 Virgata’ alba « g ®, © 6 6 6 ID \4 13 I 23 6, @ 66 & & © 6. &, @ i 2M ra 6 a hi os, od Ss) ha _ CMe ae | Me ne 1 . ozUdalO’T UBS "4 JO JIWUINS IveU UI0Iy paydeIs0j04g GONVA NIVN—VLYUVW VINVS °9P VGVAEN VUNdIS FHL UNIVERSITY OF MICHIGAN MUSEUM OF ZOOLOGY Miscellaneous Publications No. 8 The Amphibians and Reptiles of the Sierra Nevada de Santa Marta, Colombia BY ALEXANDER G. RUTHVEN WITH A DESCRIPTION OF THE REGION BY M. A. CARRIKER, JR. ANN ARBOR, MICHIGAN PUBLISHED BY THE UNIVERSITY SEPTEMBER 15, 1922 ADVERTISEMENT The publications of the Museum of Zoology, University of Michigan, consist of two series—the Occasional Papers and the Miscellaneous Publica- tions. Both series were founded by Dr. Bryant Walker, Mr. Bradshaw H. Swales and Dr. W. W. Newcomb. The Occasional Papers, publication of which was begun in 1913, serve as a medium for the publication of brief original papers based principally upon the collections in the Museum. The papers are issued separately to libraries and specialists, and, when a sufficient number of pages have been printed to make a volume, a title page, table of contents and index are sup- plied to libraries and individuals on the mailing list for the entire series. The Miscellaneous Publications include papers on field and museum technique, monographic studies and other papers not within the scope of the Occasional Papers. The papers are published separately, and, as it 1s not intended that they shall be grouped into volumes, each number has a title page and table of contents. ALEXANDER G. RUTHVEN, Director of the Museum of Zoology, University of Michigan. S CONTENTS Page eMC EiGi aE MeN rat are eta rh ANY oo Ss Recs Cornea Bd a8 5 OD TE YT eS a FS ...); miei dee... n2 69 lo ea fer 15 east pica OibittON diss Joh.) 052-21) ae Re ant ee ele 17 Ditciriee st area sa. oS ous Se See bo A 19 PescmsOn-Or TMG ElainiG. ics: .:.. 0) aoe terue Sean oes 6 21 Haarlem eoctabioires ty 802s). se. oS ar ok PON ode. 25 MOAR TICPR RCS e tas at re sc). ss 2s Ae eee ene 25 Een enn VCR MDNOSD 3 coo) esa, 3S ave «2 ee epost Ee ae 27 Peet eiIe i OAvaINASS: oo 2t lw fd. eee ee oe ee eee. 27 General Character of the Faunas, Mammalian and Avian........ 28 eC PS RMU TO Ne LS. Bk as st ee ne ere Soe ne 20 List of Localities referred to by Simons, Brown, Smith, Carriker and the University of ‘Michigan Expéditionss. 2.2.5.2 520.2.. The Reptile-Amphibian Fauna and Its Geographical Relations........ Bimannoi waerlenown. Patina... .. 5+ sa". aes hon se hoes a octee tae ative WV EciCOII Ne a Shan"... : . - safe aberheeee 2b. ooenaye eran ete We Sica ORe <3 oss ace s+. - ss «x a ee Re ete d Environmental Factors in the Vertical Distribution.............. . ie cmdeus-ef thew loud \Vorest..Pauna .-.\: gee eeies «. S. bei e Lb ae Distribution of the Desert and Dry Forest Fauna......0....... DPR SRR A St oe te... ee cee aes ee ke LL EE ps Vee i aS... a AR RR ie Lc 5 5 ce «3 Sg NA sect peng Boas Crocodilini ‘Testudinata Meet m Celie Liss te ws le le. Ole © = 6 6 6.8) © 9) samen WEe he s\)w ine es’ ’e, @)5 ‘sy ae eo (oe 0: 16 ILLUSTRATIONS Plate Frontispiece. ‘The Sierra Nevada de Santa Marta. I Figure 1. Harbor and city of Santa Marta. Figure 2. Vegetation of the plains near Santa Marta. II Figure 1. Wet forest on San Lorenzo. i) Figure Moderately wet forest on San Lorenzo. III Figure 1. Dry forest on San Lorenzo. Figure 2. Moderately wet forest on San Lorenzo, IV Moderately wet forest on San Lorenzo. V_ Figure 1. A stream on San Lorenzo. Figure 2. Lower part of the cloud forest on San Lorenzo. VI A burned area near the summit of San Lorenzo. VII Vegetation of the region about Riohacha. VIII Figure 1. Savanna at Valencia de Jesus. Figure 2. Drymobius boddaerti (Sentzen). IX Figure 1. Basiliscus barbourt Ruthven. Figure 2. Phrynonax poecilonotus (Gunther). X Figure 1. O-xybelis acuminatus (Wied). Figure 2. Drymobius rhombifer (Gunther). XI Figure 1. Eleutherodactylus carmelitae, new species. Figure Geobatrachus walkeri Ruthven. Hypopachus pearsei Ruthven. w ON Figure XII Figure 1. Eleutherodactylus carmelitae, new species. Figure 2. Cryptobatrachus fuhrmanni -(Peracca). Figure 3. Eleutherodactylus megalops Ruthven. Figures 4, 5. Eleutherodactylus cruentus (Peters). Charts Facing page 40. Altitudinal distribution of amphibians and reptiles on Mt. San Lorenzo. Page 47. Distribution of common lowland lizards in the Santa Marta region. Map Following Plate XII. The Santa Marta Mountains region. THE AMPHIBIANS AND REPTILES OF THE SIERRA NEVADA pDE SANTA MARTA, COLOMBIA By ALEXANDER G. RUTHVAN (With a description of the region by M. A. Carriker, Jr.) INTRODUCTION The explorations of the University of Michigan in the Santa Marta region have had as the principal object the study of the distribution of several groups of animals abundant in the tropics. It is believed that geography should be emphasized by museums, that the most reliable results at the present time are to be obtained by the study of the distribution of particular groups by persons acquainted with these groups systematically, and that certain problems can best be studied in particular regions. The groups which have received principal attention in the study of the Santa Marta region are the Formicidae, Amphibia, and Reptilia. Con- siderable work has been done on the Crustacea, and rather extensive col- lections have been made in other groups, but the ants, amphibians, and reptiles are the groups studied in sufficient detail to permit of the drawing of general conclusions upon the distribution. ‘This paper contains descrip- tions of the field work and the region studied and a discussion of the distribution of the amphibians and reptiles. THE Freyp Work The specific problems in the distribution of the reptile-amphibian fauna of the Santa Marta Mountains region which were to be investigated were two: to determine more exactly the nature and place of the changes in the fauna which evidently take place in western Venezuela or eastern Colombia, and to obtain more data on the responses of these groups to changes in altitude. It is probably well known to herpetologists that a number of reptiles and amphibians found in northern Venezuela do not occur in west- ern Colombia and that a number of northern Colombian species do not occur in Venezuela, but there has been no evidence of the nature and extent of the changes in the faunas nor data by which the place could be more accurately fixed. Besides being in the general region of the above- mentioned faunal change, the Santa Marta Mountains seemed to offer exceptional opportunities for studies in the altitudinal distribution of rep- tiles and amphibians. The abrupt rise from sea level to heights perennially covered with snow and the tropical location of the region promised a considerable fauna and sharply delineated distributions. The first expedition was sent to the region in 1913. The party, con- sisting of A. S. Pearse, F. M. Gaige, and the writer, spent eight weeks, Oh tis ALEXANDER G. RUTHVEN from July first, on the west end of the range. During this time collecting was done from the coast at Santa Marta to the bench at 8,300 feet (the Cerro Quemado) on San Lorenzo, the territory covered being a strip through the villages and haciendas of Santa Marta, Bolivar (1)*, La Tigrera (II}, Minca (III), Agua Dulce (IV), and Cincinnati (V), and on the lowlands at La Gaira, Cienega, and Fundacion. Subsequent to this time M. A. Carriker, Jr. made collections which supplemented the series obtained at the west end of the range and provided informacion on the fauna of the north side of the range from Don Diego to snow-line. In 1920 the writer with Mr. Gaige again visited the range, and during June obtained ad- ditional information on the fauna of the forested area on San Lorenzo, exploring the great valleys on the east side of San Lorenzo and the head- waters of the Rio Cordova. In July and August Mr. Carriker, Mr. Gaige, and the writer went to Riohacha by boat, and, securing mules from the Indians, worked entirely around the range by way of Arroya de Arenas, Fonseca, San Juan de Cesar, Valle de Upar, Valencia, and Fundacicn, arriving at Santa Marta on August 13. The localities at which collecting was done are given in Mr. Carriker’s account which follows. In the field an attempt was made to obtain data on habits, ecological distribution, and exact limits of ranges, in the belief that the most detailed geographic and ecological data are needed in the interpretation of relation- ships and distributions. Anyone acquainted with the secretive habits of the amphibians and reptiles, and giving consideration to the size of the region, the diversity of conditions, and the number of species comprising the fauna, will understand the difficulties of the task. No one can appre- ciate better than the writer the incompleteness of the results. The nature of the field work in itself limited the areas and the forms that could be studied in detail. However, it has seemed that the publication of the results is justified in several ways: The reptile-amphibian fauna was practically unknown; there is a distinct possibility that the natural conditions will be greatly changed in the near future; and certain conclusions, bearing on the general problems which were attacked, are possible from the data. Acknowledgments: It is with great pleasure, but with considerable difficulty, that the writer attempts to acknowledge the assistance received during the progress of this study. The expeditions were made possible by Dr. Bryant Walker, who has so consistently supported several of the activi- ties of the Museum. In the field the assistance and hospitality of Mr. and Mrs. Orlando L. Flye, Mr. and Mrs. M. A. Carriker, Jr.. Mr. Robert Sar- gent, and Mr. William Flye contributed so largely to the success of the work that it is not exaggerating to say that to them belongs in considerable part the credit for the results obtained. The indebtedness of the members of the University of Michigan expeditions to these persons cannot be satis- factorily expressed, but it should be pointed out that such knowledge as we have of the fauna of the Sierra Nevada de Santa Marta is in no small part due to the assistance and advice which Mr. Flye has so freely and 1’These numbers are used on the map to indicate the location of the haciendas. AMPHIBIANS AND REPTILES OF SANTA MAR?A ae kindly given to collectors, and to the careful field work of Mr. Carriker. The writer also wishes to express his indebtedness to his companions in the field, M. A. Carriker, F. M. Gaige, and A. S. Pearse. Their loyalty and indefatigable efforts to advance the work on the reptiles and amphibians, although principally interested in other groups, made it possible to obtain a large amount of data that otherwise would not have been secured, and contributed to the pleasure of the field work. The writer is under obligations to Mr. Samuel Henshaw and Dr. Thomas Barbour for the loan of material in the Museum of Comparative Zoology, and to Dr. Henry W. Fowler and the authorities of the Academy of Natural Sciences of Philadelphia for the opportunity of examining the specimens obtained by Mr. J. A. G. Rehn and ‘Mr. Morgan Hebard in 1921. He has also had considerable assistance from Dr. Thomas Barbour, Dr. George Boulenger, and Miss Joan Procter in the identification of specimens, and from Dr. Leonhard Stejneger in the settling of problems in nomenclature. The writer congratulates himself on being able to publish a description of the region by Mr. Carriker. Certainly no one is more familiar with the area, arid no one who has visited it is better fitted to describe the conditions from the biological standpoint. The writer has been able to spend much time in the field with Mr. Carriker, and fully agrees with the interpretations of the conditions given in his account, DESCRIPTION OF THE SANTA MARTA REGION (With comments on the zoological work which has been done there and a list of localities) By M. A. CARRIKER, JR. The preparation of this paper was suggested by Dr. Ruthven at the time of his second expedition to this region in the summer of 1920. He argued that the information the writer possessed as the result of personal observations over an extended period should be published as an aid to future workers as well as to those who have already made extensive bio- logical studies here. It is due to his enthusiasm and encouragement that I was induced to undertake the work, and it was with his help that the out- line of the paper here presented was prepared. The writer first arrived in Santa Marta in 1909, and with the exception of about two years spent in various parts of the interior of Colombia, has been more or less a constant resident in the region, living in the lowlands near Santa Marta and in the mountains within a radius of twenty-five miles of that city when not engaged in making trips to other parts of the Nevada. (See map following Plate XII.) As one approaches Santa Marta from the sea (PI. I, fig. 1) the beauti- ful snowcapped peaks (see frontispiece) forming the apex of the range are visible for many miles and present an almost irresistible attraction to any lover of the wild and unknown spots of the earth, but especially to the 8 ALEXANDER G. RUTHVEN naturalist, who cannot but be fascinated by the prospect of collecting in this great, isolated, almost unknown mountain mass: The region received the attention of some of the earliest naturalist-explorers who visited Colombia, and many species of animals and plants were described from the meagre collections made here, but the fauna and flora are still largely unknown. ‘To the biologist the interesting features of the Sierra Nevada de Santa Marta are that it is a lofty mountain mass, rising abruptly from sea-level and entirely separated from any other mountain range or system, that it has undoubtedly never been connected with the nearest mountain range—the eastern Andes of Columbia,—and the diversity in conditions. The diversity in ecological conditions is probably exceeded in few areas of the same size, as the range rises from the sea and from lowlands which are arid wastes, or covered with semi-arid woodland or dense humid forests; there are scores of miles of mountain slopes covered in places by a dense tropical forest, in others by semi-arid woodland, and in still others by great savannas or bare, rock-strewn slopes; and the upper regions constitute a vast solitude of paramos, extending from the timber-line up to the eternal snows. In animal and plant life the region is very rich, and there are appar- ently a large number of endemic species. Of all of the groups represented, only the mammals, birds, reptiles, and amphibians have received particular attention, and even in these groups there is still much to be discovered. The greater part of the region at all altitudes is a vast wilderness, unknown and uninhabited, and the number of biological and archaeological problems is legion. In this paper I will attempt to give, as nearly as possible, the infor- mation on the physical and biological conditions I have accumulated during the past nine years. Since the published data are meagre, this information should be useful to those who shall be able to continue the exploration and study of this remarkable region and its exceedingly rich flora and fauna, and to those who have already done some work in the range, but who do not have a comprehensive idea of the region as a whole and the relations which the different parts bear to each other and to the adjoining areas. It should also assist in preventing useless duplication of work and aid future workers to select untouched fields. It should be pointed out here that biological work in the Santa Marta range should be done at once if it is to be done at all. There is now begin- ning a colonization of the mountains which may, in a few years, wipe out the great forest areas and make such changes in the natural conditions that it will be most difficult, if not impossible, to arrive at a correct knowledge of the original composition and distribution of the flora and fauna. There are two colonization schemes on foot now in Germany, with the Sierra Nevada de Santa Marta as their objective, while great interest has been aroused in the region throughout Colombia, and a large amount of land has been taken up during the past two years. A new road has been opened, beginning at Fundacion, the present terminus of the Santa Marta Railway, and extending into the mountains a distance of 25 to 30 kilometers and to AMPHIBIANS AND REPTILES OF SANTA Marta 9 an altitude of about 6,000 feet. Other roads are being projected by the Government and by private interests, and with these once opened it is a foregone conclusion that all the forested country within reach of them will soon be reduced to pasture land and cultivated fields. Already the greater portion of the humid forest, occupying the lowlands on the west side, between Cienega and Fundacion, has disappeared, and to get at undisturbed conditions representing this habitat it is now necessary to go to the great forests of the Ariguani, about thirty miles south of Fundacion. The Field Work of the Writer: I began collecting birds in the Santa Marta district in 1911, arriving at the city of Santa Marta late in May. The first work was done at the Hacienda Cincinnati (formerly known as Valparaiso). ‘This is the largest coffee plantation in the region and lies on the northwest slopes of Mt. San Lorenzo, between 3,000 and 5,000 feet. It is surrounded by forest extending downward to the foothills and upward to the crest of the mountain, which attains an elevation of 8,300 feet at the western end. More or less continuous work was done at Cincinnati and on San Lorenzo during June and July, and in August, lower down, between Mamatoco and Cincinnati, at La Tigrera and Minca, and Agua Dulce. No more collecting was done until March of the following year, when an attempt was made to reach the Sierra Nevada proper by way of the ridge connecting it with San Lorenzo. Three weeks were occupied in this attempt, which was finally given up as impracticable, after an elevation of about 8,000 feet was reached on the main ridge leading to the paramos. There was no trail of any sort, nothing but unbroken forest, and trail had to be cut with machetés. After reaching an altitude of about 8,000 feet progress was checked by an impenetrable mass of brush and huge bromelias from four to ten feet in height. Water was very difficult to secure and birds extremely scarce, so that, after a couple of days of heart-breaking work in an attempt to pass this barrier, the trip was abandoned and the party returned to Cincinnati. After this failure to reach the Nevada, work was continued at Cincinnati and around Mamatoco until the end of April, when I returned to the States for a much needed vacation, having been in the field constantly for nearly three years. My return to Santa Marta was made in September of 1912, accom- panied by Mrs. Carriker. Some intermittent collecting was done in the vicinity of Cincinnati during the remainder of 1912 and the early part of 1913. In April, 1913, accompanied by Mrs. Carriker, who had now acquired some skill in the preparation of specimens, I again began intensive work which continued almost uninterruptedly during the remainder of 1913. The first locality worked was the valley of the Tamocal Creek, between Mama- toco and La Tigrera. Later the vicinity of Mamatoco and Santa Marta was examined, and several trips were made to Gaira and to the beach beyond. In May a trip was made to the hacienda Las Vegas, situated at an elevation of from three to five thousand feet above sea level on the northeast slopes of the Horqueta, near one of the sources of the Rio Men- ae) ALEXANDER G. RUTHVEN daguaca. ‘Two weeks were spent here and very thorough work was done. Birds were not abundant, however, and the ground was broken and the forest difficult to penetrate. The venomous snake known as “Fer-de-lance” was entirely too common to be agreeable and several narrow escapes spoiled the pleasure of collecting. After returning from Las Vegas, camp was established at Minca for two weeks, and the slopes between 1,500 and 2,500 feet were thoroughly worked, but the heavy rains interfered with collecting. From July 6, to August 3, work was done at Cincinnati, on San Lorenzo and near Mamatoco, during which time many new forms were added to the list and many loose ends of the problems of local distribution were caught up. On August 6, a trip was made to Fundacion in company with Dr. A. G. Ruthven and Mr. F. M. Gaige, of the University of Michigan, and Dr. A. S. Pearse, of the University of Wisconsin. Two weeks were spent here with splendid results. The region proved to be a most interesting and fertile field for all members of the party, and many forms new to the general region were taken. After the return from Fundacion some time was spent looking for particular forms in the vicinity of Mamatoco, Bonda, and Gaira, after which a trip was made on September 26, to Cienega Grande. The party embarked from Pueblo Viejo (near Cienega) in a big “Bongo” or dug-out canoe, with a smaller canoe in tow for shooting purposes. The first stop was made at Playa Caiman, near the northwestern corner of the Cienega, where camp was made on a narrow strip of sand beach separating the Cienega from the sea. All water was brought from near Pueblo Viejo, where it had been carried in canoes from the mouth of the Aracataca River. Life here was a constant torture by day and by night because of the hordes of mosquitoes and sand-flies, which not even a solid muslin canopy would keep out. Our only relief was when there was a brisk breeze from the sea, but it was not the season for the sea breeze, and it came but seldom. Needless to say, all possible haste was made to escape from such an unde- sirable locality, and after five days we broke camp and fled to the open Cienega in our big canoe, hoping to get at least one good night’s sleep. We did sleep, but so soundly that it was not until morning that we discovered that we had been raided during the night by vampire bats, and the whole party was covered with blood-stains from the many bites of these bats. It may seem unreasonable to the uninitiated that we could have been thus bitten and not be disturbed in our sleep, but the fact is that there is no pain produced at the time of the bite, nor indeed for some hours afterward. I had ample proof of this while collecting vampires in an old tunnel in Venezuela. I caught the bats in an insect net and then seized them care- fully by the nape of the neck and thrust them into cyanide bottles. Their skin is very loose around the neck and several times they succeeded in wriggling about and nipping my fingers. On such occasions I never knew I had been bitten until I saw the blood flowing from the wound. After a bath and breakfast we put up our sail and headed for Trojas AMPHIBIANS AND REPTILES OF SANTA MARTA IE de Cataca, a fishing village built on piles over the water at the mouth of the Aracataca River. We secured quarters in a comparatively new house and made ourselves fairly comfortable, although the odor of drying fish was very disagreeable for several days until we became somewhat ac- customed to it. It was a pretty sight to see the little fleet of fishing canoes sail away at dawn in search of the great schools of Lisa. When luck was good the canoes would return between three and four o’clock, loaded to the gunwales, and then there was pandemonium let loose in the village. Singing, shouting, whistling, and laughing, while everyone big enough to wield a knife joined in the task of cleaning them for drying, this operation usually consuming half of the night. When the catch was very great the entire night was spent in festivities. Practically all shooting in this vicinity was done from a canoe along the shores of the lagoon or up the streams entering into it, for at this sea- son of the year the shores of the Cienega are inundated for a long distance inland. Herons, terns, and ducks were abundant, while the forests were teeming with parrots and monkeys. Occasionally a Manatee was seen at the mouth of the river, but none were secured. October 3 to 14, was spent here and many aquatic forms were secured. Little or no work was done between the return from Cienega Grande and our departure for the Sierra Nevada in January of the following year. On January to, with Mrs. Carriker and one Colombian assistant, I left Santa Marta for the north coast, going by land, with pack mules, as far as Don Diego. The trail was a trail in name only and was almost impassable. ‘Three and a half days were consumed in reaching Don Diego. The trail passes through Mamatoco and Bonda, and thence along the road to Don Amo for a short distance, when it turns off to the left and crosses over into the valley of the Jordan, passing through Calabaso and Cacagua- lito, thence down the Jordan to the Rio Piedras, then down the Rio Piedras to the sea. From here the open beach is followed to Don Diego, except where detours inland are necessary to cross the rivers Mendaguaca, Gua- chaca, and Buritaca. We arrived at Don Diego at dusk, after thirteen hours in the saddle; tired, hungry, and covered with bites of black flies, sand-flies, mosquitoes, and wood ticks, the latter being still attached in myriads to our persons. A hurried rub-down with kerosene removed the ticks, after which a delight- ful sea bath somewhat restored our flagging spirits, and a good supper pro- vided by our genial host, Mon. Barbier, completed the restoration. The whole northeast coast and the hills up to 3,000 feet or more con- tain more insect pests than any place I know of in Colombia, excepting some portions of the Atrato basin, and Don Diego proved no exception to the general condition. The attacks of flies and mosquitoes can be checked to a certain extent by the use of “dopes” or nets, but there seems to be no effective relief from the ticks, which cannot be entirely avoided and are, in many places, a serious drawback to the best field work. Don Diego is a French plantation which at the time of our visit was 12 ALEXANDER G. RUTHVEN largely abandoned, and is now entirely so. It is in the heart of the heavy, humid forest which reaches to the very beach all along the northeast coast east of the Rio Piedras. A month was spent here, not because we wished to stay, but because we could not get away, since the trip to Dtbulla was to be made by sea in a large dug-out canoe, and the canoe was not forth- coming. To secure the canoe it was necessary to make the trip to Dibulla, which I did, accompanied by our assistant, and riding the two mules, while Mrs. Carriker remained in Don Diego. After two days of strenuous endeavor and the promise of an exorbitant sum, a canoe and crew of five men were secured. We embarked in the evening, with a fresh northeast wind astern, and reached our destination early next morning. Then fol- lowed a half day of frenzied packing and at 4 P. M. we embarked and attempted to put to sea. There is no harbor of any sort at Don Diego, merely an open beach, and in passing the second line of breakers we were nearly swamped and were forced to return to shore, with much of our outfit wet and damaged by the salt water. The next afternoon we made another attempt which was more successful, and we reached the open sea with only a slight wetting in the outer line of breakers. Thirty-six hours were spent in getting to Dibulla, bucking the wind and strong current. We arrived at dawn, nearly exhausted from the cramped position in the canoe and lack of proper food. No pack animals were available in Dibulla, and it became necessary to send to Pueblo Viejo for them. In the meantime we spent ten days very profitably in collecting around the hacienda La Victoria (near Dibulla), where we had been most hospitably received. No collecting had ever been done here and much interesting material was secured. Our baggage was taken to Pueblo Viejo on oxen, while we rode our own saddle animals. Owing to the slowness of the oxen we were compelled to spend a night on the road, reaching Pueblo Viejo at about 2 P. M. the second day. We pitched camp on the banks of the Rio Ancha, about half a mile from the village, thus escaping the too persistent visitors, including pigs and dogs, of the town. W. W. Brown had done considerable collecting at this place, but nevertheless much new material was secured and many new records made. Pueblo Viejo is only about 2,000 feet above sea level, surrounded by broken country, which has for the most part been cleared and has now reverted to scrub or savanna. Our old friends, the wood ticks, which were so numerous both at Don Diego and Dibulla, were much in evidence there and caused no end of annoyance. After returning from a morning’s shoot- ing it was almost always necessary to strip completely, sponge off with kerosene, and then bathe in the river to remove both ticks and kerosene, while the clothing went into a can of boiling water. From this base the valley and heights of Chirua were worked, as well as the valley of the Rio Ancha above for a long distance, and also the Macotama Valley as far up as Santa Cruz. We secured oxen from the Indians of San Miguel for the transpor- tation of our equipment to that village, a special trip having been made there AMPHIBIANS AND REPTILES OF SANTA MARTA tz for that purpose. We arrived at San ‘Miguel late in the evening, spending the night in the “guest house” of the village, and crossing the river the next morning to the plateau opposite where a camp site had been prepared in advance. The whole of the Macotama Valley, from where the road enters it below Santa Cruz up to the paramos, is largely deforested, the mountain slopes being largely unbroken savanna, except in the ravines, the only forest left being along the crest of the ridges and on the upper slopes. As a rule, what woodland was left was invariably on steep north exposure, land unsuit- able for cultivation by the Indians. We spent five weeks in camp at San Miguel, using that as a base for all work from there (5,500 feet) up to the snow line. An unfortunate, though unavoidable misunderstanding with the old chief, brought about by our failure to present him with a large jug of rum, was the cause of our not being able to secure oxen for moving our camp to higher altitudes. He even forbade us to shoot on the paramos, but we paid no attention to him, promptly going up the valley the next day to 10,000 feet, without hindrance. Virgin forest was found at San Miguel only at long distances from camp, which made it necessary to spend many nights in the field in order to get the morning shooting, sleeping sometimes in the forest on the ground and other times in deserted Indian huts. No guides could be secured from the Indians, so that we were reduced to “following our noses” until we found what we wanted. The first time I went up to Lake Macotama I started at 5 A. M. on muleback, riding up to 9,000 feet, where the mule was left and the journey continued on foot. At noon the lake was reached (15,000 feet), where lunch was eaten, after which I returned to camp, reaching it at 7 P. M. with a good bag of birds, but well tired out. Lake Macotama certainly rivals many of the Alpine lakes in beauty and grandeur of scenery. In shape like a half horseshoe, it is about three-quarters of a mile long and from 300 to 400 yards in width, and is almost surrounded by beetling crags and cliffs of nearly naked rock rising to a height of about 1,400 feet above the surface of the lake. Down on one side, less precipi- tous than the rest, dashes a little stream of water from the melting snows above. The colors of the lake are exceedingly beautiful. Along the grassy shores the water is pale green, gradually becoming darker green, then merging into blue-green, light blue, and finally into the deepest ultramarine, showing that the lake must be of profound depth. Later, when we all went up to the snow, we left camp very early, taking the two mules, Mrs. Carriker mounted on one and a pack on the other. We again left the animals at 9,000 feet, dividing the pack between the hunter and myself. We reached the lake at 4 P. M., ready to drop from fatigue, for the packs contained food for three days, blankets, guns, ammunition for three days’ shooting, camera and aneroid. Camp was. made on the lake shore under the lee of a gigantic boulder. We secured a good supply of fire-wood, ate supper, and turned in, to pass a wretchedly cold and miserable night, with the temperature down to 26° F. Next morn- ing, after a hasty breakfast, we started by moonlight at 4 A. M. to climb 14 ALEXANDER G. RUTHVEN the cliff along the course of the little stream, which seemed to be the only feasible mode of ascent to the heights above. The top was reached after an hour of hard climbing, and another lake, a small and shallow one, was discovered there, from which stretched a comparatively gentle slope up to the snows. Respiration now became more and more difficult, and halts more frequent and longer. Our native helper succumbed to mountain sick- ness and was of little use until the following day, after we had descended below 12,000 feet. At last the snow was reached and its glittering slopes ascended for some distance, to an altitude of approximately 17,500 feet. To our keen disappointment, we now discovered that further advance along this route was impossible, for here began a knife-edge ridge of jagged rock which extended nearly a mile to the slopes of the main snow peak, the pas- sage of which appeared an utter impossibility. To approach the heights above from any other direction meant a descent of at least 2,000 feet into another valley opening out toward the north, in which lay another fairly large lake. From this valley an uninterrupted slope leads up to the highest point of the Nevada. Such a detour meant at least another day, or even two, and we had insufficient food to make the attempt. We turned back and spent the rest of the day shooting in the valley below snow-line, where birds of several species were fairly common, one of which was the strikingly handsome humming-bird Oxypogon cyanolaema, which feeds largely on the flowers of the “Frailejon,” and which is peculiar to the Sierras of Santa Marta. In the little stream above the upper lake we discovered numerous individuals of a fine new amphibian, later described by Dr. Ruthven as Atelopus carrikeri. On the shores of Lake Macotama was taken a single specimen of our beautiful North American warbler, Geothlypis trichas, the first record for South America. After another disagreeable night on the shores of Lake Macotama, we started back to San Miguel, collecting along the way, and reached camp without mishap late that evening. I afterwards made another trip to the lake with my hunter, sleeping the first night a short distance below the lake, where we had a more comfortable camp, and on the second night in an Indian hut at 11,000 feet. On this trip the paramos between 11,000 and 15,000 feet were thoroughly worked and much valuable material collected. Later in April, after a few showers had warned us that the rains were near, we broke camp, sent to Pueblo Viejo for oxen (the chief still being disgruntled), and started for the coast, where we arrived in four days. From Dibulla we proceeded by canoe to Riohacha, reaching there in one night’s travel. A week was spent at Riohacha in strenuous collecting. The intense dry heat of the sandy wastes and low scrub around the town were very trying, after having descended so rapidly from the cold temperature of the high altitudes, and we found it impossible to stay out after 9:30 A. M. Fortunately, the thorny scrub surrounding the town was fairly swarming with birds, so that with three hours’ shooting enough material was secured between the two of us to keep all hands busy the remainder of the day in preparing it. A few more days of collecting would have been desirable, but a schooner AMPHIBIANS AND REPTILES OF SANTA MARTA 15 was leaving for Santa Marta, and since transportation between these places is very uncertain we could not afford to let the opportunity go by. We embarked in the evening, and at dawn of the second day reached Santa Marta, after a calm and uneventful voyage, just four months from the day we had started. Altogether it was a most successful trip, although many difficulties and delays were experienced which might easily have been avoided had it been possible for us to secure in advance even the most meagre information about the region we intended to visit. While the Indians were extremely shy and distrustful of us, they were honest and inoffensive, never molesting us in any way or coming near our camp during our absence. After the trip to the high altitudes some time was spent in revisiting certain localities on the west end of the range to supplement collections. Evidently the next work to be done was a reconnaisance of the east and south sides of the range, and in 1920 the opportunity presented itself. In July and August Dr. Ruthven, F. M. Gaige, and I went to Riohacha from = Marta in a small schooner. We worked at Riohacha for ten days, and then, with mules secured from the Indians, made a trip entirely around the Nevada, by way of Fon- seca, San Juan de Cesar, Valle de Upar, Fundacion, and Cienega. The trip was most successful as a reconnaisance, and even more than this, as detailed work was done on the reptiles and amphibians, and at the same time sufficient data was secured to clear up numerous questions on the nature of the vegetation of the southern slopes and contiguous lowlands and the distribution and origin of the fauna of that side. It will be clear from this account that while much has been accom- plished both by previous workers and ourselves, there still remains a vast amount of detailed work to be done. ‘To be worth while, collecting must be carefully done, for the diversity of conditions and the differences in altitude within short distances render collections with general data almost valueless. Previous Biological Investigations: ‘The late Mr. H. H. Smith arrived in Santa Marta in May, 18098, with his wife and two other assistants. They made their first headquarters at Bonda, where they remained for a consid- erable period, working the surrounding country for long distances. Mr. Smith himself did no collecting of birds or mammals, and probably secured but few reptiles and amphibians, confining his efforts to botanical and ento- mological work, in which he was, I understand, most proficient. All verte- brates collected were taken by native hunters employed by him, so that as a consequence we have a vast collection of skins labelled “Bonda’’ which in many instances came from places far removed from that village, both in actual distance and especially in altitude. Bonda lies in the valley of the Rio Manzanares, at about 200 feet above sea level, but near the foot- hills of the Horqueta Mountain, so that a collector starting out from Bonda might easily reach an altitude of 3,000 feet or more in a day’s shooting. Tor this reason it is impossible to use much of Mr. Smith’s data for plot- ting altitudinal distribution of species. This is especially true of his work 16 ALEXANDER G. RUTHVEN done at Bonda, although it holds good also for many specimens labelled “Valparaiso” and “El Libano.” The Valparaiso of Smith is today the hacienda Cincinnati, and El Libano is the mountain ridge lying between Cincinnati and the adjoining coffee estate of La Victoria, which descends from the Cerro Quemado of San Lorenzo. There is no doubt that Mr. Smith took every precaution to label his specimens correctly, but when native hunters are employed it is not possible for a person who is unfamiliar with the local topography to tell from a hunter’s description just where each specimen was secured. Also, the hunters themselves do not always. remember and are very apt to give fictitious localities for specimens when they have forgotten the true one. In those days the importance of altitude in determining the distribution of species was not appreciated and most collectors paid little attention to it, so that they might be working on the slope of a mountain, with headquarters at 4,000 feet, and make collections. from 2,000 to 6,000 feet or more, and give all specimens the locality of the base of operations. Mr. Smith later spent some time at Don Diego, where Mrs. Smith was very ill, and while a great amount of material was collected there much of it was lost, and no detailed report was ever made on the remainder by any institution. Mr. Smith’s work in the Santa Marta region was confined to: the following roughly outlined sections: The valley of the Rio Manzanares from Mamatoco upwards, covering the lower slopes of the Horqueta and extending northward over the Matagiro Valley and the median part of the valley of the Jordan, at Cacagualito and Calabaso; the northwest slopes of San Lorenzo, between 2,000 and 7,000 feet, between Minca and the upper slopes of the Cerro Quemado; the vicinity of the town of Cienega (some- times called San Juan de Cordoba) and southward as far as the Rio Ori- hueca, and lastly at Don Diego. It is a pity that his specimens could not have been more carefully labelled as to altitude, because a great quantity of skins were made which would now be of inestimable value in plotting the local distribution of species in the regions above enumerated. He deserves, nevertheless, great credit for what he did accomplish, as does Mrs. Smith, because she super- intended all the bird and mammal work and did much of the taxidermy herself. They worked in the region at a most difficult time, as it was during the late revolution, which caused them no end of annoyance. Also, the country around Santa Marta was much more of a wilderness then than now, with almost no roads or trails and very few settlements in the mountains. Mr. W. W. Brown arrived at Santa Marta shortly before Mr. Smith and almost at once made his headquarters at Bonda, staying with Mr.. Orlando L. Flye, who was in charge of the electric power plant at Bonda. All the material collected by Mr. Brown in the vicinity of Bonda and Santa Marta was labelled “Santa Marta,” and is, as a consequence, worse tham useless for plotting distribution, since it is actually misleading in many cases. After working at Bonda he went to the Sierra Nevada, where I have little knowledge of his work except what was given me by two men living in: AMPHIBIANS AND REPTILES OF SANTA MARTA 17 Pueblo Viejo, who had been employed by him as hunters during all his activities on the north slopes of the Sierras. Contrary to Mr. Brown’s statements that he himself collected all of his specimens, I have unquestion- able first-hand information that both in Bonda and in the Sierras, at least at Pueblo Viejo, San Miguel, and the paramos, he did little or no shooting, employing native hunters for that purpose, while he prepared the specimens. As in the case of Smith’s data, I would hesitate to use any distributional records based solely on Brown’s collections, unless supported by other data. Mr. Brown worked both sides of the Sierras, at Pueblo Viejo (then San Antonio), Chirua, Santa Cruz, San Miguel, and Macotama on the north slopes, and San Sebastian, Pueblo Viejo, and El Mammon on the south slopes, as well as the intervening paramos of Macotama, Chiruqua, and Chinchicua. Notwithstanding the fact that he passed through Dibulla and Riohacha and traversed the road from the latter place to Valle de Upar, no collecting was done in any of these regions, so that the first defi- nite information secured concerning their bird fauna was by the writer at Dibulla and Riohacha in 1914 and along the road from Riohacha to Valle de Upar and thence around to Fundacion in 1920, when attached to the University of Michigan expedition. It is to be understood that I have no reason for making these comments upon the work of Brown and Smith other than the desire that our knowl- edge of the fauna of the region should be free from the confusion caused by inexact records. Geographical Situation: The geographical boundaries used for the region discussed in this paper (cf. map) are as follows: On the north and northwest the Carribean Sea; on the northeast and east the Goajira Penin- sula, or, more exactly speaking, the Rio Rancheria (called Calancala on its lower reaches), up to where it turns abruptly northwest to enter the Sierras; from this point on the Rio Rancheria on a straight line to San Juan de Cesar, on the Rio Cesar; thence down the Rio Cesar to where it divides into two channels (Cafio Corredor and Cafio Matanza); from this point in a more or less straight line to Alto de las Minas, passing through Cam- perucho, and thence to Ariguani; from there the trail to Fundacion, the terminus of the railway from Santa Marta; and from Fundacion a line to the south end of the Cienega Grande and thence across to Playa Caiman and the sea. The line running across from the Rio Cesar to Cienega Grande is a purely arbitrary one, since it does not designate the boundary of any known habitat. However, the upper Rio Cesar and Rio Rancheria represent an actual faunal and physiographical boundary between the Sierra Nevada and the eastern Andes, and I have used no records from localities on the slopes or foothills of the eastern Andes, such as Villanueva and Manuare (included by Simmons). The lower Rio Rancheria is also an arbitrary boundary line, since it traverses the arid Goajira Peninsula and cuts through a distinct faunal area which is well represented around Riohacha. Thus we have in the mountain mass an absolutely isolated area containing its own distinctive habitats and faunal characteristics, into which 18 ALEXANDER G. RUTHVEN enter but two outside influences, that of the Magdalena basin on the south- west and that of the central plateau of Venezuela through the Goajira Peninsula on the east. The great mountain mass known as the Sierra Nevada de Santa Marta is, roughly speaking, in the form of a trihedron, with the San Lorenzo and Horqueta peaks and outlying ridges forming the northwest edge; the end of the long spur ridge which disappears in the big bend of the Rio Ran- cheria, the northeast edge; and the great southwestern buttress between the Rio Ariguani and Guatipuri, which ends at Alta de las Minas, forming the southwestern edge. ‘The vertex of the trihedron, massive central peaks, covered with perpetual snow and rising to an altitude of not less than 18,000 feet (probably more), lies somewhat nearer to the northwestern edge, The base of the north face of the trihedron is about 90 miles long, that of the southeastern face about 120 miles, and that of the western face about go miles. It will be noted that the Santa Marta Mountains have a remarkably small base for the heights attained. There is an almost complete absence of foothills between the points forming the three angles of the trihedron, the mountains rising rapidly and in many cases very abruptly from the sur- rounding plain, which is for the most part but a few hundred feet, or less, above sea level. Consequently, there are no broad valleys anywhere on the lower slopes of the mountains, and the numerous rivers rush down from the high altitudes to the plains below, carrying vast quantities of silt which is dumped into the Rio Magdalena, the Cienega Grande, and the Caribbean Sea. It is only in the Cienega Grande that an adequate idea may be obtained of the immense amount of material which has been car- ried down by these streams. In this great lagoon the detritus has extended the shore line for long distances and filled up what was evidently once a deep arm of the sea, until today it is but a shallow lagoon with mangrove swamps encroaching upon it from all sides. No less than seven large rivers drain the north slopes. On the south side are four magnificent streams, all having their source in the snow, while on the west side are six of nearly equal size. The largest streams are the Rancheria, Badillo, Guatipuri, Ari- guani, and Fundacion. The existing maps of this region are very inaccu- rate in many respects, particularly in the plotting of the rivers. They con- vey an erroneous impression of the mountain ridges separating the various drainage basins, the long, bold ridge running east into the great bend of the Rio Rancheria being poorly indicated, as well as the one, even higher and wider, which ends at Alto de las Minas, forming the southwest apex of the triangle. The mountain valleys are narrow and V-shaped, with little or no flood plain until the plains are reached, where they burst out into the surrounding lowlands and deposit great masses of boulders and gravel in parallel ridges resembling lateral glacial moraines. ‘The slopes of the mountain valleys are, for the greater part, steep, and in many instances precipitous, especially on the south sides of the ridges, which would suggest a tipping up of the strata in some remote geologic time. ‘The ridges are for the most part nar- AMPHIBIANS AND REPTILES OF SANTA Marra 19 row, even knife-edged, the crests often being but a narrow wall of rock, in places but a few feet in width along the crest. As the higher altitudes are approached, and one emerges above the ‘cloud zone,” where the precipita- tion is less, the ridges become broader and more rounded, with remains of plateaus and gentle slopes. The valleys are broader and shallower, and on the vast paramos surrounding the snow peaks are to be found great undu- lating tracts of open moorland, broad valleys, and gentle slopes, with many small, picturesque lakes. These mountain lakes were possibly caused by glacial action in the remote past, and many of them have been nearly filled with silt from above in comparatively recent times, but at the present day no true glaciers exist in the Sierras, to my personal knowledge, unless some very small ice masses are included under the term. | doubt whether any very large glaciers ever existed in the Sierra Nevada de Santa Marta, since I have been able to identify no moraines. I suspect the cause of this absence of glaciers to be the complete isolation of the range and the lack of great buttressing foot- hills and mountains such as abound in the Andes. The rapid and unopposed ascent of the superheated air from the lowlands and its direct action on the snow masses above cause a much higher temperature and higher snow- line than would otherwise be the case at these altitudes. The topography of the Sierra Nevada de Santa Marta might be con- sidered to indicate that the range is young, but this does not necessarily follow. ‘Traces of volcanic action are wanting and the exposed strata show clearly that the range has been thrust up. The exposed rock is largely granite and in the greater part of the region the subsoil is a peculiar decom- posed granite, abounding in fine particles of mica. While clay deposits are present in many districts, they are decidedly less common than the decom- posed granite. I have seen limestone in only one small district, on the road between Fonseca and San Juan de Cesar, where the road skirts the base of the foothills. This limestone area is but a few miles in length (east and west), although it may extend far out into the valley. The narrow knife-edged ridges in many places seem to be the result of great vertical masses of hard rock which resist erosion along the tops of the divides and maintain the steep slopes. Ridges composed almost entirely of great masses of boulders and solid rock are very common. In addition to the resistance of the rocks and soil to erosion, it must be remem- bered that the east, north, and west slopes of these mountains from the lowlands to the timber line are clothed with a dense forest growth, and that on the south slopes, while the forest growth is largely absent and the topog- raphy is nearly the same as on the other slopes, the ridges are inclined to be broader and more rounded. The conditions thus seem to favor the per- sistence of a youthful topography. Climate: ‘The seasons are more or less the same on the different slopes and the contiguous lowlands, although the humidity and the temperature vary greatly on the different exposures, due principally to the trade winds. The surrounding lowlands possess the usual tropical heat of these latitudes, 20 ALEXANDER G. RUTHVEN tempered on the north coast by the trade winds and sea breezes, but are very hot on the west and south sides during most of the year. Ascending the mountains, the temperature gradually lowers. Between 4,000 and 5,000 feet is an ideal climate, having a minimum temperature of about 60° F. and a maximum of 80°, where an open fireplace is a great comfort at times and woolen clothing almost a necessity, especially during the rainy season from 5,000 feet upwards. When 8,000 feet is reached the minimum drops to an average of about 54°, and it is always chilly in the shade, while in the open the direct sun’s rays are blisteringly hot around midday. At 9,000 feet, or a little above, the mercury drops to 46° F. The above temperatures are taken from San Lorenzo, and are from 4° to 5° higher than at corre- sponding elevations on the western slopes of the main Sierras. The frost line varies with the season and the exposure, but is probably somewhere near 11,000 feet. The lowest average limit of snow is somewhere near 16,000 feet, rising on the southern exposures during the dry season to 17,000 or even 17,500 feet, and dropping down to 15,000 in the sheltered valleys of the north slope. There are times during the dry season when heavy falls of snow occur as low as 13,000 feet, but this snow never lasts more than from twenty-four to forty-eight hours, usually falling during the night and melting during the following day. On the north slopes of the mountains there is a far greater humidity and lower temperatures than on other slopes at corresponding elevations. ‘The climate at Pueblo Viejo (2,000 feet) approximates that of 4,000 feet on San Lorenzo and the west slopes, while that of San Miguel (5,500 feet) corresponds closely with that of the Cerro Quemado de San Lorenzo, which has an elevation of 8,000 feet. These differences in conditions on the north slopes apparently exert a powerful influence on plant and animal life on that side, causing a cor- responding descent in altitudinal distribution. At Don Diego the heavy mountain forest extends down to the very edge of the narrow coastal plain, and the upper foothills fauna descends with it, literally squeezing out the lower foothills forms from this region or forcing them to mingle with those of the coastal plain fauna below. There are two distinct and well-marked seasons, the rainy and the dry, the former covering the months of April to November, inclusive, over the greater part of the mountains from 3,000 feet upward. Below 3,000 feet the seasons-of rain are retarded or extended by local conditions. On the northern side, from the Rio Frio around to the Rio Piedras, the rains do not begin before May or June, and usually are over by the end of October, while on the north side, from the Rio Piedras to beyond Dibulla, they begin as early as March, and often continue as late as the end of December. Even during the dry season the north exposure of the Sierras is refreshed by almost daily fog banks, caused by the condensation of the moisture carried by the trade winds, when they come in contact with the cool vegetation of the mountain slopes. The great southern exposure of the Sierras receives a much smaller amount of moisture than any of the other portions, which is divided into two periods, at least in the lowlands and foothills, much resembling the seasons of the interior of Colombia. The spring rains fall AMPHIBIANS AND REPTILES OF SANTA MARTA 21 in April and May, while the greater part of June, July, and August are nearly without rain, the heavy rains coming in September and October. The cause of the scarcity of rain on the south side is, at first glance, hard to explain, but after a careful study of the topography and prevailing winds had been made a plausible explanation was arrived at. The prevailing winds are the northeast trades, which sweep in, moisture laden, from the Caribbean. ‘The south slopes of the Sierras are, to a great extent, sheltered from these winds by the long spur ridge forming the southeast apex of the trihedron. ‘The trade winds are low-flying, not extending, as a rule, to more than a thousand feet, or less, above the earth’s surface, unless they meet with some obstacle which deflects them upward. ‘The northeast spur ridge of the Nevada forms this obstacle, and its northern slope, covered with a dense forest of massive trees, receives the first impact of the winds, condenses the greater portion of the moisture carried by them, and then deflects them upward and over it to the south slopes of the mountains, but not until they have lost the greater part of their moisture, and little or none is left to be condensed into rainfall there. Farther to the south the winds sweep into the valley of the Rio Rancheria unopposed, and thence through to the Rio Cesar Valley, the height of land between having no effect on them, and having been thoroughly heated as they traversed the long stretch of arid wastes composing the Goajira Peninsula, they are drying winds. These strong northeast winds sweep through the great valley between the Sierra Nevada and the eastern Andes, drive back the gentle, moisture-laden land breezes from the Magdalena Valley and prevent their entrance from the southwest beyond a certain point. To the north of the southwest angle of the Sierra Nevada the moisture-laden winds from the Magdalena reach the west slopes of the Sierras unopposed and discharge their moisture there, giving an abundant rainfall during the rainy season on this side and its con- tiguous lowlands. Vegetation of the Plains: ‘The vegetation of the region under discus- sion is diversified. Beginning with the lowlands, we have, on the north side of the mountains, extending from the mouth of the Rio Piedras to a point beyond Dibulla, a heavy growth of forest, consisting of large, tall trees and very little tangled undergrowth, which extends almost to the sea beach. This may be strictly termed a humid tropical forest, but it is not jungle-like in its characteristics. Beyond Dibulla the forest gradually changes, merging first into the “dry forest,” such as is prevalent around Santa Marta, and then to low, thorny scrub and cacti, prickly pear, etc. (PI. VII.) A small, slender fan palm is abundant from Dibulla to Riohacha and as far south as Barbacoas, growing most commonly in the open parts of the scrub on low spots which are flooded during the short rainy period. This palm was not observed in any other section. Near the mouth the banks of the Rio Rancheria are clothed with mangroves, while higher up, above the influence of the tide, the same class of trees and shrubs are found which are found in similar situations around Santa Marta. Following the trail from Riohacha southward to Fonseca, the vegeta- 22 ALEXANDER G. RUTHVEN tion is similar to that around Riohacha until the Rio Camarones is reached at Barbacoas, and where the foothills begin. The flood-plain of the river here is very wide in places and heavily timbered with massive trees, inter- mingled with dense undergrowth, while the low hills are covered with the semi-arid type of low woodland, abounding in shrubbery and small trees, the whole region showing the effect of a very low humidity. After leaving Barbacoas the trail soon begins to wind up into the foothills of the great northeast spur-range. The forest becomes heavier and heavier, with less and less undergrowth, until at an elevation of perhaps 3,000 feet the forest is composed entirely of tall, massive trees, with little or no undergrowth. This forest undoubtedly extends westward unbroken over the whole north- ern flank of the Sierras. ‘The trail crosses the range at an elevation of approximately 5,000 feet, then descends to the Rio Rancheria above Fon- seca. The heavy forest of the north slopes extends unbroken up to the crest of the divide and slightly over it, but there is an almost instantaneous change after crossing the crest; the dense forest of large trees gives way to an open wodland of small trees and tangled underbrush, even the species changing almost entirely. Descending the slope, the woodland becomes sparser, acacia-like forms and cacti soon appear, and there are tracts of open savanna. J,ower in the foothills and upper parts of the valley we have again the same vegetation as along the trail between Riohacha and Barbacoas. The flood-plain of the Rio Rancheria is heavily forested, like that of the Camarones at Bar- bacoas, except in the vicinity of Fonseca, where it has been deforested. However, as elsewhere noted, these forests of the river flood plains derive their moisture more from the subsoil rather than from the rainfall. From Fonseca to San Juan de Cesar the trail is close to the base of the foothills over a cactus- and acacia-covered plain, and the height of land separating the basins of the Rio Rancheria and the Rio Cesar is so inconsiderable that it is scarcely noticeable as one crosses it. There is but one small stream descending from the foothills between Fonseca and San Juan, and this was dry when we crossed it. The adjacent foothills are either sparsely clothed with low scrub or grass, or are bare and rocky slopes. From San Juan de Cesar to Valle de Upar the vegetation is practically the same, except for the forests of the Badilla River flood-plain, which is very broad and well wooded. The acacia-like shrubs and trees and cacti abound everywhere else, with some slight approach to savannas in places, especially near Badilla and above the flood-plain of the Rio Guatipuri. The flood-plain of the "Guatipuri is very extensive as one approaches Valle de Upar, with many canals and irrigation ditches, but on the south side of the river the land is higher and the arid woodland is largely in evidence. ‘This same condition continues southward “(at least along the trail) for about five miles, except that cacti and acacias are less abundant and other species of trees and shrubs begin to appear, and then the dry woodland changes grad- ually into a fairly heavy flood-plain forest which extends southwestward, with but few breaks, to a point some ten miles south of Valencia de Jesus. This is, in greater part, a magnificent lowland forest, consisting of many AMPHIBIANS AND REPTILES OF SANTA MARTA 22 large spreading trees, intermingled with smaller trees and considerable undergrowth. However, it is not properly a humid forest, since the types of trees are those of the driew woodland, developed to a more luxuriant degree, rather than those of the true humid forest, leading one to believe that the moisture, on which depends its luxuriant growth, is drawn from the subsoil rather than from the surface water of the rains. There is no doubt, however, that the rainfall is greater south of Valle de Upar than to the north of it, and these valley forests evidently extend unbroken down to the Rio Cesar and thence to the Magdalena River, because the bird fauna present here is strictly Magdalenan and very different from that to the north of Valle de Upar. This forest (along the trail) ends very abruptly at a small creek, the character of the soil changing equally abruptly, which leads me to believe that the presence or absence of luxuriant forest may be dependent to a considerable degree on the character of the soil, or, more strictly speaking, the subsoil, which I have found to be usually an extremely hard, impervious formation wherever savannas are found in the lowlands. This association of savanna and hard subsoil does not, however, hold good on mountain slopes. The savannas of the Macotoma Valley and elsewhere, I am led to believe, are the results of deforestation and repeated, systematic burning on the part of the inhabitants, rather than to any lack of humidity or nature of the soil. As mentioned above, the flood-plain forest at Valencia ends abruptly to the west, changing to scattered clumps of low trees and tangled shrubbery, intermingled with areas of savanna. (PI. VIII, fig. 1.)\ The savanna is not typical, however, as certain easily recognized and characteristic species of grasses and shrubs are absent. The whole plain from here on to Cam- perucho is very flat, trees and shrubs become less frequent and the tract of savanna larger and larger, until the woodland is confined to the immediate banks of the little streams which descend at rather frequent intervals from the nearby foothills. AAs Camperucho is approached the country becomes slightly rolling and hilly, for not far beyond here the trail crosses the last remnant of the great southwest spur-ridge. Here in this hilly country the savannas present a truly typical aspect, true savanna grasses and shrubs being present. Soon after leaving Valle de Upar the trail again approaches the very margin of the foothills which rise abruptly from the level plain, a peculiar characteristic noted from the time we left Fonseca. From here on to Cam- perucho the trail is never far from the foot of the hills which present the same barren aspect all the way across from Fonseca. Higher up savannas abound, but from about 3,000 feet down little or no grass is present, except for scattered areas, while the narrow valleys and ravines and lower slopes of the valleys are slightly wooded. Soon after entering the savanna district, south of Valencia, numerous large palms began to appear, becoming more and more numerous until we were traveling through a veritable forest of giant palms growing out of clean, open savannah. This same palm, locally known as “Coroso,” is abundant over the whole western lowlands as far as Santa Marta, growing 24 ALEXANDER G. RUTHVEN up to an elevation of about 1,000 feet, but never naturally in open savannah. In fact, the seeds will not germinate except in damp, shaded places, such as in the shade of the forest. Since these palms are so abundant over prac- tically all of the savanna areas between Valencia and Camperucho, it leads one to believe, together with other facts, that these savannas are not natural, but also the result of deforestation and burning, partly intentional and partly accidental, of a drier, more open class of woodland, which, growing over a very hard, impervious subsoil, would not readily reproduce itself after having once been destroyed. The argument that the palm would have been destroyed by such fires is erroneous, because this palm was seen in many instances burned around the base so deeply that not half of the original volume of the trunk was left, yet the trees were still luxuriant and strong and bearing as large seed clusters as the unburned individuals. After crossing the savanna-clad hills of Alto de las Minas, we dropped down abruptly into another and distinct region of forest. In some places it was damp and dense, but for the most part it was rather light and open, with much tangled undergrowth. The country is more undulating, the soil more gravelly and less fertile, conditions which continue from Alto de las Minas to the Fundacion River, except in the small valleys and on the great flood-plain of the Rio Ariguani. The forests of the Ariguani are truly magnificent, consisting of gigantic trees and a sparse undergrowth; they are dense and gloomy, even at mid-day. Typical humid forest types of birds prevail, and one is reminded of the forests of the Sinu, lower Atrato, and eastern Panama and Costa Rica. Giant creepers and lianas and the peculiar parasitic fig tree abound, although epiphytes are not particularly abundant. The foothills, from Alto de las Minas northward, are also clad with a forest which extends unbroken upward. From the Fundacion River to a point slightly north of Rio Frio the lowlands were originally clothed with luxuriant humid forest, but this has been nearly all cleared away between the lower edge of the foothills and the swamps which extend back from the Cienega Grande for many miles, while vast banana plantations and pastures have taken its place. The original forest here was in many parts very similar to that of the Ariguani, but not so heavy, I imagine, and with more undergrowth and wild plantains. Just beyond the Rio Frio there is again a very abrupt change from fertile banana farms to desolate salt plains covered with a scanty growth of acacia scrub and cacti, with a great deal of a peculiar salt-bush, common at Riohaca and between Donjaca and Gaira, near the sea beach. Soon after crossing the Rio Cordoba the same class of dry, thorny, scrubby woodland appears which was so common around Riohacha and between Valle de Upar and Fonseca. This condition continues on to Santa Marta (PI. I, fig.2) and around to the Rio Piedras, but extending here to a considerable extent up over the lower foothills to nearly 1,000 feet in many places and much more in others, especially on the ridges. The valleys of the streams are well wooded, however, and the acacias are less abundant than on the east and south sides of the Sierras. The shores of the Cienega Grande are all low and swampy and flooded for a long distance inland during the period of high water on the Magdalena AMPHIBIANS AND REPTILES OF SANTA MARTA 25 River, while the outer fringe of vegetation is largely composed of man- groves. It is a paradise for water fowl, while many shore birds are present on the sandy strip between the lagoon and the sea and on the open wastes around the town of Cienega. Foothill Vegetation: The vegetation of the foothills presents few abrupt changes, excepting those on the south side, which have already been described. The hills on the north side, from Rio Piedras to far beyond (northeast) Barbacoas, and those on the west side from Alto de las Minas to the Rio Cordoba, are heavily forested, much more heavily, however, on the north side than on the west, where the forests of ridges between the valleys have a tendency.toward the “dry forest” of the foothills around Santa Marta. (PI. III, fig. 1.) This foothill forest is composed of medium-sized trees, with much tangled undergrowth (except on the north side), and contains many deciduous trees, which remain bare during the dry season and thus give the undergrowth a chance for more extensive development. On the whole, the foothill forest is most uninteresting. It contains very few epiphytes, very few lichens, and almost no moss, the only redeem- ing feature being the many beautiful ferns in the damper, more shaded parts. It is choked with vines, brush, and bamboo grass, is difficult to pene- trate, and contains very few characteristic species of birds, mammals, or reptiles. This condition extends upward from 1,500 to 2,500 feet on the northwest and western sides, the higher elevation corresponding to the narrow ridges and the lower to the valleys. At these altitudes a gradual change takes place and the foothill woodland merges into the mountain forest. On the north slope this change is less apparent, the mountain forest, by reason of the greater humidity, extending down over the foot- hills for the most part practically to the edge of the costal plain, but the foothill area is, nevertheless, largely devoid of mosses and lichens and epi- phytes, which nowhere appear abundantly until the lower edge of the “cloud zone” is reached. Mountain Forest: At about 2,000 feet a marked change is noticed in the character of the forest. Deciduous trees disappear almost entirely, ferns become very abundant on north and west slopes, and a few bromeliads and -orchid-like plants appear high up in the tops of the trees. The trees become gradually taller, more slender, and set closer together. Certain distinctive lowland and foothill types disappear altogether, to be replaced by others. At about 3,000 feet a slender, graceful palm known as Maquenque, puts in its appearance, and the underbrush changes from wiry brush and bamboo grass to a low, broad-leaved palm and to more succulent types of shrubbery. lichens are now abundant on many species of trees, while bromeliads are more common and lower down. From 3,000 feet up to 6,000 feet there is little change in the vegetation, merely in the diminishing or disappearance of some species and the increase or appearance of others, but the types remain about the same. Between 4,000 and 6,000 feet the trees seem to attain their maximum development, and truly magnificent forest areas are present at these altitudes under favorable soil and slope conditions. The 26 ALEXANDER G. RUTHVEN Maquenque palm, so common lower down, disappears at 5,000 feet, but the low, broad-leaved species, “Cola Gallo,’ becomes more abundant, so that in many places it forms almost the only undergrowth present. On the northeast slopes of the Sierras, also on San Lorenzo and the Horqueta, there is a most unusual development of tree moss in the cloud zone. Between 6,000 and 8,000, or even 9,000 feet, these slopes are drenched with rain or cloud mist almost daily, and as a consequence this moss attains such a growth as to choke out most of the smaller epiphytic growth, only the hardier Bromelias being able to withstand it. Not only the trunks of the trees are solidly carpeted with it, but the branches and even slender twigs are completely enveloped with close-set filaments from one to three inches lgag Ea Phi ii rey Py: ) A magnificent palm, resembling the Royal Palm, is also abundant on the north slopes between 4,500 and 6,000 feet, appearing often in large groves, the great plume-like tops towering above the other forest growth. Above 6,000 feet the size of the forest trees gradually diminishes, except in sheltered valleys. Bromeliads and orchids become more abundant in individuals and species and now grow low down on the trunks of the trees as well as on the branches. Four very distinct species of bromelias were noted on San Lorenzo between 5,000 and 8,000 feet. At 7,000 feet the type of forest growth changes rapidly, giving way to trees of a hardier character, many with small, close-set leaves of dark, glossy green. They are not so tall, more branching, and less symmetrical in shape. Different types of undergrowth also appear. _Bamboo grass and bamboo cane become the dominant undergrowth in many places, often completely choking out every other variety of smaller growth. Such tracts of bamboo grass and cane are nearly impenetrable, except by the hardest kind of cutting with a macheté, and they contain little in the way of animal life. At about 7,000 to 7,500 feet and upward as far at least as 9,000 feet, on all open or exposed ridges, especially those which have been burnt, a peculiar large bromeliad with great recumbent stalks grows along the ground. (Pl. VI.) These -~ plants harbor many species of amphibians and insects and hold a consider- able amount of water in their crowns, even through the dry season. Coarse brake ferns of several species are abundant, especially where fire has passed, extending from 4,000 to 8,000 feet or more. Countless species of smaller ferns abound at all altitudes, from 1,500 feet to timber-line, from the most delicate of maidenhairs to the great tree ferns. There are also present on San Lorenzo, from 7,000 feet upward, great numbers of an exceedingly tall and slender palm, known locally as the “wax palm,” and apparently very similar to, or identical with, the palms which are so abundant in the central Andes in the vicinity of Quindio Pass. I have no recollection or note of the presence of these palms in the Sierra Nevada proper. From 7,000 feet up to timber-line there is little change in the types of vegetation, certain species of trees and shrubs disappear, to be replaced by others of similar type. The trees, however, become lower and more branching, and on exposed ridges are gnarled and twisted, while nearly all are smothered in masses of many curious forms of epiphytes, mosses, and lichens. At timber-line the change AMPHIBIANS AND REPTILES OF SANTA MARTA 27 from low forest to open moorland is sometimes very abrupt, though as a rule gradual, the trees becoming smaller and smaller and gradually giving way to shrubbery and bushes intermixed with coarse grasses. Paramo Vegetation: The immense paramos or moor-like tracts of land extending from timber-line to the perpetual snow present many curious types of plants. Naturally the prevailing vegetation is grass, of many spe- cies, but numerous varieties of bushes and shrubs are present, as well as many kinds of annual flowering plants of considerable beauty and bright colors. The most peculiar and characteristic form of the paramo is the great mullein-like plant known as “‘Frailejon,’ of which there seems to be but one species in the Sierra Nevada de Santa Marta, three very different varieties being found in the Sierra Nevada de Chita, in the eastern Andes. The leaves as well as the stalk are thick and woolly. In favorable spots it attains to a considerable size, but I never found it more than five to six feet high in the Santa Marta Mountains. In the eastern Andes it is often seen growing from fifteen to eighteen feet in height, with the trunks from eight to ten inches in diameter. The flower buds are enclosed in a mass of silky wool, of a pale greenish yellow, or golden yellow, very beautiful just as they begin to open, of large size and in clusters of four to eight. Hum- ming-birds feed almost exclusively on these flowers in season, especially Oxypogon, while the seeds furnish the bulk of the food for the many spar- rows of the paramo. Mountain Savannas: Great tracts of savanna are present on the south- ern slopes of the Nevada and on the north slope in the Macotama and Rio Ancha valleys, also to a lesser extent in thé Palomina, all more or less above 2,000 feet. The cause of these savannas has been a debated question and has not, I believe, been satisfactorily explained in many cases. It is a practical certainty that where such savannas occur above 2,000 feet and are in a zone belonging to, or surrounded by, a forest belt, such as those of the north slope, they are due to deforestation and persistent burning. The fact that in regions of extensive savannas the rainfall and consequent amount of humidity is less is an effect of, rather than a cause for, the pres- ence of the savannas, because the absence of forest growth always lessens the amount of atmospheric condensation. Colombia is a country which has been populated in certain sections for many years, and the valley south of the Sierra Nevada de Santa Marta is one of the oldest of these, it having been settled early in the sixteenth cen- tury. The town of Valle de Upar was founded before the colonization of North America began. Deforestation and systematic burning, covering a period of from one to four centuries, must inevitably destroy large areas of forest, while the natural flora to follow in such cases would be the hardy grasses which constitute the vegetation of the present savannas. Several instances tending to prove this theory have come under my personal obser- vation, covering a period of but nine years, so that if obvious results can be obtained in that short time, what might not take place in a period cover- ing several centuries. Not only have the southern slopes of the Sierra 28 ALEXANDER G. RUTHVEN Nevada been settled by the Spaniards and their descendants for nearly four hundred years, but they found there upon their arrival a large population of Indians, living largely by agriculture, the remnants of which still exist on those southern slopes and in the three valleys of the north slope which today show large areas of savanna. It is an absolute certainty that the savannas on the north and west slopes have been caused entirely by defores- tation and burning done by the Indians. I am also inclined to believe that the southern slopes of the Sierras were at some time forested to a much greater extent than at the present day and that their present barren con- dition has been brought about largely by deforestation and burning. A mountain savanna when once started, and burned over every year at the end of the dry season, will spread very rapidly, for with each burning the sur- rounding woodland is killed for distances varying with the character of the undérbrush and the steepness of the slope, the fire often running up a steep slope through the forest for miles before dying out. The Colombians as well as the Indians invariably burn over the savannas every year, just before the spring rains, to remove the hard, woody grass stems and to stimulate the growth of tender fresh grass for their livestock. It is probable that the original forest growth of the southern slopes was not dense owing to the lesser amount of precipitation there, brought about by natural topographical conditions, as elsewhere explained, and for this reason was the more readily destroyed by fire. Once destroyed, its replacement was an impossibility, owing not only to repeated burnings but to diminishing rainfall, due to the absence of woodland. Probably the most striking confirmation of the recent formation of savannas at the foot of the southern slopes, is the fact that there is an entire absence of any reptilian or amphibian fauna peculiar to natural savanna, all the forms taken on these savannas being the same as those present in the surrounding and adjacent forest or scrub. General Character of the Faunas, Mammalian and Avian: ‘Taken as a whole, the fauna, especially the bird fauna, of the Sierra Nevada de Santa Marta, is not an abundant one, as represented either by species or individuals, when compared with that of many parts of the Colombian Andes. The great group of fruit-eating tanagers, so abundant in all three ranges of the Andes, is conspicuously absent here, but three species of Tanagra being found from sea-level to timber-line. buthraupis and Comp- socoma are entirely absent, while Poecilothraupis and Ramphocelus are represented by but a single species each. The other great fruit-eating groups, the parrots, toucans and trogons, are also poorly represented from the foothills upwards. Parrots are abundant in the lowlands of the west and parts of the south side. In the mountains proper but four species of parrots, two of trogons and three of toucans are found, which is a mere bagatelle as compared with the Andes, especially the parrots. This undoubt- edly shows that the forest growth, which is very different from that even of the eastern Andes, offers little in the way of fruits and berries. On the other hand, insectivorous forms are more abundant, especially the Dendro- colaptide and Formicariide. Hummingbirds are also poorly represented everywhere, and present in small numbers, with the exception of a few AMPHIBIANS AND REPTILES OF SANTA MARTA 29 very hardy, wide-ranging forms of the lowlands. However, above 4,000 feet there are not more than about a dozen species of Trochilide, a very poor showing as compared with either range of the Andes. Monkeys are rarely seen above 2,000 feet, and very uncommonly above the costal plain. In Venezuela I have taken Mycetes up to 7,000 feet, where it was appar- ently not rare, according to local report. Among the mammals, the wild pigs and agoutis are the most abundant, ranging from sea-level to at least 8,000 feet. Tapirs are fairly common above 3,000 feet. Deer are nowhere common, except the large species found in the delta region of the Magdalena, which in reality hardly belongs to the region under discussion. Rodents and squirrels seem to be abundant nearly everywhere up to at least 6,000 feet, as well as marsupials and foxes; the latter, however, are not, as a rule, seen above 2,500 feet, although I saw a pair at Vista Nieve at 5,000 feet. A description of the amphibian-reptile fauna is given by Dr. Ruthven. Conclusion: The great inaccessibility, solitude and isolation of the Sierra Nevada de Santa Marta have been the chief causes of the fragmen- tary nature of the biological work done there, and until better means of travel and communication have been opened up there are numerous large areas which will remain unexplored and unknown. A few regions, such as Mount San Lorenzo and the Horqueta, which are easy of access, have been intensively worked along several lines, and certain accessible portions of the lowlands may easily be studied, especially the banana district. How- ever, the bulk of the region, both lowlands and highlands, is accessible for intensive biological study only by means of well-equipped expeditions with considerable time and money at their disposal. Such expeditions could, however, be easily organized and carried out by the combined efforts of several large and well-equipped scientific institutions, each one furnishing one or two experts in certain branches of biology, and bearing its corre- sponding quota of the expenses. Such expeditions, under capable leader- ship, and composed of men accustomed to work in the tropics or other wild and uninhabited regions, could obtain important results, and there would be no duplication of field work, with consequent loss of time and money. There is probably no other portion of the American tropics which today offers such a promising field for intensive biological study as the Sierra Nevada de Santa Marta. The work already done has been entirely too fragmentary and widely separated to lead to many definite general conclu- sions, especially concerning such broad problems as faunal origin and rela- tions. ‘Too much mere reconnaissance and too little intensive study of life history and distribution have been done. In the present sketch (for with our present knowledge it can be classed as nothing more than a sketch) there has been little attempt to arrive at many definite conclusions. An attempt has been made to set down the facts as known to the author, with the hope that they may stimulate further work in this most interesting field and serve as a basis for intensive study. 30 ALEXANDER G. RUTHVEN List of Localities Referred to by Simons, Brown, Smith, Carriker, and the University of Michigan Expeditions. Alto de jas Minas. Point on the trail between Valencia de Jesis and Fundacion, where it crosses the lower extremity of the great southwest spur ridge of the Sierra Nevada. Probably not more than 600 feet above sea-level, and three days’ journey from Fundacion. Ariguani, A cattle ranch of considerable extent which has been opened up on the Rio Ariguani at the point where the Valle de Upar trail crosses it. It is in the heart of the great lowland forest, about thirty miles south of Fundacion, and at no great distance from the lower edge of the foothills. Aqua Dulce, Plantation on San Lorenzo, altitude about 3,000 feet. Arroyo de Arenas. A cattle ranch and regular stopping place for pack trains and travellers journeying between Riohacha and Fonseca. It is a long, hard day’s march from Riohacha, and the first point at which good water and forage can be obtained. It lies at the edge of the flood-plain of the Rio Camarones (called Barbacoas here) and is probably about 500 feet above sea-level. Outside of the flood-plain forest the vegetation consists of low woodland of the dry forest type, acacias and cacti being conspicuous. Badillo, Once a flourishing little town, but now a miserable collection of huts, situated on the trail between Fonseca and San Juan de Cesar, and near the Badillo River. To the southwest of the town lies the heavy wood- land of the river flood-plain, while to the north the vegetation consists of dry acacia-like scrub and cacti, interspersed with patches of open savanna- like areas. The foothills are but a few miles distant. Barbacoas. A tiny hamlet at the crossing of the Rio Barbacoas, about two miles south of Arroyo de Arenas. Bolivar. A small farm and lowland station of the Santa Marta Coffee Company, from which pack mules start for the hacienda Cincinnati. It is about five and a half miles from Santa Marta, a mile and a half from Mamatoco, and about 100 feet above sea-level. The prevailing vegetation, outside of the irrigated areas, consists of low, semi-arid woodland, acacias and cacti. The nearby Jamonaca Creek Valley is rather heavily wooded and offers good collecting ground. Cacagualito, A small farm, now abandoned, at which Mr. Smith col- lected. It lies on the trail between Bonda and the north coast, by way of the Quebrada Jordan and the Rio Piedras. It is between 800 and 1,000 feet above sea-level, on the ridge separating the valley of the Matajiro and Jordan, and lies in the foot-hills section of the semi-arid coast belt. Camperucho. A small settlement of scattered houses on the trail between Valencia de Jesus and Alto de las Minas, a hard day’s journey from Valencia. It is in a rolling, hilly region, on a remnant of the same ridge as Alto de las Minas, and not more than 400 to 500 feet above sea- level. It is largely a region of savannas and scattered clumps of scrubby woodland. AMPHIBIANS AND REPTILES OF SANTA Marta ar Cerros de Caracas. A broad, well-defined mountain range of the Sierra Nevada proper, beginning just below San Miguel, at the Rio Macotama, at an altitude of about 5,000 feet, and rising abruptly to about 8,000 feet, then gradually to about 13,000 feet, where it ends in a jumble of jagged, naked pinnacles of rock, which form a barrier to further ascent to the snow-capped peaks above. The whole of the eastern end and southeastern flank has been denuded of forest, but the northwest side is steeper and for- ested to the crest. Chinchiciia, Valley of. A locality given by Simons as having an alti- tude of 6,500 feet, and situated on the south slopes of the Nevada a short distance below San Sebastian. Chinchiciia, Pass of. Pass at head of valley of same name, having an altitude of 11,000 feet. Locality used by Simons. Chirtia, The name of a large valley and the small stream which drains it. Situated on the northeast slopes of the Nevada, at an altitude of from 2,000 to 5,000 feet, and empties into the Rio Ancha at Pueblo Viejo. The valley is nearly enclosed by mountains rising on the south side to about 5,000 feet, known as the Heights of Chirta. The valley is sparsely inhab- ited by Indians, and nearly the whole of the lower portion has been de- forested in years gone by, and is now covered either with grass, low scrub, or cultivations of the Indians. Chirtia, Heights of. The mountain ridge lying on the southwestern side of the valley of Chirtia, with an elevation of about 5,000 feet. Chiriiqua, Paramo de. The immense paramo lying on the crest of the Nevada to the south of the upper Macotama River and to the east of the snow peaks, with an altitude of from 11,000 to 15,000 feet. This is doubt- less the paramo de “Curigua” of Sievers. The name Chiriqua was used by Brown, and I followed his nomenclature without taking the precaution to verify it, so that I do not know which name is the correct one. Lake Macotama, mentioned elsewhere in this report, lies at the northwest end of this paramo, at 15,000 feet elevation. This paramo may be reached either from Pueblo Viejo on the north slope or from San José on the south slope. It 1s easier of approach from the north side, where a good Indian trail leads from Pueblo Viejo over to the Macotama River, thence up that stream through Santa Cruz, San Miguel, T'aquina, Macotama, and on up to the paramo. ‘This trail is passable for oxen or a good mule, and the paramo may be reached from Pueblo Viejo in two days. Cienega (formerly known also as San Juan de Cordoba). A city of between 15,000 and 20,000 inhabitants, twenty miles by rail from Santa Marta, and situated on the sea coast, just at the northeastern corner of Cienega Grande, the great brackish lake lying between the Sierra Nevada and the Magdalena River. Cincinnati, A large coffee plantation lying on the western slopes of Mount San Lorenzo, between 3,000 and 5,500 feet. This plantation was known as Valparaiso at the time Smith and Brown collected here. It is a 32 ALEXANDER G. RUTHVEN fine collecting field, and has been visited by all naturalists who have come to Santa Marta. The plantation is about nineteen miles from Santa Marta, and is reached by the road leading out through Mamatoco, Tigrera and Minca. Concha. A small bay southeast of Santa Marta, also the short creek and valley emptying into it. There is a small stock farm there belonging to a resident of Santa Marta. It is the first indentation of the coast beyond Taganga, and the second from Santa Marta Bay. Copéi, A stopping place for travellers and pack trains on the trail between Ariguani and Camperucho, about fifteen miles south of Ariguani. It is in the heavy lowland forest. Corazones. A little hamlet near Valle de Upar, between that town and Badilla, and not far from the Rio Guatipuri. It lies just between the Guati- puri flood-plain forest and the dry scrub of the open country. Dibulla. A squalid village on the coast about half-way between Santa Marta and Riohacha, and about three miles east of the point where the trail leaves the coast for the Sierra Nevada, by way of Pueblo Viejo and San Miguel. Donama, A small farm, now nearly abandoned, lying on the divide between the head of the Matagiro Valley and the Rio Piedras, at an eleva- tion of about 1,500 feet. It is on the trail between Bonda and the planta- tion Las Vegas. It is just between the “dry forest” area of the Santa Marta coast belt and the humid forest of the north slopes of the Sierras, the valley of the Rio Piedras being largely in the latter zone. Don Diego. An old plantation (now practically abandoned) belonging to a French company, and situated on the north coast at the mouth of the Don Diego River, one of the largest rivers draining the north slopes. There is a trail leading to it from Santa Marta, going out by way of Mamatoco, Bonda, Calabaso, Jordan, thence down the Rio Piedras to the coast, and then following the beach. This trail is almost impassable through disuse and lack of repair. The place is much more easily reached by sea from Santa Marta. The region around Don Diego is heavily forested, has a heavy rainfall and is very rich in animal life. Donjaca. A flag station and passing point for trains on the Santa Marta Railway between Gaira and Cienega. It is on the coast, and lies in the semi-arid coast belt of scrub and cacti. El Libano, A name used by Mr. Smith for a point on the northwest slope of San Lorenzo with an altitude given by him of 6,000 feet. It is in the upper part of what is now known as the plantation of La Victoria, which adjoins Cincinnati on the east. Conditions are practically the same as around Cincinnati. El Lorenzo (see San Lorenzo). El Mammon. A locality at which Mr. Brown collected in the Sierra Nevada, and given by Dr. Allen as having an altitude of 8,000 feet. It is on the south slopes of the Sierras, a short distance above San Sebastian. AMPHIBIANS AND REPTILES OF SANTA MARTA 38 Fonseca. A small town on the Rio Rancheria a few miles below the point where it breaks out of the foothills into the valley lying between the Sierra Nevada and the eastern Andes. I do not think it can be more than 500 feet above sea-level. Except for the heavy woodland of the Rio Ran- cheria flood-plain, the surrounding country supports a vegetation very sim- ilar to that encountered all along the trail between Riohacha and Arroya de Arenas, viz., low open scrub of acacia-like trees and cacti. It lies on the main trail between Riohacha and Valle de Upar. Fundacién. P1. Tu fhe.) If the amphibians and riparian forms are excluded, there are few differences to be observed in the fauna of these two terrestrial habitats, probably because of the limited extent of the ravine for- ests. On the other hand, ground and arboreal species may be recognized. Aquatic environments: Streams (shallow and larger lower down and swifter but with pools in the hills) ; ponds few in number and usually with much vegetation. The fauna is without aquatic spe- cies near the foothills, except possibly Kinosternon integrum, con- sisting of amphibians with aquatic larval stages and such riparian forms as the iguana and basilisk. Habitats above 2,200 feet: Terrestrial Forest: Tropical rain forest, becoming gradually wetter from its lower limits; trees large, forming dense shade; a deep layer of forest litter. Cloud forest, in general like the rain forest, but cooler and more humid, large bromelias. (PI. II; Pl. Ill, fig. 2; Pl. IV; Pl. V, fig. 2.) Fauna divisible into three groups—the arboreal, bromeliadicolous, and ground forms. Artificial clearings Aquatic environments: Streams small and torrential, without aquatic species of reptiles and amphibians. (PI. V, fig. 1.) It will be seen that seven associations of species of reptiles and amphi- bians are recognized: the ground species, arboreal species, and aquatic species below 2,200 feet; and the ground species, arboreal species, brome- liadicolous species, and artificial clearing species above 2,200 feet. It will also be noted that these associations may be grouped into two major hab- itats—the forest below 2,200 feet (in large part scrub forest and small ravine forests and open cactus areas, dry litter with higher temperatures and low humidity) grading through rain forest above 2,200 feet into a cloud forest (dense forest, deep wet litter, large bromelias, lower. temper- atures, abundant rainfall). The aquatic habitat below 2,200 feet furnished no strictly aquatic forms. AMPHIBIANS AND REPTILES OF SANTA MARTA 43 It is clear that the bromelias must be considered a minor habitat for the groups under consideration. The Bromeliaciae are in general an impor- tant habitat in the wet forest above 4,500 feet. (PI. V, fig 2; Pl. VI.) The ‘list of species found in these plants is as diversified as similar lists for other localities, and includes earthworms, leaches, peripati, isopods, myriapods, cockroaches, spiders, Heteroptera, beetles, scorpions, insect larvae (flies, dragonflies, beetles), a salamander, and frogs of several species. However, lists of species taken in bromelias may be misleading, for all of the forms found in these plants are not, strictly speaking, bromeliadicolous. In bromelias are to be found species which are apparently confined to this habitat and those which are found here occasionally but quite as often, and in some cases much more commonly, in other situations affording conceal- ment, such as under bark, in hollow trees, and in masses of vegetable debris lodged in trees and shrubs. It would seem to be best to restrict the term bromeliadicolous to those species which for the most part breed in the bromelias, and to those which, if not breeding there, occur principally in these plants. Under this definition one salamander, Oedipus adspersus (viviparous), and two frogs, Eleutherodactylus cruentus and E. delicatus, are bromeliadicolous forms in the Santa Marta Mountains, and Eleuthero- dactylus sanctae-martae, and E. insignitus may prove to be. The distribution of the more common amphibians and reptiles found on the northwest slope of San Lorenzo is summarized in the following table. The cross-section made may be located by the towns and plantations given at the head of the chart. (Facing p. 4o.) It is to be understood that the data on the distribution of the snakes is fragmentary. This is to be expected, since collecting in this group is well known to be fortuitous in any region. The data is not without value, however, when it is considered that every species taken above 2,200 feet and not yet found on San Lorenzo below this altitude is known to occur on the lowlands in northern Colombia. This fact, with the further one that only one lizard and six frogs, a total of seven species out of forty-two common species (including the snakes), are not known from below 2,200 feet, shows clearly that the fauna has been in large part and probably entirely derived from the lowlands.* A study of the table will reveal a certain amount of zona! arrangement of the species. Only five species, three amphibians, one lizard, and one snake out of nineteen desert forms have not been found above the desert, and of these at least the snake certainly occurs higher. Similarly. of a fauna of twenty-three species found in the dry forest, fourteen forms are known to occur on the desert, and of the nine forms not taken on the desert three are snakes and six are lizards, all of which will probably be found to occur there, as all but one have been found elsewhere on the low- lands. It is evident that the fauna of the dry forest and desert are prac- 2Compare Ohaus, Ent. Zeitung, Stett., 1900, pp. 211-212, 237-240; Calvert, Ent. News, XXII, pp. 402-411; Scott, Ann. and Mag. Nat. Hist., X, pp. 424-438. 4A similar condition was found by the writer in the San Francisco Mountains, New Mexico. Bul. Amer. Mus. Nat. Hist., XXIII, 603. 44 ALEXANDER G. RUTHVEN tically the same, the ranges of but few of the species being affected at the common boundary of the more open plains and scrub forest. f There is a decided difference in the fauna roughly below and above 2,200 feet. Of a total fauna of twenty-three species in the dry forest, thirteen species do not occur above 2,200 feet, five others occur higher only in clearings, and still another is not known to breed above that altitude. so that nineteen out of the twenty-three species are naturally limited in upward range between 2,000 and 2,500 feet. The four species which range higher are snakes. It will be noted that seven of the lizards out of sixteen dry forest forms have not been found above 600 to 800 feet. These seven undoubtedly range higher and were not found between 800 and 2,200 feet because careful collecting could not be done in that region. The upper limit may be placed tentatively at 2,000 feet, because they have not been found at 2,200 feet, although some doubtless drop out well below this altitude. In brief, then, none of the amphibians of the desert and dry forest (six species) apparently breed above 2,200 feet; the only lizards of these habitats (16 species) ranging above this altitude are five which are found in artificial clearings, and it is the distribution of the snakes which seems to be little affected, four species out of six ranging higher. None of the eight forms found in the moderately wet forest, between 2,200 and 4,500 feet, are peculiar to this area. A number of forms have not been found below it in this region, but these all range above it into the cloud forest. The cloud forest fauna consists of eighteen species, of which twelve have not been found below 4,000 to 4,500 feet. These are nine amphibians, one lizard, and two snakes. Moreover, of the eight forms which enter the moderately wet forest three forms (one amphibian and two snakes) do not range entirely through the area, one (an amphibian) reaches 2,200 feet, and four (snakes) are found in the dry forest. In other words, out of eleven forms of lizards and amphibians found in the cloud forest not one occurs below 2,200 feet, and the cloud forest and dry forest faunas are thus entirely different in respect to the lizards and amphibians and overlap only when the snakes are added. As has been said, the moderately dry forest has no peculiar forms, but is composed of species which range below or above it, or in both direc- tions. It is thus to be considered as an intermediate zone in the distri- bution of the reptiles and amphibians, and the fauna of the slopes falls into three groups, distinct as to species—a desert and dry forest group, twenty- four species; a cloud forest group, fourteen species; and a group of forms of general distribution, four species (of snakes). This vertical distribution corresponds to the vertical distribution of dominant terrestrial habitats. Environmental Factors in the Vertical Distribution: It should be clearly understood by geographers that geographic data yield only indirect evidence of the environmental factors in distribution. It would be easy to attribute the apparent zonal distribution above described to temperature, to moisture, to‘'the nature of the ground cover, or other factors, but this would be pure assumption. The writer has expressed the opinion® that no 5 The Geographical Review, X, pp. 241-248. AMPHIBIANS AND REPTILES OF SANTA MARTA 45: one factor will account for the distribution of a fauna, and that zonal dis- tribution of faunas based on temperature differences probably does not occur on mountains. It should be understood that this is not equivalent to saying that there can be no such thing as zonal distribution on mountains. Indeed, it may be accepted as a corollary of the proposition that the environ- ment as a whole controls the distribution of each species, that when a large number of conditions are changed at about the same place a considerable part of the fauna may be affected. It has been shown that the dominant habitats change at 2,200 feet, and this means a change in many conditions of life. It is clear that zonal distribution occurs on San Lorenzo, and the data indicate that temperature is probably not the principal factor. As stated, nine dry forest forms have not been found on the desert at Santa Marta. Of these, eight have been found in dry forests elsewhere on the plain. If there is a difference between the fauna of the desert and ’ dry forest, the factors are not evident. Of the lowland forms which do not get above 2,200 feet, five (lizards) go a little higher and three range from 100 to 2,300 feet higher in artificial clearings. This suggests that some other factors than temperature are, under natural conditions, effective at this point in limiting the upward range on San Lorenzo of the species of this group. Of the cloud forest species in all groups (18), twelve are known to occur here or elsewhere in northern Colombia on the lowlands, which again suggests that on San Lorenzo the temperature of this area is not the only condition which limits the lower range of the species. It is to be noted that of the fourteen species known from the cloud forest and not lower than 2,200 feet, ten are amphibians which quite certainly do not lay the eggs in water, one being viviparous and nine more or less certainly known to lay eggs which produce young in the adult form, while three, possibly four, are associated with bromelias which do not descend lower. Further- more, at least one lowland amphibian seems to be limited in its upward distribution by the absence of standing water above 2,200 feet, since it occurs well above that altitude apparently without breeding. In the cloud forest, the nature of the forest, the absence of standing water and the humidity are evidently factors with which to reckon. The wet forest comes down to the coast on the north side of the mountains. Carriker (see p. 20) concludes from a study of the birds that the forest fauna on that side also comes down to sea-level, a conclusion that is substantiated by the occurrence at Don Diego of Phrynonax poecilo- notus and Phyllobates subpunctatus, species not found below 2,200 feet on San Lorenzo. ‘This lowering of the wet forest habitat is probably due in large part to a greater rainfall, and while the temperatures are also lower it is not possible that they are as low as in the cloud forest. The greater humidity and the forest are quite as likely to be critical conditons as is the temperature. There is no reason to believe that the distribution of snakes is less effected by differences in temperature than that of lizards, and yet the data, while fragmentary, show less response on the part of the snakes to the 46 ALEXANDER G. RUTHVEN differences that exist between the dry forest and the cloud forest. The principal difference in habits between the two species in the groups in this region is in food habits, the lizards being largely insectivorous, the snakes mostly carnivorous. Certainly the food supply for lizards is much more different in the cloud forest and in the dry forest than it is for the snakes. Taking into consideration the known altitudinal distribution and the habits of the species, the conclusion seems warranted that an absence of standing water and greater humidity and more shade above 2,200 feet” may be important factors in limiting the downward and upward migration, respectively, of the cloud forest and lowland amphibians, and that the more open conditions below 2,200 feet are more favorable than those in the wet forest for at least five of the lizards found below 2,200 feet, as they occur above this altitude when clearings are made. It should not be inferred that the writer denies the influence of tem- perature in limiting the vertical distribution of particular species in this region. It is only meant that particular temperatures are not clearly respon- sible for the zonal distribution. It is quite possible that when several spe- cies are stopped at approximately the same place, in some cases a temper- ature difference may be the cause, while with others other factors may be operative. This view is quite in harmony with physiological data, and on San Lorenzo other factors are probably operative, since the abrupt change from desert to cloud forest is attended by changes in many environmental conditions for animals. Relations of the Cloud Forest Faunas: ‘The fact that, while the lizards and amphibians of the cloud forest are limited in their downward range in the moderately wet forest on the northwest slope of San Lorenzo, five out of the eleven species not found below 2,200 feet are known to occur elsewhere on the lowland, is more than evidence that temperature is not the only controlling factor in the vertical distribution, but would seem to indi- cate clearly that this fauna is or has been continuous with a lowland fauna containing many of the species found on San Lorenzo only above 2,200 feet. To obtain informati6én on the origin of the cloud forest fauna, work was done at Don Diego and a trip was made entirely around the range from Riohacha to Fundacién. In the work on the east and south sides of the range the lizards received principal attention, in the belief that being more easily collected they would be most likely to yield dependable results in the short time which could be devoted to the work. The distribution of the species is summarized in the accompanying chart. From the chart it will be evident that a desert or dry forest extends from Santa Marta to Riohacha on the west, south, and east sides of the range, and that as far as the lizards are concerned it is the principal fauna of the lowlands. Of a total fauna of twenty-four forms only nine have not been taken at Santa Marta, where there is no heavy lowland forest. Of the nine forms not taken at Santa Marta four are distinctly forest spe- cies (Mabuya agilis, Tupinambis mgropunctatus, Loxopholis rugiceps, and Polychrus spurrelli), and it is believed to be significant that these species 6 Conditions unfavorable for the development of eggs laid on the land. 47 PTILES OF SANTA MARTA 4 4 wy AMPHIBIANS AND Rg ‘UOISIA DAD DIUDS ayy Ul Spapsy puvjoro] uowuios fo woungrajsiq 9 lL ST whe 6L 9L SPOOM MOT La Asis aes So) Ses oe (Gy ieee kien eae eee it fea I t eae rare ee Ts) Se eee (a eae ee ee te a a ea aes a get a eae eo ST \ I T ygea10j3 Aap puB YIESEeq Gan eee) ot el ee (i) ES lela 8 3 —¢ 9 BPOOM MOL [-------------------- [== 1 pus 4seir0jy f£1aq hi i sSpoom MOT lL qiesog it Seuaty op Bsor1y Boosuog Iedyj ep eLlLea 4993 0023 898) }4qBy pus 04 04 MO TAQ [edyouyag o¥eTd uod weyoByoTYy BYIBT BUWOT uBne UBS uofoupung ByzBy Bqyues 48 ALEXANDER G. RUTHVEN have only been taken at Don Diego, Aracataca, Fundacion, Valencia, Las Pavas, and Arroyo de Arenas. As has been pointed out, the desert and dry forest is replaced by low- land forest in the valleys of the Aracataca, Fundacion, and Ariguani, and by a dryer but still heavy forest in the valleys at Valle de Upar, San Juan de Cesar, and Fonseca, while at Loma Larga, Arroyo de Arenas, and Don Diego the forest is heavy. The occurrence of the four species named above at least indicates a wet forest element in the general fauna at the places mentioned, and since two of the cloud forest forms (Phyllobates subpunc- tatus and Phrynonax poecilonotus} have been taken at Don Diego, and since others (Lepidoblepharis intermedius and Cryptobatrachus fuhrmannt) have been taken in the Choco, it may be confidently expected that at least a part of the cloud forest fauna reaches the lowlands in the large valleys at the west end of the range and has been derived along this route from the lowland forest fauna of the Magdalena basin. The large number of forms apparently peculiar to the cloud forest suggests that this almost completely isolated habitat possesses an endemic fauna. This can be determined only when much more is known about the reptile and amphibian fauna of Colombia. The Distribution of the Desert and Dry Forest Fauna: It is well known to herpetologists that the faunas of Venezuela and western Colombia are different in many respects. The differences suggest a break in the distri- bution of a number of forms in eastern Colombia or western Venezuela. Too little is known of the distribution of most of the species to permit of the locating of the place or places where the change takes place for many species, but the distribution of the lowland forms in the Santa Marta Moun- tains region apparently throws some light on the problem. The chart (p. 47) shows that on the south side of the range many of the lizards do not range beyond certain localities in a particular direction. It is evident that since the principal range of the species is not indicated and there are many chances that the distributions are not accurately known, this summary cannot be relied upon to reveal the eastern and western limits of all of the forms included. There are, however, several forms which here approach the range of near relatives or whose range is definitely enough known to indicate the nature of the relationships of the faunas of the two regions. Many forms are common to Venezuela and Colombia. Some eastern forms, e.g. Hyla venulosa (not included in the chart), occur as far west as Fundacion, while one, Tropidodactylus onca, stops at Riohacha. Ameiva bifrontata, a Venezuelan form, is replaced between Fonseca and Valle de Upar by Ameiva bifrontata divisus, a Colombian form. One Colombian species, Phyllobates subpunctatus (not shown in the chart), does not get farther east than the Don Diego region, and is replaced by a nearly related form in Venezuela, while Ameiva maculata and Cnemidophorus lemniscatus gaigei enter western Venezuela at least before being replaced by other forms. These facts indicate that there is no sharp break between the faunas of Venezuela and Colombia, and that the eastern and western forms which enter the region are limited in range at different places. AMPHIBIANS AND REPTILES OF SANTA MARTA 49 List OF SPECIES CAUDATA Oedipus adspersus (Peters)—San Lorenzo, 4,500 to 7,000 feet; Rio Frio, 1,000 m. (U. S. National Museum); heights east of San Miguel, 6,000 to 7,000 feet (M. A. Carriker). Found principally in bromelias in the trees and on the ground; occasionally found in decaying logs and stumps or under decaying leaves. This is a viviparous species, and both young and adults were taken in the bromelias. SALIENTIA Phyllobates subpunctatus (Cope).—San Lorenzo, 2,200 to 7,000 feet ; heights west of San Miguel, 7,000 feet, and Don Diego (M. A. Carriker). A ground form generally found near water, and usually occurring in num- bers where there are small clearings grown up to grass and herbaceous plants. Although common, this species is a difficult one to collect and study in the Santa Marta Mountains. The breeding habits and tadpole have been described (Ruthven and Gaige, Occ. Papers, Museum of Zoology, Univ. of Michigan, No. 10). The eggs have not been found, but they are evi- dently laid on the land. Adult males carrying tadpoles have been found from June 4 to July 14. Venezuelan specimens examined (La Gaira and San Esteban) are P. trimitatus, as stated by Barbour and Noble (Bull. Mus. of Comp. Zool., LXIII, pp. 401-402). This species can be distinguished from P. subpunc- tatus by the fact that it either has a dark bar across the chest, or the throat, chest and belly are dark, but there are other distinctions between the forms. The only structural difference which the writer has found is in the rough- ness of the skin of the dorsal surface. Over most of the body the skin is smooth or slightly granular, but in the lumbar region it tends to be more granular, and in P. trinitatus is raised into more or less numerous and prominent warts. Of 20 specimens from Trinidad, 14 have distinct warts in the lumbar region, and in 6 Trinidad specimens and 3 Venezuelan speci- mens the warts are small and few in number. Of 18 Santa Marta speci- mens of P. subpunctatus, one has a few faint indications of warts, in others the skin is granular or smooth. Notwithstanding a general similarity, the specimens from Trinidad and Venezuela (P. trinitatus) can be distinguished from those from Santa Marta (P. subpunctatus) by differences in the coloration. In the latter the black or blackish brown lateral stripe is very distinct and is usually bordered above by a light stripe which may extend around the snout, but at any rate is usually well defined forward to the eye. In Trinidad and Venezuelan specimens the dark line is less distinct and the light band is also much less distinct and often absent. In every specimen of P. trinitatus, except the very dark ones, the lateral dark band is widened or divided pos- teriorly to the eye, involving from a fourth to three-fourths of the tympa- num and reaching the limb insertion. In P. subpunctatus the black band is a little widened on the neck and includes but the upper margin of the 50 ALEXANDER G. RUTHVEN tympanum. Specimens of P. trinitatus are frequently so dark that the dorsal coloration is obscured, but when this does not include the ventral surface the gular band is distinct. Both in P. trinitatus and P. subpunctatus the coloration of the sides consists of a variegated pattern of black or brown and light brown, dull yellow or white. In the former the dark colors pre- dominate and the light colors have the form of light spots or dots and nearly always a short streak extending forward from the groin. In one specimen there is an angular light line below the black band in front of the arm. In P. subpunctatus the light color is more extensive and usually forms a reticulated pattern or conspicuous round spots, and the pale streak from the groin is seldom as definite as in Trmidad specimens and is usually absent. Geobatrachus walkeri Ruthven.“—San Lorenzo, 5,000 to 9,300 feet. Taken under leaves on the ground in the forest and under stones and the stems of bromelias in the clearing at 8,300 feet. Very common in the latter habitat.’ (PL xX? fis: 2.) The coloration is more variable than indicated in the original descrip- tion. Most of the specimens are dark olive gray to olivaceous black above ; the belly is gray in appearance due to the combination of a blackish wash and minute white dots. Some specimens have russet dots on the dorsal surface, and a few are brick red above, vinaceous rufus in the inguinal region, and Hay’s russet on the belly. A narrow white or ochraceous tawny vertebral line may be present. Atelopus ignescens (Cornalia).—San Lorenzo, from 5,000 to 7,000 feet. A forest form, the adults of which have not been found at a lower eleva- tion than about 6,500 feet; young, recently hatched, were common in the quebrada Viernes Santo, from July 15 to July 24, at an altitude of about 5,000 feet. The specimens agree closely with the descriptions, and if they are cor- rectly referred to this form it is probable that the species will be found at lower levels in some of the large wooded valleys, such as that of the Ari- guani. ‘ihe adults were taken on the ground in the forest, two of them in copulation on June 5. The smallest immature specimens were found at the margin of a stream, but there was evidently a later migration, for after the middle of July larger young were found several rods from the water. No eggs were found. The adults are very variable in color. The belly may be light orange yellow, apricot orange, pale ochraceous salmon, or dragons- blood red; the ground color of the back varies from olive to black. Atelopus carrikeri Ruthven..—Paramo de Macotama, 8,000 to 16,000 feet (snow-line). Taken along small streams, some of them rivulets from melting snow, by M. A. Carriker. Not known from other localities. According to Mr. Carriker this amphibian has a very offensive odor. Hypopachus pearsei Ruthven.°—Fundacion. Found only under logs about an open marsh where they were rather common. (PI. XI, fig.3.) 7 Occ, Papers, Mus. of Zool., University of Michigan, No. 20. 8 Occasional Papers, Museum of Zool., Univ. of Mich., No. 28. 9 Proc. Biol. Soc. Wash., 1914, 77. AMPHIBIANS AND REPTILES OF SANTA MARTA 51 Gastrophryne ovale (Schneider) —Fundacion. Only two specimens observed, both under logs in low woods. Eleutherodactylus insignitus Ruthven.*°—Heights east of San Miguel, 6,000 to 7,000 feet (M. A. Carriker). Taken in bromelias on the ground and in trees. Eleutherodactylus cruentus (Peters)—San Lorenzo, 5,000 to 8,300 feet; heights east of San Miguel, 6,000 to 7,000 feet (M. A. Carriker). Mostly found in bromelias, either on the ground or in trees. (Pl. XII, figs. 4, 5.) The habitat of this form is evidently the bromelias. In the forest at 5,000 feet on San Lorenzo it is to be found in numbers in the trees, the bromelias at this altitude being mostly above ten feet on the trees. From about 6,000 to 8,300 feet the bromelias grow lower down on the trees and on the ground in open places, but the frog is still confined to them. It is very abundant in the large bromelias which dominate the flora of the clear- ing at 8,300 feet. The breeding habits have been described by Ruthven.* The eggs have not been found elsewhere than in bromelias. The breeding season is evi- dently long. Carriker collected eggs at San Miguel on April 24; and on San Lorenzo, in 1913 and 1920, many sets were found as early as June 9 and as late as July 25, the entire period during which the species was under observation. The observed dates of hatching are June 15, 19, 28, 30, July I, 22-25. This indicates that the eggs are laid at different times during the rainy season. The number of eggs in a set is uncertain, for it has not been determined whether or not the females lay all of the eggs in one group. Often two or more sets may be deposited closely together, but whether by the same or by different females has not been determined. Usually from four to eight eggs are laid in one cluster, for when larger numbers are found in one group it is possible to distinguish more than one set. The adults are silent and retiring during the day, concealing themselves in the axils of the leaves; but at night they are active, the “song” being a rather harsh “check, check.” Eleutherodactylus carmelitae, new species*” GP ele he i Pie SE fot) Diagnosis: To be distinguished from EF. sanctae-martae Ruthven by the following characters: eye larger, tongue entire, tympanum less than one-half the orbital diameter, nostril equidistant from eye and end of snout; and from £. megalops Ruthven by greater size and larger eye, obovate tongue, larger disks (those of third finger larger than ear), smooth or finely granular texture of dorsal epidermis, narrower interorbital space (much narrower than upper eyelid), smaller palmar and sole tubercles, and in the position of the nostril. Description: Tongue obovate, entire behind. Vomerine teeth in two short, transverse groups close together behind the choanae. Head as broad 10 Occ. Papers, Mus. of Zool., Univ. of Mich., No. 34. 11 Occ. Papers, Mus. of Zool., Univ. of Michigan, No. 11. 12. Named for Mrs. M. A. Carriker, to whom the members of the several expe- ditions are indebted for valuable assistance in the field work. 52 ALEXANDER G. RUTHVEN as long; snout rounded, with rounded canthus rostralis, longer than diame- ter of orbit; nostril equally distant from eye and end of snout. Inter- orbital space much narrower than upper eyelid; tympanum about one-fourth the diameter of orbit. Fingers free, first shorter than second; toes free; disks moderately developed, that of the third finger larger than the ear; metatarsal tubercles small. The hind limb being carried forward, the tibio- tarsal articulation reaches between the eye and nostril. Skin smooth above and below, except that the region of the tympanum is granular; a glandular - fold above the ear. Green above, with black markings consisting of two spots on the snout, a cross band between the eyes, two streaks on the occiput, a W-shaped mark on the shoulders, followed by some short, irregular bands. Sides of head paler, with black bands radiating from the eye, one along the canthus rostralis, two downward and one along the auricular fold, the first three narrowly bordered with greenish white. Sides of body pale green with irregular markings. Chin and throat pale drab with small, white spots ; lower surfaces elsewhere white, in some specimens bright yellowish-green in life. Type specimen, total length 33 mm., hind-leg 59 mm., head 12 mm. Type Specimen: Museum of Zoology, University of Michigan, Catalog No. 54,528; Quebrada Viernes Santo (5,000 feet), San Lorenzo, Santa Marta Mountains, Colombia; Juiy 16, 1920; Alexander G. Ruthven, collector. Habitat: Most often found under rocks in shallow streams, occasion- ally under leaves on the bank of streams, 5,000 to 8,200 feet, San Lorenzo. Notes on Paratypes: The paratypes show little variation. In nine specimens the ear is from one-fourth to one-third the diameter of the eye and the heel reaches the anterior margin of the eye or the end of the snout when the hind limb is extended along the side. The ground color of all specimens is green, and the dorsal markings show a tendency to form longi- tudinal bands in some individuals. Remarks: This species is apparently near Eleutherodactylus surdus Boulenger, from western Ecuador. As far as shown by the description of the latter, it differs in having an exposed, although somewhat indistinct, tympanum and a much narrower interorbital space. Eleutherodactylus megalops Ruthven.4’—San Lorenzo, 5,000 to 8,000 feet; heights east of San Miguel, 6,000 to 7,000 feet (‘M. A. Carriker). Only observed among leaves on the forest floor. On San Lorenzo very common from 5,000 to 7,500 feet; above 7,500 feet found in small num- bers; not observed in the cleared area between 8,000 to 8,300 feet. It is possible that the species is debarred from the latter habitat by the more open conditions and that it reaches higher elevations in the heart of the range. (PI. XII, fig. 3.) During the day what was apparently the song of this species was heard constantly throughout June and July, but the eggs could not be discovered. 13 Occ, Papers, Museum of Zool., Univ. of Michigan, No. 309. AMPHIBIANS AND REPTILES OF SANTA MARTA 53 With little doubt, the latter are laid under the forest litter and probably in the ground, the young hatching in the adult form. During the last few days in June tiny, evidently recently hatched, young were found in numbers in the forest. The material is very confusing, for the variations in coloration and in the glandular ridges are great, but the writer is now convinced that it represents but one species. Eleutherodactylus sanctae-martae Ruthven.4*—San Lorenzo, 4,500 to 9,300 feet; heights east of San Miguel, 6,000 to 7,000 feet, and heights east of Taquina, 8,000 to 9,000 feet (M. A. Carriker). Evidently a ground form, since 10 of the 25 specimens found were among leaves on the forest floor. One specimen taken at 5,000 feet was in a bromelia that had fallen to the ground, and four were in bromelias growing on the ground. Nothing has been learned of the breeding habits of the species. Eleutherodactylus delicatus Ruthven.’—San Lorenzo, 4,500 to 8,000 feet. In 1913 three of these tiny frogs were found among leaves on the forest floor, and in 1920 two more adults were taken in bromelias. It is difficult to determine if the species is strictly a bromeliadicolous form, for it is so small as to be easily overlooked in tearing apart the plants. Eggs were discovered in the decaying leaves and litter at the base of the outside leaves of the bromelias growing on the ground at 8,000 feet, where they are easily overlooked. They are small, the membrane is tough and stained by the decaying vegetation, and the eggs are not united into masses or glued to the leaves. The young are born in the adult stage, and before hatching have a large, thin and transparent tail like that of E. crwentus. No recently laid eggs were found: six taken on June 21, 1920, hatched the same day. In one adult the hind legs are not cross-barred. Another adult indi- vidual taken in 1920 differs from the original description in having a dis- tinct tympanum, a longer hind leg (the heel reaching the nostril), in having a dark stripe from the end of the snout to the anus, and in possessing large, flat warts at the angle of the jaw. This specimen represents an extreme variation, for the recently hatched young are like this specimen, like the type, or intermediate in respect to the characters mentioned. Ceratophrys calcarata Boulenger—Fundacion. ‘Taken in the low forest and in the open country. Several heads were found in forest pools, but living specimens were only seen at night. The species is much feared by the natives, evidently because they believe that the pugnacious habits indicate noxiousness. When disturbed the enormous mouth is opened widely and the frog snaps viciously at the disturber. Individuals experimented upon would seize a finger and allow themselves to be carried about for several minutes. The specimens agree closely with the original description, except that the tympanum may be distinct or indistinct. Pleurodema brachyops (Cope)—Santa Marta; Aracataca (Rehn and Hebard). Taken at night along irrigation ditches. 1* Occ. Papers, Museum of Zoology, University of Michigan, No. 309. 15 Occ. Papers, Mus. of Zool.. Univ. of Mich., No. a3. 54 ALEXANDER G, RUTHVEN Pleurodema pusilla (Ruthven)**°—Santa Marta to Bolivar; Gaira; Fundacion; Fonseca; Arroyo de Arenas. Found about ponds and marshes at the west end of the range and about pools in the low forest at Fundacion, on the bank of the Rio Rancheria at Fonseca, and under rocks near pools in the bed of the Rio Barbacoa at Arroyo de Arenas. Dr. Boulenger has (in litt.) referred the specimens described as Palu- dicola pusilla to Paludicola. It certainly does not belong to this genus as restricted by Méhely,”” but is nearer to Pleurodema as defined by that author. The sternum has a long style terminated by a small, slightly cleft xiphisternum. Leptodactylus ‘sipliouis (Daudin).—Fundacién. Generally distributed, occurring in the clearings, about the open marshes, and in the low forest. During the day found under logs and fallen leaves, but active at night and easily taken in numbers by the aid of a headlight. The specimens show two types of coloration, some having regularly arranged transverse spots, others a broad, pale vertebral stripe. Leptodactylus pentadactylus (Laurenti).—Fundacién; Valencia; Arroyo de Arenas; Don Diego (M. A. Carriker). Generally distributed at Fundacion, being found under logs during the day and by headlighting at night. At Valencia observed both in the forest and on the savanna, in the latter habitat under logs. Taken at Arroyo de Arenas under stones at pools in the bed of the Rio Barbacoa, and at Don Diego in the forest. The specimens have been identified by Dr. Boulenger. Leptodactylus bolivianus Boulenger. —Santa Marta to Bolivar; Fun- dacion; Valle de Upar; Don Diego. On the banks of streams and ditches. Specimens were found under logs and in dense grass during the day and by headlighting at night, the species being active at night. The material has been compared with the type by Miss Joan Procter. It may be noted here that the writer cannot separate the Santa, Marta speci- mens from paratypes of Leptodactylus insularum Barbour'® from Saboga Island in the Bay of Panama. : Eupemphix pustulosus (Cope).—Santa Marta to 2,200 feet; Fundacién ; Valencia; Don Diego. On the desert at Santa Marta found about the ditches and marshes, and on the lower slope of San Lorenzo near streams in the wooded ravines. At Fundacidn found about the forest pools and under logs in the clearings and about the grass marshes. ‘The Valencia specimens were taken in the flood-plain forest and under logs on the savanna. The Don Diego specimens were collected in the low forest. This species has the form of sternum described for E. natteri by Méhely.” Bufo marinus (Linnzus).—Santa Marta to 4,500 feet; Gaira; Funda- cion; Loma Larga (observed). ‘This species doubtless occurs in all locali- ties about the mountains. On San Lorenzo it apparently does not breed above 2,200 feet, although it is not rare between 2,200 feet and 4,500 feet. 16 Occ. Papers, Mus. of Zool., Univ. of Michigan, No. 30. 17 Ann. Hist. Nat. Musei. Hung., II, 1904, p. 213. 18 Bull, Mus. Comp. Zool., 1906, 46, p. 228. 19 Ann, Hist. Nat. Muset. Hung., II, 1904, p. 217, Pl. XVIII, fig. 9. AMPHIBIANS AND REPTILES OF SANTA MARTA 55 Bufo granulosus Spix.—Santa Marta to 2,200 feet on San Lorenzo; Gaira; Fundacion; Aracataca (Rehn and Hebard); Valencia; Valle de Upar. Found in damp places (about streams, ditches, ponds, etc.), on the desert and savanna, and in the flood-plain forest at Valle de Upar. Hyla crepitans Wied.—Fundacion; Aracataca (Rehn and Hebard); Palomina and La Concepcion (W. W. Brown); Arroyo de Arenas. At Fundacion headlighted about the open ponds and marshes, both on the ground and in bushes; at Arroyo de Arenas taken in the flood-plain forest. On August 8, 1913, the species was breeding in numbers in an open marsh at Fundacion. Hyla vilsoniana Cope.—A single specimen taken in a clearing at Fun- dacion is referred to this species. : Hyla underwoodi Boulenger.—Aracataca (Rehn and Hebard); Fun- dacién. Headlighted in an open marsh at Fundacion. Of four specimens from Fundacién two are tan with four brown stripes and two are silvery with one gray stripe on each side. Hyla venulosa (Laurenti)—A large adult taken in the forest at Fun- dacion. The specimen is indistinguishable from specimens taken in British Guiana. Cryptobatrachus fuhrmanni (Peracca).—San Lorenzo, 4,000 to 5,500 feet. Found along the streams in the quebradas, the adults among rocks and under leaves, the young clinging to rocks in the streams. (PI. XII, fig. 2.) In using the name Cryptobatrachus fuhrmanni the writer does not wish to give the impression that he is convinced Noble” is in error in referring the species to Peters’ genus Hyloscirtus. At least to the satisfaction of the writer, however, Noble has clearly established neither its generic identity with Hyloscirtus bogotensis nor its family relationships. In 1913 Boulenger identified. certain specimens from the Santa Marta Mountains as a new genus of the family Leptodactylidae, and the writer described these under the name of Cryptobatrachus boulengeri. Later, in 1920, suspecting that the specimens thus described were immature individuals and males of fuhrmanni, an effort was made to secure a series of specimens, and with this material at hand it is evident that only one form is to be recognized. The writer believes that Peracca was in error in referring the form to the genus Hyla. The fact that the sacral diapophyses are not dilated separates it from most of the forms in that genus. No characters are given in the description that makes it impossible to place it in the genus Hyloscirtus, but until a comparative study can be made of Hyloscirtus bogotensis and the form under discussion the writer prefers to consider Peters’ species as one of doubtful generic affinities. Noble places fuhrmanni in the family Leptodactylidae because it has “the cylindrical diapophyses of the leptodactylids,” “the form of the sacral diapophyses is of more diagnostic value than the shape of the terminal 2¢ Bul, Amer. Mus. Nat. Hist., XX XVII, 803-807. 21 Occ. Papers, Mus. of Zool., Univ. of Michigan, No. 33. 56 ALEXANDER G. RUTHVEN phalanges,” and “claw-shaped phalanges are to be expected in the lepto- dactylids.” In the opinion of the writer, these conclusions are not convincing. Fry*? has shown of the Leptodactylidae that “the Australian members of this family exhibit all stages between that of Heleioporus with consider- ably dilated diapophyses and the [cylindrical] condition shown in South American genera.” It remains to be pointed out that not all of the neo- tropical leptodactylids have cylindrical diapophyses,* and that among the forms usually referred to the Hylidae there are considerable differences in the form of these processes, cylindrical, slightly dilated and expanded dia- pophyses being represented. It would thus seem to be clear that the form of the sacral dipophyses is not to be depended upon as a character distinguishing two families. If the form of the sacral diapophyses is not a reliable family character, then the association of a particular shape with T-shaped or claw-shaped phalanges does not indicate the diagnostic value of the form of the terminal digital phalanx. The hypothesis that the claw-shape is derived from the T-shape, and that the former is to be expected among the leptodactylids, is interesting and will possibly be found to be true, but it must be admitted that at the present time the claw-shape prevails among the hylids with dilated diapophyses, and it is only found in the leptodactylids if such forms of doubtful affinities as evanst, goeldi, and fuhrmanni are placed in this family on the basis of that unreliable character the form of the sacral diapophyses. In view of the variations in the form of the sacral processes, it would seem either that the two families should be combined or that they should be separated on the basis of the form of the terminal digital phalanx, until more data is secured. While it is uncertain whether one or more families are represented by the species now referred to the Hylidae and the Leptodactylidae, some progress may be made in the classification by revising the genera and segre- gating the species most alike in the totality of their characters. Evansz, goeldi, and fuhrmanni are alike in having claw-shaped digits, cylindrical diapophyses with somewhat dilated cartilaginous head, bell-shaped gills in the larval period,™* and in that the female carries the eggs on the back and the young are born in the adult stage. It is proposed that these forms be united under the generic name of Cryptobatrachus until the identity of Peters’ Hyloscirtus is established. The genus may be referred to the Hyli- dae until the value of the form of the terminal digital phalanx is fully under- stood. The writer is inclined to believe that the presence of cup-shaped gills in the larval stage indicates a close relationship between Gastrotheca, Cryptobatrachus, and Ceratohyla, whether or not it is believed that the 22 Proc. Roy. Soc. Queensland, XXVII, p. 73. 23 Noble has recently described a form with T-shaped phalanges and slightly dilated diapophyses (Bul. Amer. Mus. Nat. Hist., XLII, pp. 441-443). 24 Boulenger states that “no traces of gills are to be seen in the embryo of H. goeidi,” but in the description of Hyla evansi he states that the larva has “allantois- like membranous respiratory organs. In this respect it agrees with Hyla goeldi Blgr., to which it is nearly related, and with Ceratohyla bubalus Esp.” Proc. Zool. Soc. London, X, p. 209. AMPHIBIANS AND REPTILES OF SANTA MARTA 57 three genera have diverged far enough to warrant their separation into different families. Phyllomedusa tarsius (Cope).—Aracataca; Fundacion (observed). Dense woods in flood-plain forest. A single specimen is referred to this species, although it is not typical. The hind limbs are shorter than in the specimens of P. tarsius examined, the heel extending to the middle of the tympanum instead of to the anterior border of the orbit; the parotoids are indistinct, and the white markings around the vent, characteristic of P. tarsius, are lacking. Hylella pearsei, new species”° Description: ‘Tongue subcircular, slightly emarginate behind. Head broader than long; snout as long as diameter of orbit; loreal region high, oblique; interorbital space broader than upper eyelid; tympanum indistinct, about one-third the diameter of eye. Fingers webbed at base, first shorter than second; toes two-thirds webbed; disks of fingers and toes well de- veloped, those of fingers larger than tympanum. The hind limb being carried forward, the tibio-tarsal articulation reaches the posterior margin of the eye. Skin smooth above; belly and an area below the thighs coarsely granular; a dermal fold from axilla to groin. Above pale brown with numerous small, round, dark-brown spots, a poorly-defined brown band posterior to the eye, limbs without well-defined markings; ventral surfaces pale yellow. Length, head and body, 14 mm.; length of head, 5.5 mm.; width of head, 6 mm. Type Specimen: Museum of Zoology, University of Michigan, No. 54,639; Fundacion, Santa Marta Mountains, Colombia; August II, 1913; A. $. Pearse, collector. Habitat: ‘Taken in tall grass in an open marsh. Remarks: Noble (Buil. Amer. Mus. Nat. Hist., XJ.1J, pp. 441-445) has recently placed one of the species hitherto referred to Hylella (H. buck- leyi Boulenger) with the Leptodactylidae. H. pearsei has the claw-shaped terminal phalanges of the Hylids. SAURIA Gonatodes fuscus (Hallowell) —Santa Marta to about 800 feet on San Lorenzo; Fundacion; Aracataca (Rehn and Hebard); Las Pavas; Valen- cia; Valle de Upar; Fonseca; Riohacha. Very common in the desert scrub and dry forest and on walls in towns in the localities given above, apparently less common in the valley forests. Often found in logs, but generally on trees, particularly those which are partly decayed. Eggs laid in decaying wood. Gonatodes vittatus (Lichtenstein)—Don Diego (M. A. Carriker) ; Palomina (W. W. Brown) ; Arroyo de Arenas; Fonseca. A very common species on and in logs and on trees in the forests at Arroyo de Arenas and Fonseca. 25 Named for the collector of the type specimen, Dr. A. S. Pearse. 58 ALEXANDER G. RUTHVEN - Phyllodactylus ventralis O’Shaughnessy.—Santa Marta to about 100 feet, San Lorenzo; Riohacha (observed). Only five specimens taken, of which two were in buildings in Santa Marta and Bolivar, one was found under a stone on the desert at Santa Marta (another was observed in a rock slide), and two were under the bark of standing trees in the dry forest above Bolivar. The one observed at Riohacha was in a house. Thecadactylus rapicaudus (Houttuyn)—Bolivar to 600 feet, San Lorenzo; Las Pavas; Tucurinca (Rehn and Hebard); Arroyo de Arenas; Valencia; Fonseca. Found in the dry forest on San Lorenzo and in the valley forests in the other localities recorded. During the day only taken under the bark of trees; observed at night running about over the tree trunks. Lepidoblepharis intermedius Boulenger—San Lorenzo (5,000 feet). This species was described in 1915 as Pseudogonatodes furvus** just before the writer received Boulenger’s description. The type of P. furvus Ruthven has been compared with Boulenger’s material by Miss Joan Procter, and it is her opinion that they represent the same species. It should be noted that the Santa Marta Mountains specimen is much larger than the larger of the two specimens from the Choco. Lathrogecko sanctae-martae Ruthven.2"—Bolivar to 600 feet, San Lorenzo; Tucurinca (Rehn and Hebard); Fundacion; Las Pavas; Valen- cia; Fonseca; Arroyo de Arenas. The very small size and secretive habits of this lizard make it easily overlooked. All but two specimens were found on the ground under leaves and logs. Two specimens were found in a decayed stump. During the hottest part of the day they are more active and can be seen darting about among the leaves. Anolis solifer Ruthven.**—Known only from the type specimen, Cata- log No. 6,549, Museum of Comparative Zoology, taken at La Concepcion by W. W. Brown. Anolis gaigei Ruthven.*®—San Lorenzo from Bolivar to 4,000 feet; Fundacién; Tucurinca (Rehn and Hebard); Pueblo Viejo (M. A. Carri- ker) ; Pueblo Viejo and Palomina (W. W. Brown) ; Las Pavas; Valencia; Loma Larga; Arroyo de Arenas. Usually found on the ground or on grass or bushes. While this anole has been found on San Lorenzo practically throughout the dry forest and up to 4,000 feet in the wet forest, it is apparently rather rare below 600 feet and above 2,700 feet, and above 3,000 feet is restricted to artificial clearings. It was found to be abundant from 2,000 feet to 2,700 feet. It was not a common form at Fundacion, Las Pavas, Valencia, Loma Larga, and Arroyo de Arenas, but this observation means little in view of the short time spent at these places. Anolis solitarius Ruthven.*° San Lorenzo, 4,500 to 6,000 feet; Pueblo 26 Occ. Papers, Mus. of Zool., Univ. of Michigan, No. to. 27 Occ. Papers, Museum of Zoology, Univ. of Michigan, No. 21. 28 Occ. Papers, Museum of Zoology, Univ. of Michigan, No. 32, pp. 4-5. 29 Occ. Papers, Mus. of Zool., Univ. of Michigan, No. 32, pp. 6-8. 80 Occ. Papers, Mus. of Zool., Univ. of Michigan, No. 32. AMPHIBIANS AND REPTILES OF SANTA MARTA 59 Viejo (M. A. Carriker) ; Palomina (W. W. Brown). This species is dif- ficult to collect, as it frequents the heavy forest and is apparently not abun- dant. Only one specimen was taken as low as 4.500 feet on San Lorenzo; the others observed in this region were between 5,000 and 6,000 feet. In the forest on San Lorenzo about six were observed in two months’ work, two on the ground, the others on ferns and vines to a height of 20 feet on trees. Most of the specimens taken were secured in the large quebrada of Viernes Santo, in 1920, where, owing to the width of the stream, the forest canopy is broken and there is a riparian association of low shrubs and trees. In this habitat the lizard was found on the small trees and shrubs. Norops auratus (Daudin).—Santa Marta to 2,200 feet, San Lorenzo; Aracataca (Rehn and Hebard); Fundacién; Las Pavas; Valencia; Valle de Upar; Arroyo de Arenas; Palomina and Pueblo Viejo, 8,000 feet (W. W. Brown). Common in open woods and in clearings; generally found on grass and low bushes; abundant about the open marshes at Bolivar and Fundacion. Tropidodactylus onca (O'Shaughnessy ).—This species was only found at Riohacha, and in but one place—a desert flat between the town and the Rio Rancheria. It was apparently not common. All of the specimens taken (17) were on the ground. It is very shy and at the slightest cause for alarm dashes into a hole. It is to be noted that the adpressed hind limb does not reach the border of the eye, as stated by Boulenger, but the posterior border of the ear. Polychrus spurrelli Boulenger—Aracataca (Rehn and Hebard). Taken from stomach of Leptophis occidentalis. The specimen is referred to P. spurrelli with some doubt. It corre- sponds to the description of that form and differs from P. marmoratus in the scutellation of the body, but it is like the latter in having a gular denticulation. Basiliscus barbouri Ruthven.—From Santa Marta to 2,200 feet; Gaira; Fundacion; Valencia; Valle de Upar; Loma Larga; Palomina and La Con- cepcion (W. W. Brown). Apparently common in the localities in which it was found, occurring only along the streams, where it is generally found on rocks, logs, etc., near the water, occasionally in bushes up to a height of about twenty feet. When alarmed, individuals frequently dash across the water on their hind legs, and occasionally adopt this method of locomo- tion on land. (PI. IX, fig. 1.) Leiocephalus erythrogaster (Hallowell).—Bolivar to 2,000 feet; Tucu- rinca (Rehn and Hebard); Valencia. Only observed on the ground and on logs and rocks in dry forests. The species is secretive and apparently somewhat local in distribution in the localities in which it was observed. Iguana iguana (Linneus)—Santa Marta to about 600 feet, San Lorenzo; Aracataca (Rehn and Hebard) ; Fundacién; Valle de Upar; San Juan de Cesar; Fonseca and Arroyo de Arenas (observed) ; Riohacha. Where there are deep ponds or streams (¢. g., Fundacion and Valle de 60 ALEXANDER G. RUTHVEN Upar) this lizard is generally abundant on the shores, frequenting the branches overhanging the water, from which they throw themselves when alarmed. In the regions where the streams are shallow the species is appar- ently less abundant, does not have this habit, and, while occurring along the streams, seems to be more generally distributed. At Riohacha old and young individuals were found in the desert scrub at a distance of a mile from water, in this habitat running about on the ground like the large Ameivas. Tupinambis nigropunctatus Spix.—One specimen from Don Diego (M. A. Carriker). Reported by Mr. Carriker to be not uncommon in the cacao plantation at Don Diego. It does not seem to be recorded that this lizard is a scavenger. The natives on the Demerara River, British Guiana, cap- ture them on fish-hooks baited with flesh. Ameiva ameiva maculata (Fischer)—Santa Marta to Agua Dulce; Las Pavas (observed) ; Valencia; Valle de Upar; Fonseca; Loma Larga; Arroyo de Arenas; Riohacha. On San Lorenzo found on the desert and in dry woods to 2,200 feet, above 2,200 feet only in the clearings. In the other localities taken in the flood-plain forests. This form is close to Ameiva ameiva, as stated by Barbour and Noble,” and the large series obtained shows that most of the scale characters given by these writers are not sufficient to distinguish the subspecies. The pre- anals are usually as much enlarged as in the typical form, and, while the shields on the under side of the tibia are usually a little smaller and more numerous in maculata, the range of variation in these scales is slight and overlapping in the two forms. ‘The best scale character is the width of the occipital plate, which is quite constantly much less in ameiva, being usually less than one-half the greatest width of the adjacent scale in ameiva and with few exceptions from two-thirds to the entire width of the adjoining scale in maculata. Barbour and Noble state that ameiva has two pairs of frontoparietals, but this must be an error, since in a large series two pairs occur only as an abnormality, and this abnormality is also found in maculata. Ameiwwa bifrontata divisus (Fischer).—Salamanca Coast and Santa Marta to 2,200 feet; Aracataca and Tucurinca (Rehn and Hebard); Valle de Upar; Fundacién. A common ground lizard in the localities mentioned. The subspecies A. b. divisus*® exists in the typical form from the coast at Santa Marta to Fundacion. In a large series of specimens not one has the row of granules between the supraoculars and frontals continued ante- rior to the anterior border of the third supraocular. Of a series of eight specimens taken at Valle de Upar six are like the Santa Marta specimens and two have one granule anterior to the suture of the second and third supraocular. (See p. 62.) No representatives of the species were found between Fundacion and Valle de Upar. It is possible that the Valle de Upar form will be found to be closer to typical bifrontata than to the sub- species divisus or intermediate, but this is not apparent from the material at hand. Also the writer is not certain of the identity of the Don Diego 31 Bul. Mus. Comp. Zool., LAV, p. 467. 32 See Ruthven, Occ. Papers, Mus. of Zool., Univ. of Michigan, No. 2, 1913. AMPHIBIANS AND REPTILES OF SANTA MARTA 65 specimens. Of the two specimens in the collection from that locality, one is like divisus and the other is bifrontata. Since both types of scalation are found at Riohacha, the specimens are provisionally identified as bifrontata. For remarks on the relationship of the two forms see infra. Ameiva bifrontata Cope.—Badillo; Fonseca, Arroyo de Arenas; Rio- hacha: Don Diego. Very common about Riohacha and the other localities given. A ground form found both in the woods and on the desert. One individual was seen to run on its hind legs. At Riohacha about one-half the specimens taken (21 out of 50) have the first and second supraoculars entirely separated by granules. In 25 the granules between the frontal and supraoculars extend farther forward than the suture between the second and third supraoculars, and in four the granules end at that suture as in divisus. (See p. 62.) These differences are not sexual. Whether these specimens should be referred to bifroutata, as is done here, or are to be considered intermediate, can be determined only by the examination of a large amount of Venezuelan material to dis- cover just how variable is bifrontata. From a very small series of the typical form from Venezuela the writer believes that it does not constantly have the first and second supraoculars separated. The status of the material from Don Diego, Arroyo de Arenas, Fonseca, and Badillo is not clear, but since one of two specimens from Don Diego and one of three specimens from Fonseca are typical bifrontata, while the two taken at Arroyo de Arenas, one from Fonseca and one from Badillo, have the inner row of granules continued farther than the suture of the second and third supra- oculars, the logical conclusion is that the lizards in these localities are closer’ to bifrontata than divisus. There can be little doubt that the two forms intergrade in the Santa Marta Mountains region, whether the lizards at the east end of the range are typical bifrontata or intermediate. The distribution of characters is shown in the following table. Distribution of the characters of Ameiva bifrontata and Ameiva bifrontata divisus in the Santa Marta Mountains region. Beyond To suture of suture of Entirely Superior granules, supraoculars supraoculars around last 3 continued anteriorly: 2and 3 2and 3. supraoculars Subspecies Santa Marta . . : 9) as PunCACiOnN 2 « I=... 9 Ametva b. divisus Valle de Upar . 6 2 Eccitlates Sots a Lee I I Mest amncera tee. dais a: I I I Arroyo de Arenas... . 2 Ametva bifrontata- iohacha’. 3. 2 5 4 25 ay Won Dieeo 25... « I I Cnemidophorus lemniscatus gaigei Ruthven.**—Salamanca Coast, Gaira (observed), and Santa Marta to 2,200 feet; Tucurinca and Aracataca (Rehn and Hebard) ; Fundacién; Las Pavas (observed) ; Valencia; Valle 38 Occ. Papers, Mus. of Zool., Univ. of Mich., No. 16. 62 ALEXANDER G. RUTHVEN sory B2 2 O o LEA e Saws ANH 6. BY nis CGI Riohacha Head scalation of Ameiva bifrontata, A. bifrontata divisus and intermediate specimens in the Santa Marta region. AMPHIBIANS AND REPTILES OF SANTA Marra 63 de Upar; Badillo; Fonseca; Loma Larga; Arroyo de Arenas Riohacha ; Palomina and La Concepcion (W. W. Brown) ; Don Diego (M. A. Carri- ker). Common on the lowlands all about the range, preferring open habi- tats, but not uncommon in the more open parts of the flood-plain forests at Valencia and Valle de Upar. Although a ground form, it readily climbs about in low bushes in search of food. Leposoma dispar Peters——Bolivar to 600 feet; Las Pavas; Tucurinca (Rehn and Hebard). Found among leaves in the dry forest, except that two specimens were in damp leaves in a marshy place on the bank of the Tamocal River. Loxopholis rugiceps Cope—Fundacion ; Las Pavas; Valencia. At Fun- dacién found in some numbers under logs and in the grass about an open marsh and along the river; at Valencia two specimens found among leaves and under detached bark in the valley forest. The males alone have pores, five on each thigh, and usually four, some- times 5, in the preeanal region. The chin shields are constantly one anterior and four pairs, the members of the first two pairs being in contact. The color of the males is richer (the dorsal parts blackish, the under surfaces more or less red), the females being brownish above and whitish below. Tretioscincus bifasciatus (Dumeril).—Bolivar to 2,000 feet; Tucurinca (Rehn and Hebard); Fundacién; Las Pavas; Valencia; Valle de Upar; Fonseca; Arroyo de Arenas; Don Diego. A common form both in the dry and valley forests. Found among fallen leaves, on and in logs, and on the trees. In many specimens the color of the tail is a brilliant blue. Bachia bicolor (Cope).—Bolivar to 600 feet; Tucurinca (Rehn and Hebard) ; Loma Larga; Palomina (W. W. Brown). Under logs and rocks in the dry forest. There are two species of Bachia in the Santa Marta Mountains region, one of which is the same as Heterodonium bicolor. The specimens taken in the localities given above have the scales in 28 (29) annuli, and have from 44 to 46 scales between the oceiput and the tail. The hind limbs are undivided and are decidedly larger than in the specimens which have been referred to the following form. ‘There are four digital tubercles on the fore limbs in all specimens. It is possible that this species is synonymous with B. flavescens, but if so the definition of the latter must be changed, for the Santa Marta Mountains specimens have fewer scales than given by Boulenger for favescens, and the dorsal scales are distinctly imbricate. Bachia dorbignyi (Dumeril and Bibron).—Valencia; Valle de Upar; Fonseca. Found under logs in the dry forest at Fonseca and in dry parts of the flood-plain forests at Valle de Upar and Valencia. The species is a very active one. The specimens referred to this species resemble those identified as B. bicolor, but differ in having the scales in 26 annuli, in having 46 to 48 scales from the occiput to the base of the tail, in having the hind limb more reduced, and in having but three digital tubercles on the fore limbs. It will 64 ALEXANDER G. RUTHVEN be noted that the scale counts are those of B. tridactylus, but it is clearly not that form, since the dorsal scales are overlapping and there is no inter- parietal plate. Gymnophthalmus sumichrasti (Cope).—Bolivar to 600 feet ; Tucurinca (Rehn and Hebard) ; Valencia; Badillo. The specimens, five in number, were taken on the ground under leaves, except that’ one of two specimens captured at Bolivar was found in the grass ina pasture. They are secretive and very difficult to capture, for they are mostly active only when the sun is shining, or at least are best seen then, and are usually observed slipping about among the leaves. The scutellation does not conform to the description given by Boulen- 34 ger** as shown by the following table: Longitudinal Scale between Scalesat Scales on poste- Locality—Sex scale rows headandtail base of tal rior fourth of tail Bolivar, male 4) 3.) 13 34 smooth keeled Bolivar, female... 15 38 keeled Walencia, male. . 2. 13, 15 34 smooth Valencia, male. ... . 15 34 smooth keeled Badillo, female .. . 13,515 38 smooth It will be seen that the numbers of.scales on the median dorsal line are those of G. quadrilineatus, and that the number of longitudinal scale rows may be either 13 or 15 or intermediate. In the specimens with both 13 and 15 rows an extra row is added on the sides about half-way between the fore and hind limbs. In four specimens the scales on the base of the tail are smooth as given for G. quadrilinedtus, but in one specimen they are dis- tinctly keeled. In the two specimens in which the tail is intact the scales of the distal fourth of this appendage are keeled. The coloration is as described by Boulenger, except that the ventral scales may bear round, black spots. Amphisbaena fuliginosa Linneus.—A single specimen taken on San Lorenzo at an altitude of 4,000 feet, crawling along the ground in the forest at daybreak. A second individual was seen in the same locality. Mabuya agilis (Raddi)—Don Diego; Arroyo de Arenas; Valencia; Las Pavas. Only found in regions of dense forest, but there taken in open as well as shaded places. SERPENTES Helminthophis petersii Boulenger—Fundaci6n; Valencia. Taken under logs in dry forest at Fundacion and in the flood-plain forest at Valencia. The two specimens secured have 22 rows of scales. Dr. Boulenger has advised me that he was in error in ascribing 20 rows of scales to the species, the type specimen having 22 rows. Leptotyphlops macrolepis (Peters).—A single, slightly mutilated speci- men found dead in the dry woods at Bolivar is referred to this species. It differs from the description given by Boulenger®® in that the second labial reaches the eye, but it differs from L. myopica in the proportions and in 34 Catalog of Lizards, British Museum, II, p. 428. 35 Catalogue of Snakes, British Museum, I, p. 60. AMPHIBIANS AND REPTILES OF SANTA MARTA 65 having six lower labials. The diameter of the body is contained 41% times in the length, the length of the tail 13 2/3 times. Epicrates cenchria (Linneus).—Fundacion. A single specimen taken near Fundacién was secured from a native. Constrictor constrictor (Linneus).—San Lorenzo, 2,200 feet. Taken in the forest. Reported as frequently seen at an elevation of about 2,200 feet and lower. The characters given below are those of C. constrictor, not C. impera- tor. Dorsal scale rows 81, supraoculars 24, ventrals 247, subcaudals 55, rows of scales across forehead 18, dark cross-bands 20. Drymobius boddaertii (Sentzen)—Two hundred feet above Bolivar (observed) to 5,500 feet, San Lorenzo; Valencia; Palomina and La Con- cepcion (W. W. Brown). A common ground snake in the wet forest on San Lorenzo; generally distributed above 2,200 feet, below this altitude only seen in ravines in the dry forest. At Valencia found in the flood-plain forest. The only food found in the stomachs examined was frogs. (PI. Will, fig: 2.) Drymobius rhombifer (Ginther).—San Lorenzo from 3,000 feet to 5,500 feet; Don Diego (M. A. Carriker); Palomina and La Concepcion (W. W. Brown). A common ground snake in the wet forest. One speci- men was taken from an egg in the stomach of a Lachesis lanceolatus. The stomachs examined contained frogs, Prostherapis subpunctatus, and Eleu- therodactylus sp. (PI. X, fig. 2.) The differences in the number of subcaudals is great, the extremes being 84 and 103. Phrynonax poecilonotus (Gtnther).—San Lorenzo, 4,500 feet; Don Diego (M. W. Carriker). The specimen taken on San Lorenzo was ina coffee tree; the one from Don Diego was found on the ground in a dense cacao grove. The San Lorenzo specimen had eaten a partly fledged bird. Peracca®* has shown that the characters used to distinguish P. poecilo- notus, lunulatus, fasciatus, and guentheri are not sufficient. The Santa Marta Mountains specimens are like those hitherto referred to P. fasciatus in that the dorsal scales are weakly keeled, the preocular is in contact or narrowly separated from the frontal, and the upper labials are eight. The dorsal scale formula is 21-19-17-15-13 and 21-23-21-19-1'7-15-13-(11). CEL. UX, figs 2.) Drymarchon corias melanurus (Dumeril and Bibron).—San Lorenzo, 600 feet to 4,000 feet; La Concepcion, Palomina, and “Macotama or San Miguel, 6,000-8,000 feet” (W. W. Brown). On San Lorenzo found in the wet forest, and although but few specimens were seen, the number of cast skins would indicate that the species occurs in some numbers. Chironius carinatus (Linnzeus).*7—Santa Marta; Palomina and La 36 Bol. Mus. Torino, No. 253, pp. 6-8. *7 Kitzinger Syst. Rept., pp. 29 and 31, specifically designates Linne’s Coluber carinatus as the type of his genus Chironius. Dr. Leonhard Stejneger (7m litt.) has called my attention to the fact that C. carinatus has Coluber chironius Donndorf, Zool. Beytr., vol. 3, p. 209, as a synonym and is consequently type by tautonymy. 66 ALEXANDER G. RUTHVEN Concepcion (W. W. Brown); Valle de Upar (observed). Of the two specimens seen one was in a river, the other on the bank of a stream. Leptophis occidentalis (Giinther)—One specimen from Aracataca (Rehn and Hebard). In the stomach of this specimen was found the lizard referred to Polychrus spurrelli. Leimadophis melanotus (Shaw ).—Bolivar to 5,500 feet, San Lorenzo; Fundacion; Arigauni; Valencia; Arroyo de Arenas; San Miguel, 5,000 feet (M. A. Carriker). Taken on the edge of marshes and ditches at Bolivar, in the wet and dry forests in other localities on the lowland, in the forest at 5,500 feet, under the bark of a stump at Fundacién (one), and on the savanna at San Miguel. One specimen had eaten a frog. Leimadophis albiventris Jan.—Pueblo Viejo, 8,000 feet (W. W. Brown). A single specimen is referred to this species. Lampropeltis nicropholis Cope-——The writer has examined the speci- men from Cacagualito, Colombia, recorded by Griffen** and examined and referred to this species by Dr. F. N. Blanchard. Leptocalamus torquatus Ginther.—A single specimen taken in a log in the dry forest at Valencia. The specimen conforms to the description of the type in having a broad nuchal collar which was yellow in life. The head and the region just behind the collar are dark brown, the rest of the dorsal surface a brownish olive, the ventral surface pearly white. The portion of the rostral visible from above is about one-third of its distance from the frontal. Atractus irridescens Peracca.~—San Lorenzo, 4,000 to 6,500 feet; San Sebastian (W. W. Brown). Apparently not uncommon but a secretive form between 4,000 and 6,500 feet; usually found under logs and stones. One specimen has 15 dorsal scale rows, and 9 have 17 rows. The upper labials are 7 in every specimen; the lower labials are 7 in 4 speci- mens, 7 and 8 in four, 8 in one, and 6-7 in one. The ventrals vary from 152 to 169, the subcaudals from 22 to 34. The coloration is very variable. The dorsal coloration in some specimens is brownish black with small trans- verse pale brown spots, in others dark brown with small, obscure black spots, and others (young) are red with prominent black spots and cross bars. A large specimen from San Sebastian is olive brown with black spots and cross bars. The ventral surface may be immaculate white or dull white irregularly spotted with black or with two broad black bands. The irridescence is strongly marked in the young and medium-sized specimens. The writer is not sure of the identity of the specimens referred to, particularly of those from San Sebastian, but has given them this name as they conform to the descriptions of Peracca*® and Boulenger.*® The varia- tions of the species in this genus have not been determined, and the char- acters used to distinguish the forms have evidently not been properly evaluated. 88 Mem. Carnegie Mus., VII, p. 176. 39 Bol. Mus. Torino, No. 252, 1806. 49 Proc. Zool. Soc. London, 1913, 1035. AMPHIBIANS AND REPTILES OF SANTA MARTA 67 Imantodes cenchoa (Linnzus).—A single specimen from Aracataca (Rehn and Hebard). Leptodeira albofusca (Lacépéde).—Bolivar; Fundacidn; Fonseca. Found near irrigation ditch at Bolivar, about a marsh at Fundacién, and in the flood-plain forest at Fonseca. A nocturnal species usually found under logs, under the scales of palms, etc., during the day. One specimen had eaten a frog. In six specimens the dorsal scale formula is I9-2I-19-17-15, and in three it is 21-19-17. For remarks on the relations of this form and L. annulata see infra. Leptodeira annulata (Linnzus).—Arroyo de Arenas. A single speci- men taken at night on a tree trunk in the flood-plain forest. The dorsal scale row formula is 19-17-15. The ground color is brown- ish white with numerous broad, black-edged, brownish spots about one and one-half scales apart descending to the first row of scales. After examining a series of specimens from South America, Griffen** has arrived at the conclusion that 1. annulata is a synonym of L. albofusca. The writer does not concur in this opinion. It is true that many of the characters do not, because of overlapping, indicate two forms; but in the number of dorsal scale rows there is an apparent difference at least which is not brought out in Griffen’s counts. When two species are as much alike as these two, the maximum number of scale rows means little. It is the average for the entire body that is significant. In ZL. albofusca the scales are usually 19-2I-19-17-15 or 21-19-17 in the females and usually 19-17- 15-(13) in the males. From the number of specimens of L. annulata recorded with 19 rows the usual formula is evidently 19-17-15. The two forms can be considered identical in the dorsal lepidosis only if it is revealed after a study of large series that the scale formulae are the same. Clelia petola (Linnzus).—Don Diego (M. A. Carriker) ; La Concep- cion (W. W. Brown). The specimen taken at Don Diego was on the ground in a cacao plantation. In all of the specimens the dorsal coloration is principally red, the tip of each scale is tipped with black, and the black cross bars are present only on the anterior one-third of the body and distinct only on the anterior one-fourth. Oxybelis acuminatus (Wied).—Bolivar to 800 feet, San Lorenzo; Fundacion; Arroyo de Arenas; “Macotama or San Miguel, 6,000-8,000 feet” (W. W. Brown). Common in trees and bushes at Bolivar, and taken in the same habitat at Fundacion and Arroyo de Arenas. Known locally as the bejuca snake. (PI. X, fig. 1.) The scale formula is 17-15-13, 15-17-15-13, and I5-17-15-13-II. Erythrolamprus aesculapu (Linnzus)—San Lorenzo from 4,500 to 6,000 feet; Palomina, La Concepcion, Pueblo Viejo at 8,000 feet (W. W. Brown). A common ground form in the wet forest. A specimen taken on San Lorenzo had eaten two small individuals of Atractus irridescens. 41 4nn. Carnegie Mus., XI, pp. 321-326. 68 ALEXANDER G. RUTHVEN Another had eaten two frogs, Eleutherodactylus cruentus and Geobatrachus qwalkert, All of the specimens have black annuli in pairs, the number on the body varying from 11 to 14. The subcaudals are 44 to 64, the ventrals 186 to 200. Tantilla longifrontale (Boulenger).—San Lorenzo, 4,000 feet; Palo- mina (W. W. Brown). The San Lorenzo specimen was taken on a rock in a stream. Tantilla semicinctum (Dumeril and Bibron).—Valle de Upar; Bonda (a specimen in the Museum of Comparative Zoology, from the H. H. Smith collection). The Valle de Upar specimen was in dry woods. Tantilla melanocephala (Linneus)—Fundacion; Valencia. Two specimens taken under logs, one in a swamp, and one in low forest. Stenorhina degenhardtui (Berthold).—Valencia; Arroyo de Arenas. Only two specimens taken, both on the ground in dense woods. The specimens have the color described as variety A by Boulenger,** and there is a broad, irregular black band on the median ventral line. Micrurus mipartitus (Dumeril and Bibron)—San Lorenzo, 5,000 feet; San Sebastian (W. W. Brown). The single specimen found on San Lorenzo was among leaves in heavy forest. The specimens (6) all have more than the maximum number of white rings given by Boulenger,** the number being from 69 to 75. Micrurus dumerilit (Jan) —Macotama and La Concepcion (five speci- mens collected by W. W. Brown) ; Fundacion (1). The Fundacion speci- men was found under a log on the margin of an open swamp. The material is puzzling. ‘The specimens correspond in many ways with the descriptions of Jan and Boulenger, and, as these writers had but one specimen each, it is to be expected that a series, such as is at hand from the Santa Marta region, will reveal variations. The only constant differences observed are in the size of the eye, which is three-fourths, not two-thirds, of its distance from the mouth, and in the relative length of the frontal and parietal scales, the former being shorter than the latter. The Fundacion specimen is in coloration similar to those of Jan and Bou- lenger, and the larger one of Werner.** In the others the black bands are about half as wide as in those mentioned, the laterals are poorly defined and are often, and the middle ones occasionally, interrupted ventrally. This coloration is apparently represented by Werner’s smaller specimen. In some specimens from La Concepcion the postnasal is broadly in contact with the preocular, in others the suture is short, and in still others these scales are completely separated. The triads on the body vary from 9 to 14, the ventrals are 180-200, and the subcaudals are 31 to 50. It is the opinion of the writer that all of these specimens must for the 42 Catalogue of Snakes, British Museum, III, p. 230. 43 Catalogwe of Snakes, British Museum, III, p. 431. 44 Abh. Konig. Bayerischen Akad., Bd. XXII, pp. 382-383. AMPHIBIANS AND REPTILES OF SANTA MARTA 69 present be referred to M. dumerilu (Jan), of which M. colombianus Grif- fen* is a synonym. Micrurus hollandi (Griffen)—There is a single specimen in the col- lection of the Museum of Comparative Zoology, labelled “Santa Marta, Colombia, lowlands, Dr. Sellards,’ which corresponds closely with the description of M. hollandi Griffen (loc. cit., pp. 218-219). The head is entirely black except for two yellow spots on the parietals. Sibynomorphus mtkani (Schlegel) —La Concepcion (W. W. Brown). Bothrops lanceolatus (Lacépéde).—La Tigrera (600 feet) to 5,000 feet, San Lorenzo; Fundacion; Palomina (W. W. Brown). A nocturnal form said to be not uncommon up to 4,500 feet. Only one specimen observed as high as 5,000 feet on San Lorenzo. In the stomach of one specimen was found an egg of Drymobius rhombifer. In all of the specimens the keels of the dorsal scales are nearly or quite as long as the scales. Bothrops lansbergi (Schlegel).—Bolivar; Fonseca. Among leaves in dry forest at Bolivar, and under log in low woods at Fonseca. The dorsal scale formula is 25-23-21I-19, and in one specimen, a male, there are 25 rows only on the neck. Crotalus terrificus (Laurenti).—San Sebastian (W. W. Brown) ; Boli- var (reported by M. A. Carriker). The dorsal scale formula is 27-29-27-25-23-21-19. CROCODILINI Crocodilus acutus Cuvier.—One specimen on the ocean beach at Don Diego (M. A. Carriker) ; one seen on the bank of the Fundacion River at Fundacion. Caiman sclerops (Schneider).—A single specimen taken in a stagnant lagoon in the low forest at Don Diego (M. A. Carriker). TESTUDINATA Testudo denticulata (Linnzeus).—Santa Marta (reported) to 600 feet, San Lorenzo; Fundacién; Copei. The species is much more widely dis- tributed than indicated by the above list of localities. It apparently prefers dry forests and probably occurs in this habitat everywhere on the west, south, and east sides of the range. Shells were observed between Arroyo de Arenas, Fonseca, San Juan de Cesar, Valle de Upar, Valencia, and Las Pavas. The species is assiduously hunted for food, and the shells are utilized as basins. Kinosternon integrum (Le Conte).—A single specimen taken in an open marsh at Fundacion. 45 Mem. Carnegie Museum, VII, 216-217. ALEXANDER G. RUTHVEN PLATE 1 Figure 1. Harbor and city of Santa Marta as seen from the entrance to the bay. Figure 2. Vegetation of the plains near Santa Marta. The vegetation of scraggly trees, cacti and other plants in some places tends to grow in clumps; in others it is more evenly distributed. Santa Marta MovunTAINsS PLATE I ALEXANDER G. RUTHVEN PLATE I Figure 1. Wet forest at 4,500 feet on San Lorenzo. Figure 2. Moderately wet forest below 4,500 feet on San Lorenzo. The abrupt transition to the dry forest at 2,200 feet may be seen at the low peak just to the left of the center of the picture. Santa Marta Mountains PiaTe II COCKAYNE BOSTON ALEXANDER G. RUTHVEN PLATE III Figure 1. Dry forest, at 1,500 feet, on San Lorenzo. The trees have been largely removed and are replaced by an artificial dry savanna. Figure 2. Detail of the moderately wet forest, at 4,000 feet, on San I,orenzo, showing the nature of the undergrowth. II] 42V1d SNIVLNONOJ VLUVI, VINVS sa 8 - yar at » Mil 3 i page alt A on fe aie ee eT Ct my ALEXANDER G. RUTHVEN PLATE IV The moderately wet forest, at 4,000 feet, on San Lorenzo, illustrating the size and nature of the trees. The wild fig is conspicuous in both pictures. Santa Marta Mountains “> er tt: srt a = Di sa i“ : wt BL « wa, >A S r if ey) ALEXANDER G. RuruveNn PLATE V Figure. 1. A stream above 2.200 feet on San Lorenzo. The valleys are V-shaped, and the stream beds are very steep, so that the streams are torrential and character- ized by many falls. Figure 2. View in the lower part of the cloud forest on San Lorenzo, showing the abundance of bromelias on the trees. SANTA Marta MowuntTaIns PLATE V » ee) is Be eS ‘ il ita hod j * \ ‘ re ered +) pte Nea i ¢ a 7 7 ro a e : _ ais 2 \ F a » » - e Pep *% * a ~~ s . mf ae ‘ ‘ a a oe 4 Soret j - of =. =) y h ig Pa » LTT Fil ‘A al ee sce shed - ait a yas wa ae | oe : py “ee er OR RY Ter i ee ee eee ne a ere ee ee oe te ee re ee re ; ‘ - aA «ae T EX ss ite ~ cee Ns 3 : : ae ) ; Ce ALEXANDER G. RUTHVEN PLATE VI A burned area near the summit of San Lorenzo, the “Cerro Quemado.” The dominant plant is a gigantic bromelia. ale od rr” oad RS 8.7 LA Awd SNIVLINAOJ VLUVI, VINVS yee ee (ee ee ee T - j ALEXANDER G. RuTuVEN PLATE Vid Vegetation of the region about Riohacha. Over large areas the giant cacti and scraggly trees are mostly confined to mounds higher than the surrounding flats, but on the ridges and terraces with gravelly or stony soil, the dry forest is more evenly distributed. The river valley has an open forest of large trees. Santa Marta MountTAINns PLaTE VII COCKAYNE BOSTON ALEXANDER G. RUTHVEN PLATE VIII Figure 1. Savanna at Valencia de Jesus. Between the clumps of trees and cacti the vegetation consists principally of grasses. Figure 2. Drymobius boddaertii (Sentzen). SanTA Marta Mountains PLATE VIIT COCKAYNE BOSTON ALEXANDER G. RuUTHVEN PLATE Ex Figure 1. Basiliscus barbouri Ruthven; photographed in a stream on San Lorenzo. Figure 2. Phrynonax poecilonotus (Ginther), taken on San Lorenzo. This heavy-bodied snake climbs about in trees by using the tail as a prehensile organ. Santa Marta MowntTaAINns PLATE = Sie jas SB a man a A a a a ALEXANDER G. RUTHVEN PLATE X Figure 1. Oxybelis acuminata (Wied), taken near Santa Marta. An arboreal species which has such a slender body that it can move directly through the branches of the trees at an astonishing speed. Figure 2. Drymobius rhombifer (Gunther). San Lorenzo. Santa Marta MountTaIns PLATE X ft at: | Ue uy i . cP raat eA, ay vi y ae qinila oe eel . ab i fl , ly a VEX te if fe 7 ‘) 7 rf i a! Vise). ALEXANDER G. RUTHVEN PLATE XI Figure 1. Eleutherodactylus carmelitae, new species, x1%. Figure 2. Gcobatrachus walkeri Ruthven. About 2 1-5 times natural size. Figure 3. Hypopachus pearsei Ruthven. Santa Marta MovwuntTAINs PLaTE XI a ae Nn eri x. COCKAYNE BOSTON ALEXANDER G. RuTHVEN PLATE XII Figure 1. Eleutherodactylus carmelitae, new species. Figure 2. Cryptobatrachus fuhrmanni (Peracca). Female with eggs in position. San Lorenzo. Figure 3. Eleutherodactylus megalops Ruthven. San Lorenzo. The large eyes are well shown in this picture. Figures 4, 5. Eleutherodactylus cruentus (Peters). Adult and eggs. The eggs are in position on the leaf of a bromelia. Santa Marta MountTains Piate XII COCKAYNE BOSTON ¥e ist 7, (es a) a ' 1 * .atlhy ll TF ua eee eran (Pie |) ~ ‘ 4 ; ©? ware a vee ’ ¢ Arroya de Arenas 7 N faca R ao F Arace, SOUTH AMERICA Fram a Survey by MTF A.ASimons : ih om if a ae er Ag fy UNIVERSITY OF MICHIGAN MUSEUM OF ZOOLOGY Miscellaneous Publications No. 9 Notes on American Species of Triacanthagyna and Gynacantha BY E. B. WILLIAMSON ANN ARBOR, MICHIGAN PUBLISHED BY THE UNIVERSITY PULMe; 19228 ADVERTISEMENT The publications of the Museum of Zoology, University of Michigan, consist of two series—the Occasional Papers and the Miscellaneous Publi- cations. Both series were founded by Dr. Bryant Walker, Mr. Bradshaw H. Swales and Dr. W. W. Newcomb. The Occasional Papers, publication of which was begun in 1913, serve as a medium for the publication of brief original papers based principally upon the collections in the Museum. The papers are issued separately to libraries and specialists, and, when a sufficient number of pages have been printed to make a volume, a title page, table of contents and index are sup- plied to libraries and individuals on the mailing list for the entire series. The Miscellaneous Publications include papers on field and museum technique, monographic studies and other papers not within the scope of the Occasional Papers. ‘The papers are published separately, and, as it is not intended that they shall be grouped into volumes, each number has a title page and table of contents. ALEXANDER G. RUTHVEN, Director of the Museum of Zoology, University of Michigan. NOTES ON AMERICAN SPECIES OF TRIACANTHAGYNA AND GYNACANTHA (ODONATA) By E. B. WiLLIAMSON During the past seventeen years many dragonflies from tropical America have been added to my collection. From time to time, as material in certain groups became of sufficient interest, reports on this material have been pub- lished by others as well as myself. In this paper it is proposed to study the specimens of the genera indicated in the title. During our field trips no effort has been spared to make our collections of these striking insects as complete as possible. These various trips may be briefly summarized as follows. I collected in Guatemala and Honduras in 1905 and 1909. B. J. Rainey, my father, L. A. Williamson and I collected in British and Dutch Guiana and in Trini- dad in 1912. In 1916-1917, Jesse H. Williamson and I, as members of the University of Michigan—Williamson Expedition to Colombia, collected in that country. In 1920, H. B. Baker, W. H. Ditzler, Jesse H. Williamson and I were the members of a collecting party for the Museum of Zoology, Uni- versity of Michigan, in western Venezuela. The last three named devoted most of their time to collecting dragonflies, and Mr. Baker, who was more interested in other groups, brought in some specimens which are properly credited to him in each case. Jesse H. Williamson collected in Peru in 1920, and in Florida in 1921. For all the assistance I have had, both from the members of the various parties and from the University of Michigan, I wish to express my deepest thanks. Mr. Rainey, Mr. Ditzler and Messrs L. A. and J. H. Williamson not only joined these expeditions at their own expense but gave all their time and strength in the field to collecting dragonflies, and their efforts and coop- eration were the largest factor in the success and pleasure of these collecting trips. The Venezuelan collection especially was so rich in aeshnines of the two genera Triacanthagyna and Gynacantha, embracing as it did no less than twelve species, that it was felt the material on hand from all the trips could be profitably reported on. This study was begun with the idea of making such a report but early difficulties in the determination of certain specimens compelled me to borrow material from other collections and to study carefully all the species in the New World. Acknowledgements. Through Dr. Calvert I borrowed valuable material from the Philadelphia Academy of Natural Sciences (abbreviated as A. N. S. in the text), and from him, with the kind consent of Mr. Henshaw, I also obtained material belonging to the Museum of Comparative Zoology (ab- breviated as M. C. Z. in the text) which Dr. Calvert had borrowed for study. 4 E. B. WILLIAMSON Dr. Calvert also turned over to me his notes and keys on American species of these genera, loaned me some otherwise unavailable literature, and has given me valuable advice and criticism. Through Mr. Herbert Campion I have been able to study one of the males in the British Museum, determined by Kirby as G. subviridis. Through Miss Bertha P. Currie, I have obtained the material of these two genera in the United States National Museum (abbreviated as U. S. N. M. in the text). Mr. W. T. Davis (abbreviated W. T. D. in the text) has kindly loaned me all his material in these genera. In the same way, Mr. Hugo Kahl of the Carnegie Museum (abbreviated Carn. Mus. in the text) has loaned me all the specimens of these genera in the collections at Pittsburgh. Through Dr. C. H. Kennedy I have obtained material belonging to the Ohio State University (abbreviated O. S. U. in the text). And Professor Needham has loaned me the specimens in the collections at Cornell. To Dr. Ris I am indebted not only for the loan of material without which this paper would be much more incomplete than it is, but he gave me permission to describe one species which he had already detected as new in his collection, and he also kindly loaned me some litera- ture and has furnished me with various notes. Mons. Severin kindly sent me Martin’s types of G. satyrus, and the specimens labelled trifda in de Sely’s collection. ‘These specimens are discussed in detail on pages 8-11. A review of the pages which follow will show how incomplete my study would have been without the unselfish cooperation of all these friends. Dr. Kennedy made all the drawings for this paper, and Miss Mina L. Winslow all the photographs of the wings. THE GENERA TRIACANTHAGYNA AND GYNACANTHA I have not given the question whether the American species discussed in this paper should be included in one, two, or three genera any particular study and I have no well founded opinion in the matter. My knowledge of related species in other regions than the American is very limited, and, of the American species aratrix, for which Forster erected Selysiophlebia, I have seen no specimens. However, in the key to species which follows I have indicated at least some characters by which two groups, previously recognized by other authors, may be rather satisfactorily defined, and I am considering these groups as genera. As to the relationships of these two groups with old world groups, as I have stated, I have no opinion. In Am- erica I think two genera exist, regardless of their world relationships and the names by which they should be known. And Selysiophlebia remains to be studied. A factor that seems to me to have weight in this decision is the fact that in Triacanthagyna two groups have developed, somewhat paral- leling groups in Gynacantha. In Triacanthagyna we have the paler septima with its unconstricted abdomen and the darker brighter trifida group with the male abdomens constricted. In Gynacantha, somewhat resembling the septima group of Triacanthagyna, we have several species of the nervosa group, while a number of darker and more brilliant species may represent the trifida group. At the same time it is not implied that Triacanthagyna has the wealth of specific adaptions which we find in Gynacantha. The two TRIACANTHAGYNA AND GyYNACANTHA 5 genera may be recognized by the following characters: Upper piece of the arculus equal to or larger than the lower piece, Rs forked, separated from Rspl by at least three cells; Sc ending at the nodus ; median space free ; mem- branule of hind wings reduced, with only a slight extension on the anal margin; eyes in contact for a greater distance than the antero-posterior dimension of frons and vertex combined; males with the anal triangle and auricles well developed, and no dorsal teeth or well developed carinae on segment 10; females with two or three long inferior spines on 10. Constricted and unconstricted abdomens.—These terms have been used rather loosely and indefinitely or relatively by authors. The third segment of the abdomen is the region involved. On the second and succeeding seg- ments, except the last two or three, on each side of each tergite two distinct longitudinal carinae can be detected. One of these is the lateral carina and below this, on the margin, is the ventral carina. Both may be armed with small denticles. On the second segment of the male the ventral carina parallels the extreme margin of the genital fossa and often bears denticles of various forms. The lateral carina on the same segment in its anterior part is modified into the auricle. (n the third segment the relative position of the lateral and ventral carinae can be accurately described and figured, and serves as a measure for the degree of constriction of the segment. If the carinae are widely separated and approximately parallel the segment is unconstricted. If the carinae fuse at about the level of the transverse carina and, anterior to that point, are joined or are parallel and in close juxtaposi- tion the segment is constricted. If the carinae approach at the level of the transverse carina so they are separated by a distance possibly half the dis- tance of separation near the apex of the segment, the segment may be defined as slightly or somewhat constricted. By defining this character in terms of the position of the carinae it is believed ambiguity or uncertainty in the future can be avoided, and especially in cases where the abdomen is not clearly constricted or not constricted. The trifida group (Triacanthagyna): This group of four species has given me the most difficulty in this study. All of them had been confused under one name trifida, and to add to the confusion Triacanthagyna needhami is a synonym of the true trifida. The females so far known, strange to say, are more easily recognized than the males. In the latter the appendages are so similar that in three of the species I detect no difference and in the fourth, caribbea, differences can be detected, probably, only by direct comparison with other species. Collections south of the equator will probably reveal additional species as practically all of the material seen by me is from more northern localities. Although the species are so similar I am nevertheless convinced, after many hours spent studying them, that the four are distinct. The nervosa group (Gynacantha): Inthe same way in Gynacantha, the nervosa group offers some serious difficulties. In the trifida group no varia- tion within a species in the form of the male abdominal appendages was detected. But among certain species of the nervosa group, especially nervosa, bifida, crocetpennis, litoralis and interioris, the superior appendage seems somewhat flexible. Flattened out, there is a distinct angle on the inner 6 E. B. WILLIAMSON edge where the abrupt widening takes place. But if this angle warps or rolls up, as it seems to do frequently, then the angle is lost, and the passage from the basal stem of the appendage to the expanded apical portion is less abrupt. For example I have specimens of nervosa, which so far as I can detect, agree well with both of Ris’ figures (35), figure 19 representing G. bifida and figure 20, G. nervosa. As in the trifida group body colors here are practically uniform in the various species, or, if differences exist, these are not well enough known to be of any service at this time. But fortunately in the males of the nervosa group, contrary to the condition found in the tri- fida group where all the males have constricted abdomens, constriction of the abdomen of the males has progressed in different degrees. And at the same time some striking wing patterns have been developed or have survived within the group, a thing which does not exist in the trifida group. By the use of these two characters, form of abdomen and color of wings, a fairly satisfactory classification of the material before me has been possible, but there are some individual cases which present problems I cannot positively solve at this time. See text under G. litoralis. Eyes and flight: Dr. Calvert (9) has written an interesting account of the relative size of the eyes in various insects, especially in dragonflies, wit) reference to the varying amount of light in the different environments in which the insects spend their lives, and he points out the great relative advantage of the eyes of Gynacantha over some other species. In such pre- daceous insects as dragonflies keenness of sight and power of flight might be expected to vary directly. In Gynacantha, for example, it is probable that the increased size and efficiency of the eyes, make possible the swift and often erratic flight, which would otherwise often terminate fatally. In dark forests many genera of agrionines with eyes relatively very small, compared with Gynacantha, live in great numbers, apparently living in such environ- ments more successfully even than Gynacantha. Their eyes are amply keen enough to detect food and mates, and to avoid the pitfalls spun by design- ing spiders. In fact, were they gifted with all of the powers of sight and flight of Gynacantha, and at the same time remained as small as they are, it is possible they might not survive in the environment where they now flourish. For Gynacantha escapes the ever present danger of spider webs not by keenness of sight but by sheer bulk and momentum. It is unusual to find a fully adult Gynacantha with no bits of spider webs attached to wings or body, strands which would have proved the undoing of a Heteragrion or a Palaemnema with equally unguarded flight. In fact Gynacanthas hawking in the evening may often be observed colliding violently with grass or sedge stems and leaves, and once a nervosa struck a motionless insect net handle with a resounding whack. Crepuscular flight is probably only an adaptation to the food supply and many aeshnines often practice it, though it seems more confirmed in certain species of Gynacantha and Triacanthagyna than in other genera such as Coryphaeschna and Aeshna. In Aeshna I have seen certain sun-loving species hawking late in the evening when it was so dark they were visible only against the sky or against the reflection of the sky in the pool over which they were flying. At this time their flight was as TRIACANTHAGYNA AND GYNACANTHA 7 swift and as well controlled, apparently, as is possible for the larger eyed Gynacantha. Generalizations on the habits of flight of aeshnines are dangerous even when a large amount of data is available, but it may be mentioned that the equally large-eyed Neuraeschnas, on the single occasion I observed them, were crepuscular with swift and well controlled flight. I believe such crepuscular flight as is found in Gynacantha and Triacanthagyna and possibly other aeshnine genera is a specialized habit related to the food dupply, some approach, under favorable conditions, to which is seen in Coryphaeschna, Aeshna, Boyeria, and probably other aeshnines, Tholymis and Pantala among the libellulines, and Neurocordulia among the cordu- lines. Certainly it is an adaptation found chiefly among the aeshnines and possibly not present at all in the Zygoptera, though certain species are noticeably most active late in the afternoon. Ecological notes: That there is some sensitive adjustment to their en- vironment is indicated by the spotted distribution of even the most widely ranging species of Gynacantha. For example, Palma Sola, Venezuela, is in a very level heavily forested region and lies about 35 meters above sea level. By railroad it is about 37 kilometers from Tucacas on the coast. Here in the evenings nervosa flew literally by hundreds or thousands. About 30 kilometers farther inland on the railroad lies the plantation Boqueron, at an elevation of about 125 meters. In the vicinity are many beautiful streams of different sizes and character. There are extensive pastures and corn- fields but closely surrounding these are forests as dense and of the same general character as the forests about Palma Sola. The landscape is varied with some low hills and the general result is a region apparently much richer and certainly much more attractive than the region about Palma Sola, but during eight evenings at Boqueron we saw not a single nervosa though we looked diligently for them. Cristalina lies far inland in Colombia, near Puerto Berrio, which river port is about 164 leagues above Barranquilla. The elevation of Cristalina is about 320 meters. In general the topography roughly suggests the topography about Boqueron, though there is less flat forest. In fact it differs from Boqueron about as Boqueron differs from Palma Sola. But at Cristalina we found nervosa flying in about the same numbers as we found them at Palma Sola. Why were they absent at Boqueron? ‘There is also the factor of seasonal distribution to consider. We may expect this to become more evident as one leaves the equator. It was conspicuously evident at 15-16°N. at sea level at Puerto Barrios, Guate- mala, where 7. septima was abundant in May and June and entirely wanting, that is, there were no great crepuscular flights, in January and February, when none was seen. Generalizations as to where and when these aeshnines may be found are therefore dangerous. They are essentially forest insects, though the forest may be low and chaparral-like; they require moist earth for ovipositing, and rains or overflow must provide water for the larvae; species and individuals seem most numerous at comparatively low elevations above sea level; and they are essentially tropical. 8 E. B. WILLIAMSON ORGANIZATION OF RESULTS This paper originated in my efforts to identify properly specimens in my own collection and its purpose is to facilitate identifications by other stu- dents. It makes no attempt to summarize completely the extensive litera- ture on these species and it should therefore be regarded as supplemental to the other literature. To make it as useful as possible for the purpose for which it has been prepared different parts of the paper are designed as a check on other parts. For example, in the keys to species the least possible use has been made of venational characters and of male appendages. The tabulation of the venational characters and the figures of appendages there- fore will serve as checks on determinations made by the key. In the text under each species are given data on size, form and size of female appendages where known, and in some cases additional data designed to facilitate identi- fication of material, and notes on habits, colors in life, etc. As a check on my own determinations I have listed all the material studied by myself, but I have not summarized all the records gathered by others, since as stated above this paper is merely supplemental to the other literature. The abbreviations used in designating various collections have been mentioned above in the paragraph under acknowledgments; E. B. W. refers to my own collection. Dr. Walker’s use of certain letters in designating abdominal spots, as explained by him (see under his name in the biblio- graphy) has been frequently used in this paper. All measurements are in millimeters, and the length of the abdomen as given is in every case exclusive of the appendages. Notes on colors of living specimens must not necessarily be considered complete. For example, failure in any case to mention pseudopupillae must not be taken as evidence that these are lacking. Field notes are often made under unfavorable circumstances, and I have often found myself regretting the lack of detail or lack of definiteness which the notes sometimes show. In some cases I have slightly re-edited these notes but in no case has anything been added to them. In the field only a small hand lens was available for examining material. Later studies have been made with a Zeiss binocular, using the number 2 eyepiece and the F 55 and A, objectives. SPECIMENS IN THE DE SELYS COLLECTION, FoLLowinc MartTIn’s REVISION, UNDER THE LABELS TRIACANTHAGYNA TRIFIDA AND GYNACANTHA SATYRUS Several months after this paper had been completed and sent to Dr. Ruthven for publication Mons. Severin kindly sent me for study the thirty- two specimens in de Selys collection under the label T. trifida and the six male types (not seven, as stated by Martin) of G. satyrus. Six species are included in this lot of thirty-eight specimens. In order to avoid as much as possible the rewriting of several parts of my manuscript, already long since completed, these specimens are considered one by one in this section of my paper, and only the changes rendered imperative in the text are made. These changes have to do with Triacanthagyna (Gynacantha) satyrus Martin. TRIACANTHAGYNA AND GYNACANTHA 9 The study of this material, sent me by Mons. Severin, tends to confirm my opinion expressed under 7. septima in the text, that septima is really a synonym of obscuripennis. Incidentally, the matter I could not understand, i. e. why de Selys’ collection contained only a single male of this widely dis- tributed and common species, is explained; there are seven specimens under the labels trifida and satyrus. The study of this material also causes me to think I have included two species under T. ditzleri, but at this time adequate characters for separating the two are not discernible. The small specimens from Central America and northern South America will probably be found to be specifically distinct from the similar but larger specimens from southern Brazil. One of these males is referred to in the text under 7. ditzleri. Other specimens are present in de Selys’ collection under the label trifida. I have designated as the type of ditzleri a male of the smaller northern and better known form. A solution of the problem of the status of the two forms awaits adequate material from southern Brazil. An examination of the following report on these thirty-eight specimens will impress the student, I think, with at least two things. It shows again, what should now be well known, that differentiation among species in at least certain aeshnine genera has progressed along such lines that specific characters are not conspicuous or obvious to the student, however effective they may be in nature. Following from this, the greatest care and discrim- ination in the determination of species is necessary. Incidentally it may be remarked, the description of new species in such groups, based on a single female or on any other number of specimens which the particular case renders inadequate, is not science but is an affliction. Thirty-two specimens in de Selys’ Collection labelled Triacanthagyna trifida Unless otherwise mentioned, in addition to labels indicated, each speci- men bears the following label: Collection Selys. Triacanthagyna trifida Rb. Revision Martin, 1909. ‘Triacanthagyna trifida Rb. 1. male | Honduras | Gynacantha trifida? R. male (in de Selys’ hand) | C’est espece different Forster (and on the reverse) il n’est pas trifida (all in Forster’s (?) hand)|. The specimen is 7. septima. male.|C1. 2./7|. The specimen is 7. septima. . female.| Mexique Salle|28|. The specimen is 7. septima. . female.|female 151|Boctes (?)|6|. The specimen is T. septima. . female.|151 female |14/16|. The specimen is 7. septima. . female.|Para, Schulz (?) (in de Selys hand).|10|. The specimen is T. septima. An - WRN 7. male.|Copa Cabana|113|32|. By its hamules it is T. ditzleri. In the anal loop three cells in the anterior row in both hind wings. This is the larger Brazilian form provisionally referred to ditzleri in this paper. 8. female.|Botafogo|14|. This is probably the female of the larger Brazilian form referred to 7. ditzleri. In the anal loop there are two cells in the anterior row in both hind wings. g. female.|Botafogo|21|._ Same remarks as under number 8. 10 E. B. WILLIAMSON 10. male.|Bocas (?) Sta. Cather. (?)|8].. Same remarks as under Number 7, except that the anal loop in one wing has three cells in the anterior row while the other wing has two. it. male.|Bates|143 A143 var. A minor|11]. The specimen is typical T. ditzleri. 12. female.|Bates|38|12|Gynacantha praedatrix Bates. Amazones (in de Selys hand). The specimen is typical 7. ditzlert. 13. male.]Panama (in de Selys’ hand)|9|._ The specimen is typical T. ditzleri. 14. female.|Coary. Amaz. Sup. (in de Selys’ hand)|17|. The specimen is typical T. ditsleri. 15. male.|Bates|51|30|._ The specimen is typical T. ditzleri. There are three cells in the anterior row of cells in the anal loop in the right hind wing. 16. female.|Dusem (?) 15/2 (in de Selys’ hand)|24]. The specimen most closely resembles 7. ditzleri, but has the anal loops unusually large with three cells in the anterior row of cells. I believe it is ditzleri but identification is not positive. Vena- tion shows it is certainly not trifida. In size and degree of constriction of the abdomen it is ditslert. 17. male.|Small rectangle of silver paper|Sta. (Cruz, Bolivia|3|. Hamules conceal- ed by glue or shellac; eyes are gone and the specimen is badly faded, but the legs and appendages are those of 7. caribbea, to which species, in the absence of more and better material, this single specimen is referred. 18. male.|25 (?)|Para St.|4!. Head and superior appendages gone, but the legs and hamules clearly show it to be 7. caribbea. 19. male.|Para. Schulz|23|trifida. petite sous race (Forster's (?) hand)|. This is a specimen of JT. satyrus and doubtless belonged to the same lot of specimens as that from which three of the types of Gynacantha satyrus were selected. 20. male.|C1.]13].. This specimen, which lacks the last seven abdominal segments, is Gynacantha adela. 21. female.|Bocas (?) Sta. Cath.|a small rectangle of orange colored paper|14!. The specimen lacks all its legs but one and the apices of the appendages are gone. I think it is certainly Gynacantha adela, certainly it is not a Triacanthagyna. 22. male.|male.|a small rectangle of gilt paper|Cuba|5|trifida Ramb.|A. trifida|Type Gynacantha trifida Rb.| Stock label, described in introductory sentence, with the ad- ditional word type. The eyes are gone, the abdomen is slightly damaged, and the left superior appendage is gone. This is one of Rambur’s types of trifida. 23. female. Similar to Number 22, and also labelled type. The eyes are gone and all of the abdomen but a few fragments of two or three basal segments. 24. male. Similar to Number 22, also labelled type, and carrying in addition the following label in de Selys’ hand; Triacanthagyna trifida R. Type de Rambur, Coll. Serville, male, Cuba. The eyes are gone and only the two basal abdominal segments, themselves damaged, remain. 25. male.|Cuba|22|. The specimen is 7. trifida. 26. male.|Cuba|25|Triacanthagyna trifida. Rb. Collection Selys|. Lacks the stock label described in the introductory sentence. The specimen is 7. trifida. 27. male.|Cuba|20| specimen is 7. trifida, —(?) trifida (and on the reverse) Forster dixit|. The 28. male.|Cuba|18]._ The specimen is T. trifida. 29. male.|S. Domingo|29|Triacanthagyna trifida Rb. Collection Selys|. Lacks the stock label described in the introductory sentence. The specimen is 7. trifida. his and numbers 20 and 31 doubtless belonged to the same lot as one of the specimens selected as one of the types of Gynacantha satyrus. 30. male.|Santa Domingo|26j113|Triacanthagyna trifida R. St. Dominque male (in de Selys’ hand)|. The specimen is 7. trifida, TRIACANTHAGYNA AND GYNACANTHA II 31. male.|S. Domingo]|31|Triacanthagyna trifida collection Selys|Triacanthagyna trifida Rb.|. The specimen is 7. trifida. 32. female.|Jamaique Deby|15|Triacanthagyna trifida R. female Jamaique (in de Selys’ hand)|. The last seven abdominal segments are gone. In size, venation, and constriction of abdomen, the specimen is 7. trifida, The six (not seven as stated by Martin) types of Gynacantha satyrus in de Selys’ Collection These all belong to the genus Triacanthagyna. Each specimen bears the two following labels in addition to labels indicated in each case: Collection Selys Type. Gynacantha satyrus M. Révision Martin 1909. Gynacantha satyrus Mart.; and a red bordered label on which is printed TYPE, and on which is written Gynacantha satyrus Mart. 1. male.|Pilanqui (?) Ecuador (?)|2|. The specimen is 7. septima. 2. male.|S. Domingo|6|. The specimen is 7. trifida and is probably one of the same lot as number 29, 30 and 31 above, properly labelled T. trifida. 3. male.|Yurimaguas, Perou!56|1|. The specimen is 7’. satyrus. 4. male.|Para. Schulz|3]. The specimen is 7. satyrus. Numbers 4, 5, and 6 of this lot and number 19 above doubtless belonged to the same lot of specimens. 5. male.|Para. Schulz|4|Boite 46]. The specimen is T. satyrus. - 6. male.|Para Schulz|5|. The specimen is T. satyrus and | have designated it as the type of the species. ALPHABETIC List OF NAMES PROPOSED IN THE Two GENERA Triacanthagyna caribbea, new species. Triacanthagyna ditzleri, new species. Triacanthagyna needhanu Martin. See text under T. trtfida. Triacanthagyna obscuripennis Blanch. See text under 7. septima; not in the key to species. Triacanthagyna (Gynacantha) satyrus Martin. Triacanthagyna septima Selys. Triacanthagyna trifida Rambur. Gynacantha adela Martin. Gynacantha (Selysiophlebia) aratrix Forster. See text under G. chelifera; not in the key to species, Gynacantha auricularis Martin. Gynacantha bifida Rambur. Gynacantha caudata Karsch. Gynacantha chelifera McLachlan. Not in the key to species. Gynacantha convergens Forster. Gynacantha croceipennis Martin. Gynocantha ereagris Gundlach. Gynacantha gracilis Burmeister. Gynacantha interioris, new species. Gynacantha jessei, new species. Gynacantha jubilaris Navas. See text under G. membranalis. Gynacantha klagesi, new species. Gynacantha laticeps, new species. Gynacantha limai Navas. See text under G. convergens. Gynacantha litoralis, new species. Gynacantha martini Navas. See text under G. adela. 12 E. B. WILLIAMSON Gynacantha membranalis Karsch. Gynacantha mexicana Selys. Gynacantha nervosa Rambur. Gynacantha robusta Kolbe. See text under G. bifida. Gynacantha (Aeshna) subviridis Selys. See text under G. auricslaris. Gynacantha tenuis Martin. Gynacantha tibiata Karsch. Species are arranged in the text in the same order as in the tabulation of vena- tional characters as follows: T. septima, ditzleri, caribbea, trifida, satyrus, G. laticeps, chelifera (and aratrix), adela, convergens, tenuis. caudata, tibiata, jessei, auricularis, klagest, ereagris, mexicana, nervosa, bifida, croceipennis, litoralis, interioris, gracilis, and membranalis. Key to American Species of Triacanthagyna and Gynacantha 1. Two rows of. cells between Mr and M2 beginning under the stigma; fork of Rs near the proximal end of the stigma in the front wing (except in trifda) and more basal in the hind wing than in the front wing. Third joint of penis in ventral view elliptical, without membranous margins, from slightly shorter than to less than half as long as the second joint, and only slightly wider; ligula rela- tively long and narrow, in ventral view subequal in width; superior appendage with a short narrowed base, apically blade-like, with parallel or subparallel mar- gins to near the minutely toothed apex, the blade without carinae or processes; posterior part of hamular process relatively low and therefore appearing more nearly horizontal, with no deep sulcus entirely across the process dividing it from the anterior part, which is relatively narrow. Ventral process on abdominal seg- ment 10 of female three-pronged. So far as observed crepuscular, flying in the evening and rarely in the morning; flight very erratic. Triacanthagyna. 2. 1... Two rows of cells between Mi and M2 beginning at or proximal to the stigma in the hind wing and usually in the front wing; fork of Rs distinctly basal to the stigma in the front wing and more basal in the hind wing (except in ara- trix). Third joint of penis in ventral view roughly triangular with a large expanded membranous basal margin on each side, equal to or slightly exceeding the second joint in length, and nearly or quite twice as wide; ligula relatively shorter, wider caudad, not subequal in width in ventral view; appendages variously shaped; posterior part of hamular process high and vertical or subvertical, sharply divided by a deep sulcus, which extends across the process, from the anterior part which is broad and thin. Ventral process on abdominal segment 10 of female two-pronged. Habits varied, when crepus- cular, flight less erratic and often in regular beats at a uniform height. Gynacantha. 6. 2 (1). Legs entirely pale; thorax without definite dark markings; abdomen pale; anterior edge of frons seen from above convex. Male with abdomen not con- stricted at segment 3; the opposing hairs on the blades of the superior appendages reduced in length from the base to the middle of the blade, the apical half with only jocdinany (short ‘hairs: 5.225523 or... canis ete cent rsemeelels vanbue este ars lelafes Be septima. 2.’ Legs more or less dark, in tenerals the legs are pale but the apices of the femora and bases of the tarsi dark to black; thorax with definite dark markings; abdomen dark; anterior edge of frons seen from above more or less angled. Male with abdomen constricted at segment 3; the opposing hairs on the blades of the superior appendages about equally long and numerous the entire length Oki they dla Gif ccs tvewsreceie src cies cxerces dese caueia testers cho fo peueucucroremenoustebecanonne trifida group, 3. 3 (2’). Anterior row of cells in anal loop usually consisting of two cells; second and third femora similar in color. Hamular process relatively small and short, less than .4 in length, equal to much less than one-third the distance from its TRIACANTHAGYNA AND GYNACANTHA 13 posterior edge to the anterior end of the median sulcus of the anterior lamina; superior appendage with the sides of the blade beyond the narrowed base parallel. Female with abdomen very slightly constricted, the space between the lateral and ventral carinae distinctly narrowed at the level of the transverse carina; appendages about as long as the last three segments............ ditzlert, 3. Anterior row of cells in anal loop usually consisting of three cells. Hamular process larger, more than .4 in length, equal to more than one-third the distance from its posterior edge to the anterior end of the median sulcus of the anterior LERTEMI LN cA Magar Nixes Neve Sec oe td iol ho ehahe. sreveeriate ond cite lhe Macros ae aie ae teeta he a eile 4. 4 (3'). Second and third femora dissimilar in color. Hamular processes with the mesal edges diverging posteriorly, less than .6 long; superior appendages with the sides of the blade beyond the base slightly converging posteriorly as seen in supero-internal view. Female with abdomen not constricted, and the appendages gSliinihe Saorctieie Weer Woe IEE eee GesMIGMG, co oobconnccasepangsdcuec carthbea. , 4. Second and third femora similar in color. Hamular processes with the mesal edges subparallel; superior appendages with the sides of the blade beyond the base parallel. Female of the single species known with the abdomen constricted at segment 3 and the appendages as long as the last three and one-half segments. ME aN so = grt oath ake al ahedeess Lido WS Stats ER I Soders ShECT GAs alee 5. 5 (4). Hamular process less than .6 long; margins of genital fossa without spines Hemalesas) describeds in paracnap lie eee eer. prey lots tetera trifida, 5’. Hamular process about .7 long; posterior border of genital fossa with a largy patch’ of black spines or teeth. Female not*known: .......2.:.->.2-4->. satyrus. 6 (1'). No antenodal crossveins of the first or second series, basal to the first thickened antenodal, present (rarely a single crossvein may be seen in one gi) mn yas ey ehes unl epee se Bencayc: View, ay turd Gyahep eval ccewe.<. Vee Meee AMES PORT ee Toe hs 6’. One or more antenodal crossveins of the second series, and rarely of the first series, basal to the first thickened antenodal, present in front wings or hind wings or both; venation complex, two rows of cells between M; and My, im- mediately following the loop in Ms; size large, abdomen 55 or more, hind wing 55 or more, and stigma front wing 5 or longer; colors bright and contrasting; wings with a colored basal area extending to nearly the first antenodal at least; femora light, clear reddish brown to almost black with the apices dark to black, tibiae darker, the dorsal surface sometimes more or less light yellowish brown; abdomen greatly enlarged at base, constricted at 3. In the male the ventral carina on abdominal segment 2, posterior to the point of convergence, concave, meeting the lateral carina at an acute angle; spines of anterior lamina direct- ed caudo-ventrad. Apparently on the wing throughout the day............... dt did S A ACID CETTE Oe En ee, HIRE ATs Gals Ao cies gracilis group. 27. 7 (6). Small to medium insects (abdomen 32-54, hind wing 32-54); (dull to brilliant coloration, legs variously colored; stigma rarely 4.5 long, in which cas7 the auricles of the male are greatly enlarged or there is a median apical tubercle on the sternum of abdominal segment I in both sexes; venation simple to com- plex; wings hyaline to tinged yellowish brown, rarely with very restricted brown at hase. In the male the abdomen is constricted at segment 3 (except in laticeps) and the ventral carina on 2, posterior to the point of convergence, 1s concave, meeting the lateral carina at an acute angle; spines of anterior lamina varicusly Ine CtCU Man ADIES sViALIOUS fev siete sks (sacs sere alercle eersiee cksierever ere Several groups. 8. 7. Medium to large (abdomen 43-62, hind wing 42-57, stigma front wing 4 long or longer), dull colored, brownish insects, with green if present dull or pale; sides of thorax brown with at least four usually distinctly defined brown to black spots or areas as follows: one surrounding the metastigma, a spot above 14 E. B. WILLIAMSON the metastigma, a spot at the upper end of the second lateral suture, and a spot or stripe posteriorly on the latero-ventral carina; legs pale colored; venation complex, two rows of cells between Ms and M, immediately following the loop in M,; wings hyaline to tinged yellowish brown or with a brown or brownish longitudinal stripe near the costal margin; sternum of abdominal segment 1 with- out a median apical tubercle. Auricles of male not greatly enlarged; the ventral carina on abdominal segment 2, posterior to the point of convergence, straight or convex, never concave, meeting the lateral carina in a more or less rounded angle; spines of anterior lamina directed caudad and slightly ventrad. So far as observed, flying for a short time in the evening just before dark or rarely ON (alewele \iColotals, (alblicaboyee Hla GEN) nemeimiod daca oomutdo obo on te nervosa group. 16. 8°(7).. Legs pale -colored,. yellowish’.or reddish).....%).202.a/anci 20m sine ot eae seb ee = 9. 8’. Legs black or black and yellow, bases of femora more or less pale................ ah ete, Hit ae Berns be erie er Lae eM ony RCI LOST compe 5 aga A Soe tibiata group. 15. g (8). No distinct brown stripe on the first lateral suture; anal loop normally separated by two or three rows of cells from the posterior wing margin. .... 10. t g’. A distinct brown stripe on the first lateral suture; anal loop separated by a single row of cells from the posterior wing margin. .......... caudata group. 14. 10 (9). Smaller species, hind wing less than 45 in length. Auricles of male in lateral view not extending caudad to the level of the transverse carina at mid- height. In the female the lateral and ventral carinae on abdominal segment 2 SU APAELay & ste nae cre rese ee aye ouster ecnip wna us sheer ore exe lees rahe Rieke tater ealeeetey nate ea ne 10. Larger species, hind wing 45 or more in length. Auricles of the male very large, expanded, in lateral view extending caudad far beyond the level of the transverse carina at mid-height; spines of anterior lamina not directed more ventrad than caudad. In the female the lateral and ventral carina on abdominal segment 2 diverging anteriorly, and the abdominal appendages shorter than the LaSts'tywic SCOMIENIES:. daisies aod welnb minuets bo. waatoaile rome cnteae auricularis group. 13. 11 (10). Spines of anterior lamina of male directed more ventrad than caudad. etna GanOt asc illllgnas setts = seine irs cusolieapeekereiee ey me DUE AS facet BAN. eR De Lo Be res Ud 11’. Wings not brown spotted at base. Male with spines of the anterior lamina directed more caudad than ventrad, and two cells between anal loop and anal triangle in first row posterior to A. Female with wing bases hyaline and four cellsepastentom tom. and spromimaletonthevanalel Oops setter rien etettet te adela, 12 (11). Base of wings hyaline. Abdomen of male not constricted at 3 and in the female the distance between the lateral and ventral carina on the same seg- ment less narrowed than in convergens. In the male one cell between the anal loop and the anal triangle in the first row posterior to A, and in the female three cells in the first row posterior to A and proximal to the anal loop; the SIAMESE SPECIES, 2s Olas s wssipies xe pele oeincleci iat Ree al Ceeteh ee tek Reta a kee laticeps. 12’. Base of wings tinged brown nearly to or slightly beyond the first antenodal. Abdomen of male constricted at 3, and in the female the distance between the lateral and ventral carinae on the same segment narrowed at the level of the transverse carina to about one-half the width posterior to this point. In the male two cells between the anal loop and the anal triangle in the first row posterior to A, and in the female four cells in the first row posterior to A and proximal Por catial OOPs' eis cls ee ease ecto one Sieh ceria Stes ece) Cates a eee tear nea convergens. 13 (10). Auricles of male large, not brilliantly colored ventrally, the black edge not produced anteriorly on the ventrum to the level of the hamular process; no teeth on the ventral carina of segment 2. Female with abdomen not con- stricted at 3; lateral carina on 2 uniformly low throughout its length. auricularts. TRIACANTHAGYNA AND GYNACANTHA 15 13. Auricles of male larger, brilliantly colored ventrally, the black edge produced anteriorly on the ventrum to the level of the hamular process; ventral carina on 2 with a row of scale-like teeth anterior to the point of convergence. Fe- male with abdomen constricted at 3; lateral carina on 2 elevated at the point Gorrespending to the’ auricle in the male? ... meer i ssa tess oka. cs canes dee klagest. 14 (9). Male usually with a row of small scale-like teeth on the ventral carina of the second segment anterior to the point of convergence; the lateral carina on the same segment, posterior to the level of the apex of the auricle, black or dark brown. Female with the lateral and ventral carinae on abdominal segment 3 only very slightly converging just anterior to the level of the transverse GNP. ct letcs cage EhCuI OM RO ETON ear PER cic a DN RR caved Sh Cog Se I tenuts. 14. Male without teeth on the ventral carina of abdominal segment 2; lateral car- ina on the same segment, posterior to the level of the apex of the auricle, pale Sreed te Neral Ge TOE HMO WIN. 5705 < ond. oew bi A EN O IA fot Bane caudata. 15 (8). Tarsi of four posterior legs striped yellow dorsally; sternum of abdomin- al segment I with a low median posterior tubercle; abdominal appendages yel- low. Apical abdominal segments of male yellow or yellowish. .......... tibiata. 15’. Tarsi of all the legs black or with the merest trace of yellow; no tubercle on sternum of abdominal segment 1. Apical abdominal segments and append- AGES OLeimMale Macks wemale mots /kiMoOwiit 5. ern teeters aleieeertciiaicts ota eiciere Jessel, 16 (7'). A wide black stripe covering the posterior third or more of the metepimer- on; wings hyaline or with the merest trace of color near the costal margin. Male with the abdomen greatly constricted at segment 3. Female with the lateralecanina sone sSectnlents 2) GiStin clly black eer eleteet aie mexicana. 16’. No distinct wide black stripe on the metepimeron. Lateral carina on seg- ment 2 on the female not distinctly black. Males, paragraph 17; females, paragraph 22. 17 (16'). Abdominal segment 3 greatly constricted as seen in dorsal view, the lateral and ventral carinae fused or in juxtaposition at the level of the trans- verse carina and for a short distance anterior to that point. .................. 18. ' 17’.. Abdominal segment 3 slightly or not constricted as seen in dorsal view, the lateral and ventral carine separated throughout their length .................20. 18 (17). Wings with a longitudinal stripe of color near the costal margin from Nase ovapex, darker basally, shading out apically 252... 6.25 04. 222. - mterioris. Pemiies without. a costal colored ‘Stripe .... .....jnqsts adele Do deends sod aeewieee 19. 19 (18’). Smaller, hind wing 42-43, venation less complex, two rows of cells between the anal loop and the hind margin of the wing. Cuba and the Baha- (TESS G.0.0-08o 0 AEG COERCION oo Mie bec ea cine pee eee ae ereagris. , 19’. Larger, hind wing 48-50, venation very complex, three rows of cells between the anal loop and the hind margin of the wing. South American............litoralis. 20 (17’). Wings hyaline or uniformly brownish tinged; lateral and ventral carine on abdominal segment 3 approaching at the level of the transverse carina, but distinctly separate and diverging anteriorly from this point............ nervosa, , 20’. Wings with a distinct stripe of color near the costal margin; lateral and ventral carine on abdominal segment 3 widely separated throughout their length and only slightly approaching anterior to the level of the transverse carina .......2r. 21 (20'). Color on wings confined to near the costal margin......... slelsisiele OPO? 16 E. B. WILLIAMSON 21’. Wings with the entire membrane suffused with yellowish............ croceipenmis. 22 (16'). Abdomen slightly constricted at segment 3, the space between the lateral and ventral carinze narrowed at the level of the transverse carina to about one-half or less the maximum width between them posterior to this point. ...23. 22'. Abdomen not constricted at segment 3, the space between the lateral and ventral carine slightly or not narrowed at the level of the transverse carina......... 24. 23 (22). Smaller, hind wing 43-46, venation less complex, two rows of cells between the anal loop and the hind margin of the wing. Cuba and the Bahamas. .ereagris. r 23’. Larger, hind wing 55, venation more complex,three rows of cells between the anal loop and the hind margin of the wing. South America. ................ litoralis. DAN 22.) WINES Ny daline cor siniionmly, (iNGld: «5 was s acinbileagieeec acetone nervosa. 24'. Wings with a distinct dark stripe near the costal border.................. 25. 25 (24’). Wing membrane suffused wih yellowish or brownish, the costal stripe bounded posteriorly by R........... Mec ie ae A hata oy che ae ive croceipennts. ane) Wings Jyaline iexcept.the costal Stripe....2...<. 0% 5+ sine oe ean eounuele eeeeimete 26. 26 (25’). Costal color yellowish, darker to brownish at base, bounded posteriorly Diya Sokesiec. oye dysesis copes Absiocb tele co ea an ee WIENER Ee OGRE Fe Ree eee eee bifida. 26’. Costal color darker, browner and wider, bounded posteriorly by M.. .interioris. 27(6'). Colored basal wing area reduced, occupying less than half the median space and bounded posteriorly by A; T-spot on frons well marked. Male with the ventral carina on segment 2 with few to no teeth anterior to the point of convergence and a few posterior to the same point and with seven or eight denticlés: ion the: saunicles sc. ce (eee eects ae Mee See ORC etree ene gracilts. 27’. Colored basal wing area extensive, occupying all or nearly all the median space and extending posteriorly beyond A, especially in the hind wing; frons above black or obscure, no evident T-spot. Male with the ventral carina on segment 2 with a row of ten or more teeth anterior te the point of convergence and none posterior to this point, and with only three or four denticles on the auricle Bea eH ako hak ss Atakcsedts labora nateme een areca Bees as eteiiake taltrisien Mhonstets Leber aieuancsetere ceetere ers membranalts. DISCUSSION OF SPECIES? Triacanthagyna septima Selys 6, 7, 30 (female is ditzlert), 11, 12, 23, 25, 26, 36, 40. Abdomen male 38-45, female 42-48; hind wing male 34-41, female 37-43; superior appendage male 4.2-5, female 5-5.2; stigma front wing 3-3.5. Dr. Calvert has called attention to the variation in the number of cells in the anal triangle in this species. And equally surprising is the variation in both sexes in the number of rows of cells between M, and Mspl in the hind wing and the number of cells in the anterior row of cells in the anal loop. Of fourteen males from Puerto Barrios, six have four rows between M, and Mspl, seven have three, and one has four on one side, three on the other; of twelve females from the same locality nine have four rows and three have three rows. Moreover the four-rowed condition is not confined to specimens "Numbers following each species refer to the citations in the bibliography. TRIACANTHAGYNA AND GYNACANTHA 17 from Guatemala, specimens from as widely separated localities as Mexico and Ecuador showing the same character. The larger number of specimens have three cells in the most anterior row of cells ot the anal loop, but speci- mens with two cells are numerous, and asymmetrical specimens are not rare. With few exceptions there is a single row of cells throughout between M, and Ks but in a few cases there is a single double cell. In the males there are usually two cells just posterior to A between the anal loop and the anal triangle, rarely one; and in the female there are four cells between the anal loop and the wing base. In the male the sternum of abdominal segment 1 may be smooth or may have as many as twelve small black spines and the spines on the ventral posterior margins of 2 vary froma single row of six or seven spines on each side to a small patch of as many as sixteen spines. In con- trast with these variable characters, there are three denticles on each auricle, one of these concealed on the inside of the auricle and anterior to the other two. The wide variation in many characters, the great variation in size, and the wide distribution of the material before me, led me to expect to find two or more confused species, but I have been unable to recognize more than one species. At the same time I have been unable to identify the species ob- scuripenmis. It is rather remarkable that septima is represented in the large de Selys collection by a single male. As recorded by Martin obscuripennis is represented in the same collection by two males and two females. The appendages of obsciipennis as figured by Martin seem relatively shorter and heavier than those of septima, but characters for separating the two species are not very definite. Much of the material before me has been studied by Dr. Calvert and I follow him in referring it all to septima but it seems to me not improbable that that name is really a synonym of obscuripennis. ‘The abdominal appendages of the female of septima are elliptical with a low median longitudinal dorsal carina and the extreme apex acute. My notes contain the following references to living colors: Males, Bolivar, Colombia: Eyes dark greenish brown above, gray beneath; thorax largely bright light green with few light brown markings, dorsum with a semicircular area covering the lower part; abdomen light brown, markings green and black. Female, Santa Marta, Colombia: Thorax above rich light brown, with a wide curved stripe on the mesepisternum green, wider and yellowish below, sides largely green; abdomen clove brown, markings obscure yellowish and black. Several well preserved males are colored as follows: On dorsum of thorax a lower central triangular or semicircular area, and an antehumeral stripe, which) tends to fade out in the surrounding green. pale brown; sides of thorax and abdominal segments 1 and 2 and base of 3 to or beyond the transverse carina, largely green. Dorsum and sides of abdomen brown, green as follows: 1, D spot; 2, AD narrow, MD large, joined with AML, PDs separated in the midline above, and on side PD and PL not joined; 3, AL, ML, MD, PD and PL present, the two latter not joined; on 4 and progressively posteriorly the spots are smaller and more obscure, MD, the best marked, disappearing after 8; extreme base of 2-8 dark ringed. 18 E. B. WILLIAMSON At Puerto Barrios the large palm swamp adjoining the town had been . cut off prior to 1909 for some distance back from the coast, and over this area in May and June, as darkness came on in the evenings and as darkness gave way to light in the mornings, septima came from the nearby forest to dart back and forth, high in the air, or near the ground, in countless thous- ands. But because of their erratic flight the collector found himself netting only an occasional specimen where dozens were apparently awaiting the sweep of the net. At Bolivar the following note: “This spectes the most erratic flier | have ever seen, and possibly the most difficult to catch. Along road after sunset. Few flying where there were many last evening. Began flying at 5:45 p. m.” At Puerto Berrio, “Flying along the railroad track below the hotel about 6:10 p. m.,—almost dark.” At Cienega, “A male flew into a lighted car at railroad station at 7:15 p. m.—dark outside.” At Palma Sola, “Flies along railroad track and in clearings in the woods just after sundown; very erratic especially in clearings; in town along railroad tracks associated with Gynacantha nervosa.” At Tachira, “Flying along railroad after sundown.” Their erratic flight is due to sudden darts after their prey which they detect at some distance. In clearings and along roads or trails the sky above them is a clear bright field and I have often seen septima rise several feet to seize its prey. In fact, so far as I can recall, they always strike from below. I have never found them flying in the mornings except at Puerto Barrios. It seems improbable that early evening and early morning flying occur gen- erally, for the usual great differences in day and night temperatures might be expected to prevent this. In my field notes the reference to septima as the most erratic flier I have ever known should doubtless be modified. Other species of Triacanthagyna are probably as swift and erratic, and, after repeated observations one after- noon, I give the palm for abrupt and erratic and perfectly controlled flight, among all the dragonflies I know, to Tholymis citrina. Material examined: Mexico (A. N. S. 1 male, 1 female; U. S$. N. M. 1 female); Guatemala (Van. Patten, 1 female M. C. Z.), Puerto Barrios (May 25, 26. 28, and 30, and June 23. 1909, 16 males, 12 females E. B. W.), Santa Tomas (May 29, 1909, 1 male, EK. B. W.), San Felipe (Hay and Maxon, February 22, 23, 1905, 1 male, 1 female, U. S. N. M.); Costa Rica, Guapiles (P. P.'C., July 13, 1915, 1 female, A. N. S.); Colombia, Santa Marta (December 16, 1916, 2 females, KE. B. W.), Bonda, Dept. Magdalena, 250 feet (H. H. Smith, July and August, 2 males, 1 female, Carn. Mus.), Bolivar, near Santa Marta (December 21, 24, 25 and 26, 5 males, 3 females, E. B. W.), Cienega, Dept. Magdalena (February 26, 1917, 1 male, E. B. W.), Rio Frio, Dept. Magdalena (January 7, 1917, 1 male, EK. B. W.), Rio Neuvo, Dept. Bolivar, on Magdalena River, between Magangue and El Banco (Jan- uary 22, 1917, 1 female, FE. B. W.), Puerto Berrio, Dept. Antioquia (Jan- uary 31 and February 21, 1917, 2 females, E. B. W.) ; Ecuador, (F. Campos R., 1 female, Cornell), Guayaquil (F. Campos R., 4 males, 2 females, A. N.S., 1 female, Ris), El Salado (F. Campos R., 1 male, 1 female, A. N. S.), S. Rafael (F. Campos R., 1 female, A. N. $.), Babahayo (F. Campos R., TRIACANTHAGYNA ANT GyYNACANTHA 19 1 female, A. N. S.); Bohzvia, Province del Sara, Dept. Sta. Cruz, 350 m {José Steinbach, February, 1913, 1 female, Acc. 5076, Carn. Mus.) ; V enesu- cla (Appun, 1 male, M. C. Z.), Puerto Cabello (1 female, M. C. Z.), Palma Sola, Dept. Falcon (March 4-10, 1920, 10 males, 5 females, FE. B. W.), Tachira, Dept. Falcon, about 364 m (April 11, 1920, 1 male, E. B. W.); Dutch Guiana (1 female, O. 8. U.; 1 male, U. S. N. M.; 1 female, M. C. Z.; 1 male, 1 female, Ris), Paramaraibo (K. Mayo, 1 female, A. N. S.); French Guiana, Cayenne (S. M. Klages, February, 1917, 3 males, 1 female, Acc. 5873, Carn. Mus.) ; Brazil, Para (February 6, 1901, 1 female, Ris; C. F. Baker, 1 female, EK. B. W.), Santarem (A. H.,Fassl, July, 1920, 1 female, Ris), Manaos (‘Miss H. B. Merrill, 2 males, 2 females, U. S. N. M.), Bom Jesus de Itabapoana, Rio Janiero (Zikan, March 26, 1906, 1 female, Ris) ; Trimdad, San Juan (March 2, 1912, 1 female, E. B. W.); Cuba (Poey, 1 male, 1 female, M. C. Z.). See also discussion above of specimens in the de Selys Collection. Triacanthagyna ditzleri, new species 6 (Chapada male as trifida), 10 (April female and August male as trifida, September female as septwma). Abdomen male 36-40, female 41-43; hind wing male 33-36, female 37-42; superior appendage male 4.8, female 6-6.9; stigma front wing 3.4-3.6. Male.—Living colors, from notes on the Palma Sola male, March 6, 1920. (Descriptive notes in parentheses made from preserved specimens.) Eyes above dark green, in front on either side an oblique line of bright blue, narrowly bordered on its upper edge with black especially at the superior end of the line; lower half of eyes pale green with a slight brownish cast and one conspicuous black and five or six smaller brown pseudo-pupillae. (Face green, paler and duller below, darker to black above, anteclypeus brown; frons above in front black, stem of T-spot obscure or wanting; frontal vesicle black; occiput small, green.) Rear of head greenish yellow, a broad black band from foramen to occiput. Prothorax brown, anterior lobe blue, elevation of posterior lobe dark brown, the free edge narrowly greenish blue. Thorax brilliant grass green, marked with rich dark brown, almost black on the dorsum; dark mid-dorsal area nipple-shaped, narrower above, the edge on either side at midheight concave then widening out and convex below. Antehumeral stripe wide, almost as dark as the dorsal area, shading out above, especially anteriorly, and very narrowly joined above with the middorsal area; two lateral narrow brown stripes, the first the wider and forked above; interalar spots green. Abdomen above rich reddish brown, almost black, apex of 7 and 8 and 9 paler. In a male from La Fria the thorax above and the abdomen above were entirely black, not brown and not paler apically. Abdominal markings bright green as follows: 1 with D interrupted, L present; 2 with AD taper- ing from base to apex, AML large, MD narrow and triangular, PD narrowly divided in the mid line, PL and PD not joined, PL about the same size as PD on either side, both small; (two or three denticles on each auricle; 20 E. B. WILLIAMSON margins of genital fossa without spines or teeth, a small patch of short bristles on the marginal prominence opposite the posterior end of the ligula) ; on 3 AL triangular, reaching about half the distance to the transverse carina ; 3-7 each with MD small, triangular, narrowly separated, and PD small and elliptical, both MD and PD progressively smaller posteriorly, almost gone on 7, on 8 MD represented by mere dots; ML, and PL apparently wanting or very faint on 3 and posterior to that segment; AL on 4 and posteriorly inconspicuous or wanting. (Wings hyaline, costa dark reddish brown to black, stigma dark brown, ventral surface paler and yellowish; venation black or nearly so. ) (First femora black, green behind the full length except the extreme apex, first tibiae and tarsi dark brown; middle and hind legs similar in color, femora and tibiae light to dark reddish brown, each femur paler toward the base and black at apex, each tibia dark to black at base and apex, tarsi black or nearly so.) Female——Based entirely on preserved material. Face yellowish brown, yellower below, darker and obscurer above with only faint tinges of green. Head above similar to the male. Thorax similar to the male, the middorsal dark area more nearly tri- angular in shape, the dark colors reddish brown, paler, and the green less vivid. Abdomen, like the thorax, paler than in the male, only the apical three or four segments apparently black or much darker than the paler basal segments. D and L present on 1, the latter large; 2 with AML, MD and PL, joined into a long wide longitudinal bar just above the lateral carina; on 3 AL, and ML joined, and the other spots, MD, PD and PL, present; 4 with the five spots, AL, ML, MD, PD and PL all present, though small; these spots can all be detected on 5 and 6, and in life doubtless some of them are continued farther posteriorly. Appendages regularly elliptical, widest about the middle, maximum width .9-1.0, a low median longitudinal dorsal carina, apex regularly rounded, extreme apex acute. Wings and legs as in the male. Of fourteen wings of males, twelve had two cells in the most anterior row in the anal loop, and two wings had three cells; of fourteen wings of females nine had two cells and five had three cells. In the males there were invariably two cells just posterior to A between the anal loop and the anal triangle; and in the females there were in every case four cells posterior to A from wing base to anal loop. Of both sexes twenty-three front wings had a single row of cells throughout between M, and Rs, three wings had a single double cell, and two wings had two-three double cells; in the twenty- eight hind wings all had a single row of cells between M, and Rs. Ditzleri is the smallest and possibly the handsomest of the trifida group, and the smallest specimens are smaller than the smallest septima I have seen. However, the male from Blumenau, referred to this species, is larger than the measurements given above, the abdomen being 42.5 and the hind wing 4o. The wings are slightly tinged brownish; there are two cells in the anterior ‘ TRIACANTHAGYNA AND GYNACANTHA 21 row in the anal loop. The second and third legs are darker than usual, the reddish brown of the femora being darker with the apical black more ex- tensive. But I believe the species is certainly ditzleri. Several evenings at Palma Sola, returning home about sunset, walking on the railroad track through the practically continuous forest which sur- rounds the village, suddenly before us, near at hand and as far as we could see down the track, would appear the widely darting forms of Triacantha- gynas. Even when seen in numbers each individual's flight was too erratic and independent to permit calling the assemblage a flock. One evening they were especially numerous. We tried striking at individuals, and we ran down the track beating back and forth with the net, but two of us doing our best failed to catch a single one. Returning later one evening I ran along the track striking right and left as rapidly as possible and thus by sheer good fortune netted two individuals, a male each of ditzleri and of caribbea. The single specimen taken at Puerto Barrios was associated with 7. septima. I had been catching specimens of septima since about sundown and when ditzleri was captured it had grown so dark I could see individuals only when they came between me and the clear sky. I think all the other specimens of ditzleri we captured were taken in the forest where, when flushed, they usually flew a short distance and alighted on some twig or stem, or even tree trunk, in characteristic aeshnine position with down-hanging abdomen. Material examined: Guatemala, Cayuga (Schaus and Barnes, April, 1 female, August 27, 1 male, September 16, 1 female, A. N. S.; first two recorded, Calvert 10, as T. trifida, the female of September 16 as septima), Puerto Barrios (June 23, 1909, 1 male, E. B. W.) ; Colombia, Puerto Berrio, Dept. Antioquia (January 31 and February 8 and 21, 1917, 3 males, 3 fe- males, K. B. W.) ; Venezuela, Palma Sola, Dept. Falcon, (March 6, 1920, I male, EF, B. W.), El Guayabo, Dept. Falcon (April 22, 1920, 1 male, E. B. W.), La Fria Dept., Falcon (April 12, 13, 16 and 17, 1920, 4 males, 2 females, FE. B. W.) ; British Guiana, Bartica (H. S. Parish, May 28 and June 1, 1901, 2 males, O. S. U.) ; Dutch Guiana (1 male, 1 female, O. S. U.; 1 male, det. by R. Martin as trtfida, Ris; 1 male, E. B. W.), Paramaraibo (Miss K. Mayo, 1 female, A. N.S.) ; Brazil, Para (C. F. Baker, 1 female, E. B. W.), Tapajos, Amazon, Monte Christo (A. H. Fassl, May, 1920, 1 female, Ris), Porto Alegre, Rio Grande de Sul (J. D. Haseman, January 21, 1909, 1 male, Acc. 3768, Carn. Mus.), Blumenau, Santa Catharina, (1 male, FE. B. W.), Chapada (1 male M. C. Z., recorded, Calvert 6, as trifida). See also discus- sion above of specimens in the de Selys’ Collection. Type male and allotype female, La Fria, Venezuela, April 17, 1920, EF. B. W. Named for William Howard Ditzler, who, as a member of the University of Michigan Venezu- elan Expedition collected, among many other dragonflies, the type of this handsome species. Needham’s Figure 3, Plate XXXIX. A Genealogic Study of Dragon- Fly Venation, is probably of the wings of this species. 22 E. B. WILLIAMSON Triacanthagyna caribbea, new species 6, 10, 11, all in part. See text. Abdomen male 41-45, female 44-47; hind wing male 39-45, female 42-46; superior appendage male 5.4-5.7, female 6.9; stigma front wing 3.5-4.2. Male.—Living colors, from notes on the Palma Sola male, March 6, 1920. (Descriptive notes in parentheses made from preserved specimens. ) Very much like T. ditsleri; head above darker green, the black line, border- ing the blue bar across the eyes, is itself narrowly bordered above with light green, and the blue bar itself below passes through darker blue to black ; rear of head with very nearly white replacing the greenish yellow. Prothorax almost white, posterior lobe darker. Thorax less brilliant green but still very bright; dark markings reddish brown, less contrasted with the green as compared with 7. ditzleri, but still very distinct, the dorsal dark area more widely connected with ante- humeral stripe, thus more isolating the dorsal green. Abdomen brown, much paler than in 7. ditzleri; on 2 PL, is large, quad- rangular, twice as wide as PD above it, and joined with the auricle, not so joined in 7. ditzleri; (three denticles or three and the vestige of a fourth on each auricle; one to nine, usually four to six, small teeth of varying sizes on the margin of the genital fossa on the prominence opposite the posterior end of the ligula) ; on 3 AL larger, reaching the transverse carina; 3-8 with MI, present as small pale yellowish brown spots; MD and PD still con- spicuous on 7; on 8 as conspicuous as on 7 in T. ditsleri; PD represented on 9 by small spots. (Wings hyaline to brownish tinged, costa dark reddish brown to almost black at apex, stigma reddish brown, beneath lighter yellowish brown, vena- tion black or nearly so.) (First and second femora black or nearly so, paler at base and green behind, the two similar in color, thus different from ditgleri where the second and third legs are similarly colored; third femora black, or nearly so, thus distinctly darker than in 7. ditzleri; tibiae black beneath, reddish brown above; tarsi black. ) A teneral male, Santa Marta, December 14, 1916, in life was colored as follows: Eyes above dark brown. Dorsum of thorax bright brown, stripes bright green and distinct; sides green with brown markings in definite pattern. Abdomen brown, markings black (at the extreme apex of each segment from 3 posteriorly to about 7), and (spotted) obscure yellow. (In this specimen the third femora are dark brown and the contrast between the second and third femora is as well marked as in adults. ) The following notes on liv:ng colors were made on an apparently adult male, Bolivar, Colombia, December 21, 1916: Colors of thorax bright green and rich brown in distinct and sharp pattern. Abdomen brown with green and black markings. Female—Based on preserved material. Face light yellowish brown, darker above, with only faint traces of green; frons above dark dull green shading into black anteriorly, with the stem of the T-spot obscure or want- ing, as in the male. TRIACANTHAGYNA AND GYNACANTHA 23 Thorax and abdomen as in the male, to judge from preserved material, the colors possibly slightly paler, especially those of the abdomen. Append- ages similar to ditzleri in shape, basal narrowing slightly longer. Wings and legs as in the male. Of fourteen wings of males twelve had three cells in the most anterior row in the anal loop, and two wings had two cells; of fourteen wings of females all had three cells. In thirteen wings of males there were two cells just posterior to A between the anal loop and the anal triangle, and in one wing there were three cells; in the fourteen wings of females there were in every case four cells posterior to A from wing base to anal loop. Twenty-two of the twenty-eight front wings had one row of cells throughout between M, and Rs, four wings of females had a single double cell, and one wing of each sex had two double cells; in all twenty-eight hind wings there were invariably a single row of cells. In a dry brush-choked gully near Santa Marta we found slightly teneral specimens of this species in December. Crawling about through the almost impenetrable spiny vegetation the collector flushed occasional individuals, a few of which it was possible to capture in his fingers. Other specimens were taken soon after sunset flying along the road at Bolivar and other specimens from various localities were flushed in forests. Caribbea, in the form of the superior appendages of the male, differs from the other three species of the trifda group, but this difference is slight and will be detected probably only by those having specimens of the several species for comparison. The green color of the posterior surface of the second femora, which thus resemble the first femora and differ markedly from the third femora, is obvious in well colored specimens of both sexes. In ten- eral individuals this green is replaced by gray and the color is less con- spicious as the entire femur is paler in such specimens. But in these cases the third femora are darker than the second, so the dissimilarity between the second and third femora still holds. In trifida Rs forks more basally than in other species of the group and this can be more readily detected in the front wing (because of the greater proximity of the fork to the stigma) than in the hind wing. In figure 146, Martin 25, tvifida, Rs in the front wing forks basal to the level of the last postnodal crossvein, and in the hind wing basal to the level of the stigma a distance equal to or greater than the length of the stigma. A comparison of Martin’s figure 146 with figure 3 of caribbea in this paper will show this difference, which however, like the differences in appendages, is difficult of precise definition. See last paragraph under T. ditzlert. Material examined: Me-vrico (1 male, 1 female, U. S. N. M.) ; Guatemala, Cayuga (Schaus and Barnes, forest, August 23, I female, A. N. 5S.); Colombia, (Turbo, Mack, 1 male, M. C. Z.), Santa Marta, Dept. Magdalena (December 13, 14 and 16, 1916, @ males, 9 females, E. B. W.), Bolivar, near Santa Marta (December 21, 23, 26, 1916, 2 males, 1 female, FE. B. W.) ; Puerto Berrio, Dept. Antioquia (February 8, 1917, 1 female, E. B. W.); Venezuela, Caracas (Rolle, 1910, 1 female, Ris), San Esteban, Dept. Cara- bobo (February 8, 1920, 1 female, E. B. W.), Palma Sola, Dept. Falcon 24 E. B. WILLIAMSON (March 4, 6 and 9, 1920, 1 male, 3 females, 1 female shot by H. B. Baker in forest, E. B. W.); Boqueron, Dept. Yaracuy (March 16, 1920, 1 male, E. B. W.), Aroa, Dept. Yaracuy (March 14, 1920, 1 female shot by H. B. Baker in forest, E. B. W.) ; Dutch Guiana (Thorey, 1 male, M. C. Z.); Brasil, (Heyer, 1 male, labelled “Gynacantha elata Hagen”, and 1 male labelled “Aeschna augusta Hagen,’ M. C. Z.), Bom Jesus de Itabapoana, Rio Janeiro (Zikan, July 27, 1905, I female, Ris). The single specimen from Guatemala is recorded as trifida in Calvert 10; and the five males and one female from Mexico, Colombia (Turbo), Dutch Guiana, and Brazil (Heyer, M. C.-Z.) are recorded as the same species in Calvert 6 and It. See also discussion above of specimens in the de Selys Collection. Type male and allotype female, Palma Sola, Venezuela, March 6, 1920, E. B. W. It should be noted that some of the above material is teneral and other- wise in a bad state of preservation, making determinations difficult, and the female of T. satyrus is not known and may be confused as a female of this or other species. For these reasons the records for Mexico, Guatemala, and Rio Janeiro, Brazil, are open to question. Triacanthagyna trifida Rambur I, 2, 5, 6 (Chapada male is ditzleri; Mexico, Colombia, Dutch Guiana and Brazil [Heyer] specimens are caribbea), 8 (is satyrus), 9 (is satyrus), 10 (male and April female are ditzleri; August female is caribbea), 11 (see 6 above), 12, 13, 16, 20, 23, 25, 26, 34, 35, 360, 39; all references in the literature to specimens outside the range of T. trifida, as indicated in the material studied in this paper, are probably to some other species than trifida. Abdomen male 42-45, female 47-50; hind wing male 41-43, female 44-47; superior appendage male 6-6.3, female 9-10.6; stigma front wing 3.3-4.2. I have seen only preserved material. The coloration is essentially the same as in 7. ditsleri, but the dark dorsal thoracic area, in those specimens where the pattern is discernible, is triangular, with the lateral edges nearly or quite straight, not nipple-shaped as in 7. ditzleri. The wings are hyaline to yellowish brownish tinged. Twelve wings of males and twelve wings of females each had three cells in the most anterior row of the anal loop. In the males there were invariably two cells just posterior to A between the anal loop and the anal triangle, and in the female there were four cells posterior to A between the wing base and the anal loop. In three front wings of males one front wing of a female and two hind wings of males there was in each case a single double cell between M, and Rs; in all the other wings there was a single row of cells throughout. For a note on the basal forking of Rs see text under T. caibbea. See also last paragraph of text under T. ditzleri. In the specimens examined there were no spines or teeth on the ventral margins of abdominal segment 2 in the male, and the auricles were armed each with three denticles or with three and the vestige of a fourth. Append- ages of female elliptical, a low median longitudinal carina, of nearly uni- form width throughout after the gradually tapering basal third; apex rather abruptly rounded, extreme apex acute, maximum width 1.3. TRIACANTHAGYNA ANL GYNACANTHA 25 Material examined: California (1 male, 1 female, A. N. S.); Georgia, Billy’s Island, Okefenoke Swamp (June, 1912, I male, Cornell) ; St. Simons Isl. (Sept., Oct., 1910, 1 female, Cornell) ; Florida, Lakeland (W. T. Davis, November 8 and 10, 1911, 2 females, W. T. Davis), Hastings (1 female, Cornell), St. Augustine (C. W. Johnson, 1 male, A. N. S.), Miami (1 fe- male, Cornell) ; Cuba, (Poey ,1858, 1 male, 1 female, M..C. Z.), Havana (C. F. Baker, 1 male A. N. $.), Guanajay (Palmer and Riley, May 6, I male, 1 female, U. S. N. M.), Baraca (Aug. Busck, September, 1901, I male, U. S. N. M.) ; Hayti (Dr. Abbott, 1 male, A. N. S.), Samana (Frazar, 2 males, 1 female, M. C. Z.), Tortuga, N. W. Hayti (W. L. Abbott, July, 1917, 1 female, A. N. S.) ; Jamaica, Bath (Mrs. Swainson, 1 male, 1 female, A. N.$.). See also discussion above of specimens in the de Selys Collection. Triacanthagyna needhami Martin is a synonym of T. trifida unless an- other species, which I have not seen, inhabits Florida. Triacanthagyna satyrus Martin 8, 9, under both titles as trifida, 25, 26. Abdomen male 42-43,? female 43; hind wing male 39-42,? female 41; superior appendage male 6-6.3,? female 7.8; stigma front wing 3.6-4. Male——Colored like T, ditzleri except that on abdominal segment 3 MIL and PL are evident in the preserved material, and these spots may have been present on segments posterior to 3 during life. Like 7. ditzleri it is darker than I’. caribbea, and the following note was made of the La Fria male at time of capture. Thorax and abdomen above black, not brown, and not paler apically in the case of the abdomen. The frons above is dark to entirely black. Dr. Calvert has kindly given me the foliowing notes on the living colors of the Costa Rican male taken by him. Eyes brilliant green, darker above. Frons dark brown with an ill-defined superior black spot. Clypeus and labrum pale green, the latter more yellowish; labium pale brown but with some pale green on the middle of the lobes. Thoracic dorsum dark brown with a cuneiform, sharply marked, green antehumeral stripe, diverging from above from its fellow of the opposite side. Sides of thorax bright green, a narrow brown stripe on the first and second lateral sutures. Abdomen black, the following green; on I a transverse posterior dorsal, and on each side a transverse posterior lateral stripe, the three not confluent; on 2 each side a transverse anterior stripe, a narrow median transverse line, three posterior spots arranged in a transverse line, a mid-dorsal line and the auricles ; on 3 a basal spot each side; on 3-7 each with a pair of transverse triangular spots just posterior to the transverse carina; on 3-5 a pair of transverse posterior spots. Pectus and under side of abdomen pale flesh color. A teneral female from Rio Janeiro, Collection Ris, may belong to this species. and the above measurements of the female are based on this speci- men. Although pale and faded it has plainly the leg colors of the trifida group and the third abdominal segment is constricted as in T. trifida. ‘The 26 E. B. WILLIAMSON slightly shorter appendages alone distinguish it from the geographically widely separated 7. trifida. In the three males available and in the single female provisionally refer- red to this species there are invariably three cells in the anterior row of cells in the anal loop; in the male there are two cells posterior to A between the anal loop and the anal triangle, and in the female there are four cells posterior to A between the wing base and the anal loop; in the front wings of males, one wing has a single double cell between M, and Rs. two wings have three double cells, and one wing has four double cells; in all the other wings of both sexes there is a single row of cells throughout between M, and Rs. The auricles of the second segment of the male are each armed with three denticles or with three and the vestige of a fourth. Material examined: Costa ‘Rica, Banana River (P. P. C., forest, upper reservoir, November 9, 1909, one male, A. N.'S.) ; Venezuela, La Fria, Dept. Tachira (in forest, April 17, 1920, 1 male, KE. B. W.); British Guiana, Bartica (H. S. Parish, May 27, 1901, 1 male, O. S. U.). See also discussion above of specimens in the de Selys Collection, where Martin’s six types of satyrus are considered in detail. As explained there, these types were re- ceived for study long after this paper was completed. During the prepara- tion of this paper Mons. Severin kindly sent me one of the types of satvrus, but this one happened to be the Santo Domingo specimen. On the basis of this examination satyrus was reduced to a synonym of trifida in my paper, and I described my material as a new species. With the examination of all six of Martin’s types it is now possible to save his species and I have design- ated one of these specimens as the type of satyrus. Gynacantha laticeps, new species Abdomen male 33-33.5, female 35-38; hind wing male 32, female 34-37; superior appendage male 4.5; stigma front wing 2.25-3. Male and female—Labium and face dull pale yellow, slightly darker above, and occupying most of the frons above, which has a darker greenish shade, so the brown T-spot is reduced and inconspicuous; frontal vesicle dark brown; occiput greenish yellow; rear of head dull pale yellow, black margining the eyes above, and this joined with a broad black bar to the foramen. Prothorax pale brown or greenish brown, the posterior edge yellowish green. Thorax green; if any markings are present in life they have not survived post-mortem changes. Abdomen not constricted, brown or yellowish brown, marked with pale (green or yellow or both) and black; 1 and sides of 2 apparently largely green, I darker above apically ; 2 with AD very narrow and extending from the base to and across the subapical transverse carina, but ending before the extreme margin of the segment and interrupted at the median transverse carina; a basal trace of AD on 3; MD present as a triangular spot on either side on 2-7; sides of 2-7 slightly if any paler than the dorsa, no definite TRIACANTHAGYNA AND GYNACANTHA 27 pattern of dark and pale in the dried material; median transverse carina, subapical transverse carina, and extreme apex of 2-7, and to a lesser extent of 8 distinctly black ringed; 8-10 and appendages brown, slightly darker than the segments basal to them. Margins of genital fossa of 2 without teeth or spines; auricles small, armed with five to seven denticles. Wings hyaline; costa, venation and stigma brown or yellowish brown to nearly black; the stigma in the females is lighter yellowish, due probably to immaturity, and the posterior vein, especially, is conspicuously black. In the two males and two females, of all the wings, there is, in each wing, a single row of cells throughout between M, and Rs; in all the wings the loop in M, is greatly reduced and in the wings of the two males and one of the females there is, in each wing, a single row of cells throughout between M, and M,; in the other female in each front wing there is one double cell and in each hind wing there are three to five double cells. Legs pale dull yellow, unmarked, the first femora slightly darker above. The two females show a marked difference in size of all parts but I cannot separate them on any other character. Material examined: Brazil, Minas Geraes (I. Rolle, 1914, 2 males, 2 females, Ris), type male and allotype female of this material. I am able to describe this species through the kindness of Dr. Ris who loaned me the four specimens I have seen, and who expressed the opinion, when the speci- mens were sent to me, that they represented an undescribed species. Gynacantha chelifera McLachlan Ph AS. OT Gynacantha (Selysiophlebia) aratrix Forster TA 5. 20: Unfortunately I have not seen specimens of these species. They are about of the same size (abdomen 40-43, hind wing 35-37, Martin, 25), and the descriptions and figures of appendages indicate very closely related if not identical, species. In fact, with this evidence alone before me, I regard aratrix as a synonym of chelifera. Gynacantha adela Martin 25, 20, 35. Abdomen male 44-45, female 44; hind wing male 42-43, female 43; superior appendage male 5.1; stigma front wing 2.7-3. The margins of the genital fossa of the second abdominal segment of the male are armed on either side, just anterior to the point of convergence, with three to six small scale-like teeth in a single row; the auricles are each armed with six or seven denticles. In one of the males from Bolivia in the left front and hind wing M,, distal to the loop in M,, is switched anteriorly one row of cells, resulting in three rows of cells between M, and M,. This 28 E. B. WILLIAMSON is the only case I have seen in all the wings studied in which a main sector is so switched. The appendages of the single female are broken. Material examined: Peru, Campamiento, Colonia del Perene (W. T. M. Forbes, June 12, 1920, 1 female, Cornell) ; Bolivia, near Coroico, Yungas (W. J. Gerhard, May 2, 4, and 12, 1899, 3 males, A. N. S.); Brazil, Minas Geraes (Rolle, 1914, 1 male, Ris). See also discussion above of specimens in the de Selys Collection. Navas (29) describes a male from Sao Paulo as a new species, G. mar- lini. Huis figures of appendages indicate that the inferior is relatively slight- ly longer than in specimens of adela seen by me, but definite characters for separating the two species are not evident and it is probable martini is a synonym of adela. Gynacantha convergens Forster 15, 17, 20, 35. Abdomen male 42, female 44-45; hind wing male 38, female 41-41.5; superior appendage male 4.8; stigma front wing 3.3. I follow Dr. Ris’s determination of this species. The margins of the genital fossa of the second abdominal segment of the male are without teeth or denticles. The auricles are each armed with five to seven denticles, of which the two or three most anterior ones are very small. Between M, and Rs in the front wing there is one double cell in two male wings and one female wing, two-three double cells in one female wing, three double cells in one female wing, and four double cells in one female wing; in the hind wing there are no double cells in two male wings, one double cell in two female wings, one-two double cells in one female wing, and two-three double cells in one female wing. Appendages of both females broken. Material examined: Bolivia, Province del Sara, Dept. Santa Cruz, 450 m (José Steinbach, 1918, 1 female, Acc. 6443, Carn. Mus.) ; Argentina, Yuto, Jujuy 450 m (P. Joergensen, April, 1911, 1 male, 1 female, Ris). Navas (32) describes a single female from Italiba, Brazil, as Gynacantha limat. From the very nature of the case positive identification is impossible, but I believe /imai will prove to be a synonym of convergens. Gynacantha tenuis Martin 25, 20. Abdomen male 38.5-44, female 41-47; hind wing male 39-41.5, female 38-46; superior appendage male 4.2-4.4, female 3.2-3.6; stigma front wing 3.2-3.4. Specimens in the M. C. Z. labelled tenuis by Hagen make the determina- tion of this species certain. In Martin’s description (25) his figure 178 is certainly tenuis. In the last paragraph of his text he says “L/aile de l’in- dividu figuré porte exceptionnellement deux rangs de cellules sous la boucle anale.’ Presumably this remark applies to the specimen of which the appendages are figured, figure 179. It is doubtful if figure 179 is really tenuis as it does not agree well and really’much more resembles auricularis. TRIACANTHAGYNA AND GYNACANTHA 29 Moreover Martin’s description of the female appendages as ‘“‘extrémement longs” is hardly applicable, and there is the absence, in the description of the male, of the broad mid-lateral brown thoracic stripe, so I am inclined to believe more than one species is included under tenuis in the de Selys Collection. The wings are hyaline to lightly brown tinged. In the male the margins of the genital fossa on abdominal segment 2, anterior to the point of con- vergence, have on either side from none to thirteen small scale-like teeth. When these teeth are present in numbers they are in a single irregular row. Posterior to the point of convergence there are none to two similar teeth. The auricles on the same segment are each armed with five to nine denticles. The abdominal appendage of the female is linear widening very slightly from the base to about five-sixth the length from which point it tapers symmetrically to the acute apex; maximum width .36-.4. The following color notes were made from the freshly killed Tachira male. Eyes above dark green, shading below into paler green and grayish with black pseudopupillae. Prothorax pale flesh brown, hind lobe brown. Mesepisternum with inner half green, outer half brown, a rich brown meso- thoracic collar, continuous with narrow middorsal brown, this collar nar- rowly bordered in front with green; mesepimeron, metepisternum and metepimeron green; a dark brown, almost black stripe between the mese- pimeron and metepisternum, and a brown stripe between the metepisternum and metepimeron. Three anterior interalar sclerites on each side and those between these, green or greenish; the others blue. Abdominal segment 1 brown, pale basally, darker apically; 2 brown above, a longitudinal mid- dorsal green stripe wider basally and at midlength, and passing apically into a wide transverse trilobed blue band, on either side a basal spot and a median triangular spot, green; 3 and 4 with MD present, small and green; 3 with PD present and blue; 3-8 with ML present, greenish or yellowish; AL, present on 3, blue, reaching the transverse carina; 8-9 with a dorsal median yellowish spot on either side. The following notes were made on the living colors of the female taken at Cristalina. Eyes dark green, almost brown above, paler to greenish gray beneath. Dorsum of thorax green, divided in the middorsal line and bordered below with brown; the dorsal green fades out behind into a wide brown antehumeral stripe; behind this, on the mesepimeron, is a pale greenish stripe three-fourths as wide as the antehumeral brown stripe; behind the green stripe is a very distinct dark brown stripe one-third as wide as the pale stripe anterior to it. Abdomen brown, green and black markings small, not evident on 7-10. As we observed tenuis at La Fria it was on the wing through the day. It frequented dense forests and was usually found hunting about, with apparently relatively weak flight and much poising, in knee-high vegetation, usually of rather thick growth. In such locations it frequently alighted on vertical stems within a few inches of the ground. In the same forest we found Triacanthagyna ditslerti and Gynacantha tibiata, gracilis and mem- branalis. ‘Thinking there might be some twilight fliers, we waited one 30 FE. B. WILLIAMSON evening, near some favorable looking openings in the forest, till it was quite dark without seeing a dragonfly. In this forest palms were dominant, with some exogens, many of large size and a few giant ceibas, and more rarely a small wild cacao, and many vines and scattered areas of heliconias of varying extent. Material examined: Colombia, Cristalina, Dept. Antioquia (February 16, 1917, 1 female, E. B. W.);' Peru, Iquitos (H. S$. Parish, May 11, 1920, I male, E. B. W.), El Encanto, Rio Caraparana, Putumayo Dist. (W. T. M. Forbes, August 25, 1920, 1 male, Cornell) ; Venezuela, Tachira, Dept. Tachira (April 6 and 8, 1920, 2 males, 1 female, E. B. W.), La Fria, Dept. Tachira (April 12-18, 1920, 19 males, 4 females, E. B. W.) ; French Guiana, Tamanoir, Mana River (S. M. Klages, May and June, 1917, 2 males, 3 females, Acc. 6008, Carn. Mus.) ; in the M. C. Z. are 3 males and 3 females; one female bears the label “Essequibo (British Guiana), Schneider’; two females each bear the label “Chapada (Brazil)”; two males bear the label “America,” one an additional label “Charp”; a large blue label, ‘tenuis, n. sp. Hagen,’ and a label “Gynacantha Rambur’’; the remaining male has a single label “176.” Gynacantha caudata Karsch 22, Abdomen male 40-41; hind wing male 40-41; superior appendage male 4.2-4.3; stigma front wing 3. As compared with Karsch’s description of a single male the following may be noted. Face is greenish above. Thorax very similar to that of tenuis, the brown outer part of the mesepisternum wider than the green inner part, and also the dark stripe between the mesepimeron and mete- pisternum wider than in tenuis. Abdomen apparently brown, extreme apex of 1-7 and 10 black; 1 with L, well developed and D probably present; 2 with AD, MD, PD, AML, and PL present, AMI, developed basally well onto the dorsum, PD and PL narrowly separated; margins of genital fossa without spines or teeth, auricles with seven to nine denticles, the anterior ones very small, increasing in size posteriorly; 3 with AL, MD, and ML, clearly present, all reduced, MD and ML joined, both narrow, probably apical spots are present in life on this and succeeding segments, but these cannot be detected on any of these segments in the single preserved specimen before me; 4-6 with AL, MD, and ML present, MD and ML, separated or very narrowly joined, AL small and narrow, not produced posteriorly in its lower part; AL the only spots to be detected on 7. Material examined: Ecuador, San Rafael (F. Campos R., 1 male, ESS NOS. Ve Gynacanthya tibiata Karsch 6, 9, ‘10, 22, 25, 26, 27, 36. Abdomen male 41-50, female 51-53.5; hind wing male 41-50, female 50-53; superior appendage male 5-5.4, female 7-7.5; stigma front wing 3-4.5. TRIACANTHAGYNA AND GYNACANTHA 31 This is the most brilliantly colored of the American Gynacanthas. In keeping with this it is an especially alert and wary species apparently not at all crepuscular in its habits. Dr. Calvert’s maie (9) was taken during the day (before 4 p. m.), and at La Fria it was observed from about 9 a. m. to 4 p.m. Several were seen but in spite of our best efforts we succeeded in taking only a single male, though two days, after the capture of this single male, were practically given to search for this species. At La Fria the heavy tropical forest north of town covered the nearly level surface in a great unbroken stretch of dense verdure. Through this forest, at the season we collected there, odonate life was widely distributed with no con- centrating at suitable spots such as streams or permanent pools. When odonate life is so scattered in the temperate regions, even where agriculture or natural conditions permit easy access to all parts of the dragonfly domain, it is a matter of common knowledge that the capture of specimens is difficult or well nigh impossible. To appreciate some of our difficulties in the search for tibiata, instead of the varied landscape of an agricultural region with its regular fields and wood lots and its section-line roads, where the collector searches possibly in vain for Ophiogomphus, imagine the unbroken and unknown expanse of an almost impenetrable forest with its somber twilight and brooding silence. In such a forest we found ftibiata. One would suddenly appear alighting on a bush twig twenty feet away. A movement would be made towards it and it disappeared as quickly as it had come. The collector resumes his aimless wandering, an hour passes before another one is seen and the experience repeated, or night may come without another one having put in an appearance. The four of us in several days saw less than a dozen specimens. At another season conditions might well be very different. When the rains have started little streams through the forest, and when some of the muddy depressions have become ponds, the capture of tibiata may be a less difficult matter. When we saw them they ranged freely through the forest, flying at varying heights and resting on twigs from four or five feet to ten or twelve feet high without any of the effort at concealment which Gynacantha so often shows and which Dr. Calvert has well described (9, p. 315). Sometimes at rest the abdomen was more nearly horizontal than vertical, and at such times the yellow tipped abdomen and the position strongly suggested a gomphine rather than an aeshnine. Dr. Calvert has described the living male from Peralta, March 23, 1910, as follows: Eyes bright green above, yellowish green below, posteriorly narrowly edged with blue; three horizontal rows of pseudopupillae visible in profile view. Frons above blue on each side of the ‘T-spot, green anteriorly, as also are clypeus, lips, basis of mandibles, thorax, abdominal segment I, basal half of 2 and much of 3-5. Blue as follows: rear of head inferiorly, ‘interalar region of hing wings, auricles, posterior half of 2, and each side of the base of 3, which latter merges gradually into the green. Dark brown or black as follows: rear of head superiorly, a trans- verse stripe just behind the auricles, and another apical one on 2, a mid- dorsal line, an apical transverse stripe, and a transverse stripe at the median carina on 3-7, and a transverse stripe at five-sixths the length on a2 E. B. WILLIAMSON 3-6; 8 and 9g pale reddish brown with a pair of transverse green streaks near the base of each; 10 and appendages yellow; 6 and 7 are greenish anteriorly merging into brown posteriorly. Legs black, first arid second femora inferiorly, and second and third tibiae superiorly, pale green.. When this insect was flying the chief color effects were the green of the head and thorax, and orange at the end of the abdomen due to a blending of the red of 8 and 9g and the yellow of 10 and the appendages. It would poise in the same spot in the air for a minute or so at a time and at five to ten inches above the ground which was a somewhat muddy piece of grassy road. Twice I struck at it and came very near it, but it flew swiftly away only to a short distance, soon returning to a spot near where I| had struck at it. (Many male Anisoptera, following copulation, will return to the spot where they captured their ovipositing mate, certainly associating the place with the capture, and again seeking the female there —E. B. W.) The following notes on living colors were made on the male captured April 13 at La Fria: Eyes bright dark green above, paler im trent, bluish posteriorly, (not rear of head), and gray beneath with one black and several brown pseudopupillae. Labrum almost white, face pale bluish, darkening above; frons above blue, T-spot black; rear of head above narrowly shining black, a broad black band to the foramen, remainder very light blue, darker above, white below. Front lobe of prothorax almost white, middle and hind lobes brown, the latter narrowly green edged behind. Thorax bright green, a small brown area below on the mes-and metepimeron ; middorsal carina narrowly dark reddish brown; interalar sclerites green, except the two posterior median ones and the one on either side between these two on the wing base, which four are bright blue. Abdominal segment 1 green, light brown above on the apical half; 2 green below and beneath the auricles, above green at base, a longitudinal median bar, green in front and shading into blue behind where it passes into a broad transverse bar of blue which is bordered posteriorly with black ; auricles bright blue above; above each auricle is a broad brown area not quite divided by a transverse green bar which is slightly wider above; anteriorly this brown area shades out into the basal green; (none to four small denticles in a scattering row on the converging margins of the genital fossa; auricles with five to six denticles) ; abdomen passing progres- sively posteriorly from a rich brown and bright green coloration to a dull orange color on 8 and 9, and light yellow on 10 and appendages; 3-7 with a very narrow longitudinal middorsal brown stripe which disappears pro- gressively posteriorly; 3-10 beneath yellow; 3 on either side at base blue, remainder above green, shaded posterior to the transverse carina with light reddish brown, on either side a small triangular median and a larger rounded subapical bright green spot; 4 similar, but lacking the basal blue and the posterior brown shading more extensive; 5-7 similar to 4, the area posterior to the transverse carina brown, and therefore progressively posteriorly occupying more of each segment, all with four green spots as described for 3, except the apical spots which have disappeared on 7; 8 light TRIACANTHAGYNA AND GYNACANTHA 33 reddish brown or dull orange with only a trace of the median green spots and the apical spots wanting; sides of 3-8 light golden brown, shaded with darker and with green; base of 3 blue; a distinct green spot below, posterior to the transverse carina on 3-9, faintest on 9. Material examined: \/e.vico, Colima (Rolle, 1913, 1 male, Ris), Cordoba (F. Knab, December 24, 1907, 1 female, U. S. N. M.) ; Costa Rica, Cimarones (C. H. Lankester, April 8, 1 male, A. N. S.), Ontario Farm (C. H. Lanke- ster, September 16 and 17, 2 males, A. N. S.); Venesuela, La Fria, Dept. Tachira (April 13, 1920, 1 male, E. B. W.). Gynacantha jessei, new species Abdomen male 47; hind wing male 43; superior appendage male 5.4; stigma front wing male 3.0. Male.—The following description is based on the single known speci- men. Notes on living colors were made at the time of capture and these notes form the basis of the description, other notes based on the well pre- served specimen being included in parentheses. Eyes above very dark blue, shading into paler dull blue at midheight, the lower third gray. Frons above light yellowish brown, T-spot large, face light greenish brown, labrum blue (labium dingy greenish yellow; labrum dingy blue at base shading out to dingy orange at the margins; anteclypeus dingy greenish yellow; face green in front, bluish on either side next the eyes; frons above and frontal vesicle black, the frons on either side with a narrow transverse blue bar, which is a dorsal continuation of the lateral blue of the face, thus forming a T-spot, with the cross-bar in the midline almost as wide as long, and a stem as wide as long; occiput small, greenish, darker in front, paler and clearer behind; rear of head pale yellowish, narrowly shining black above, and a broad band from this to the foramen). (Front lobe of prothorax pale, middle lobe brownish, hind lobe brownish, pale margined behind. ) Thorax bright green with the following narrowly lined brown: middorsal carina, humeral and first and second lateral sutures; the lower end of the middorsal brown stripe has two cross bars of the same color, the inferior one of these is the longer and has the ends turned upward; metepimeron with the upper third blue, shading below into green; interalar spots between the front wings green, between the hind wings blue; (in the dried specimen the green and blue are fairly well preserved but the brown sutural stripes are no longer evident). Abdominal segment 1 rich brown, apex black, on either side a subapical transverse broad green bar which ends dorsally at the point which marks the inferior termination of the posterior carina; below, the bar curves forward along the ventral margin of the segment and almost reaches the base of the segment; 2 black marked with bright blue as follows: a trans- verse basal ring, narrow in the middorsal line but widening rapidly on the sides and continuing with the blue on the auricles which are broadly edged with black, a median longitudinal bar, arising in the basal ring and ending posteriorly at the level of the posterior carina, a triangular spot on each 34 EK. B. WILLIAMSON side just anterior to the median carina, and a subapical transverse band on either side (the above description of 1 and 2 applies equally well to the oreserved specimen; 2 beneath, anterior to the lateral carina and the auricles largely green, auricles beneath largely black with a broad bar of blue, a dark spot on the margin of the genital fossa opposite and adjacent to the spine of the anterior lamina, margin of the genital fossa, anterior to the point of covergence, very narrowly brown and armed with a single row of seven to nine low scale-like teeth; auricles with eight denticles, the anterior ones very small, increasing in size posteriorly) ; 3-10 above very dark brown to black; 3 has AL, blue, large and shading out above, and MD, PD, ML, and PL green and small; 4 and. 5 have AL, reduced and the four green spots reduced; 6-8 have AL, greatly reduced to a narrow transverse basal line and PL, wanting; PD is almost gone on 6 and does not appear on 7 and 8, and on these two segments MD and ML, are greatly reduced; on the posterior segments spots MD, PD, and PL are green or bluish green but ML, is duller being greenish or bluish yellow; 8-10 and appendages black (all the above markings are more or less discernible in the preserved specimen ; beneath the abdomen is brown or yellowish brown). (Wings hyaline, costa dark, stigma reddish brown, venation black or nearly so; anal loop short and rounded, separated from the anal triangle by two cells; one row of cells throughout between M, and Rs in all the wings.) (Legs entirely black except as follows: first femora broadly green or blue behind for the entire length, and the merest streak of yellow above at the base of the middle tibiae.) Material examined: a single male, the type, taken near Puerto Berrio, Dept. Antioquia, Colombia, January 31, 1917, by Jesse H. Williamson, for whom this beautiful species is named; specimen in coll. E. B. W. In the account of our collecting trip to Colombia (Univ. of Mich., Mus. of Zool., Misc. Publ. No. 3) on the bottom of page 13 and top of page 14, I have described the forest where this specimen was taken. J. H. W. found it hanging on a “vine in the shade in this deep forest. Jessei is a close but very distinct relative of tibiata. Whether it will be found to have the wide distribution of the latter species remains to be seen. Tvibiata has not been taken in Colombia but it has been taken in Venezuela in the nearby Catatumbo River basin and there is no reason why these two hand- some species may not at some future date be found in the same forest. Gynacantha auricularis Martin 23 (as subviridis), 25, 26. Abdomen male 50-52, female 51-55; hind wing male 49-51, female 51-54; superior appendage male 5.4, female 4.2-4.8; stigma front wing 4.2-4.8. The following notes are based on the living colors of the male taken at Palma Sola. Eyes above green, lighter than in T. ditzleri and T. caribbea, shading out into lighter brighter green in front and below into light green- ish yellow with three or four large brown pseudopupillae. Rear of- head TRIACANTHAGYNA AND GyNACANTHA 35 pale flesh, narrowly black above, with an irregular black bar to the foramen. Prothorax flesh or very light brown, hind lobe darker. Thorax above green, a narrow interrupted half collar of light yellowish green, bordered above with a brown transverse stripe three or four times as wide which is narrowly continued up the middorsal carina to spots below the antealar sinus; on either side a broad indefinite brown antehumeral stripe which fades out above in the green, and is darker below; mesepimeron light greenish brown, greener above, shading out to brown below; mete- pisternum and metepimeron pinkish flesh brown, traces of green about the stigma; beneath the same pinkish flesh; sclerites between front wings bright grass green, between hind wings brilliant evanescent blue. Abdom- inal segment 1 flesh, apical third or half brown above; 2 brown above, a narrow longitudinal median green line which passes posteriorly into a broad blue subapical ring, the extreme apex black; basally a narrow transverse green line on either side, widely separated from the median green line; posterior to mid-length on either side a narrow transverse green stripe or line, narrowly bordered black, and less widely separated than the basal lines from the median line; auricles bright blue above, edged with black (margins of genital fossa without denticles or spines; auricles with seven to nine denticles, with several long black hairs on the dorsal surface near or between the basal teeth) ; 3 with a large lateral basal blue spot on either side, and traces of pale light brown spots at the transverse carina, base and apex shaded into darker brown; 4 brown, shading into darker at base and apex, a minute green spot on either side at the tranverse carina; 5-7 brown, shading into darker apically and a darker line at the transverse carina; 8 and 9 slightly darker brown; Io slightly paler than 8 and 9 except at apex; at the transverse carina, on either side of 8 trace of a pale spot. Female abdominal appendages narrow, of nearly uniform width through- out, about .5 wide, narrowed on the inner margin in the basal seventh, the apical seventh tapering to the acute apex, the tapering mostly on the inner edge, the outer edge being relatively straight throughout the length of the appendage. Aeschna viridis Rambur was described from a single male with the last seven abdominal segments gone, and “sans indications de patrie.’”” De Selys (Rev. des Od.) stated that it is an exotic species very different from A. viridis Eversmann, but there is nothing to indicate that de Selys regarded it as a Gynacantha. The next mention of the species is by Kirby (23) who records specimens from Para and Tapajos, referring them to Gynacantha. At my request these specimens were studied by Mr. Campion who kindly sent me one for study. He reports that there is no reason to think any of these specimens were ever seen by de Selys. The male loaned me by Mr. Campion is G. auricularis Martin. If it should be shown that these speci- mens in the British Museum, or any of them, had been compared with Rambur’s type, now lost (Martin, 25), and determined by de Selys as subviridis, then auricularis would become a synonym. However the descrip- tion of subviridis (viridis) seems to exclude this identification and we have in addition Mr. Campion’s opinion that de Selys never saw the Para and 26 E. B. WiLLIAMson Tapajos specimens in the British Museum. The last opinion, that of Ris (36), that subviridis is a synonym of Coryphaeschna luteipennis is as plausible as any. With the loss of the imperfect type, which lacked a locality label, further discussion of the application of the name is futile. The only auricularis we saw alive came flying down a little creek near Palma Sola at the noon hour, circled once or twice the little pool at the edge of which we were seated eating our lunch, and then alighted on some small tree rootlets on the underside of an overhanging washed out creek bank. Here it was impossible to use a net and the dragonfly was carefully stalked and caught in the fingers. Material examined: Costa Rica, probably Ontario Farm (C. H. Lankes- ter, September 18, 1919, 1 female, A. N. S.); Venezuela, Palma Sola, Dept. Falcon (March 6, 1920, 1 male, E. B. W.) ; British Guiana, Bartica (H. S. Parish, May 21, 1901, 1 male, O. S. U.); French Guiana, Pied Saut, Oyapok River (S. M. Klages, Nov., 1917, 1 male, Acc. 6111, Carn, Mus.) ; Brazil, Para (C. F. Baker, 1 male, 1 female, U. S. N. M., 1 female, E. B. W.), Val del Can, Para (Hagmann, May 19, 1901, I female, Ris), Chapada (Gyn. “ft.” number 195, 1 male, M. C. Z.), Tapajos (1 male, British Museum, determined by Kirby as subviridis). Gynacantha klagesi, new species Abdomen male 47-49, female 50; hind wing male 47-48, female 50; superior appendage male 6, female 5.4; stigma front wing 3.6-3.0. Male.—Labium yellow, greenish at the side; face green, yellowish below and against the eyes, darker above; frons above brown in front for more than half its length with a brown bar to the black frontal vesicle, the pale area on either side of the bar light greenish gray; occiput green or yellowish ; rear of head yellowish, narrowly shining black above, with a broad bar of brown or black down to the foramen. Prothorax light brown, posterior border green. Thorax bright green above, middorsal carina and margins of the antealar sinus black or brown, the carina narrowly bordered on either side with brown; a short, broad antehumeral. stripe or area occupying the outer half and lower two thirds of the mesepisternum, darkest along the anterior edge parallel to the carina, and shading out on all the other edges; sides of thorax bright green, a pale brown line on the humeral suture, a short brown line above on the first lateral and a darker brown line on the second lateral suture. Interalar wing spots green anteriorly and blue posteriorly; spots beneath the front wings green, beneath the hind wings blue and yellowish. 3eneath pale flesh. Abdomen brown to black, the apical segments darkest; 1 brown, L large, yellow, shading into blue above, D apparently not present, extreme apex dark to black; 2 with AML yellow below and blue above, auricles below bright yellow with greenish shadings, above blue, broadly surrounded with dark brown or black, the blue not continuous with AMI; AD and MD narrow, green in color, the latter apparently continuous into PD which is joined with PL to form a broad blue transverse bar; transverse carina and TRIACANTHAGYNA AND GYNACANTHA 37 extreme apex black; nine to twelve scale-like teeth in a single row on the posterior half of the converging margins of the genital fossa; eleven to fourteen denticles on the auricles, the most anterior ones the smallest, in- creasing in size posteriorly, the posterior five or six about equal in size; 3 with AL, blue, MD and PD small, apparently green in life, no other spots evident in the preserved specimen; AL and MD small but present on 4-8, ML, not quite so plain but fairly distinct traces of it in all of the same segments ; 9-10 and appendages dark to black; 3-7 light brown beneath, each segment darker at apex; 8 pale beneath, apparently light green in life. Wings hyaline, costa dark brown, stigma brown surrounded with darker veins, venation black or nearly so; one row of cells throughout between M, and Rs; in the four front wings examined the loop in M, is single rowed between M, and M, in two wings and with a single double cell in the other two wings; in the four hind wings likewise the loop is single rowed in two wings and with a single double cell in the other two wings. Legs light reddish brown; first femora darker, especially at apex, and with a broad green or gray stripe behind; second and third tibiae yellow dorsally. Female.—The single specimen is not fully matured and the color is not well preserved. The head is similar to the male. The prothorax and thorax are faded to a uniform unmarked light reddish brown. Abdominal seg- ment 1 brown, darker above posterorly, apically ringed with black; 2 with only MD, PD and PL, evident, the first apparently green, the last two clearly blue, but the abdomen is too faded to determine the color pattern certainly; 1t probably does not differ materially from the male. However AL can be discerned on 3-8, and MD and ML on 3-7; as in the male PD, if present in life, has disappeared due to postmortem changes. The lateral carina on 2 is black and at the level of the transverse carina is elevated in the nearest approach to an auricle attained by the female of any American Gynacantha. The ventral carina on the same segment is also characteristi- cally curved at its posterior end in a manner suggesting the form in the male. At the anterior end of the ventral carina and just above it is a brown spot. The abdominal appendages are linear to about five-ninths the length, where they widen slightly in a symmetrical expansion which tapers sym- metrically to a rather abruptly acute apex; the expanded portion with a maximum width of about .5 and with a low median dorsal longitudinal keel. Wings similar to the male, stigma paler, due doubtless to less mature condition; one row of cells throughout between M, and Rs; in both front wings and one hind wing the loop in M, is single rowed between M, and M,, and in the other hind wing there is one double cell and one forked cell. Legs as in the male. Material examined, French Guiana, Tamanoir, Mana River (S. M. Klages, May and June, 1917, 2 males, t female, Acc. 6008, Carn, Mus., type male, June and allotype female, May). This fine species is named for S. M. Klages, whose collections have added much to our knowledge of neotropical insects. 38 E. B. WILLIAMSON Gynacantha ereagris Gundlach 1%, Abdomen male 43-44.5, female 45-46; hind wing male 42-43, female 43-40; superior appendage male 6; stigma front wing 3.5-4. The above data is from Calvert. I have seen no additional specimens and have nothing to add to his discussion of this species. It has been taken only in Cuba and the Bahamas. Gynacantha mexicana Selys 6:70; "25. 26: Abdomen male 49-52, female 49-51; hind wing male 46-48, female 47-50; superior appendage male 6-6.6, female 7.5-9; stigma front wing 4.6-5.1. Wings hyaline to brown tinged, slightly to distinctly brown at base between C and R-++M, darker between Sc and R+M, as far as or slightly beyond the first antenodal; more or less distinctly yellowish between Sc and R-LM to the nodus, and between C and R beyond the nodus; this longi- tudinal stripe usually most evident in wings otherwise hyaline, but even in some hyaline wings no trace of it exists. In the male there are nine to fifteen scale-like teeth in a row on the ventral carina of segment 2 anterior to the convergence of the margins of the fossa, and teeth are absent or there may be as many as five on the prominence, near the apex of the segment, where the lateral and ventral carinae meet. There are four or five denticles on each auricle. The abdominal appendages of the female have the outer edge relatively straight- ened and the inner curved, the widest point slightly distal to the middle, so the basal half of the appendage is more tapering than the apical half; a very faint longitudinal median dorsal carina; maximum width 1.5; extreme apex acute. The appendage is thus distinctly different from such other species as nervosa and interioris. The following notes on living colors were made from specimens col- lected at Bejuma and Palma Sola. Male——Eyes above dark brown (or dark green), almost black, with a small (or large) green reflecting area in front on either side, which green area may be wanting in the female; below light brown (with a slight yellowish cast), with one large black and about five dark brown pseudopupillae; rear of head pale dull yellow, narrowly black above, with a broad band of black to the foramen. Prothorax light flesh brown, middle lobe with a bluish cast. Thorax above bright (or dull) olive green, clouded with brown; a narrow rich brown mesothoracic half collar which is bordered below by a slightly narrower, narrowly interrupted, bright yellowish green half collar; above the brown half collar the thoracic green is brightest and clearest, and on its outer edges are traces, more or less, of a dorsal brown stripe which is definite only opposite and immediately adjacent’to the green; mesepimeron above green, below brown; metepisternum brown, restricted green above; metepimcron TRIACANTHAGYNA AND GyNACANTHA 39 brown, a trace of yellow at the extreme upper edge; four spots on the side and a latero-ventral stripe black; sclerites between wings bright blue (or those between front wings green). Abdomen reddish brown, 1 paler with an apical narrow transverse blue bar, bordered very narrowly behind with black; 2 with a similar wider blue bar (PD) and narrow green MDs, black borderéd behind, reaching or not to the auricles which are brown, edged with black; 3 with AL small, blue; 3-8 with MD and PD present and small, green in color, smaller and paler progressively posteriorly, PD on 7 and MD and PD on 8 yellowish, 9 with PD only; 3-8 and 10 each with extreme apex black ringed; 3-7 each nar- rowly black edged on the lower lateral margin, PL, more or less distinctly present. Female similar, only slightly duller, thorax above lighter, and AL, MD and PD on 3-7 light dull yellow (probably green in some cases). In flight and environment this species so far as I have observed resembles G. nervosa though it is possibly more wary and slightly more erratic. A Georgetown I took a single female flying alone at twilight in the Botanical Gardens. At Bolivar it was flying at twilight with other species. At Palma Sola it flew along the railroad tracks in the evening with G. nervosa. Between the northeastern part of the town of Bejuma and the river, so called, lying to the north are a number of artificial depressions, some with shallow pools of water, formed by excavations for clay. There is no adja- cent native forest, but there are scattered low bushes, and, at short distances, some trees along the river and in nearby coffee plantings. The mud about the pools is irregularly roughened by the sharp hoofs of visiting burros, and in their deep tracks we found the females of mexicana ovipositing. Yellow breasted flycatchers, resembling kiskeedees, at dusk from nearby bushes watched the pools for the dragonflies. No attempts were made by the birds to capture the swift flying males but when a female alighted and descended into one of the burro tracks there was usually an observant bird ready to take advantage of her helpless position. Alighting over the depression the bird seized the dragonfly at her work several inches below the surface. No other species of Gynacantha was found at these clay diggings. Along the grade for the railroad Y at Fundacion are a number of shallow, muddy pools formed by excavations for the railroad grade. Here we found mexicana flying at dusk, coming from over the adjacent pastures to patrol the pools, returning again to the wider range of pastures. Material examined: Colombia, Bolivar, near Santa Marta (December 21 and 24, 1916, 3 females, EF. B. W.), Fundacion, Dept. Magdalena (Janu- ary 9 and I0, 1917, 2 males, 1 female, FE. B. W.), Ecuador, San Rafael (F. Campos R., 1 male, 1 female, A. N. S.), Babahoyo (F. Campos R., t male, A. N.S.) ; Venezuela, Palma Sola, Dept. Falcon (March 7-10, 1920, 5 males, 3 females, KE. B. W.), Bejuma, Dept. Carabobo (February 12-23, 1920, 7 males, 5 females, EF. B. W.), El Guayabo, Dept. Zulia (April 20, 1920, 1 female, FE. B. W.) ; British Guiana, Georgetown (February 18, 1912, FP temale BW.) : Brazil; Para (C. F. Baker, 1 male; FE. B: W.). 40 E. B. WILLIAMSON Gynacantha nervosa Rambur 6,-7,.11,. 12, 19, 2D, 23, 25, 20; 31, 35,90. Abdomen male 50-54, female 52.5-57; hind wing male 47-54, female 5-50; superior appendage male 6.5-7.5, female 5.4-6.3; stigma front wing 4.5-5.7. Wings hyaline to brown tinged; in some cases, and this seems especially true of tenerals or younger individuals, irrespective of locality, there is a faint yellowish longitudinal stripe, which suggests the wing coloration of bifida, but in such specimens of nervosa the yellow line is not darkened and more conspicuous basally as it is in bifida. The margins of the genital fossa of segment 2 in the male are armed on each side, anterior to the point of convergence, with eleven to nineteen scale-like denticles, mostly in a single row, about twelve being the most usual number ; and the auricles are armed with five to eight denticles, five or six being the usual number. The female abdominal appendages are linear at the base, the apical third or more slightly widened, suggesting in a way a modified male appendage of this group; the expanded portion with a dorsal median longitudinal carina or low keel; maximum width .6; extreme apex acute. Dr. Calvert has kindly furnished me with notes on living colors of two Costa Rican males. Where these descriptions are not in accord the differ- ences shown are indicated in parentheses. Eyes olive green above, very pale brown below (dark brown .above, bordered anteriorly with a line of pale green, below which the eye is pale brown) with distinct black pseudopupillae ; face and lips very pale brown; frons superiorly pale green with a black T-spot; vertex black with a pair of very small greenish spots; occiput pale bright green; rear of head very pale green, eye margins black. Dorsum of thorax pale green, sides with some green but mostly (also pectus and legs) pale brown (most of sides of thorax and pectus very pale lilaceous ). Dorsum of abdomen dark brown with small (pale) green dorsal spots (or transverse lines) as follows: a dorsal apical transverse green stripe on I; a middorsal longitudinal green line on 2; three transverse stripes, all inter- rupted middorsally, at anterior end, middle, and posterior end of 2; 3-8 similar to 2, lacking the longitudinal green stripe, but the transverse stripes mere lines, the middle stripe at the transverse suture (the apical spots or interrupted stripes absent or almost so on 7 and 8); 9 with (without) apical spots; 9 and 10 with (without) traces of the anterior spots; ventral surface of abdomen pale lilaceous (probably usually brown, E. B. W.), a pale brown longitudinal stripe on ventral part of each tergite of 2-8. The following notes were made on living colors of specimens taken at Palma Sola. Male.—Eyes above dark green, a blue transverse line in front on either side, shading below into pale yellowish brown with six to eight or more brown pseudopupillae; rear of head narrowly black above, remainder light dull yellow. Prothorax flesh colored, middle lobe brown. Thorax above with a very narrow light yellow mesothoracic collar, this collar bordered above and below with light reddish brown, dorsum green, rather dull, indefinite antehumeral brown areas fading out above; mesepimeron and 52 TRIACANTHAGYNA AND GYNACANTHA 4l metepisternum brown, greenish above, metepimeron paler; sides of thorax with small distinct brown or black spots; beneath light flesh brown, whitish; interalar spots green. Abdomen brown; 1 pale at base with a brown basal spot above on either side and a narrow subapical bluish green transverse band, the extreme apex black; 2 with a widely interrupted narrow transverse basal green band, a less interrupted oblique green band at the transverse carina, which carina is black, a wider uninterrupted but narrowed, subapical blue band, the extreme apex black, and a narrow longitudinal median green stripe which posteriorly passes into the apical blue; (J. H. Williamson noted on a Palmdale, Florida, male: spots between wings and on basal ab- dominal segments, green) ; auricles above bluish gray, edged black; 3-8 each with a very small basal spot on either side, minute on 7 and 8; 3-9 each with a small triangular spot on each side at the transverse carina, these spots basal on 9; 3-8 and 10 narrowly black at apex, 10 with a small basal median spot; lower lateral margins of 3-9 pale, this pale area interrupted on each segment at the tranverse carina and by a downward projection of the dorsal brown between the transverse carina and the apex, thus dividing the pale lateral area of each segment into three areas or spots, the posterior one of these areas becomes progressively smaller posteriorly and wanting on 6 and posterior to that segment. Female.—Similar to male but duller; lower lateral edge of abdomen paler, the posterior one of the three pale areas on each segment present on 3-7, and the sides below of 8-10 largely pale. The following less detailed notes were made from slightly teneral males taken at Santa Marta: eyes above and thorax above greenish brown; narrow obscure short brown thoracic stripes on mid-mesepisternum; thorax below and behind brown; wing bases and markings on 1 and 2 blue. Abdomen clove brown, markings black and light dull yellow. And of an adult male taken at Bolivar I have the following note: thoracic dorsum greenish brown, sides light brown; abdomen darker brown; 1 and 2 marked blue, other seg- ments marked black and yellowish. G. nervosa seems essentially crepuscular in its flight. Where the species occurs abundantly the numbers on the wing and in sight at once, the mobile active flight, and the rapidly coming darkness of the tropical night combine to form a scene to fire the imagination. The collector has spent the day in the forest closely surrounding the little village of thatched huts which he knows as “home,” and among the hundreds of dragonflies flying about woodland pools and streams no nervosa have been seen. At sundown, muddy, wet and tired he returns home. Suddenly a large brown dragonfly goes with undulating flight down the village path before him. In a bit of near-by garden another with lower more direct flight appears. Then three or four are seen circling about a thatched hut and at once the garden and path are alive with interweaving forms and the flight is on. They come from everywhere, the air is filled with them, some fly erratically, others patrol regular beats, apparent spots of greater density lure the collector from one point to another. As suddenly as they appeared, only a few are seen, and then they are gone, and the disappointed collector with possibly only two or three specimens in his bottle, realizes that the twenty to thirty 42 E. B. WiILLiAMSON minute flight is at its end, and that he will not see nervosa again for twenty- four hours. ; That the flight is not always confined to the evening however is shown by a note by Klages at Cayenne, French Guiana: “Flies at dusk and at dawn only; captured in mangrove swamp.” Neither is the flight confined to villages, but so far as I have observed numbers are found only about clear- ings or extensive open places, and it is probable that in such places, in _ proximity to houses, cattle, grasses and freshly hewed timbers, nervosa finds its most abundant food supply. Abundance of food supply may be a factor in determining the number of flights in each twenty-four hour period or the amount of activity during the day. The number of specimens recorded below from Cayenne, where two flights a day were observed, and from Palma Sola, where there was a single flight, may possibly indicate a differ- ence in abundance of nervosa on the wing and a difference in the abundance of the food supply. In dark places in the forest with presumably different insect prey one might expect to find nervosa on the wing throughout the day and our observations seem to bear this out. In the forests about Palma Sola, where most wonderful evening flights occurred, the four of us, ranging the forest every day, never saw a nervosa till the evening flight began. At Puerto Berrio, where we did not see nervosa about the town, we took a single male at 9 A. M. one day flying in brush in dark forest about five kilometers from town. J. H. W. noted at Palmdale, Florida: nervosa found in darkest part of cypress grove in creek bottom, flying about bases of trees, or hanging up from three to five feet above the ground. As might be expected from its habits at certain times and places of flying closely about buildings, nervosa occasionally and apparently accidentally, enters such build- ings and is rarely entrapped and captured there. At Cristalina, Colombia, about noon one day I saw a Gynacantha fly out from some brush far ahead of me along the creek. It fluttered along in a helpless manner and in attempting to return to the brush, fell into the water from which I picked it. It was a male of G. nervosa and on the dorsum of abdominal segments 2 and 3 were six small white eggs. This specimen was sent to Miss Currie at Washington, and the eggs were identified as those of some diptera, apparently a tachinid. There is no previous record of tachinids being parasitic on dragonflies. ‘The same day another nervosa, apparently ill also, was seen fluttering through the brush with hanging abdomen, but I lost sight of it and did not capture it. A male, also taken at Cristalina, had in its mouth an insect identified by Mr. McAtee as a cicadellid. I have observed nervosa ovipositing on two occasions. At Maraquita, Colombia, at the edge of town there was a much used water tap, the overflow water being drained in an artificial ditch with steep dirt sides. About 6 P. M. several females were observed ovipositing in the soil on the banks of this ditch. In Trinidad a female was taken about noon ovipositing in the damp but hard earth of a wet-weather stream bed in low forest. Stauro- phlebia in larger numbers were ovipositing at the same place. Material examined: California (1 female, A. N. S.); Florida (W. H. Finn; ‘Coll Cy Vo Riley a imales US) ON iS) Cre tenale amie roe. Paradise Key, Everglades of Dade Co. (C. A. Mosier, November, 1917, 1 TRIACANTHAGYNA AND GYNACANTHA 43 male, U. S. N. M.; shot in dense hammock 18 miles S..W. of Paradise Key, H. S. Barber, March 11, 1917, 1 female, U. S. N. M.), South Jacksonville (W. T. Davis, November 3, 1911, 2 females, W. T. D.), Lakeland (W. T. Davis, November 8, 1911, 2 females, W. T. D.), Punta Gorda (W. T. Davis, November 15, 1911, 1 female, W. T. D.), Gulfport (G. A. Reynolds, June, 1914, I male, Ris), Palmdale (J. H. Williamson, April 5, 1921, 3 males, I female, E. B. W.), St. Petersburg (Mrs. Chas. C. Deam, October 6, 1907, I male, E. B. W.), Miami (S. N. and M. C. Rhoads, January 16-24, 1899, 3 males, 2 females, E. B. W.) ; Guatemala, Santa Lucia (February 1, 1905, I male, 1 female, E. B. W.); Costa Rica, Liberia (J. F. Tristan, room in schoolhouse, January 12, 1910, 1 male, A. N. S.), Caché (C. H. Lankaster, caught in office, March 19, 1910, 1 female, A. N. 8.) ; Canal Zone, camp at Empire (Lieut. G. C. Dunham, 1917, 1 female, U. S. N. M.); Colombia, Puerto Colombia, Dept. Bolivar (December 11, 1916, 1 male, E. B. W.), Santa Marta, Dept. Magdalena (December 13-16, 1916, 4 males, E. B. W.), Bolivar, near Santa Marta, (December 21-26, 1916, 2 males, 4 females, E. B. W.), Don Diego, 100 ft., and Bonda, 250 feet, Dept. Magdalena (H. H. Smith, 1 male, 2 females, Carn. Mus.), Rio Frio, Dept. Magdalena (January 7, 1917, 1 female, E. B. W.), Fundacion, Dept. Magdalena (Jan. 9 and 13, 1917, 1 male, 1 female, E. B. W.), Puerto Berrio, Dept. Antioquia (February 21, 1917, 1 male, E. B. W.), Cristalina, Dept. Antioquia (February. 11-20, 1917, 20 males, 9 females, KE. B. W.), Mara- quita, Dept. Tolima (February 4 and 5,.1917, 2 females, E. B. W.) ; Ecuador, Babahoyo (F. Campos R., 1 male, 3 females, A. N. S.), Guayaquil (F. Campos R., 4 males, 2 females, A. N. S., 2 males, 1 female, Ris) ; Bolivia, Province del Sara, Dept. Santa Cruz, 350-450 meters (José Steinbach, I male, 3 females, Accs. 5076, 5574 and 6443, Carn. Mus.) ; Puerto Suarez (José Steinbach, November, 1908-January, 1909, 1 female, Acc. 3842, Carn. Mus.) ; Venezuela (Appun, 1 male, M. C. Z.), Palma Sola, Dept. Falcon (March 4-10, 1920, 4 males, 13 females, E. B. W.), La Fria, Dept. Tachira (April 12-16, 1920, 7 males, 4 females, EF. B. W.), Tachira, Dept. Tachira (April 10, 1920, 1 female, E. B. W.) ; British Guiana, Rockstone (February I, 1912, 1 female, E. B. W.) ; Dutch Guiana, Paramaraibo (K. Mayo, 1 male, A. N. S.); French Guiana, Cayenne (S. M. Klages, March, 1917, 3 males, Acc. 5897, Carn. Mus.), Tamanoir, Mana River (S. M. Klages, May, 1917, 2 males, Acc. 6008, Carn. Mus.) ; Brazil, (1 female labelled “Mus. Berol,” and “robusta,” M. C. Z.), Chapada (1 male, M. C. Z.), Cachoeira (1 female, MC. Z:);Trinidad, Baracon, Chaquanas (‘March 7; 1912, 1 female, E. B. W.); Cuba (Poey, 1 female, M. C.-Z.), S. Diego d. 1. Banos (Palmer and Riley, April, 1 male, U. S. N. M.); Hayti, Samana (Frazar, 1 male, M. C. Z.); Jamaica (C. W. Johnson, 1 female, A. N. §.). Gynacantha bifida Rambur 5, 6, 25, 26, 31, 33, 35, 36, 30. Abdomen male 51-52, female 53-54; hind wing male 51-52, female 54-55; superior appendages male 7.2, female 6; stigma front wing 5.1-5.7. ‘In the male there are eleven to thirteen scale-like teeth in a single row on each side of the converging margins of the genital fossa of the second 44 E. B. WILLIAMSON abdominal segment. The auricles are armed with five to seven denticles. The female abdominal appendages are narrow, slightly widening on the inner edge from the base to about two-thirds the length where the maximum width is about .6; the median keel on the apical third low and inconspicuous ; apical third tapering symmetrically to the apex, with the extreme apex acute. Dr. Ris (35) has studied this species carefully and places G. robusta Kolbe as a synonym. Material examined: Brazil (through Scheider, Berlin, 1 female, Ris), Minas Geraes (Rolle, 1914, 1 female, Ris), Bom Jesus de Itabapoana, Rio Janeiro (Zikan, January 21, 1905, 1 female, Ris), Salto Grande, Rio Par- anopanéma, Sao Paulo (J. D. Haseman, October 21, 1908, 1 male, Acc. 3202, Carn. Mus.), Porto Catherina de Santa Leopoldina (2 males, EK. B. W.) ; Argentina, Yuto, Jujuy, 450 meters (Joergensen, April, 1911, 1 male, Ris). Gynacantha croceipennis Martin , 24, 25, 26, 36. Abdomen male 55, female 60; hind wing male 54, female 57; superior appendage male 7; stigma front wing 6. I have seen only the material from Peru and Bolivia in Dr. Ris’ collection, carefully studied by him (36). The young female from Pozuzo is the only Peruvian specimen; it is smaller than the others, abdomen 55 and hind wing 55 (not 52, as stated). Teneral specimens of bifida some- times show some yellowing of the wings and it is not impossible that large series of bifida would show that species to be as variable as nervosa 1s known to be in this character. The same thing may be true of all the nervosa group, including imterioris to which species I am inclined to refer this Pozuzo female, though I have seen no other specimens referred to that species as darkly tinged as the Pozuzo female. The specific distinctness of croceipennis is open to serious question. In the males, wing color, a dubious character in this connection, alone separates it from bifida; and in the females the same unreliable character alone separates it from bifida and interioris. At the same time but Little material is known and better char- acters may be discovered later. Because of its geographical range and the greater elevation at which it occurs as compared with bifida, the name should stand for the present, though it is not improbable that a condition analagous to that found in certain species of Hetaerina is duplicated here, for we know some Hetaerinas of wide range, which, in the mountains, at- tain a size and a depth and extent of wing coloration unrivalled by the same species in lower situations. Gynacantha litoralis, new species Abdomen male 50; hind wing male 48; superior appendage male 6.6; stigma front wing male 4.8. Male.—Labium and face pale yellowish brown; frons and frontal vesicle black, a pale brown median transverse bar on either side of the frons, wider than in imterioris, and with its fellow of the opposite side defining a con- TRIACANTHAGYNA AND GyNACANTHA 45 spicuous T-spot; occiput yellow or greenish yellow; rear of head as in interioris. Prothorax and thorax as in interioris. Abdomen faded, apparently similar to that of interioris but darker, so the lateral carinae on 4-7 are not conspicuously darker than surrounding areas ; the four lateral brown spots on 2, described in imterioris, not as distinct as in that species ; auricles each with six denticles; about ten scale-like teeth in a single row on each side of the genital fossa anterior to the point of convergence; PL, apparently wanting on 5-7 and other lateral spots appar- ently reduced on these segments, but these details cannot be certainly de- termined from the dried specimen. Appendages brown. Wings slightly clouded at the extreme base, costa and stigma light yellowish brown, the end and posterior veins of the latter darker; venation generally a light reddish brown, giving the wing a ruddy apearance at certain angles. Legs similar to those of iterioris, the tarsi less darkened apically. The following notes were made 9n the recently captured male: Eyes above dark green, blue across the middle at the level of the frons, below drab. Dorsum of thorax pale green over a limited area, shading out above and laterally, sides dark flesh or light brown with shadings of green beneath the wings; spots between front wings green, between hind wings blue. Abdomen dark brown, nearly black, 1 and 2 paler, (sub) apical blue rings on I and 2 and an interrupted basal blue ring on 3; dull drab lateral spots on 3-9; beneath pale brown, darkest on 4-6. Flying at twilight along a dry ditch near the botanical gardens. Dr. Calvert has seen the single male and in his opinion it represents an undescribed species. He compared it with ereagris which has simpler venation and superior appendages less widened apically. Material examined: Dutch Guiana, Paramaribo (February 22, 1912, a single male, the type, E. B. W.). With some question I have referred also to this species two males and two females from Manaos, Brazil, collected by Miss H. B. Merrill, and in the U. S. N. Mus. One of these specimens bears the following note: Lake near Manaos, Santa Maria, February 22, 1908; first three segments of abdomen with spots of sky blue; eyes blue; sides of thorax green and brown; abdomen brown. In general appearance these four specimens are strikingly different from the type from Paramaribo, but I have found it impossible, with the limited material, to satisfactorily separate them. The Brazilian specimens are larger; abdomen male 53, female 56; hind wing male 52, female 55; superior appendages male 6.7; stigma front wing 5.4-5.6; and the wing venation is much more complex. For example, in addition to characters indicated in the tabulation of venational characters, in the Brazilian specimens there are two rows of cells between 'M, and Rs adjacent to the forking of Rs; in the type there is one row of cells throughout between M2 and Rs. In the Brazilian specimens, moreover, the wings of all are uniformly brown-tinged throughout and the veins are dark colored. To this character and to the larger size are due the striking dissimilarity in general apearance of the 46 E. B. WILLIAMSON specimens from the two localities. At the same time the differences in complexity of venation, the coloring of the wings and the size offer no specific characters and I have been unable to detect others. Gynacantha interioris, new species Abdomen male 54-56, female 54-57; hind wing male 51-52, female 54-56; superior appendage male 7.2-8.1, female 7.5; stigma front wing 5.4-6.3. Male and female——Labium and face pale yellowish brown; more or less darker above; frons above and frontal vesicle black, a pale brown median transverse bar on either side of the frons which thus, with its fellow of the opposite side, defines a large conspicuous T-spot; occiput greenish yellow; rear of head light brown or yellowish brown, above narrowly shining black adjacent to the eyes with a brown or black bar to the foramen. Prothorax light brown, the middle and hind lobes indefinitely shaded with brown, the front lobe almost white. Thorax brown, darker to almost black with or without greenish re- flections on the middorsum; metastigma surrounded with black, a small brown spot above it, another brown spot on the mesepimeron at the upper end of the second lateral suture, and a more or less distinct spot posteriorly on the latero-ventral carina. Abdomen brown; I with a subapical dorsal transverse blue bar, bordered behind and below with black; 2 with AD the length of the segment, and MD and PD present, apparently all blue in life, sides obscure brown, bright colors not evident in dried material, in the male a brown spot just above the auricle near its middle, a round spot posterior to this a distance slightly more than the diameter of the anterior spot, a brown area below this posterior spot in the angle of the auricle, and a brown spot near the ventro-anterior margin of the segment; auricle blue above, black edged, armed with five to eight denticles; the ventral margin of the genital fossa, anterior to the point of convergence, armed with a single row of eleven to thirteen scale- like teeth; 1-8 narrowly black ringed at apex; 3-8 with MD and probably ML present on all, and 3-7 apparently with AL, ML and PL present on each, MUL especially large and distinct; in life these spots are probably greenish yellow or, in the case of the lateral spots, yellowish; lateral carina black on 4-7; appendages brown or black. Appendages of female, narrow at base, slightly more than the apical half expanded, suggesting a modified male ap- pendage of this group, expanded portion with a dorsal median longitudinal carina or low keel; maximum width .8-.9; apex tapering, acute. Wings hyaline to brownish tinged, with a longitudinal dark stripe, especially conspicuous in the hyaline wings. Costa light yellowish brown; the stigma the same color, darker above, the enclosing end and posterior veins black; venation black or nearly so. First femur dark brown, shading to black at apex, a broad pale stripe behind; second and third femora pale reddish yellow; tibiae and tarsi pale yellow, paler than the second and third femora, the tarsi dark to black at the apex adjacent to the dark reddish brown tarsal claws. TRIACANTHAGYNA AND GYNACANTHA 47 J. H. W. noted of the male taken by himself on June 6 at Campamiento, Colonia del Perené, elevation 680 meters: Twilight flier at the crossing of the trail and Quebrada Repressa, 5:50 P. M., rare and difficult to catch. The female we took at Tachira, elevation 364 meters, was flying erratically along the railroad track when it was nearly dark. Material examined: l’enezuela, Vachira, Dept. Tachira (April 10, 1920, 1 female, KE. B. W.) ; Peru, Campamiento, Colonia del Perené (J. H. William- son, June 4-6, 1920, 3 males, 1 female, EK. B. W.), San Ramon (J. C. Bradley, June 16, 1920, I female, Cornell), Yurimaguas (H. S. Parish, April 3, 1920, 1 female, E. B. W.); Brazil, Santarem (S. M. Klages, May, 1919, I female, Acc. 6324, Carn. Mus.). Type male and allotype female, Campamiento, Peru, June 5 and June 4, 1920, respectively, E. B. W. For a possible additional record see text under croceipennis. Gynacantha gracilis Burmeister 3, 6, 8, 9, 10, 11, 20, 25, 26, 38. Abdomen male 60-62, female 58-62; hind wing male 57-62, female 57-63; superior appendage male 6.4-8.2, female 8.2-9.4; stigma front wing 5-6.2. Female appendages with the outer edge relatively straight, inner edge widening from the slender base to about two-thirds the length of the ap- pendage where the maximum width is about 1 or 1.1, from which it tapers gradually to the acute apex; a low median dorsal longitudinal carina is more or less evident. The following notes on living colors were kindly given me by Dr. Calvert: Male, forest near highest reservoir, upper Banana River, Costa Rica, No- vember 9. Eyes ranging from blue above through green adjoining the genae, yellow to reddish brown along the postero-inferior margin. Frons brown with a superior black spot; nasus olive; labrum and external surface of mandibles yellowish; labium pale brownish green. Thorax chiefly grass green with an ill-defined brown antehumeral stripe, a narrow brown stripe on the second lateral suture, and a superior brown stripe on the mete- pisternum; pectus pale brown. Abdominal segment 1 luteous with a posterior dorsal blackish spot; 2 dark reddish brown, a middorsal line, a transverse median line and a posterior transverse stripe blue, and all interrupted ; auricles blue above, black margined; 3-10 dark brown, almost black; 3 flesh colored laterally on the basal two-thirds, with a basal blue spot on each side; a fainter blue spot just posterior to the median transverse carina on each side of 3-8, becoming fainter on each successive segment; ventral surfaces of 3-10 pale reddish brown with a trace of blue near the middle of each. Legs dark red, knees and tarsi black. Material examined: No locality (Mus. Berol., 1 female, M. C. Z.); Costa Rica, probably Ontario Farm (C. H. Lankester, September 15 and 18, 1919, I male, 1 female, A. N. S.) ; Ecuador, Quevedo (F. Campos R., 1 fe- male, A. N. S.); Bolivia, Rio Yapani, 650 m., Dept. Santa Cruz (José Steinbach, 1914-1915, 1 female, Acc. 5574, Carn. Mus.); Venezuela, La Fria, Dept. Tachira (H. B. Baker, April 12, 1920, 1 male, E. B. W.); British Guiana, Kartabo, Bartica Dist. (W. T. M. Forbes, October, 1920, 48 E. B. WILLIAMSON 2 males, one “in shady path,” Cornell); Dutch Guiana (Mus. Berol., 1 male, M. C. Z.); Brazil, Tapajos, Mte. Christo and Barreiras (A. H. Fass], May, 1920, 3 males, Ris), Rio Janiero, Bom Jesus de Itabapoana (Zikan, November 23, 1904, 1 female, Ris). Gynacantha membranalis Karsch 6, 9, 20, 21, 25, 26, 27, 30 (as jubilaris), 36, 37. Abdomen male 55-63, female 61.5-65; hind wing male 55-60, female 59-64; superior appendage male 6-6.5, female 6.9-7.2; stigma front wing 5.5-6.6. The auricle in the male has only three, more rarely four, denticles, and between the two posterior (basal) teeth and on the dorsal surface of the auricle are a number of long black hairs. Superior appendage of female narrow, the outer edge nearly straight, widening very gradually on the inner edge from the base to beyond the middle where the maximum width is about -9; from this point tapering gradually in the apical third to the acute apex; a low dorsal median longitudinal keel usually about half as long as the appendage itself and with its extremities about equally distant from the base and the apex of the appendages or more extended on the apex. Among the females, and especially among the more recently emerged of these, are individuals with the costal area of the wing colored dark brown, the color continuous basally with the usual basal colored area; this brown longitudinal stripe is bounded posteriorly before the nodus by R and after the nodus by M,. Between females with this region darkly marked and females with the same area hyaline, all degrees of coloring exist. Of the Guacimo, Costa Rica, male, Dr. Calvert noted: “Eyes in life brilliant metallic green.” The single male from Rockstone, British Guiana, may owe its dark color to its recently emerged condition, though the blue color on segment 2 was bright and clear. My brief notes made on this specimen are: Eyes and thorax above black, sides of thorax black and light brown; abdomen above black; sides of 3-6, below the transverse carina, pale; 2 with the auricles and spot above bright clear blue. The thorax of this specimen showed no trace of green which was the dominant color in the Venezuelan specimens described below, and the dorsum especially was a deep velvety black. The specimen is fresh, but apparently mature; equally recently emerged specimens from Venezuela are green, and I believe the difference is racial rather than individual. The following notes were made from a San Esteban male: Eyes dark green, almost black above, gray below. Prothorax brown with a short transverse green bar near the posterior border. Mesepisternum brilliant green with a black bar, giving the effect of two dorsal green stripes on either side of the dorsum; black bar darkest below, shading out above; mesepimeron slightly but very little duller than the mesepisternum; a black bar on the second lateral suture, shading out over the metepimeron to brown, with the upper and posterior edges of the sclerite green. Segment 1 largely green, dorsum apically brown; 2 black, auricles above, a narrow transverse basal line on either side, a narrow longitudinal middorsal line, and a large transverse subapical spot TRIACANTHAGYNA AND GYNACANTHA 49 on either side, bright blue; sides of 3 below from base to slightly posterior to the transverse carina duller blue; 3-5 each with a small triangular blue or green dorsal spot on either side at the transverse carina, the two spots on each segment very narrowly separated; otherwise the abdomen above and on the sides is dark brown or black. Femora rich reddish brown, black at apices. The following notes were made from a San Esteban female: Eyes very dark brown above, gray beneath. Mesothorax and metepisternum green, the last gray behind along the suture; dorsum of thorax darker green with a black bar on each side and black on the middorsal carina and border- ing the antealar sinus in front; metepimeron black, gray behind and below; black between the front wings, posteriorly blue spotted between the wings. Segment 1 light brown, dark above apically; 2 largely black above with a blue subapical spot on either side and traces of a longitudinal middorsal stripe; 3-6 with a small spot on either side at the tranverse carina, these spots progressively smaller posteriorly, and the spots on each segment very narrowly separated. Membranalis is a wide ranging and adaptable species. In Venezuela, after collecting it on the high rocky quebradas back of San Esteban, associ- ated with such things as Hetaerina, Cora, and Heteragrion, we were sur- prised to find it equally at home about the small scattered muddy spots in the heavy low-lying forests about La Fria, where its dragonfly associates were such things as Lestes, Metaleptobasis, and Orthemis. The San Esteban specimens were all freshly emerged, and their recent advent into aerial life may account for the frequently observed high soaring, back and forth over some quebrada, which opened a path through the forest for their flight. On February 5 at 4 P.M. several were seen soaring high above the reach of an insect net and an effort was made to bring them down with our revolvers loaded with dust shot. We crippled two, but they darted downward into the forest and we could not find them. The next day J. H. W. at the same place, flushed a specimen, which he easily caught, and which proved, by the shot marks, to have been one of our targets of the day before. On February 9, we observed that membranalis was on the wing at all times from about 9 A. M., when we reached the quebrada above Las Quiggas where we spent the day, till 4 P. M. when we left. About 3:30 P. M. possibly eight or ten were patrolling a stretch of quebrada about two hundred feet long where an extensive land slide had opened the stream to the sun. Five of these were captured, one male and four females, and all were recently emerged. On several occasions we observed males, apparently in search of females, flying about rocky pools of crystal water high up the quebradas back of San Esteban. And at La Fria a male taken near a pool of the consistency of batter or gravy, lying in a great expanse of level forest, had parts of the abdomen and wings coated with a wash of light reddish earth, probably due to an attack on an ovipositing female. The male taken at Rockstone, British Guiana, was captured just after sundown patrolling a stagnant pool in a mud-bottomed creek in which the water 50 E. B. WILLIAMSON had ceased to flow. At this stream and nowhere else have we found membranalis and Staurophlebia associated. As Ris (36) points out, G. jubilaris Navas is a synonym of membranalis. Material examined: Costa Rica, Guacimo (in forest, P. P. Calvert, June 6, 1909, I male, A. N. S.), Alajuela (D. E. Harrower, August 2, 1915, 1 male, A. N. S.) ; Panama (Hassler, 1 male, M. C. Z.) ; Colombia (Appun, 1 female, M. C. Z.), Sta. Fe de Bogota (Lindig, 1 male, M. C. Z.), Rio Negro, East Colombia (A. H. Fassel, 1911, 1 male, 800 m., 1 female, 500 m., Ris); Bolivia, Prov. del Sara, 450 m. (J. Steinbach, 1 female, Acc. 4547, Carn. Mus.) ; Peru, Iquitos (1 female, M. C. Z.; W. T. M. Forbes, August 4, 1920, I male, Cornell); Venezuela (Appun, 1 male, M. C. Z.), San Esteban, Dept. Carabobo (February 1, 6, 8 and 9, 1920, 3 males, 9 females, RAB W.). lLasEna: Dept. Tachirai40 ms CApril12, 13) 05,18 andeen, 1920, 3 males, 1 female, and 1 male and 3 females by H. B. Baker, E. B. W.); British Guiana (1 male, Cornell), Bartica (H. S. Parish, May 15, 1901, I male, O. S. U.), Rockstone, (February 2, 1912, 1 male, E. B. W.); French Guiana, Tamanoir, Mana River (S. M. Klages, May, 1917, 2 males, 1 female, Acc. 6008, Carn. Mus.) ; Brazil Benevides, Para (S. M. Klages, October, 1918, 1 male, 1 female, Acc. 6174, Carn. Mus.). Nore In the U. S. N. M. is a male Gynacantha labelled Crowley, Louisiana, July 17, 1911, E. S. Tucker, collector. This is very close, if not identical to the widely distributed oriental G. hyalina. Miss Currie feels certain the locality label is correct. Mr. Tucker has no recollection of the specimen. If it was really taken in Louisiana it is probable it was brought to this coun- try as an egg or larva and it is improbable that an undescribed Gynacantha inhabits Louisiana. BIBLIOGRAPHY The following papers not cited in Kirby’s Catalogue have been con- sulted in the preparation of this paper. Numbered titles contain references to American species. Unnumbered titles contain more general references to genera or to other than American species. C. S. BRIMLEY 1. Notes on the Odonata and Other Insects of Lake Ellis, North Carolina. Ent. News, March, 1906, pp. 81-85. ; HorrEeNSE BUTLER 2. The Labium of the Odonata. Trans. Am, Ent. Soc., XXX, 1904, pp. I11-134, Puinie P, CALVERT 3. Burmeister’s Types of Odonata. Trans. Am. Ent. Soc., XXV, 1808, pp. 27-104. 4. Odonata from Tepic, Mexico, with Supplementary Notes on those of Baja, California. Proc. Cal. Acad. Sci., I, 1899, pp. 371-418. TRIACANTHAGYNA AND GYNACANTHA 51 ean A Contribution to Knowledge of the Odonata of Paraguay. Anal. Mus. Nac. Bs. As., VII, 1899, pp. 25-35. 6. Biologia Centrali-Americana, 1901-1908, pp. 17-420. Ent. News, XVII, 1906, p. 102. Contributions to a Knowledge of the Odonata of the Neotropical Region, Exclusive of Mexico and Central America. Annals of the Carnegie Museum, VI, 1909, pp. 73-280. 8. Notes at meeting of the Ent. Soc., Acad. Nat. Sci., Phila, May 28, 1914. Ent. News, XXV, 1914, pp 479-480. N 9. A Year of Costa Rican Natural History. Macmillan, 1917. 10. Odonata Anisoptera from Guatemala. Ent. News, XXX, 1919, pp. 31-38. 11. Gundlach’s Work on the Odonata of Cuba. Trans. Am. Ent. Soc., XLV, I91Q, PP. 335-392. HERBERT CAMPION The Percy Sladen Trust Expedition to the Indian Ocean in 1905, No. XXVII. —Odonata. Trans. Linn, Soc. of London, XV, 1913, pp. 435-440. Grorck H. CARPENTER 12. A Contribution Towards a List of the Dragonflies of Jamaica. Jour. Inst., Jamaica, II, 1896, pp. 259-263. The Geographical Distribution of Dragonflies. Scien. Proc. R. D. S., VIII, 1897, pp. 439-468. T. D. A. CocKERELL Descriptions of Tertiary Insects. Am. Journ. Sci., XXVI, 1908, pp. 69-75. A Dragonfly Puzzle and Its Solution. Ent, News, XIX, 1908, pp. 455-459. 13. Two Fossil Insects from Florissant, Colorado, with a Discussion of the Venation of the Aeshnine Dragonflies. Proc. U. S. Nat. Mus., 45, 1913, pp. 577-583. F’, FORSTER Odonaten Aus Neu Guinea. Term. Fuz., XXIII, 1900, pp. 81-108. 14. Neotropische Libellen, II. Insekten-Borse, XXII, 1905, pp. 2-5. Libellen von Tonkin. Wien. Ent. Zeit., XXIV, 1905, pp. 19-24. 15. Neue Aeschniden. Ann. Soc. Ent. Belg., LII, 1908, pp. 213-218. 16. Beitrage zu den Gattungen und Arten der Libellen. Jahrb. Nassau. Ver. Nat. Wiesbaden Jahrg., 62, 1909, pp. 211-235. 17. Beitrage zu den Gattungen und Arten der Libellen. III. Arch. Nat. Jahrg., 80, 1914, pp. 59-83. JUAN GUNDLACH 18. See Calvert, 11. Huco Kant, 19. A list of the Odonata Collected on the Isle of Pines by Mr. J. L. Graf in 1910, and by Mr. G. A Link in 1912-1913, now contained in the Carnegie Museum, Ann. Carn. Mus., X, 1916, pp 519-526. 52 E. B. WiLLIAMSON F. Karscu 20. Kritik des Systems der Aeschniden. Ent. Nach., XVII, 1891, pp. 273-290. 21. Acht Neue Aeschniden. Ent. Nach., XVII, 1891, pp. 305-313. 22, Neue Odonaten von Ecuador. Soc. Ent., VI, 1891, pp. 1-7. Die Insecten der Berglandschaft Adeli im Hinterlande von Togo (Westafrika). 1893, pp. 1-266. Odonaten. Abh. Senckenb. Nat. Ges. Frankfurt a. M., 25, 1900, pp. 211-230. W. F. Kirsy 23. List of the Neuroptera Collected by Mr. E. E. Austen on the Amazons etc. Ann. and Mag. of Nat. Hist., XIX, 1897, pp. 598-617. L. KRvUGER Die Odonaten von Sumatra. II. Fam. Aeschniden. Stett. Ent. Zeit, 1898, pp. 267-331. René Martin 24. Desc. d’Odonates Nouveaux. Ann. Soc. Ent. Fr., LXVI, 1897, pp. 589-504. Deux Espéces Nouvelles du Genre Heliaeschna. Notes from the Leyden Museum, XXVIII, 1906, pp. 221-223. Odonates de la Guinée Espagnole. Mem. Soc. Esp. Hist. Nat., I, 1907, pp. 421-432. Odonates de la Nouvelle Guinée Britannique. Bull. della Soc. Ent. Ital., LX, 1908, Pp. 195-207. | 25. Coll. Zool. de Selys. Aeschnines. 1900, pp. 1-223. 26. Genera Insectorum. Odonata. Fam. Aeschnide. Subfam. Aeschnine. 1911,. pp. 1-34. Voyage de Ch. Alluaud et R. Jeannel en Afrique Oriental. Odonata. 1915, pp. 21-50. Rosert McLacHLAN 27. On Some Odonata of the Subfamily Aeschnina. Ann. and Mag. Nat. Hist., XVII, 1895, pp. 409-425. Considerations of the Genus Tetracanthagyna Selys. Trans. Ent. Soc. Lond., 1898, pp. 439-444. RicHarp A, MutTKowSsktI 28. Catalogue of the Odonata of North America. Bull. Public Mus. Milwaukee, IQIO, pp. I-207. R. P. Loncrnos Navas, S. J. 29. Neuropteros del Brazil. I. Tres Esnidos (Odonatos) Nuevos. Rev. Mus. Paul., 8, 1911, pp. 467-481. (Seen by me through the kindness of Dr. Calvert.) 30. Neuropteros Nuevos 0 Poco Conocidos (quinta serie). Mem. Real. Acad. de Cien. y Artes de Barcelona, XI. 27, 1915, pp. 1-26. (Seen by me through the kindness of Dr. Calvert.) 31. Neuropteros Sudamericanos. Broteria, Serie Zoologica, XIV, 1916, pp. 14-35.. TRIACANTHAGYNA AND GYNACANTHA 53 32. Neuroptera Nova Americana. Mem. della Pont. Acad. Romana, II, 1916, pp. 71-80. 33. Algunos Insectos Neuropteros de la Argentina. Physis, III, 1917, pp. 186-196. James G. NEEDHAM 34. (And Maude H. Anthony). The Skewness of the Thorax in the Odonata. Journ; Ny) Y... Ent; Soc, XI; 1003; opp: 117-125. A Genealogic Study of Dragonfly Wing Venation. Proc. U. S. Nat Mus., XXVI, 1903, pp. 703-764. New Dragonfly Nymphs in the United States National Museum. Proc. U. S. Nat. Mus., XXVII, 1904, pp. 685-720. F, Ris Die Fauna Sudwest Australiens. Odonata. II, 24, 1910, pp. 417-450. Libellen von Sintang, Borneo. Ann, Soc, Ent. Belg., LV, 1911, pp. 231-255. Die Odonata von Dr. H. A. Lorentz’ Expedition nach Sudwest-Neu-Guinea 1909 und einige Odonata von Waigéa. Resultats de l’Exp. Scien. Neerlandaise a la Nouvelle-Guinée, IX, Zoologie, 1913, pp. 471-512. 35. Neuer Beitrag Zur Kenntniss der Odonatenfauna yon Argentina. Mem. Soc. Ent., Belg., XXII, 1913, pp. 55-102. Neuer Beitrag Zur Kenntnis der Odonatenfauna der Neu Guinea-Region. Resultats de l’Exp. Scien. Neerlandaise a la Nouvelle-Guineé, XIII, Zoologie, 1915, pp. 81-131. 36. Libellen (Odonata) aus der Region der Amerikanischen Kordilleren von Costarica bis Catamarca. Archiv fur Naturgeschichte, 82, 1916, pp. 1-197. YNGvE SJOSTEDT Wiss. Ergebn. schwed. Zool. Exped. Kilinandjaro, dem Meru., 14. Odonata. 1909, pp. I-52. 37. Wiss. Ergebn. schwed. Ent, Reise des Herrn Dr. A. Roman in Amazonas, 1914-1915. Odonata. Ark. for Zoologi, 11, 1918, pp. 1-54. R. J. Tintyarp On the Genus Austrogynacantha with Description of Species. Proc. Linn. Soc. N. S. W., XXXIII, 1908, pp. 423-431. On the Study of Zoogeographical Regions by Means of Specific Contours. Proc. Linn. Soc. N. S. W., XXXIX, 1914, pp. 21-43. 38. Life Histories and Descriptions of Australian Aeschnine. Linn. Soc. Journ. Zoology, XXXIII, 1916, pp. 1-83. The Biology of Dragonflies. Cambridge Zoological Series, 1917. HERMAN WILLEM VAN DER WEELE 39. Morphologie und Entwicklung der Gonapophysen der Odonaten. Tijds. voor Ent., XLIX, 1906, pp. 99-198. E. M. WALKER The North American Dragonflies of the Genus Aeshna. University of Toronto Studies, Biological Series, 1912, pp. 1-213. C. B. WiILson 40. Notes on Dragonflies of Jamaica. Johns Hopkins Univ. Circ., 1911, pp. 1-5. 54 FE. B. WILLIAMSON TABULATION OF VENATIONAL CHARACTERS The tabulation of venational characters based on the following males: Triacan- thagyna septima, 5; Puerto Barrios, Guatemala, 1; Rio Frio and Bolivar, Colombia, 2; Palma Sola and Tachira, Venezuela, 2: ditzleri, 5; Puerto Barrios, Guatemala, 1; Puerto Berrio, Colombia, 1; Palma Sola, El Guayabo and La Fria, Venezuela, 3: caribbea, 5; Santa Marta and Bolivar Colombia, 3; Palma Sola and Boqueron, Venezuela, 2: trifida, 5; California, 1; Georgia, 1; Guanajay, Cuba, 1; Hayti, 1; Bath, Jamaica, 1: satyrus, 3; Banana River, Costa Rica, 1; La Fria, Venezuela, 1; Bartica, British Guiana, 1: Gynacantha laticeps, 2; Minas Geraes, Brazil: aratrix, Martin 25, figure 182: adela, 3, Yungas, Bolivia: convergens, 1, Jujuy, Argentina: tenuis, 5; La Fria, Venezuela, 3; Iquitos, Peru, 1; Tamanoir, French Guiana, 1: caudata, 1, San Rafael, Ecuador: tibiata, 5; Colima, Mexico, 1; Ontario Farm and Cimarones, Costa Rica, 3:> Wa Fria, Venezuela, I; jesset, 1, Puerto Berrio, Colombia: auricularis, 3; Palma Sola, Venezuela, 1; Bartica, British Guiana, 1; Oyapok River, French Guiana, 1: klagesi, 2, Tamanoir, French Guiana: ereagris, 2, Havana, Cuba: mexicana, 5; Fundacion, Colombia, 1; Palma Sola and Bejuma, Venezuela, 3; Para, Brazil, 1: nervosa, 5; Bolivar and Cristalina, Colombia, 2; Palma Sola and La Fria, Venezuela, 2; Rockstone, British Guiana, 1: bifida, 3; Porto Catherina de Santa Leo- poldina, Brazil, 2; Jujuy, Argentina, 1: croceipennis, 2, Rio Songo, Bolivia: litoralis, 3; Paramaribo, Dutch Guiana, 1; Manaos, Brazil, 2: interioris, 4; Peru, 3; Santarem, Brazil, 1: gracilis, 2; Ontario Farm, Costa Rica, 1; Kartabo, British Guiana, 1: membranalis, 5; Rio Negro, Colombia, 1; San Esteban and La Fria, Venezuela, 2; Rockstone, British (Guiana, 1; Tamanoir, French Guiana, T. In the tabulation each number is the per cent of the total number of wings examined. Some numbers are followed by letters, which are explained on page 55. TRIACANTHAGYNA AND GYNACANTHA 55 Explanation of Letters in Venational Tabulation a. One present in one left front wing. b. In one wing a minute fourth cell formed by a forking, near its distal end, of the crossvein. c. Two wings each with one crossvein forked. and three wings each with a single double cell. d. Two for one cell’s length ‘in two wings and two for two cell’s length in one wing. e. One wing with two rows only one cell’s length, and one wing with double and single cells interspersed. f. One front wing with an antenodal of the first series basal to the first thickened antenodal. g. Obviously an abnormal condition. h. This is the Paramaribo specimen,—the type. i. One present in one right hind wing. j. Abnormal. k. Left front wing with added basal crossvein in both series; absent in right front wing and left hind wing; right hind wing with an added crossvein in the second series. 1. In both cases anal loop apparently malformed. m. In two wings there are some single cells between the anal loop and the hind wing margin. n. In the left front wing and left hind wing of one specimen M:, beyond the loop of Ms, is diverted or switched anteriorly one row of cells, resulting in three rows of cells between Ms; and My. This is the only case observed in all the wings examined in this study in which a main sector is so diverted. o. One row of two cells. p. Two for one-two cell’s length. q. Abnormal. r. Abnormal. s. In each case the fourth cell is small, interpolated. t. In one wing the cubital space is two cells wide for a distance of three cells. u. In one wing a few rows of three cells. 56 GYNACANTHA TRIACANTHAGYNA SUDUDAGUL IU SYIIDAS S140149 JUL syv4o}t] s1uua J1aI049 cat ate A DSOQAIU DUDILY IU SIUBDIAI 1893] 4 S14D]NIAND tassal D4D1G14 DIDpNDI simua} SUIS AIQUOI DIapv Y144DAD SGaIv] SNANIDS Nie! DIQGQépd Sao, a TEP Duindas E. B. WILLIAMSON VENATIONAL TABULATION Antenodals front wing SISA ae NEL ee Pe Aire Sse AP ERP ee 5 te Savi’ at ager te is tas Male 20), Wie énuen tet a * »~m-WH “WwW ah, eae Blke wethteh ca Sleek te . 3 Nie eo Ear a as | CC ie ee o> ar) es ei) ve ota ts el mnN - so Sa a) ot cep Ks ay 5s ee “8 6 ~ iad hit ce . Hea series |e) Acennce s) ote Ar Cit EN, . . . . . . . . Slr . . . -_Om ~ . oe el er eea re eae’ Cr ee — ee . . eae “eA re etna ey, lea! ite Cy Cte Yrl tel . Cee? Cie Lt eo ee acnieh 2a, ane “aN wy a ie 3 . ° eameiaeuren Une . en. | a ae 2 a eee Te is oe es Cae eo i oe ce.) Melrte me Om J hey Nommetae a . ea nus,) . . = ©) 66 o) ke eee ele ae 7 = sae . “AQ -:1NM io era e. . 2 Seer erat . . . «10 OO ae ey le 8; o* Sete . . . . (Rae s 2 a: te vtar be et oii be Fo) eae “in oe, ee IN = 8 © 8 et amet heres ea cane yes Coie fe) 25 See yee te oe EO 10)»: 50 30 30 25 20 a | -an 5 Sewn A 8 ft OON oe he th. oe . © te, 168) et aries s . 3 g : cuits cn tench Vol scenes of re) ey ae ay ite!) fe eo), WS ey fe, © ie AD o> a 5 a ei he al vee) | Sy) ete «> . . . Ou OF AG Oo i} ° i] _ . Ste ieeics sO 3 oe fe fe) heat on 5 oe 8 Pin Sire sgl ate ae ee CS BN ct ate Pepe: Se) . « . . . SA Ow Oe Fo NORA CR ac ata “ON; oy onl eterotens ay pe piel te Saks mre: ore ey (ate 02. ai Th net he lel ate BAe mom * ree ete etm oe we & . mS Ue wo. we on JOP iie) [es v0 . ° Oo. x et ie . Be Sw Ste ‘A N Cree Mery oe Cd Bil odaben ake al fis Onis . <) ex, 8 Of ase . . N mane mh Xe ° . Seoaer OWN x Pan oie oe mi bce PO et "6 Cie ec . ee . ee oe he CV o> its 2 ip a a ear: = ae . mit Ae. . Cot a? re . = os se) ie S86, vs re: oye Cee ct . a) a . 8 eye a) nie ce) ae Cat ' e Sei es * * oe Sue! en 6 Co Cay * se eee ere a A ace «* fe “ . 57 VENATIONAL TABULATION Postnodals front wing TRIACANTHAGYNA AND GYNACANTHA , Antenodals hind wing Sele) (8 ee: els ere eee «ee wee sere levee, ‘elwte Jeise OI 8 eee eee eee eee eee . . ee eee eae eee . 6z . cele) sie el piece wsie eae ee “ee Sie. ers Qz . eee eee eee eee . . eee . o* . lz fz ceo ai wie -‘e ere: OI eee ge Syetet elev. fest eee OI Veta eters: gI eee see eee eee vee OOoI A cee Oz of of eee PI eee so eer (0 WSs Olufe® Ie tie” te ete Cz ce Ol eee Ol oz ZI ep 0 pe” 6 2%” @.e%™ eo eta” ve efe cz © 9tG” a) ofa? vie ie Sales. of Tr 58 GYNACANTHA TRIACANTHAGYNA SYDUDAQULAUt SYIIDAT S140149]U1 syp1oqy situa qia20AI ty oi | DSO21IU DUDIIV IM SUS DIAI 1835 D] 4 SUADINIVAND tassal 04D1qQ14 DIDPNDI simuay SUIBAIQUOI plapr X1A4DAD SGAIUD] SNAKYDS py} DaqqiivI 14912}1P puiig das E. B. WiLLIAMSON VENATIONAL TABULATION Postnodals front wing (con.) mmmwmM au eur enolate . “ NRHN ° eo) tants = ° Se Ch pe a . wae Ll = = el a) re . oe Oo -1n Dhece aes Geact Nnm-aA ee 5 20 70 ro) . ay gle ° : N De 40 . . . rites . of ae . oe Sees ce . ee . . oy ogee So, ie e a, = Cities ol) Hott ews fat LiMo ° Cac . o 5 . cars ce 5 NRoOnnN a Cee BS A ete HOH He . ce . ay uke . eh ya eiete into! te e < e she fe se) ae rence . . OO} AW oy Ie 5 . el. 18, 3) ee . O . Si es 8) tele) vue eo) e . . ol he en on Ke 5 oe in) hie ° eure | e . o . * *. . . eine 5 : Sn et SPO: RO Chae CeO as . oF 0 fo) (oy Oo ee owls . . Qi ene oy -cefmie 5 . e . . e . *. e . . . . e eo) emg etter hte Petipa The ae cd . ane . . . . . . . . . . . . e op! Kole: ye Cer One Ce -OF &Q . . co) ie luc! go Ge nO) ch GES . sh ie siverue! ae Le O20, thao . . she” ge) ewe fe el ace agen fe . ors ate ne 8 hel?) ce. one me . _ Sr ee Oe O te oO od Sie ja, ia] Pome ve Ce cA OM IG . S ai wie shel lomo OF eat OO ° . awe showy ab/uuia, ite . Oe Gao . Gy a5 ete vey cet eet he rears . eer de . owas Ike! Remaye ete sigs io Wie, Met ie Cece) ch ae e aias . . . . . . . . . . . — : . Ste olen te . oh 80 at ou Me eit ie Chaets On fw . q. An oa orl kohtov mel Ke . . . ore . hue eo; ie . e Sue . heeclad he 0 * . . . Cea . wl ike! ais ke: har So . . . oe . . 33 10 Postnodals hind wing 14 Pe! 2 ee ee SY og es es oo oe ee” al "=. 266 eae 5. . ao) : ate . . oe . . . 10 50 Re Alo ee =O) ° ° ons 4 StS _ >RES arc :§ : : : Oma TOean 20nni 10 Ge) 0 fe) . siete apy a ite) oO Cueneic oe 5 * -NOmO Br. 0/ les See ae em OM 2 ee seer ee 78 i: meee ec Nn wr te oe) soo? fet ie N - % ee © tay) an ve NnNOnNM - ortevis) Helis i non ee 2, shite: Wiese 2) ey fel 2 9 eo Tiel ra stite NOR e She te ate ae ooo eee e a OR tt) oF el (eo) fo) 0 5 ty a pp e eS OH ober gel’ ae 59 TRIACANTHAGYNA AND GYNACANTHA , VENATIONAL TABULATION Basal antenodals Distal thickened antenodal in front wing the hind wing the Distal thickened antenodal In (con. ) — -- ? Postnodals hind wing OOI OO! oo! Sz Gz OOI OOI OOI JOO! OOI oo! ool OO! OO Ol of oo! . of of (0) 4 oor oOo! OOor oOo! eee eee eee oe Oo! es yuasqe yuosoid YyII YjOI y16 Y8 yz 419 yas Yjcr YVII yjOr TB) WB 439 If in ing cells le front w Number triang in B. WILIJAMSON cells i triangle hind wing E. VENATIONAL TABULATION Number ¢ 60 N . = io} ° eae boa? -_ ia) ie) eee eee £9 yZ1 os oS os oS ee | CC RiGke sisi) Lelejen (ajete. alee) 0,018) Hele. 6 smilie (eumier (aise ‘siele) (ele © “dr0.m.) eae + ©8088 e888 #88 #08 COTO) SON CER SEU e/ Or dee ere e880 ¢82 ©8808 ©88 ©8808 ©8688 #88 880 ©0886 868 #88 #88 © 88 #88 2 8 e 2 ee eee 888 2 eee eee OI A eee LX oe eee eee oe aye) (e686; Mee -\8ys\ a (e\ere). .6. 8,60 ere eee eee oe eee eee tee . eee OL see wee ZI Owe wee eae ois (ea CRC CRC ONC OC Car. To eee «ee #88 #88 Opie (apne, s0e) 918 see ° N Lae) 4 ise) (se) 2 v2) . ise) = <= 2 N ro) $ wn 0 IN ae] N ise] = ° ° Lea +r 2 2) 5 N wn . ee nN ~N . ~ . Ne) . fo) . N eos: nN . »<) : Loe) . Ge) % qut 1sa3D] Id VD puydas RQR:: S14014a STUB DIAI SLdDIN: DIDpNDI snatyos ppyly DIQQuAdDI ou a stunua diazd0s2 SUIBAIQUOD + 8: | SyDUDAQuiam UIE EnnEE En eT VHINVOVNAD VNADVHINVOVIEL op | £2 aa. a) So | & Number supratriangular crossveins in front wing Number supratriangular VENATIONAL TABULATION crossveins in hind wing TRIACANTHAGYNA AND GYNACANTHA to front cubito-anal basal in wing Number crossveins subtriangle "OS "OG 108 =**" 08 af0) -""> Se EE aoa. @ oem) (elele” ‘ea 8) ieee SORE eKarte” 16ers) Teh's) e010 ele: “(elie a. sw) vie ee nie (6) sie oS CI ZI eee ZI a) a), @ eee eae See ees cee eee ae eee eee eee eee eee cee oS os fa we) Berd) jellere CC Chae OCLs Ai SHC 0), Ciee) Siem (eels Sime siae is ee mer eee ce yZI CZ ZI or ob eae eae wee eee eee eee see eee eee eee eee “. or “Zt se, YO, On Oy "Geta Or 80g fe Zp ane ote eee eae ey. “ys OT Oz oD (eh, nes, oS oZ eae ov eee eee see eae oe ep ia eee ec Tat ‘oe OO eee wee oS 0z os OO n £Q OOI cz OOI Sree, caiehel eres evans: elieis “eles Mauure K“enfese: BOlbie Calleve selects os Clee eae wwe (wiles. CZ one S018, 1B OueL TOV Seno le Memle se inija) a) 9's 106, eles ZI ee} “Chee eyes ees ieee wens. eine | (ejhie eee ee Si Sc eee eee eee eee Oras LI Cet (val Pe eee eee eee eee eee ene eee £9 LI eee oS eee OI Uti Cz ce CIO 0 f-4 eae ov ere eee eee eee eee CZ os 0g oS eee CZ ge DICT ¢ 55 w= * 00 eee ere eee eee ZI BS eo we Oe af Ob: 2105) ota? ie sOOL1O5.4 O01 22> OO) fg oo1 of 9g eee ee ee eee OI os eee eee 2 @ ee =>a% © ere ENT AS me) os ZI MPnNOon OOH 9 xt nO Timo ROO 62 GYNACANTHA TRIACANTHAGYNA S1]DUDAG Ua Ut SIpIDAs SIAO1A I] Ut SYDLOJY SIMU G1aIOAI ppyig DSOQAIU DUDIIXIUL S1AGDIAI 1Sa5D] 4 STAD] NIWND tassal DyD1qQuy DIDpNDI sunua4 SUIFAILUOI D]apv X14 DAD $GII1]D] SNANJDS py} DIqgiAwa 142]2]1P DMydas E. B. WILLIAMSON VENATIONAL TABULATION Number of cells Subtriangle hind wing oa! in anal triangle J 100 100 100 100 70 I00 I00 100 100 100 100 90 I00 100 100 100 100 50 100 70 100 100b 100 100 BOrent 10 50 100 100 100 100 50 100 100 I00 I00 100; 100 90 100 100 I00 100 I00 I TORO MILO si once crossed 100 90 90 90 100 open 8 Subtriangle front wing 100 100 100 I00 100 I00 100 I00 I00 I00 100 100 I00 100 100 100 I00 open I0 : once crossed 90 100 90 100 100 twice crossed ... cubito-anal crossveins basal to subtriangle in hind wing Number . oe . 50 100 33 50 50 60 60 Ol an 25 S re 60 100 BO teen . . . 50 100 50 100 40 ae Claern: 17 17 30 20 40 40 67 70 80 60 60 Or noOnN 63 Number of cells in anal loop anal and posterior between wing margin VENATIONAL TABULATION Number of rows of cells TRIACANTHAGYNA AND GYNACANTHA loop \ posterior to stigma in oOo NR jo} N ————— in fork of Rs front wing pone Number of rows of cells eee ZI OOO) COU 0.0 ei (a ae) os ele -* oe . oe eee eee 8 ee wae eee ee 8 ee eee Oz eee fei 4 ore fee ae Cale els 0.0/0 ‘eo ae (6.6 » (6 Us ones eLe8| ais) Jae eae Cheer eh Veer ier hie yey Chey 61 ton eee eee eee see eee see sae see see see see Bye) (s eee eee eee eee eee eee see sae sae QI Sy oS A sae eee eee eee eee eee eee wae sae Ono eee eee eee eee eee eee eee eee one ZI . oe oe v 88 OOI oS OOI oS oZ eee eee see eee “Wt Wa) Ct a) exeve oS see eee Cerra) aLere eee nee Che re ¢ eee eee eee eee Pact) eee eee Cetin eee ene Coach} eee eee os oS eee eae eee Cur ter) sae ee sere epuENe Clo 05 ae OS Ole OOLUMIOONOOL OOL OOl ass) TOS) SOUL TOON OOL LOO MOOT OOlOOl S Bs eee ee eee a Lsrcmisch eR GOT GOL MEO Petes Sate rae ORF sie Beare I os eelem sole) lalae of OhA 3 COG OOD. D010, 000. “C-01) 0.010 ODG OO. OD GOR) C00 Prt “aisoave schtet! Cheteoe cer) oS OOI OOI OOI oS oZ one ie @ eee eee eee "ee wee eee eee eee eee ce ha) eee eee eee “oo era) 7 Oe OF O01 0S) O01 DS 66° oO1, OF O01 Zo O01 «°° ZO “OOr1 O01 ONT 06 eee 88 e woe tive we eae SERS Gy oS OI 09 eee ¢¢ eee eee eee eae eae soe eae Pies bie asthe Sisley rot Mate VIN teal MN pre age SAR seni Ute Stok aby gaat lee ey Secret pue z NMNnMn ww 64 GYNACANTHA TRIACANTHAGYNA E. B. WILLIAMSON VENATIONAL TABULATION : Number of rows of Maximum number of Maximum number of cells in fork of Rs rows of cells between rows of cells between posterior to stigma in Rs and Rspl in hind wing Rs and Rspl in front wing hind wing = aS SS > OF SUDUDAGM aut Ls Pay ee epee) = se ea sylapas Sores as eI Re saber tt soe ones eee) 8 te site _o “ S1bO1da put a eee ee ciao! aor Ghee SYDALOFY sit ce Sa BS ¢ hie . . ~ _ . SMU G1aI0AI reese ae ary ae ea A 3m: ae ; Eocene 54 Ge : on - aire All c Dpyiq cc Lee 4s moO: me DSO21aU Abeta ts Pe Genera eee DUDINX IU > Siac aioe R dee oat an - eee ne > . WwW ° SIS DIAI oe ee da Awe B.S . soi 2 fe Oe a _ om | A 6 ° Woiimeg s2smy. IAAT: ae Beare nes . . L onl . - . aie ae ™~™ oN . SLADINIVAND BNR a at - eRe pars & ; eae ° esha 2 : assal By = Prac LTE Se ieee ‘ear Ad - a . . . . Z 4 5 oO care ds ees DjD1G14 Steines Ses ee 18S : () (S) oc : : ey wate me Se DjvpNDIMIH: - Re Lt Rare . . . | ol . . . _ . . s . . . . . [e) . . SINUI4 Sat 8 : Seeks S: . _ : . . . . el) telly fe an et, ve 816 Biateus sua8szqu0g «8 2 t: Sid: EU ere loelsele od eed Mr «© « - -oeo . . Se tele AS 6 56 2 crags X 144040 eu 3: Pose ee 4 fe} ae hoes ° othe Vs 5 ite sda © Bort ete 8 os fe ate mo. on IN 8 6 Nm - snakyos SOR eps s sett G a Om - py | ie aue : Bo ee oie toe paqquny 2 1B: eae sie (ene 2 . = Sb he . > . . ~- 5 uayep i:88i: oe ae iz Meee a ° P : on fo} wv its piuidas ae a Be BORE oie Ge tae re oo Into N tino nN Nort ao} = as) is] TRIACANTHAGYNA AND GYNACANTHA 65 VENATIONAL TABULATION Number of rows of cells ; Maximum number of between M3 and M4 Maximum number of rows of cells between following the loop in M4 rows of cells between M4 and Mspl in front in front wing M4 and Msp! in hind wing wing (FS SS SS eos oe) (eho) 1.2 OO : g ee: :%&, ay [AWW mw Ae “225 183 . _— - foe) . ae: aa 32 : 8 :@8 eco oS 728 tn ne oe 7 Be && : 8 78 :§ SB :8 zo, Be 8 8 : g g 8 788 28 8: 8: 8: $s aR 22 8 S ; or . = i] ° om oe o& 8 : S3 a a . 4 oe 8: AC eae ne g eee g g +e g § 23.8 8 : aah 22 8 SR: BR g BR BR HA co INO N om t+ IN0 GYNACANTHA TRIACANTHAGYNA E. B. WILLIAMSON VENATIONAL TABULATION Number of rows of cells between Cu: and Number of rows of cells Number of cells in Cu, posterior to the between M3 and M4 the anterior row of triangle in the hind following the loop in M4 eth cells in the anal loop wing in hind wing ae a Sar meen aay ( SYDUDAGULa UL oe es ees oe . 2. Le | SyvAs -O9 Bo 2 ee WIV. -nNn ° ents A o . . . _ o StOlsaquUr a8 ale eS eee . = 78 aa) S1DA041] Obes 2S ae . He ° smua q1aI 049 ie O98 “9 e : ee & _——_ ppyiq ei: : Dod AAS oe oH pso2sau ee a: ae} Pl | . — . = DUDIIVIUL et: ice Se 5S 5 — a ps SUB DIAI ae 7S eS : N . - =~ . an 1sa3D]Y OHO AG S es < . Lol =! . om I fo} ao) SUADINIVAND on 3 eS 7 . iol - zassal te) 8 8 . . in nh CI Lal 5 . fo) pjD1quJ se ro) 8 . . 4 . | oe | _ : 5 2 DjDpNdD?I vee 3 8 ar 2 - fe) . : e), 2) simua} oe 2 +9 ; et . . . . [e) [e) Suds sa2uodr a S : 8 ate eaed -Q : ° one pjapp -O : 8 mS c . 8 . [=) : (o) . 4144 DAD Bee Sie 3 pa eo >) DY Hess fe) L sae BR: : : ; Songer is SNANYDS a Sas g g ome = p= Dpytt} Bish Q g: = . rary _ . DIQGADI Shee 3 8 : S) iS) 149]2]1P Oye 8 8 Lal s =~ al a: , pimiydas G5) ¢ 2 : 2 2 ° ; ¢ N Os N we NN I and 2 67 TRIACANTHAGYNA AND GYNACANTHA VENATIONAL TABULATION Na N ON Gua oe oe Soma = om- oe ee Enos. Cs OSs Zia COL PEE: S42 H ge Of <0 05. fg =" is H Oa-S adn fo! 0) o. . + eee Foe cS wD o8 FE ase ers ae EUS Bae 2.2 6 2 | oor oor oor oor oor AE he Ool see see oS eee LI we *** QOOI OOI OS OOI £8 OOI 06 OOI OOI OOI OOI OOI OOI OOI OOI OOT OOI OOTI OOI OOI OOI OO! OOI 09 ot OV. eee ov OOI 09 BUuIsI}s yO J]ey [eIstp Japun ee PUISIS JO yey yeurxoid Japun ze PUISIS 0} yextmoid 10 je BUISI}S JO J[eYy [estIp 1apun PUISI}S JO yey yeurxoid r9pun one BULSI}S 0} jextuo1d 410 ye 68 E. B. WILLIAMSON PLATE 1 Wing photographs by Miss Mina L. Winslow. Figure 1, Triacanthagyna septima, male, Palma Sola, Venezuela, March 5, 1920; Figure 2, 7. ditzleri, male, Puerto Berrio, Colombia, January 31, 1017; Figure 3, T. caribbea, male, Santa Marta, Colombia, December 14, 1916; Figure 4, Gynacantha laticeps, male, Minas Geraes, Brazil. “an ea ae CY a Sak aa ee ee oe ee aa i J tVIg VHINVOVNAD GNV VNAOVHINVOVIEL itl = i We - . > 2% ° 4 ates ‘ ‘ oF - : . ee SE, alah. : 4 ne = ‘ Le. >; a vie ; -S te % a a Bf = an =z i ; e 5 —s = — = : <5 a : \. ik a - - a 7 Se = a. = =e - ¥ < 7 — = as i Ae a. | +i ‘ve _ gt a i jv eve i of ri ae 5 4 5 7 ‘ j s) : » 1 . *\ { i 4 , om 7 ne ' at 70 E. B. WILLIAMSON PLATE, II Wing photographs by Miss Mina L. Winslow. Figure 5, Gynacantha tenuis, male, La Fria, Venezuela, April 12, 1920; Figure 6, G. mexicana, male, Bejuma, Venezuela, February 17, 1920; Figure 7, G. nervosa, male, Palma Sola, Venezuela, March 6, 1920. TRIACANTHAGYNA AND GYNACANTHA aes s TTY SEU Tits 1 AT TS ET x Prare TT 72 E. B. WILLIAMSON PLATE III Wing photographs by Miss Mina L. Winslow. Figure 8, Gynacantha interioris, male, Colonia del Perené, Peru; Figure 9, G. membranalis, male, La Fria, Venezuela, April 21, 1920; Figure 10, G. gracilis, La Fria, Venezuela, April 12, 1920. TRIACANTHAGYNA AND GYNACANTHA PLAt® [II LED TET aS A sit oper rrr : \ ss dtnivaennton Seg SOM os ue ms i i STTTiTt bee. Ts raga wats Suerte ee —_— ade tel irr BELL ee SURAT Sear ers Weerom” ATALITT TOILE eee | ? : Ae LA\s ; 4 eenaweeun LOT rH4 > IP =~ ‘ _ A SaaS = « « e hw t ay 7 7 4 ~ = . _ a i aC 4 j Fa om the S . i ey Z 7 (jue £ > A, 4 ~" . . == : i i Hy ; 7 ' m1 ° > aa Was 7 i 7 ©.,"4 a, > Sale * 3 : 7 r : ; ° rs) ‘ : ; : ; ; ¥ ’ f , G —- mi é = - , = ; ’ 5 1 7 ss : f . et zh >a %) ‘ ~~ ~~ ? : 2 A i ; ' ie aa = 74 E. B. W1nLIAMSON PLATE IV Figures 11-15, ventral views of third abdominal segments; Figure 16, ventral view of second abdominal segment; Figures 17-20, hamular processes. All drawings by Dr. C. H. Kennedy. Figure 11, Gynacantha nervosa, male, Palma Sola, Venezuela, March 6, 1920; Figure 12, G. nervosa, female, Palma Sola, Venezuela, March 4, 1920; Figure 13, Triacanthagyna ditzleri, allotype female: Figure 14, T. trifida, female, Samana, Hayti, M. C. Z.; Figure 15, 7. caribbea, allotype female; Figure 16, Gynacantha klagesi, allotype female, to show elevated lateral carine; Figure 17, T. ditzleri, type male; Figure 18, T. caribbiea, type male; Figure 19, 7. trifida, male, Samana, Hayti, M. C. Z.; Figure 20, 7. satyrus, male, La Fria, Venezuela. TRIACANTHAGYNA AND GYNACANTHA PrAtE UV | mM nie i V7 CO) a Z } vy 76 E,. B. WILLIAMSON PLATE. V. Figures 21-29, male abdominal appendages in dorsal and lateral views. All drawings by Dr. C. H. Kennedy. Figure 21, Triacanthagyna septima, Rio Frio, Colombia, January 7, 1917; Figure 22, 7. ditzleri, type; Fig. 23, T. caribbea type, figure at left, appendages in supero-internal view; Figure 24, 7. trifida Samana, Hayti, M. C. Z.; Figure 25, 7. satyrus, La Fria, Vene- zuela; Figure 26, Gynacantha laticeps, type; Figure 27, G. chelifera, after Figure 174, Martin 25; Figure 28, G. aratrix, after Figure 183, Martin 25; Figure 29, G. convergens, Yuto, Jujuy, Argentina, April, 1grt. PLATE V TRIACANTHAGYNA AND GYNACANTHA [= > » NNN fives 78 EK. B. WILLIAMSON : PLATE VI Figures 30-38, Male abdominal appendages in dorsal and lateral views. All drawings by Dr. C. H. Kennedy. Figure 30, Gynacantha adela, near Coroico, Yungas, Bolivia, May 4, 1899; Figure 31, G. tenuis, La Fria, Venezuela, April 16, 1920; Figure 32, G. caudata, San Rafael, Ecuador; Figure 33, G. tibiata, La Fria, Venezuela, April 13, 1920; Figure 34, G. jessei, type; Figure 35, G. auricularis, Palma Sola, Venezuela, March 6, 1920; Figure 36. G. klagesi type; Figure 37, G. gracilis, Mte. Christo, Tapajos, Brazil, May, 1920; Figure 38, G. membranalis, San Esteban, Venezuela, Feb. 9, 1920. PLATE VI TRIACANTHAGYNA AND GYNACANTHA TH D>) Wi jes 80 E. B. WILLIAMSON PLATE Vil Figures 39-45, male abdominal appendages in dorsal and lateral views. All drawings by Dr. C. H. Kennedy. Figure 39, Gynacantha mexicana, Fundacion, Colombia, January 9, 1917; Figure 40, G. ereagris, Havana, Cuba, Baker; Figure 41, G. nervosa, Palma Sola, Venezuela, March 6, 1920; Figure 42, G. bifida, Porto Catherina de Santa Leopoldina, Brazil; Figure 43, G. croceipennis, Rio Songo, Bolivia, Fassl; Figure 44, G. litoralis, type; Figure 45, G. interioris, type. PLATE VIL TRIACANTHAGYNA AND GYNACANTHA Vv om) * i : i ’ . : hs { > a) ep nd eal ot ewe ee Pou ae * ’ . ’ | A e i dest 2 or = + 9 a a f ts y 5 ~ a = 4 ; a ‘ my i a Bird GRIGGS 2} shbrb: do persrown® OUTLINE MAP OF NORTH DAKOTA lo Steele = a yf Cedar Fs |_Hettinger Sad E 104° 103° 102° 101° UNIVERSITY OF MICHIGAN MUSEUM OF ZOOLOGY Miscellaneous Publications No. 10 A Preliminary Survey oi the Bird Life of North Dakota BY NORMAN A. WOOD ANN ARBOR, MICHIGAN PUBLISHED: BY THE. UNIVERSITY JULY 2, 1923 ADVERTISEMENT The publications of the Museum of Zoology, University of Michigan, consist of two series—the Occasional Papers and the Miscellaneous Publi- cations. Both series were founded by Dr. Bryant Walker, Mr. Bradshaw H. Swales and Dr. W. W. Newcomb. The Occasional Papers, publication of which was begun in 1913, serve as a medium for the publication of brief original papers based principally upon the collections in the Museum. ‘The papers are issued separately to libraries and specialists, and, when a sufficient number of pages have been printed to make a volume, a title page, index, and table of contents are sup- plied to libraries and individuals on the mailing list for the entire series. The Miscellaneous Publications include papers on field and museum . technique, monographic studies and other papers not within the scope of the Occasional Papers. The papers are published separately, and, as it 1s not intended that they shall be grouped into volumes, each number has a title page and, when necessary, a table of contents. ALEXANDER G. RUTHVEN, Director of the Museum of Zoology, University cf Michigan. A PRELIMINARY SURVEY OF THE BIRD LIFE OF NORTH DAKOTA By Norman A. Woop The field studies upon which this paper is largely based were carried on during the summers of 1920 and 1921. The investigation was made pos- sible through the cooperation of the State Biological Station of the Univer- sity of North Dakota and the Museum of Zoology of the University of Michigan. It is a part of the program of a biological survey of the State of North Dakota, as undertaken by the State Biological Station and carried on under the direction of Crystal Thompson, Curator of the Station Museum.’ Dr. R. T. Young, Director of the Station, has provided, through the Sta- tion, most of the funds neéessary to carry on the work and has rendered every assistance in his power to make it a success. The Museum of Zoology of the University of Michigan has contributed traveling expenses to and from the field and the time of the investigator. In 1920 work was carried on from July 12 to August 25, chiefly in the Devils-Stump Lake region, and a short trip was made to the Turtle Mountains. In 1921 headquarters were made at the Station from May 2 until July 25. During this time several trips were made about the state. On June 1 I went to Cando, in Towner County. From here, at the invitation of Mr. Elmer T. Judd, an automobile trip was made across the county to the Can- adian boundary and west into the Turtle Mountains, returning June 6 to Cando for a few days. On June 13 I went to the Bad Land country by way of Fargo and the Northern Pacific Railroad to Medora, Billings'County. A few days were spent atMedora and a week at the Peaceful Valley Ranch, four miles north, owned by the Hon. Carl Olsen, who provided a trip through the Petrified Forest. On June 24 I went to Mr. William Shunk’s ranch on the Sheyenne River near Anselm, Ransom County. A trip to Williston, ;Williams County, was made on July 5, and on July 15 another to Bottineau, Bottineau County. Here Mr. Alfred Eastgate, a noted guide and ornithologist, drove me over the prairies and up into the western part of the Turtle Mountains. July 21 to 30 was spent along the Red River at Fargo, Grand Forks, and Grafton. The collections of the Agricultural Col- lege, Fargo College, and the University were examined, and the large private collection of Mr. William Williams and his son at Grafton. Some field work was done at each of these places also. Travel of more than 2,000 miles, by rail and automobile, within the state gave much insight into the character 1T wish also to acknowledge the critical work of Crystal Thompson, who revised and edited this paper and to whom much credit is due for its systematic arrangement. 4 NorMAN A. Woop of the country upon which so much of its bird life depends, and the follow- ing list is the result of close observation and field work, as well as the rec- ords, published and unpublished, that I have been able to secure. It should be understood that the list is far from complete, especially as to the distribution of species, but I hope that it may prove a basis for much future work along these lines. ACKNOWLEDGMENTS In compiling this list I have not only copied from the published data but have been given many valuable records by the following persons: Mr. Alfred Eastgate of Bottineau; Mr. Elmer T. Judd of Cando; Mr. William Shunk of Anselm; Mr. Metzger of Williston; Mr. D. R. Ducke of James- town; Mr. William Williams and Mr. Henry Williams of Grafton; Mrs. Lucia Olsen and Hon. Carl Olsen of Medora; Hon. Lewis Crawford of Sen- tinel Butte; Professor Daniel Freeman of Fargo; Professor C. C. Schmidt of Grand Forks; Professor R. T. Young of Grgnd Forks; Professor George Miller of Fargo; Mr. Russell Reid of Bismarck; Captain Lonnevik of Devils Lake; and Professor Howard Simpson of Grand Forks. Previous WorK AND WORKERS Although North Dakota is one of the newer states, parts of Dakota Ter- ritory ‘have been visited since the early part of the eighteenth century by noted travellers and ornithologists. The Missouri River, since it furnished easy access to the interior of the state, was the route taken by many of them, and the frontier forts furnished favorable camping places. The first expedition to visit the Dakota territory was that of La Veren- drye, who with a party of 52 persons left Fort La Reine, below the mouth of the Assinaboine, October 18, 1738. They arrived at the Mandan Village, lat. 48° 12’, December 3. They remained: here ten days, returning to Fort La Reine February 10, 1739. This expedition was made at a season of the year when little animal life was to be found in the region. The only animal mentioned is the bison which furnished food for the party. The Lewis and Clark Expedition was the first scientific party to oe the region. Starting from St. Louis on May 14, 1804, they reached the 46th parallel (North Dakota) on October 15, 1804. Here they built Fort Mandan and spent the winter. Leaving the ‘fort on April 7, 1805, they journeyed up the Missouri River to the Montana line, which they reached July 27, 1805. Many birds were observed and some were noted in the daily journals. ‘These early North Dakota records have been placed in the state list in regular order. ’ Prince Maximilian of Wied visited North Dakota in 1833, going by boat from St. Louis and reaching Fort Clark on June 18, 1833. He arrived ‘at the mouth of the Yellowstone on June 24 and spent two weeks at Fort Union. He then continued up the river to Fort MacKenzie, returning to Fort Union on September 29. Here he worked until October 30, when he returned down Brirp Lire or NortH DAKora 5 the river to Fort Clark, where he spent the winter. His journals contain some notes on birds which are recorded in the list. John J. Audubon visited the region in 1843. Leaving St. Louis, April 25, he journeyed up the Missouri River by steamboat, reaching Fort Union on June 12. At Fort Union he collected many new birds and made draw- ings of birds and mammals. August 16 he started with his party down the river in a forty-foot barge, “The Union,” reaching St. Louis, September 10. A daily journal was kept in which he mentioned seeing many species of birds. ‘These records have been added to the list under each species noted. J. A. Allen, as naturalist for the Northern Pacific Railroad Exploration, worked overland from Fort Rice (June, 1873), on the Missouri, along the Heart River and across the Little Missouri into Montana. He returned in September through the same general region. His report on birds consists of several local lists and an annotated list of 118 species. This list is of exceptional importance because the observations were made before the coun- try was broken up and settled and because this region is perhaps the least well known of any section of the state. Dr. Elliot Coues, as ornithologist of the Hayden Survey, went by boat from Moorehead down the Red River to Pembina in the latter part of May, 1873. He collected there for nearly a month, then worked along the inter- national boundary line to the Turtle Mountains and Mouse River, making numerous collections and notes covering 183 species of birds. In 1895, from late March until August, a party consisting of Dr. L. B. Bishop, W. H. Hoyt, John Shaler, N. L. Bigelow, and Elmer T. Judd col- lected many specimens of birds in Towner and Rolette counties. The type of the Dakota song sparrow was taken at Rock Lake, and the type of Hoyts’ horned lark was collected at Cando in Towner County by Dr. Bishop, who described and named both. The notes on species taken here (some 220) were later supplemented by others made by E. S. Bryant, who collected at Dry Lake, Freshwater Lake, and Devils Lake in 1892, 1898, and 1906. Since that time Mr. Judd has added some 20 species and in 1917 published a list of 255 species. During the last of May and the first half of June, 1901, Mr. A. C. Bent visited the Devils-Stump Lake region for data on the life histories of North American diving birds. He was accompanied by the Rev. H. K. Job and Dr. L. B. Bishop. Mr. W. L. Stockwell, Superintendent of Public Instruction, issued, April 15, 1904, a list of birds that had been observed in the state of North Dakota. This list of 339 species, compiled by C. C. Schmidt, includes all the names of birds sent to him from all observers. It includes many species, such as the sooty tern, willet, Guadalupe wren, and Kentucky and prothonotary warblers, that are geographically almost impossible in the state. During the summer of 1913 Mrs. Florence Merriam Bailey visited the Devils-Stump Lake region and studied its bird life, which she has graphic- ally and beautifully described in a series of articles in Condor. 6 NorMAN A. Woop In 1917 Mrs. Bailey returned to Dakota and visited all of the larger lakes and many of the ponds and sloughs. The results of this trip are like- wise published in Condor. In April, 1919, Professor Daniel Freeman published “A Bird Calendar of the Fargo Region.” ‘This list of 180 species is the result of several years’ observation by Professor Freeman and his students. Some of the species have a little data and some annotations, but the list contains few actual records. Professor C. C. Schmidt of the University of North Dakota published in 1921 a list of the birds of North Dakota in a book entitled “Nature Study and Agriculture.” This is a popular list without records or dates of occur- rence for the species. Although the task of compiling a list jlike the present is often tedious, it has not been an irksome one to me, as the field work on which it is based has carried me thousands of miles across the level valleys, the “Prairie Plains” and the “Great Plains” of the state. I recall the wonderful “Buttes” of the Bad Lands along the Little Missouri in “Roosevelt Land,” and the big, muddy Missouri that carried the pioneer ornithologists through the state. While I have been compiling the records and notes‘I have been constantly reminded of the many days spent by the lakes and ponds or wandering over the big, dry prairies. However dry the text of this list may seem, back of it is a living Dakota with its wonderful and interesting bird life, worthy of much future study by bird lovers. To these I hope this list will be a help and an inspiration. GENERAL PHYSIOGRAPHIC FEATURES AND DISTRIBUTIONAL AREAS OF NortH DAKOTA While I have visited many parts of the state, my time in each one was too limited to permit me to obtain complete data on many of the species, and little has been published on this subject in the state lists. The problem of distribution is the hardest one given to the bird student, since species are not strictly confined to any one area, but are continually extending their range and passing from, one area to another, as the character of the country changes. The state is divided geographically into three main areas, and the character of these areas in a great measure ‘determines the characteristic bird life of each. The discussion of the physiography of the state has been taken largely from “Topographic Features and Geological Formations of North Dakota,” by A. G. Leonard,? and “Physiography of the Devils-Stump Lake Region, North Dakota,” by Howard FE. Simpson.* The land surface included within the state may be regarded as formed of three plains rising one above the other. The lowest of these is the broad *Third Biennial Report of ‘State Geological Survey of North Dakota. Bismarck, 1904. 3 Sixth Biennial Report of the State Geological Survey of North Dakota. Bis- marck, 1912. 3rRD Lake or NortHi DAKOTA 7 Red River Valley with an elevation of from 800-1000 feet. This is bordered on the west by a higher Drift Prairie Plain rising from 1,200 to 1,600 feet above the sea. Still farther west and occupying nearly one-half of the state is the elevated Coteau du Missouri with a surface which rises from 1,800 to 2,700 feet and more ‘above sea level. The state thus repre- sents a considerable range of relief, the lowest point in the northeastern corner being 789 feet above the sea; the highest yet determined, the summit of Sentinel Butte, over 3,100 feet. The Red River Valley The Red River Valley is not a true valley but an old lake plain, north- ward along the axis of which flows the Red River of the North. The plain is remarkable for its larger level areas and for the fertile soil-lacustrine deposits of silts and clays found on the floor of the ancient glacial-marginal lake, Lake Agassiz. The portion of the plain which lies within North Dakota has a breadth of thirty to forty miles, except at the south, where it narrows to ten miles, and an elevation of from about 800 feet at Pembina to about 975 near Wahpeton. The boundary between the Red River Valley and the Drift-Prairie Plain is an escarpment so abrupt and rugged in the north as to receive the name of Pembina Mountain. The wooded and dissected character of! the bluff accents the contrast with the old lake floor. Near the middle of the state the escarpment fades into a gentle and inconspicuous slope and becomes somewhat more marked again as the south state line is approached. Through- out the entire distance across North Dakota the escarpment bordering the Drift-Prairie |Plain on the east rises 300 to 500 feet above the Red River Valley floor; in some places the slope is abrupt, in others gentle, but always it is conspicuous in this country of low relief. The Red River seems to be the dividing line between several eastern and western species, although Coues (1878, p. 546) says: The bird fauna of this region is decidedly eastern in character, and of the Pembina region, the only western trace observed was clay-colored sparrows and Brewer’s black- bird. The whip-poor-will, pileated woodpecker, crested flycatcher, phoebe, wood duck, indigo bunting, and screech owl rarely pass westward of the Red River and its valley, while of the western species, the magpie, burrowing owl, Arkansas kingbird (which Dr. Coues did not find in this valley in 1873) now are common as far east as the Red River itself. Several species that Dr. Coues found common in this valley in 1873 have now almost dis- appeared from the region, and one, the passenger pigeon, has become extinct here as elsewhere. The Drift-Prairie Plain The Drift-Prairie Plain extends from the Red River Valley escarpment on the east to the Coteau du Missouri escarpment on the west. The western escarpment runs from northwest to southeast, passing near Kenmare, Minot 8 . NorMAN A, Woop and Steele. ‘The plain varies in width from about 200 miles at the north to 100 miles at the south and has a general elevation of from 1,500 to 1,800 feet above sea level. It has a gradual, gentle slope eastward from the Coteau du Missouri and southward from the international boundary to the South Dakota line. This double slope determines the direction of the drain- age and causes the main streams to take a general southeasterly course. The topography of the plain varies from gently undulating to hilly, the differences being due almost entirely to the original disposition of unmodified glacial drift upon a nearly level plain. A few'low, well-rounded hills, Sully’s Hill, Devil’s Heart and Blue Hills, partially concealed by drift, are rem- nants of older continuous formations now all but eroded away. Other ‘less conspicuous hills are to be found stretching across the prairie in a north- west to southeast direction. ‘The plain is cut by a few abnormally deep and well-defined valleys, such as those of the James and Sheyenne rivers. Hun- dreds of small lakes, marshy areas, coulees and swales dot its surface. In marked contrast to the general topography of the plain are the Turtle Mountains and the Mouse River Valley. Each of these resembles one of the other physiographic regions of the state, and their location within the border of the Drift-Prairie Plain and their peculiar relation to one another still further emphasize the transitional character of this middle plain. The Turtle Mountains are a rough, moraine-covered tableland lying mid- way on the Canadian boundary line. ‘They have an area of 600 to 800 square miles and rise, mesa-like, 400 to 800 feet above the surrounding plain, their margin forming a gentle ‘but conspicuous escarpment on all sides. Lakes abound, and this upland is on the whole well timbered and well watered, but poorly drained. The Turtle Mountains seem to be an isolated residuum of strata elsewhere eroded back to the Coteau du Missouri. This comparatively small but distinct habitat, well wooded and watered, furnishes a suitable breeding habitat for several species rarely found else- where in the state. These are Grinnell’s water thrush, Philadelphia vireo, yellow-bellied sapsucker, mourning warbler, black and white warbler, and hairy woodpecker. The sandhill crane and great blue heron, ruddy and canvas-back ducks, and the western great horned owl also breed here. The Mouse River Valley is a glacial lake plain, similar to that of the Red River Valley, the floor of which was formerly covered by the waters of Lake /Souris. This plain lies between 1,100 and 1,600 feet above sea level and drains northward through the Mouse River. One of the most striking features of the Drift-Prairie Plain is the Devils-Stump Lake basin. The basin extends from the southern ‘slopes of the ‘Turtle Mountains and the Canadian boundary southward to a series of prominent hills lving between Devils and Stump lakes and the Sheyenne River. The eastern and western boundary lines are indistinct, but the area of the entire drainage basin is about 3,500 square miles. There is a gradual slope throughout the basin southward to Devils and Stump lakes. The fall is so slight that drainage is poor and small lakes and ponds abound. The coulees are few and shallow and rarely contain running water except in Birp LiFe oF NortH DAKOTA 9 wet seasons. formerly these coulees and the chain of lakes connected by them emptied considerable water into Devils Lake through Mauvaise Coulee and by several converging coulees ‘into both the eastern and western arms of Stump Lake. The headwaters of Mauvaise Coulee, the most important drainage line of the basin, were gathered beyond the international boundary line, and in its course southward it drained the Sweetwater chain of lakes by Lake Irvine, through which it passed, and entered Mauvaise Bay of Devils Lake as a large and permanent stream. ‘Today no surface streams flow into either Devils Lake or Stump Lake except very minor flows during spring thaws and after excessive falls of rain. The waters of Devils and Stump lakes are decidedly alkaline, as are those of nearly all the other lakes in the general region. The notable exceptions are Sweetwater, Court, and Fort Totten lakes, which are fresh. Devils Lake is characterized by broad, shallow and irregular bays con- nected by “narrows.” On the north ‘and south sides of Devils Lake, near the middle, is considerable hardwood timber. The woods come close to the lake and on the southern side are limited chiefly to the northern slopes of Sully’s Hill. This central area, the Devils-Stump Lake region, furnishes a very dis- tinctive habitat and is occupied as a breeding ground by thousands of water- loving birds. Some of the more characteristic are the double-crested cormo- rant, grebes, common and black terns, Franklin’s gull, white-winged scoter, pintail, redhead, goldeneye, blue-winged teal, mallard, and baldpate. The upland plover was formerly common, as is now the purple martin, tree swal- low, Wilson phalarope, and belted piping plover. South from this central area the Drift-Prairie is contracted to less than one-half of its northern width,and contains the Sheyenne and James rivers with their valleys. These different areas give a great variety of habitats and all have some characteristic breeding birds. The Great Plains On the western border of the Drift-Prairie Plain rises the similar and even more abrupt escarpment of the Great Plains plateau known as the Coteau du Missouri. This plateau occupies fully one-half of the state and its irregular surface varies in elevation from 1,800 to 2,700 feet above the sea. The relief is due to the effects of erosion, and only in the eastern sec- tion has 'the surface resulted from ice action. By far the most interesting feature of the unglaciated region is the famous Bad Lands, typically devel- oped along the Little Missouri. The true Bad Lands, that is, the very rough areas that are difficult to travel through, are confined to the vicinity of streams. Back from these for six miles the land is a rolling plain and is net. “bad,” The soft cretaceous clays and sands have been carved by running water into a multitude of steep-sided hills, isolated buttes and an endless variety of fantastic forms. The erosion is greatly facilitated by the sparseness of 10 NorMAN A. Woop vegetation, the slopes being almost bare. The rainfall in the region is light, but every shower is highly effective in washing away the clays and sands. The Missouri River flows in a valley from one to several miles in width and 200 to 400 feet in depth. It has a broad flood plain, above which lie one or more terraces or flats, and back of these rise the bluffs. The Little Missouri, Cannon Ball, Heart and Knife rivers, chief tributaries of the Missouri, have all cut their valleys from 100 to 400 feet below the upland plain and have more or less well-developed flood plains bordered by terraces. There are frequent wooded areas along these streams. East of the Missouri River, in the north, the characteristic species are the lark bunting, McCown’s longspur, and Say’s phoebe, while the southern part is much like the Drift-Prairie, as many eastern species, such as the bobolink and dickcissel, follow up the valley of the Missouri. In the Bad Lands along the Little Missouri the many buttes and high, dry plateaus furnish a barren habitat for desert horned larks, western lark sparrows, magpies, and long-billed curlews. The groves of cottonwoods along the river furnish suitable habitats for a few distinctly western species, such as Bul- lock’s oriole, black-headed grosbeak, lazuli bunting, long-tailed chat, Arctic towhee, and the mountain bluebird. West of the Bad Lands are the high sage brush plains with their characteristic bird, the sage grouse. This is the only North Dakota breeding ground of the species. North Dakota is a state that has developed very rapidly, and in the last twenty years there have been great changes in the character of the country, which also affects its bird life, and direct changes in the bird life as well. The breaking up of the prairies has greatly restricted the range of ‘many characteristic species (as the long-billed curlew, upland plover, and Baird’s sparrow). From various causes many lakes, ponds, and sloughs have dried up, causing a great restriction of the habitats and nesting grounds of the numerous waders and waterfowl that formerly occurred in immense num- bers. Many of the most interesting species are now rare. The trumpeter swan, the sandhill and whooping cranes are no longer common and, with the upland plover and long-billed curlew, are on the way to extinction in the state. While man with his civilization drives away or destroys some species, others, the prairie hen, bobolink, dickcissel, robin, bluebird, and martin, fol- low and increase with the settling of the land. The cause of these changes in bird life is the rapid settlement of the country, which changes the prairies into cultivated farms, the ponds jand sloughs into hay meadows or grain fields. Some waders, like the killdeer and spotted sandpiper can adapt them- selves to these changed conditions and thrive and increase in the settled country; but others, like the upland plover and the long-billed curlew, can- not seem to accustom themselves to the presence of man or to occupy his fields of grain or hay, and thus are on the way to an early extinction in the state. Brrp LiFE oF NortH DAKOTA 11 List oF SPECIES 1. Aichmorphorus occidentalis (Lawrence). Western Grebe. Davis, 1898, p. 1; a large colony (about 200 nests) breeding at Devils Lake the first of June. JUDD, 1917, p. 5; a common breeder at Freshwater Lake, Ramsey County. BatLey, 1918, p. 173; fifty-two at one time on Creel Bay, Devils Lake. Bernt, 1919, pp. 8-9; east central North Dakota, Devils Lake region: egg dates, forty-six records, May 18 to July o. A specimen in the Museum of Zoology Collection was taken at Sweet- water Lake, Ramsey County, July 21, 1900, by Charles L. Cass, and a set of four eggs was taken there by him, May 23, 1900. Another set of four was taken at the lake by E. S. Bryant, May 30, 1906. This species still breeds in many of the marshy lakes and ponds. At Devils Lake several were seen July 15, 1920, and also in May, 1921. 2. Colymbus holbelli (Reinhardt). Holbcell’s Grebe. Jurp, 1917, p. 5; a common migrant in the prairie region, breeding in ponds and lakes of the Turtle Mountains. Skins of an adult male, female, and five juvenile birds in the Museum of Zoology Collection were taken at Sweetwater Lake, Ramsey County, June 13, 1900, by Charles L. Cass. The Oological Collection contains a set of six eggs, taken there May 23, 1900. The writer saw a pair with .downy young at Sweetwater Lake, July 18, 1920; others at Stump Lake, July 24; at Lake Upsilon, Turtle Mountains, August 2, and several at Devils Lake, August 12. In late June, 1921, many pairs with young were seen in ponds along the Northern Pacific Railroad, and in July along the Great Northern Railroad. The grebes are of little economic importance, since they live on fish and aquatic insects and plants to such a large extent that their flesh is not good for food. 3. Colymbus auritus Linneus. Horned Grebe. Cours, 1878, p. 657; the horned grebe was found breeding in the Red River region: on the 20th of June, 1873, a set of four newly-laid eggs was taken from one of the prairie sloughs near Pembina: newly-hatched young were in ponds about the base of Turtle Mountains the latter part of July. Jupp, 1917, p. 5; common, migrant; nesting in Turtle Mountains. Bent, 1010, p. 27; egg dates, fourteen records, April 6 to July 7. Several specimens were noted in the Turtle Mountains, July 30, 1920, and several were seen during July and August at Devils Lake. The species was seen for the first time at Devils Lake in 1921 on May 3. 4. Colymbus nigricollis californicus (Heermann). FEared Grebe. Barrp, 1858, p. 897; records specimens taken September 17, 1856, at Fort Berthold. ITAYDEN, 1863, p- itr) Oe quite rare: two srecimens collected in *8=6 hetween Fort Union and Fort Berthold. Covers, 1878, p. 657; breeding abundantly in Turtle Moun- tains. JUDD, 1917, p. 5; abundant summer resident: breeding in colonies in all ponds of the prairie region that retain water all summer: one colony contained forty odd nests in a small patch of grass about one rod wide by a few rods long. BENt, 1919, 12 NorMAN A. Woop p. 28; quotes Goss, 1883, who found, the eared grebe breeding in communities. The first colony found was in a small lake in northern Dakota. There were twenty-five nests on an area of 10 by 20 feet, and the nests were small. ‘Three males in the Museum of Zoology Collection were taken at Sweet- water Lake, August 17, 1900, by Charles L. Cass. A female was taken at Stump Lake, June, 1897, by Alfred Eastgate. A set of eight eggs was collected June 11, 1900, at Sweetwater Lake, by Cass, and a set of six was taken at Devils Lake, North Dakota, May 28, 1898, by E. S. Bryant. The species was common at Devils Lake from July 18 to August 109, 1920, and from June 8 to July 20, 1921. 5. Podilymbus podiceps (Linnzus). Pied-billed Grebe. Cours, 1878, p. 658; not as common as the last two species. Jupp, 1917, p. 5; common, breeding in the deeper waters; earliest record April 27, and eggs were found hatching the first week in June. BEN’, I9QI9, p. 40; in the North Dakota sloughs, in 1901, this species was nesting abundantly, in company with canvas-backs, redheads, ruddy ducks, and coots. It was common on Devils Lake in 1920 and 1921. 6. Gavia immer (Brunnich). Loon. Jupp. To17, p. 5; common summer resident of the Turtle Mountains: breeding in the larger lakes. BuisuHop, 1921, p. 367; describes this as a new subspecies called lesser black-hilled loon. ‘Type. adult female. No. 13235. Carpenter Lake, Rolette County, North Dakota, July 13, 1905. L. L. B., collector. The skin of a male in the Museum of Zoology Collection was taken at Stump Lake, Nelson County, July 19, 1905, by H. A. Shaw. Another immature bird was taken at Lake Upsilon in October, 1919, by Mr. D. R. Ducke. ‘The writer saw individuals at Lake Upsilon from July 30 to August 7, 1920, and again on June 4, 1921. 7. Larus argentatus Pontoppidan. Herring Gull. ScHMIDT, 1904; all sections of the state as a migrant. SCHMIDT. 1920, p. 284; migrant and summer resident. Jupp, 1917, p. 5; a straggler during early spring migrations. FREEMAN, IQ19, p. 10; observed in flight April 6 and 8; a transient visitant. Mr. Alfred Eastgate tells me it is common in migration and breeds at Chase Lake Reserve, Stutsman County. 8. Larus californicus Lawrence. California Gull. Scumipt, 1904; nesting in central and eastern North Dakota. Jupp, 1917, p. 5; occasionally seen in summer; one shot in June, 1895, while flying over Rock Lake, Towner County, the first positive record for this region. Ripcway, I919, p. 622; breeding at Stump and Devils lakes, North Dakota. Brent, 1921, p. 131; one at Devils Lake, North Dakota, April 24. On July 25, 1920, a few individuals were seen on the Stump Lake Bird Reserve. A large gull was seen on Devils Lake in May, 1921, which may have been the herring gull or this species. Birp Lire oF NortuH Dakota 13 9g. Larus delawarensis Ord. Ring-billed Gull. Cours, 1878, p. 655; two specimens shot from a flock om September 12, 1873, near Mouse River, Dakota. HorrMan, 1882, p. 404; noticed daily and usually in pairs; found a poorly mounted specimen in the Great Lodge of the Arikaras, where it was used in one of the ceremonies. RipGwAy, I919, p. 625; breeding at Devils Lake and Stump Lake. BEN’, 1921, p. 133; on two islands of Stump Lake Bird Reserve, found breeding in colonies of about a hundred pairs in May and June, 1gor. Jbid., p. 140; egg dates, North Dakota, forty-eight records, May 9 to June 22; twenty-four records, May 31 to June 15. A set of three eggs is in the Oological Collection, Museum of Zoology, which were taken at Devils Lake, North Dakota, May 17, 1892, by James R. Craegue. A colony of this species bred at the Stump Lake Bird Reserve in 1920. At Devils Lake they were first noted in 1921 on May 10; on May 27 and June 13 a hundred or more were seen. There were a few about Devils Lake all summer, and a colony no doubt nested on a small island near Minne- waukan Bay. to. Larus franklini Richardson. Franklin’s Gull. Cours, 1878, p. 655; a specimen was shot in the Turtle Mountains, July 30, 1873, but no breeding colonies were found. R1ipGwaAy, 1910, p. 642; breeding in the Turtle Mountains and at Devils Lake. BEN’, 1921, opposite p. 171; a fine photograph of a pair on the nest, taken in North Dakota by H. K. Job. The skin of a male, collected at Sweetwater Lake, Ramsey County, June 5, 1900, by Charles L,. Cass, and a set of two eggg taken there June 8, 1896, by E. S. Bryant, are in the Museum of Zoology Collection. An abundant summer resident of all the lake regions of the state. It breeds on the islands of both Stump and Devils lakes, where hundreds were seen each day. On July 14, 1920, more than a thousand were assembled on a stony point of Minnewaukan Bay. More than half of these were young of the year. On July 25 hundreds were seen at Stump Lake, and at Lake Irvine, July 27. On August 25 a flock of thousands was seen feeding on the prairie, undoubtedly on grasshoppers. A small colony occupied Bird Island in Devils Lake in 1921. This is one of the most important of the North Dakota birds. It is very abundant and annually destroys countless numbers of the most injurious insects—crickets and grasshoppers. 11. Larus philadelphia (Ord). Bonaparte’s Gull. Jupp, 1917, p. 6; but few stragglers seen during the spring and summer months. BAILEY, 1918, p. 27; a few of this species September 1 at Sweetwater Lake, Ramsey County. It was not seen by the writer in July or August, 1920. At Devils Lake on June 8, 1921, a flock of about two hundred appeared, but all passed on after a few days. Many of these were still in the winter plumage. 12. Sterna forsteri Nuttall. Forster’s Tern. ScHMIDT, 1904; nesting in all three sections of the state. Jupp, 1917, p. 6; an occasional specimen taken. 14 NorRMAN A. Woop I did not positively identify this species, although many terns were seen daily. Mr. Alfred Eastgate tells me it nested at Kavanah Lake, Ramsey County, in 1898. 13. Sterna hirundo Linneus. Common Tern. BENT, 1921, p. 133; a large colony at Stump Lake, June 15, 1901; p. 244; egg dates, North Dakota, thirty-six records, May 31 to July 15. The skins of an adult and two young in the Museum of Zoology Collec- tion were taken at Devils Lake, July 8, 1919, by Crystal Thompson. This species is an abundant summer resident\of the Devils and Stump lakes region and the writer visited large colonies on sandy islands in both lakes. The Devils Lake colony on July 15, 1920, contained only nests! with full sets, many of them fresh. Nearly all nests of the Stump Lake colony had: young, some able to fly, on July 25. 14. Hydrochelidon nigra surinamensis (Gmelin). Black Tern. Cours, 1878, p. 656; a colony of this species was found breeding at Pembina in Tune, 1873, and others were seen later at the Mouse River. Cougs, 1897, p. 39; June *8. 1843, Audubon saw a black tern which, no doubt, bred there, at Fort Union, Wil- liams County. A large flock, comprised of hundreds of both adults and young, was seen at Lake Upsilon from July 30 to August 7, 1920. The species was seen at Devils Lake, May 23, 1921. It is an abundant summer resident in all suit- able habitats throughout the state and feeds chiefly on small fish and‘ crus- taceans. 15. Phalacrocorax auritus auritus (Lesson). Dotble-crested Cormorant. CouEs, 1878, p. 655; observed ou the Red River near Pembina late in May. Jupp, 1917, p. 6; tolerably common migrant, no records of its breeding except in Devils and Stump lakes. Barttey, 1918, p. 175; records a visit to the colony noted above on the bird rocks at Devils Lake. BENT, 1921, p. 133; visited in June, 1901, a colony of about seventy-five pairs on an island in Stump Lake. A juvenile skin in the Museum of Zoology Collection was taken at the Devils Lake colony on July 15, 1919. ‘This was the last year this species nested on Bird Island. Another juvenile in our collection was taken at ‘the Stump Lake Bird Reserve, June 23, 1919. ‘This was collected by Crystal Thompson, who also collected nests and birds for:a habitat group in the Biological Station Museum at Devils Lake. A set of two eggs collected May 30, 1896, at Sweetwater Lake, Ramsey County, by E. S. Bryant, is in the Museum of Zoology Oological Collection. It was a rare summer resident as late as 1918, when Crystal Thompson found a small colony nesting on Bird Island. On July 15, 1920, I counted thirty-three old nests and saw one adult bird. This colony left the lake per- haps because of the scarcity of fish and because it was disturbed by campers. On July 25, 1920, the colony on the Bird Reserve at Stump Lake was vis- ited and thirty-five adults and young were counted. The young were nearly Birp Lire oF NortH DAKOTA 15 full grown, but many of them were unable to fly, though they swam and dove like seals. ‘This is the same colony visited by Bent in 1901, when it numbered 150 birds. 16. Pelecanus erythrorhynchos Gmelin. White Pelican. CouEs, 1878, p. 654; an old female was shot from the steamer as we neared Pem- bina, and I heard of one or two others shot on the Red River in May. The species was only once again observed, near Mouse River early in September. Cours, 1893, p. 172; October 18, 1804, a pelican was killed, in Emmons County, by the Lewis and Clark Expedition, and on October 19 they saw pelicans in a large pond in the same county. Maximilian records seeing large flocks of pelicans October 11 ,1833, near Fort Union. THWAITES, 1906, p. 199; kort Union, October 16, 1833, white pelicans passed in large flocks. Jupp, 1917, p. 6; common migrant, said to have been found breeding at the east end of Devils Lake. Batry, 1918, p. 29; records a flock of about sixty on Sweetwater Lake, Ramsey County. In 1920 an old resident told me that 'many years ago a few nested at Pelican Bay, Devils Lake. Five were seen July 15, 1920, at Minnewaukan Bay. On May 10, 1921, six came into Creel Bay, and on July 19, forty were observed on a small island in Minnewaukan Bay. Residents of the central part of the state record that a few breed there. 17. Mergus americanus Cassin. Merganser. Jupp, 1917, p. 6; sometimes shot by hunters in the fall; none were seen by the writer in the summer. The Museum of the University of North Dakota has a mounted speci- men labelled Bismarck. Mr. J. D. Allen, of Mandan, mounted a male sent in from Bismarck, which was taken March 7, 1921, by W. E. Cleveland. The merganser occurs in small colonies in timber in the region of Devils Lake. A few males were seen in the Turtle Mountains. Sets of this species were completed from about May 26 to June 1, 1921. In one colony most of the eggs hatched from June 20 to 25. 18. Mergus serrator Linnzus. Red-breasted Merganser. Jupp, 1917, p. 6; tolerably common fall migrant. None were seen alive, but a fine mounted male is in the Williams Collec- tion at Grafton, dated April 7, 1919. ‘There is also one in the University of North Dakota Museum, labelled Devils Lake. ‘There is a mounted male in the Williams Collection taken at Mandan, April 5, 1905, by J. D. Allen. 19. Lophodytes cucullatus (Linnzus). Hooded Merganser. Cours, 1874, p. 585; breeds in North Dakota. Cours, 1878, p. 654; this is the only species of the genus actually observed. Specimens were taken at the Turtle Mountains, August 5 and 7, and at Mouse River, October 1, 1873. ALLEN, 1875, p. 68; met with near the head of Heart River about July 1, 1874, and also in September, 1874, but not seen elsewhere. Jor, 1899, p. 162; in 1890, about the first of June a pair were seen swimming in the Sheyenne River. BEN’, 1901, p. 331; one shot in Steele County, Jung 12, 1901, and six seen in Nelson County, June 15, 1901. Axsort, 1916, p. 3; found one hooded merganser breeding, the only fish duck observed. Jupp, 1917, p. 7; saw one flock of young in the Big Coulee. BArLEey, 1919, p. III; saw a female hooded merganser at Sweetwater Lake. FREEMAN, I910, p. 10; frequently seen in the Red River, May 13 to October 21. 16 NorMAN A. Woop There is a mounted male without data in the State Agricultural College. The collection of Mrs. Olsen, of Medora, contains a female skin taken on the Little Missouri River by Carl Olsen in 1920. Mr. Russell Reid informs me that on October 15, 1919, a female was shot at Bismarck, and that another was collected at Steele, Kidder County, November 2, 1921, by H. A. Sinish. 20. Anas platyrhynchos Linneus. Mallard. Cours, 1878, p. 649; flappers about a week old were seen at Pembina, June 20, 1873, on which date adult females and two young were collected. HorrmMan, 1882, p. 404; saw flocks upon the ponds, but the species was not common at any time in Sep- tember. Cours, 1893, p. 275; April 16, Lewis and Clark saw many mallards on some small creeks (in Mountrail County). Cours, 1897, p. 112; July 24, 1843, Mrs. Cul- bertson gave Audubon six young mallards which she had caught by swimming for them in the Missouri River at Fort Union. Bern’, 1901, Pl. IV; photograph of nest and ten fresh eggs. from Stump Lake, Nelson County, May 31, 1901. Jupp, 1917, p. 7; a set of eggs taken from a stack of wheat, twelve feet from the ground. There are mounted specimens in all of the state collections and an adult male in the Olsen Collection taken at Medora, October, 19109. An abundant summer resident, nesting in all suitable habitats. On July 15, 1920, a nest with twelve eggs was found out on the prairie, far from any large body of water. This species is of much economic importance. It furnishes sport and much valuable food. 21. Anas rubripes Brewster. Black Duck. Cours, 1897, p. 155; August 21, 1843, Audubon saw many old and young black mallards (near the mouth of the Little Missouri). Jupp, 1017, p. 7: Alfred Eastgate, of Stump Lake, and ex-State Auditor H. L. Holmes, of Bathgate, have each a mounted specimen. The species is of very rare occurrence in North Dakota. 22. Chaulelasmus streperus (Linneus). Gadwall. Cours, 1878, p. 650; abundant in the Turtle Mountains, August 5, 1873. Cours, 1807, p. 107; July 21, 1843, Audubon saw a single gadwall duck. These birds are found in abundance on the plains where water and rushes are present: p. 155; August 21, saw many young and old gadwalls (near the mouth of the Little Missouri River). Bent, 1901, Pl. V, No. 1, Pl. IV, No. 2, photograph of two nests found at Stump Lake, June 15. R&EED, 1904, p. 71; records a set of eight eggs collected June 19, 1808, on a low island in Devils Lake, Benson County, North Dakota, by E. S. Rolfe. A skin in the Museum of Zoology Collection was taken at Minot, October 5, 1910, by Robert P. Stark, and another, an adult male, was taken at Grafton, Walsh County, May 6, 1913, by Henry Williams. Many broods of young were seen at Devils Lake from July 15 to August 7, 1920, and at Stump Lake, July 25, 1920. It is a common summer resi- dent, nesting in thick grass about the larger ponds and lakes. 23. Mareca americana (Gmelin). Baldpate. Cours, 1878, p. 650: abundant breeder; young, still unable to fly, were found until the middle of September. Jos, 1899, p. 164; on June 27, 1808, found twenty nests of the baldpate on the islands of the Stump Lake Bird Reserve. BENvT, 1901, Brirp Lire or NortH DAKoTA 7, Pl. V, Fig. 2; Pl. VI, Figs. 1 and 2, p. 336; photograph of three nests at Stump Lake, Nelson County, June 15, 1901. REED, 1004. p. 72; a set of eggs from Lac aux Morte, Ramsey County, collected by E. S. Bryant. Jupp, 1917, p. 7; arriving middle of April. A not uncommon summer resident, nesting about the ponds and lakes. Many broods of young were seen from July 15 to August 24, 1920, at Devils and Stump lakes. 24. Nettion carolinense (Gmelin). Green-winged Teal. ALLEN, 1875, p. 67; a few pairs met with during the breeding season and a few small flocks seen in September. Covers, 1878, p. 650; thousands enter the country from the north in August; found no nests, but probably a few bred. Horrman, 1882, p. 404; frequent occurrence at Fort Berthold, Mclean County, September. Jupp, 1893, p. 382; May 4, shot a green-winged teal near Cando, Towner County. Jos, 1899, p. 163; met the green-winged teal on two occasions in Steele County, May 11, 1808, and on June 20 at Rolla, Rolette County, near the Turtle Mountains, found a female and her brood of eight young. Brnv, 1902, p. 1; near Rolla on June 20, 1898, Mr. H. K. Job found a female and eight juveniles. Jupp, 1917, p. 7; a few pairs nest each season, but it is a rare breeder. FREEMAN, 1919, p. 10; fairly common both ‘spring and fall. Barry, 1920, p. 69; at Island Lake two beautiful green-winged teal were brought in. I saw very few at Devils Lake from May to August, but at Odessa Nar- rows there were two females with broods of young on July 23, 1920. One female and a juvenile bird were taken for the record. 25. Querquedula discors (Linneus). Blue-winged Teal. Cours, 1878, p. 651; very abundant in August. One taken August 22, 1873, Mouse River. HorrmMan, 1882, p. 404; rather common at Fort Berthold in September. Cours, 1893, p. 275; April 16, 1805, Lewis and Clark report seeing blue-winged teal in some small creeks, Mountrail County. Cours, 1897, p. 176; August 2, 1843, Bell saw young blue-winged teal in a pond in the first prairie near Fort Union. Bern’, 1902, p. 2; found a nest and ten fresh eggs June 12, 1901, on Goose River, Steele County. ‘There is a male at Mandan taken June 19, 1921, and there are mounted birds in all of the state collections. An adult bird taken August 12, and an immature one taken August 15, 1900, at Sweetwater Lake by Charles L. Cass, are in the Museum of Zoology Collection. This is the most common breeding duck of North Dakota, nesting all over the state where there is water. A nest with eight fresh eggs was found in tall grass at the side of a road on July 18, 1920, and broods of young were seen on the lakes from June to August. The species is of much economic importance, because of its great abun- dance, and also its use for sport and food. 26. Querquedula cyanoptera (Vieillot). Cinnamon Teal. There is an adult male in the Williams Collection at Grafton, taken at Mandan, October 10, 1902, by J. D. Allen. The Agricultural College Col- lection contains a male taken at Maza, Towner County, in 1901, by W. L. Murray. This must be considered a straggler or an accidental visitor only. 18 NcrMAN A. Woop 27. Spatula cvypeata (Linnzus). Shoveller. ALLEN, 1875, p. 68; two or three shot near the head of Heart River in Septem- ber, 1873; no others observed. BEN, 1902, Pl. II, Fig. 1; photograph of nest and eleven eggs taken in Steele County, June 7. RED, 1904, p. 73; records a set of ten eggs, collected on Graham’s Island, Devils Lake, by E. S. Bryant, May 28, 1899. Jupp, 1917, p. 7; this species arrives about April 15, nesting in May. The Museum of Zoology Collection contains three skins taken at Sweet- waier Lake by Charles L. Cass: a female June 13, another June 14, and a male June 27, 1900. Puy, ine dae ok VE : [his is a very common summer resident, nesting in thick grass not far from water. 28. Dafila acuta (Linneus). Pintail. Cours, 1878 p. 650; two specimens taken June 20, 1873, at Pembina. Bent, 1902, Pl. II, Fig. 2; photograph of nest and eight eggs found in Steele County, June 10. REED, 1904, p. 73; set of ten eggs from Graham’s Island, Devils Lake, June 15, 1900, collected by FE. S. Bryant. FREEMAN, IQIQ, p. 10; a Fargo record of March 17. Many adults and young were seen at most of the lakes visited. On July 14, 1920, a female with nearly full-grown young was seen at Devils Lake. ‘This is one of the earliest migrant ducks, and is a common summer resi- dent, nesting in old stubblefields and on the prairie often far from any water. An important economic species. 29. Aix spénsa (Linneus). Wood Duck. ALLEN, 1875. p. 68; more or less frequent on the Missouri near Fort Abraham Lincoln and Fort Rice, June, 1873, but was not met with elsewhere. Barrp, 1858, p. 786; one in the National Museum Collection, taken at Heart River, September 22, 1856, by Dr. Hayden. Horrman, 1882, p. 404; common at certain seasons of the year. A single male seen in the possession of an Indian. Jos, 1899, p. 162; noted but once, a male in the Goose River. Jupp, 1917, p. 8; I know of but one specimen taken here. A male in moulting plumage was shot in August on Rock Lake. FREEMAN, 1919, p. 10; uncommon, May 14. There are two mounted males in the Fargo College Collection with no data; one in the Agricultural College Collection with no data; one in the University of North Dakota from Pembina County; and one in the Williams Collection taken at Grafton, May 8, 1908. This is a very rare duck in North Dakota. 30. Marila americana (Eyton). Redhead. Jos, 1899, p. 165; found redhead eggs in a, canvas-back nest. Brn’, 1902, Pl. III, Figs. 1 and 2; a nest taken in Steele County, June Io, 1901. There is a male in the Museum of Zoology Collection taken at Sweet- water Lake, September 14, 1900, by Charles L. Cass. This species was seen frequently on Devils and Stump lakes from May to August. Adults and young were noted in the Turtle Mountains on August 5, 1920. It is not a rare summer resident, nesting only on the larger lakes, in grass and rushes that grow in from one to three feet of water. It is one of the most important ducks for sport and food. Birp Liré of NortH DAKOTA 19 31. Marila valisinéria (Wilson). Canvas-back. Cours, 1878, p. 652; Turtle Mountains in July, 1873. Jos, 1800, p. 164; eleven pairs in one slough on May 21, 1898. Benv, 1902, Pl. IV, Figs. 1. and 2; photographs of nest, taken in Steele County, June 8. Bartey, 1919, p. 189; saw a female with nine nearly grown young on Sweetwater Lake, Ramsey County. There is a skin of a female in the Museum of Zoology Collection, taken at Sweetwater Lake, September 14, 1900, by Charles L. Cass, and a set of twenty eggs collected there by him, June 13, 1900. Another female ‘in the Museum of Zoology Collection was taken on May 18, 1913, at Stump Lake, by H. V. Williams. A few females were seen at Devils Lake in August, 1920, two broods at Sweetwater Lake on July 18, 1920, and others at the:Turtle Mountains, August 5 and 7, 1920. It is rather a rare summer resident, breeding in the grass and rushes of the water holes and coulees. It is considered superior to all other species for food, but has been much over-rated, as it is no better than the redhead. 32. Marila marila (Linneus). Scaup Duck. Jos, 1899, p. 163; considerable numbers in a large, reedy slough not far from Devils Lake in May and June. Jupp, 1917, p. 8; a common migrant. The species was seen only rarely at Devils Lake in May, 1921. 33. Marila affinis (Eyton). Lesser Scaup Duck. BENT, 1902, p. 105; the center of its abundance seems to be the Devils Lake region, but it was also found in Nelson and Steele counties. Jupp, 1917, p. 8; first nests found about June 15. A common summer resident, nesting in all the larger lakes and sloughs. This duck was very common on Devils Lake in May and June, 1921, but the only nest known to me was found near Bluebill Lake, Turtle Mountains, on August 5, 1920. It was in a low meadow, a third of a mile from the water, and contained eight eggs, nearly ready to hatch. 34. Marila collaéris (Donovan). Ring-necked Duck. Cours, 1878, p. 651; a specimen was seen in Dawson’s Collection. BEN’, 1902, p. 167; Mr. Job found a nest of the ring-necked duck in the Turtle Mountains, June 14, 1898. Dr. Bishop flushed two or three ducks from their nests in Nelson County in 1901, which he supposed to be ring-necked ducks, but none were taken. Jupp, 1917, p. 8; one pair seen in Big Coulee. Found breeding in the Turtle Mountains. There is a mounted male in the Williams Collection taken at Grafton, April 20, 1903. This is a rare summer resident. A pair was seen at Devils Lake, July 13, 1920. 35. Clangula clangula americana Bonaparte. Goiden-eye. BrYAnT, 1804, p. 183; a nest of ten eggs and one of thirteen at Devils Lake in 1892. Both nests were in elm trees, one fifteen feet and the other twenty-five feet from the ground. Jos, 1899, p. 164; breeding in small colonies in the timber region of Devils Lake. Brnt, 1902, p. 169; on June 1, 1901, found five nests of the American volden-eye in hollow trees at Devils Lake. Jupp, 1917, p. 8; found it breeding in the Turtle Mountains. 20 NorMAN A. Woop This is a rather rare summer resident, nesting in the wooded region about Devils and Stump lakes, and also in the Turtle Mountains. On May 15, 1921, aynest was found in the oak woods on the west side of Creel Bay, and on May 20, 1921, there was another in a cavity of an old oak tree on the Rock Island Military Reservation. ‘This nest was only fifteen feet above the ground and contained seven fresh eggs. On July 2, 1921, an adult female with twelve small young were swimming about on Minnewaukan Bay. 36. Charitonetta albeola (Linnzus). Buffle-head. Cours, 1878, p. 653; among the commonest species after the fall migration, and probably nests in Northern Dakota, as at Turtle Mountains young birds were found in July, 1873. A pair was seen at Devils Lake, May 10, 1921. 37. Harelda hyemalis (Linneus). Old-squaw. Jupp, 1917, p. 9; E. S. Bryant claims to have found one dead on the northwest shore of Devils Lake. The only specimens seen by the writer were a pair in the Agricultural College Museum, taken at Grafton, September 10, 1918, and Mr. Henry Williams reports that they are very rare in North Dakota. 38. Histrionicus histrionicus (Linneus). Harlequin Duck. OBERHOLSER, I919, p. 26; records a specimen, which he refers to the Pacific form. Histrionicus h, pacificus. This specimen, a mounted male, taken on the Missouri River at Mandan, September 23, 1912, by J. D. Allen, is in the Williams Collection. 39. Oidemia americana (Swainson). Scoter. This species must be considered as a very rare straggler in North Dakota, and only one record is known to the writer. Mr. Henry Williams reports a male taken at Stump Lake, Nelson County, September 17, 1922. 40. Oidemia deglandi Bonaparte. White-winged Scoter. Jos, 1899, p. 163; June 27, 1808, found eight nests on group of small islands in Stump Lake, which contained fourteen, thirteen, ten, ten, seven, six. one, and no eggs. The last was a new nest ready for eggs. This shows that the bird is a very late breeder. BEN?T, 1902, p. 171: one nest on June 15, 1901, contained five eggs; another, also on Stump Lake Bird Reserve, on June 22, contained twelve eggs. BAILEY, 1916, p. 55; eight birds at Stump Lake, June 28, 1913. JUDD, IQI7, p. 9; straggler to our lakes and ponds; a summer resident on Devils and Stump lakes. BAILEY, 1918, pp. 25 and 26; records two from Sweetwater Lake, Ramsey County, and half-grown young there late in August, 1917. The species was observed at Devils Lake on August'15, 1920. It was first noted in 1921 on May 25; on May 31 ten birds were seen on the lake. Dr. R. T. Young and the writer found a nest with twelve eggs almost ready to hatch at Stump Lake, July 25, 1920. This nest was built in a thicket of gooseberry bushes on the highest point of one of the islands of the Bird Reserve. The female was very tame and fluttered only a few feet away from the nest. Brrp Lire oF NortH DAKOTA 21 Devils and Stump lakes seem to be the most southern breeding grounds of the species and the only region in the United States where nests have been found. 41. Erismatura jamaicensis (Gmelin). Ruddy Duck. Cours, 1878, p. 654; common and breeding in suitable localities throughout the region. At the Turtle Mountains it was nesting in numbers in the pools where the young in late July and early August were still unable to fly. Several were taken there, July 28 to 30 and August 7, 1873. Jos, 1899, p. 164; in one slough on May 21, 1898, numerous ruddies in pairs were swimming among the reeds. BENv, 1902, p. 172; found five nests of the ruddy duck in two sloughs in Steele County in June, 1901. JUDD, 1917, p. 9; very common; breed on all the big lakes, where they arrive the latter part of April. There is a skin of an adult female in the Museum of Zoology Collection taken June 25, 1900, at Sweetwater Lake, by Charles L. Cass, and there are mounted specimens in several bird collections in North Dakota. Several individuals were seen at Stump Lake, July 24 and 25, 1920, and a few at Lake Upsilon, Turtle Mountains, July 30, 1920. ‘They were noted at Devils Lake from May 5 to July Ig, 1921. 42. Chen hyperboreus hyperboreus (Pallas). Snow Goose. Cours, 1878, p. 649; abundant during migrations. In the fall none were seen until October. Jupp, 1891, p. 169; snow geese were most abundant after October 17, 1890, near Cando. Cours, 1803. p. 203; Lewis and Clark reported that a great number of _ brant passed up the river, April 9, 1805, some of which were perfectly white except the large wing feathers. (Coues says these were snow geese.) JUDD, 1917, p. 9; flocks of thousands are a common sight from the last week in March to May 1 and a few small flocks stay until June 1. The big flight in the fall occurs about October 10, but some stay until the lakes freeze over. BAILEY, 1920, p. 106; saw a snow goose the last of September at Sweetwater Lake. I did not see this species alive. There is a mounted specimen in the University of North Dakota Museum, taken at Devils Lake, and also one at the Agricultural College. Another specimen was taken at Bismarck, Novem- Per 11, 1921, by E. J. Erbe. 43. Chen hyperboreus nivalis (J. R. Forster). Greater Snow Goose. Jupp, 1917, p. 9; specimens whose size would apply to this subspecies are shot at all times from the large flocks of the lesser snow goose. There is a specimen without data in the University of North Dakota Museum. Another, taken at Grafton, April 13, 1907, was seen in the Williams Collection. 44. Chen cerulescens (Linneus). Blue Goose. Jupp, 1893, p. 382; on May 4, shot a blue goose near Cando, Towner County. Jupp, 1917, p. 9; rare in migration; one to four are usually seen in nearly every large flock of snow geese. FREEMAN, IQI1Q, p. 11; observed fairly often; date March 24. There is a specimen in the Agricultural College Museum collected at Grafton, October 27, 1913, by H. V. Williams. Another was taken from a large flock at Grafton, March 30, 1906, by William Williams. 22 NorMAn A. Woop 45. Chen rossi (Cassin). Ross’s Goose. Jupp, 1917, p. 9; one shot April 22, 1893. Heard of others being seen, but none came to notice. I have been able to find no other record of this species in the state. 46. Anser albifrons gambeli Hartlaub. White-fronted Goose. Jupp, 1891, p. 169; some shot near Cando, October, 1890. There is a mounted bird in the Williams Collection which was taken at Grafton, April 4, 1906. None were observed during the field work. 47. Branta canadensis canadensis (Linneus). Canada Goose. Cours, 1878, p. 649; saw several broods of goslings on the Missouri River in June, 1874. HorrMan, 1882, p. 404; immense numbers were observed throughout September at Fort Berthold, Mclean County. Cours, 1893, pp. 192, 250, 269, 270; Lewis and Clark reported large flocks of geese and brant passing south at Fort Mandan, November 10, 1894. On March 31, 1805, several flocks of geese, the first ones seen, flew up. By April 10 the geese were feeding near Fort Berthold. On April 13 a lake and adjoining creek flowing into the river were named Goose Lake and Goose Creek, because of the large number of geese feeding on them. They built their nests in the tops of the lofty cottonwood trees. Coues says these were Canada geese, and he confirmed the above statement when he passed this point in the river in 1874. Cours, 1897, p. 7; Audubon saw geese and goslings near the mouth of the Cannonball River, Morton County, June 5, 1843. BENT, 1902, pp. 173, 174; reports quite a number of Canada geese breeding in North Dakota. Found one nest on June 2, 1901, in a large slough in Nelson County; another which the young had left on June 10, in Steel County. REED, 1904, p. 84; records a set of five eggs from Ellingers Lake, May 18, 1806, collected by Edwin S. Bryant. Jupp, 1917, p. 9; common breeder, noted as early as March 8 and as late as November 20. At present this might be considered a rare breeder in the state. Captain Lonnevik, of the Rock Island Military Reservation, says that it is still nest- ing at Sweetwater Lake. On August 12, 1920, a family of five were seen on Devils Lake. A small flock flew over Devils Lake May 4, 1921. 48. Branta canadensis hutchinsi (Richardson). Hutchins’s Goose. Jupp, 1892, p. 314; saw this species near Cando. Jupp, 1937, p. 9; common migrants, which usually feed by themselves in flocks of ten to fifteen. A mounted specimen taken at Sweetwater Lake, Ramsey County, October 10, 1915, is in the Williams Collection. The species seems to be rather rare in the state. 49. Branta canadensis minima Ridgway. Cackling Goose. Jupp, 1917, p. 9; two specimens shot in 1894 and one in 1895, by Dr. L. B. Bishop, were assigned to this subspecies. A mounted specimen is in the Museum of Zoology Collection, labelled Sweetwater Lake, October 10, 1915. 50. Branta nigricans (Lawrence). Black Brant. The species has not been recorded in lists of North Dakota birds. ‘The only specimen seen was an immature bird in the Williams Collection, It was taken at Lakota, Nelson County, October 26, 1917. 3rRD LiFe OF NortH DAKOTA 23 51. Olor columbianus (Ord). Whistling Swan. CouEs, 1807, p. 5: Audubon says on June 5, 1843, he saw two swans light on the prairie near the mouth of the Cannonball River: p. 10; saw two on June 8 near Mandan: p. 154; on August 17 saw twenty-two swans near Fort Union. Jupp, 1917, p. 9; one specimen was brought in in the fall of 1894. There is a mounted specimen in the Biological Station Museum taken at Stump Lake by Alfred Eastgate, and one in the University of North Dakota Museum taken at Devils Lake. According to residents, they occur during migration at the lakes. Mr. Allen, of Mandan, mounted two birds from Fullerton, Dickey County, taken October 20, 1921, by George Paulson. 52. Olor buccinator (Richardson). ‘Trumpeter Swan. HAypeEn, 1863, p. 175; seen at certain seasons of the year in large flocks through- out the northwest. A few breed in the valley of the Yellowstone. Cours, 1878, p. 648; observed on a few occasions in Dakota late in September and the first half of October during the migration. It appears to pass chiefly at night. Jupp. 1803. p. 382; not uncommon in migration. Jupp, 1917, p. 11; tolerably common migrant as early as April 13 and the latter half of October. They seldom fly in greater num- bers than four to six. In the museum at the University of North Dakota is a specimen labelled Stump Lake, North Dakota, A. J. Hutchins. This bird stood by the side of a whistling swan and was much larger and taller; its wing measured twenty-six and a half inches and it had twenty-two tail feathers, while the other swans in the collection had each twenty tail feathers and wings from twenty to twenty-one and a half inches. A trumpeter swan (No. 17779) in the collection of H. K. Coale is labelled North Dakota, November, 1891.+ Mr. Alfred Eastgate told the writer that it bred at Rock Lake, Towner County, in 1895. No live birds were seen. 53. Plegadis guarauna (Linnzus). White-faced Glossy Ibis. The only records found are a specimen in the ‘Museum of Zoology, labelled North Dakota, 1882, and a mounted specimen in the Agricultural College Museum, taken at Mandan by J. D. Allen in 1912. Mr. Williams told me he saw a bird of this species at Amidon, Slope County, in May, 1920. 54. Mycteria americana Linneus. Wood Ibis. Mr. Williams, of Grafton, mounted a bird of this species in the spring of 1900. It was taken in a big slough east of Glasston, Pembina County. 55. Botaurus lentiginosus (Montagu). Bittern. CovEs, 1878, p. 646; rather common on Mouse River in September, 1873. One was taken on September 2, another September 8. Jupp, 1917, p. 11; very common, Small numbers were seen in nearly all of the regions visited. There is a specimen in the Museum of Zoology labelled Devils Lake, 1900, Charles L,. Cass. 56. Ixobrychus exilis (Gmelin). Least Bittern. Jupp, 1917, p. 11; two seen by Mr. Bishop at Rock Lake, June 20, 1895. Also reported from Devils Lake. 4 Auk, 1915, Vol. XXXII, pp. 84-85. 24 Norman A. Woop No living specimens were seen, but there is a mounted bird in the Agri- cultural College Museum at Fargo with no data. Mr. Williams, of Grafton, says it is very rare and he has seen but one, near Glasston, Pembina County, during the past twenty-five years. 57. Ixobrychus neoxenus (Cory). Cory’s Least Bittern. Mr. Alfred Eastgate is authority for a record of this form, at Graham’s {sland, Devils Lake, 1904. 58. Ardea herodias herodias Linneus. Great Blue Heron. Cours, 1878, p. 646; observed during passage down the Red River in 1873. ALLEN, 1875, p. 67; a single specimen seen on Heart River, the only representative of the family noted on the journey. HorrMan, 1882, p. 403; singly or in pairs this species occurs along the Missouri River during the early portion of September; none noticed after the tenth. Jupp, 1917, p. 11; rare; only now and then seen in recent years; breeds in the Turtle Mountains. The species seems to be quite uncommon in the state. There is a mounted specimen in the Biological Station Museum probably taken near Devils Lake. The Williams Collection contains one labelled Grafton, September 30, 1920. One in the Olsen Collection was collected at Medora in the summer of 1920. On August 2 and August 4, 1920, a family of four were seen at Lake Upsi- lon, Turtle Mountains. 59. Nycticorax nycticorax naevius (Boddaert). Black-crowned Night Heron. Cougs, 1878, p. 646; one individual seen during the passage down the Red River, in 1873. Bryant, 1894, p. 183; colony at Lake Alice, Ramsey County, in the spring of 1892. JuDD, 1917, p. 11; tolerably common, breed in colonies in trees about Lake Alice, also in heavy grass sloughs and thick bushes at Rock Lake. Earliest arrivals, April 13. A set of six eggs in the Museum of Zoology Collection was taken at Sweetwater Lake, Ramsey County, on May 31, 1896, by E. S. Bryant. They were not rare about Twin Lakes, Ramsey County, August 18, 1920, and later were seen at Devils Lake and a small mud pond on the Rock Island Military Reservation. 60. Grus americana (Linneus). Whooping Crane. Cours, 1878, p. 646; white cranes were frequently observed in the Mouse River country in August, September, and October, but always at a distance; there is no reason to doubt that they breed there. Assott, 1880, p. 984; a flock of three were met July 6, 1879, near Pembina. HorrmMan, 1882, p. 404; frequently seen flying north- ward about the middle of September at Fort Berthold, McLean County, North Dakota. Jupp, 1891, p. 169; saw white cranes October 5, 1890, near Cando. Cougs, 1893, p. 207; April 11, 1805, Lewis and Clark record seeing some cranes, the largest bird of that kind common to the Missouri and Mississippi, perfectly white, except the large feathers on the first joint of the wing, which are black, just above the mouth of the Little Missouri River in McLean County. THwartss, 1906, p. 183; on September 22, 1833, Maximilian saw a small flock of whooping cranes near Fort Union. Jupp, 1917, p. 11; tolerably common migrant, probably nests in the Turtle Mountains; earliest arrivals, April 13. Mr. Russell Reid, of Bismarck, reports one shot near Leroy, Pembina County, fall of 1919; two small flocks seen at Bismark, one September 16, 1920, and the other April 8, 1922. Birp Lirk oF NortaH DaKxoTa 25 There are several mounted birds in the state; one in the University of North Dakota Collection, taken in Cavalier County; one in the Williams Col- lection, taken at Lakota, Nelson County, in 1919; two in the Agricultural College Collection, with no data; one in the collection of Clarence Hale, shot at Glasston, Pembina County, in 1899. Mr. Williams furnished the data for a specimen taken by W. R. Ross at Calvin, Cavalier County, April 20, 1907; one at Sherwood, Renville County, by A. J. Clark, September 19, 1908, and one at Hamilton, Pembina County, by D. D. Warren, April 12, 1912. Mr. Alfred Eastgate tells me that the species bred commonly until 1908 in Nelson County. A flock of seven whooping cranes was seen from the car window, stand- ing at the edge of a small pond near the track, just east of Medina, Stutsman County, on June 24, 1921. They flew up as the train passed, making a great spread of black and white plumage. 61. Grus canadensis (Linneus). Little Brown Crane. ALLEN, 1875, p. 67; large flock seen at the crossing of the Little Missouri, Sep- tember 15, 1873, circling high in the air over the camp; the only time the species was observed. (Allen noted this under the species canadensis, but these birds may have been mexicana.) Jupp, 1917, p. 11; one pair shot out of a flock of ten or twelve, April 18, 1895, now in the collection of H. L. Bigelow, Boston, Massachusetts. There is a specimen in the Williams Collection taken near Grafton, September 30, 1920. In the Agricultural College Museum is an individual from Custer Flats, 1912, and in the Fargo College Collection is one from Goodrich, Sheridan County, 1909. 62. Grus mexicana (Miller). Sandhill Crane. CULBERTSON, 1851, p. 118; “June 13 one of the pilots brought me a very young sandhill crane, which I put in alcohol, and I intend preserving the skeleton of a larger one, which was not in a condition to skin.” (Missouri River, between Forts Clark and Berthold.) Cours, 1878, p. 646; commonly observed after leaving Pem- bina; it appears to breed over the whole region in prairie country. HorrMan, 1882, p. 404; more frequently seen than the whooping crane. Jupp, 1891, p. 169; saw sand- hill cranes near Cando, October 5, 1890. Cours, 1897, p. 9; Audubon saw a young sandhill crane near Heart River, Morton County, on June 6, 1843; p. 95; on July 16 there was a sandhill crane about two years old in a grassy bottom near Fort Union. THWAITES, 1905, p. 199; October 16, 1833, Maximilian records large flocks of cranes passing the boat near Fort Union. Jupp, 1917, p. 11; common, breeds in Towner County in isolated sloughs that contain more or less water. The fall flights are often very large and in some localities they do some damage to grain. earliest migrants noted April 6 and a full set of eggs June 9. BAtry, 1920, p. 67; the most notable bird found in the Turtle Mountains was a sandhill crane. This species was not seen alive. There are mounted specimens in the state collections, and J. D. Allen, of Mandan, mounted one from Steele, Kidder County, taken October 6, 1921, by M. Hogue. Mr. Russell Reid says it does not seem to be rare during migrations: on April 16, 1921, eight large flocks were seen, about five hundred birds in all. Residents told me there were a few still breeding in the big sloughs, that more were seen in migration, but they are no longer a very common species. Formerly 26 NorMAN A, Woop they were of some economic importance, furnishing sport and food for hunters. 63. Rallus elegans Audubon. King Rail. Jupp, 1917, p. 11; Bryant claims to have seen one at Freshwater Lake. There is a specimen without data in the Agricultural College Museum. This was the only bird seen. 64. Rallus virgimanus Linneus. Virginia Rail. ALLEN, 1875, p. 67; met with once or twice in June, 1873, near the Heart River. BENT, I90I, p. 330; Virginia rails nested in short grass about the edges of sloughs. Jupp, 1917, p. I1; not common and noted only in fall migrations. FREEMAN, I9QI9, p- 11; has photographs of individuals found in Island Park at Fargo. The species was not seen in the field. ‘There is a specimen in the Agri- cultural College Collection taken at Grafton, June 3, 1910, by Henry Williams. 65. Porsana carolina (Linneus). Sora. Cougs, 1878, p. 647; observed during the migration in September along the Mouse River. BEN’, 190i, p. 330; nested in short grass about edges of sloughs. Jupp, 1917, p. Il; very common, nearly every slough containing water in June has a pair or more nesting. BAILEy, 1918, p. 67; common in the sloughs about Sweetwater Lake. FREEMAN, IQTO. p. II; very common, May 14. The species is more often heard than seen. One was observed at Dry Lake, Ramsey County, July 18, 1920, and one at Stump Lake, July 25, 1920. 66. Coturnicops noveboracensis (Gmelin). Yellow Rail. REED, 1904, p. 105; set of ten eggs collected June 4, 1901, in Benson County, by Rev. P. B. Peabody. Axssort, 1916, No. 94, p. 4; “Resting against a boulder, reveling in the prolonged sunset, a metallic ‘click’ penetrates the verdure directly in front. The author has several imitators. The birds are ventriloquists and you can’t disturb nor dislodge them. ‘Take two stones the size of black walnuts and bring them together squarely. This will convey to you a proper interpretation of the notes. They issue from throats of the little yellow rail, the feathered mouse of the north- ern swales.” ABsBort, 1919, p. 97; visited the yellow rail coulee in the Choctaw Basin, Benson County, North Dakota. The Rev. P. B. Peabody, of Blue Rapids, Kansas, in a letter of Novem- ber 16, 1921, says yellow rails are found rather commonly, breeding in the Sheyenne coulees of Benson County, North Dakota, in late May and early June. Eight to fifteen eggs are usual; never, so far as known, more. The nests are usually found in fine grass where water is one to four inches deep. The bird is abundant locally, and rather surpasses, in numbers, both the other rails, Rallus virgitianus and Porzana carolina, occurring in the same area. 67. Fulica americana Gmelin. Coot. Cours, 1878, p. 647; extremely abundant from the Red River to the Rocky Moun- tains. HorrMan, 1882, p. 404; frequently noticed along the Missouri River between Bismarck and Fort Berthold until September 20, 1881. Jupp, 19017, p. 12; breeds in numbers in all large sloughs and many of the smaller ones. They are very prolific layers and replace stolen sets; at least three thousand eggs were taken by Indians around Rock Lake in the early summer of 1895. Birp LiFe or NortH DAKOTA 27 There are two sets of eggs in the Museum of Zoology Collection; one lot of eleven taken at Larimore, Grand Forks County, June 10, 1896, by Alfred Eastgate, and another of six collected at Sweetwater Lake, Ramsey County, May 23, 1900, by E. S. Bryant. 68. Lobipes lobatus (Linneus). Northern Phalarope. Jupp, 1917, p. 12; tolerably common during spring migration, not noted in the fall. FREEMAN, IQIQ, p. I1; seen once on September 2. At Devils Lake this species was one of the earliest migrants from the north in 1920. On July 15 a flock of nearly one hundred were feeding on the water and several were taken for records. In 1921 a few were seen on May 17, and the number gradually increased until May 25, when a big wave of them came in, and I estimated that there were a thousand or more in Creel Bay alone. Thousands more arrived each day until, on May 30, there must have been from eight to ten thousand on the bay. ‘They were very tame and fed all around the boat on the numerous aquatic insects. Most of them left on June 2 and 3. On July 19 hundreds were on the lake again, the first arrivals from their summer homes in the north. 69. Steganopus tricolor Vieillot. Wilson’s Phalarope. Cours, 1878, p. 636; breeds throughout the country from the Red River to the Rocky Mountains. At Pembina breeds about reedy pools and prairie sloughs in June. REED, 1904, p. 100, a set of eggs from Larimore, Grand Forks County, May 30, 18907, collected by T. F. Eastgate. Jupp, 1917, p. 12; breeding in numbers around Rock Lake, Towner County. Nests were not far from the water on level, moist ground in a thick growth of grass. First arrivals were noted May 1. Bary, 1918, p. 65; photograph of nest and bird at Sweetwater Lake, Ramsey County, 1917. This is a common summer resident, breeding in suitable localities. The earliest spring arrival was May 26, 1921. During August, 1920, many flocks came and fed in the mud flats of Creel Bay, Devils Lake. 70. Recurvirostra americana Gmelin. Avocet. Cours, 1878, p. 635; not observed in the Red River region. S. C. C., p. 415; quite common, Bismarck, April 18, 1887. THwarrEs, 1905, p. 186; on September 28. 1833, near Fort Union, Maximilian records seeing the avocet with its strange upturned bill. Aszorr, 1916, p. 2; opposite this page is a photograph of a set of eggs im situ and another of the same showing young just hatched. These were evidently taken in North Dakota, but no data is given. Jupp, 1917, p. 12; a rare species, but undoubt- edly breeds. A female taken May 15, 1895, by Dr. Bishop contained an egg fully formed. The species was not seen in the field. Mr. Reid reports that it is fre- quent on some of the larger alkali lakes and sloughs northeast of Bismarck, and was noted at Chase Lake, Stutsman County, and was quite common at Lake Isabel, Kidder County, August 7, 1921. Mr. Alfred Eastgate tells me he saw avocets near Larimore in the summer of 1921. There is a mounted female in the Agricultural College Collection taken at Kenmare, Ward County, May 4, 1913. Two others, taken at Minot by W. S. Mum- mery, and one from the Turtle Mountains, October 5, 1917, by J. Kusler, 28 NorMAN A. Woop are also in that collection. The Olsen Collection contains a specimen taken in April, 1921, by Carl Olsen at the Little Missouri River. Mr. Allen, of Mandan, has mounted a bird taken at Dawson, Kidder County, October 3, 1921, and another from Windsor, Stutsman County, August 16, 1921. 71. Himantopus mexicanus (Miller). Black-necked Stilt. The only record we have of this species is of a bird seen by Mr. Williams at Grafton during the spring migration of 1905. 72. Philohela minor (Gmelin). Woodcock. Cours, 1874, p. 473; a woodcock shot at Fort Rice, Morton County. A rare migrant or straggler in North Dakota. A resident hunter reported that he secured three from a flight in October, 1920, near St. Johns, Rolette County. 73. Gallinago delicata (Ord). Wilson’s Snipe. Cours, 1878, p. 638; abundant in the Mouse River region in September, 1873, when several specimens were taken the 16th and 27th. Jupp, 1917, p. 12; moderately common migrant, very common from July to October, 1893, in wet, grassy sloughs. Since found breeding in Turtle Mountains in some numbers. FREEMAN, IQIQ, p. II; fairly common migrant, May 7. I did not find this species in the field, but saw a mounted bird in the Williams Collection at Grafton, taken there May 3, 1904, by Henry Williams. 74. Macrorhamphus griseus scolopaceus (Say). Long-billed Dowitcher. Cours, 1878, p. 638; during the fall migration in September the birds were numerous in the Mouse River region. Jupp, 1917, p. 12; moderately common, irreg- ular migrant in late summer. There is a mounted bird in the Williams Collection taken at Stump Lake, May 9, 1913, by Henry Williams. On May 31, 1921, I saw a bird of this species in a large flock of sand- pipers on a sandy point on Creel Bay, Devils Lake. Again, on July 24, a couple were seen on a small lake near Grafton in company with sanderling and lesser yellowlegs. 75. Micropalama himantopus (Bonaparte). Stilt Sandpiper. JupDD, 1917, p. 12; a small flock seen May 27, 1895, and another flock or two the last week in July, migrating south. There is a mounted specimen in the Agricultural College Collection taken at Stump Lake, May 9, 1913, by Henry Williams. A skin in the Museum of Zoology Collection was taken at Sweetwater Lake, August 13, 1900, by Charles L. Cass. Considering the rarity of this species, I was fortunate in finding a flock of about thirty on July 15, 1920. These were feeding on the mud shore of Minnewaukan Bay, Devils Lake, and all were still in the dark barred breed- ing plumage. A few others were seen at Bluebill Lake in the Turtle Moun- tains on August I, 1920. On May 31, 1921, a flock of ten or a dozen in the Birp Lire oF NortH DAKoTa 29 breeding plumage were noted on the shore of Creel Bay, Devils Lake; these were the only ones seen in the spring migration. 76. Tringa canutus Linneus. Knot. The only knots known by the writer to have been taken in North Dakota are a bird in the Williams Collection, collected at Stump Lake, May 19, 1913, and another in the Agricultural College Collection, taken at Grafton, May 30, 1919, by Henry Williams. 77. Pisobia maculata Vieillot. Pectoral Sandpiper. Cours, 1878, p. 640; common in the fall migration. First seen the latter part of July on the pools at the base of Turtle Mountains. Two specimens were taken July 28, 1873. JUDD, 1917, p. 12; abundant migrant, arriving the last week in April. At Turtle Mountains this species was common from July 30 to August 7, 1920, in all the ponds and lakes visited. In the spring my first record was on May 27, 1921, when ten were seen on the beach of, Devils Lake. In the fall migration it was first seen July 21, 1920, at a mud pond in the Rock Island Military Reservation. 78. Pisobia fuscicollis (Vieillot). White-rumped Sandpiper. Jupp, 1917, p. 12; most abundant migrant of the sandpiper family, first seen April Io. There is a mounted specimen in the Agricultural College, taken at Ken- mare, Ward County, June 1, 1913, by H. E. Peck. The species was observed in the field but once, on July 2, 1921, on the shore of Minnewaukan Bay. ‘The only two seen were taken. The species is apparently becoming quite rare in the state. 79. Pisobia bairdi (Coues). Baird’s Sandpiper. Cours, 1878, p. 640; during the fall migration in the month of August, one of the most abundant sandpipers in Dakota. Small flocks were found along the Mouse River, where specimens were taken August 21-30, 1873. JUDD, 1917, p. 12; common migrant; arrives third week in April. The first spring record was May 9, 1921, at Devils Lake, and more were recorded daily until May 23, when more than two hundred were seen. These gradually disappeared and by June 1o all had gone. The first fall record was July 14, 1920. After this date it became common at Devils Lake and was seen frequently at several other lakes and ponds. In 1921 ten were seen at Devils Lake on July 18, after which it became very common. 80. Pisobia minutilla (Vieillot). Least Sandpiper. Cours, 1878, p. 640; taken July 30, 1873, at Turtle Mountains. Jupp, 1917, p. 13; fairly abundant migrant, first arrivals May 2, staying three or four weeks. This species was seen first on May 10, 1921, at Devils Lake, and by May 28 two hundred or more were there. They had all gone by June 3. On the return migration a few were noted on July 2 at Devils Lake, and on July 27 they were quite abundant. 30 Norman A. Woop 81. Pelidna alpina sakhalina (Vieillot). Red-backed Sandpiper. Jupp, 1917, p. 13; fairly common migrant in 1895; first arrivals May 2. There is a specimen in the Williams Collection, taken at Grafton, May 16, 1910, by Mr. Henry Williams. In 1920 none were noted in July and August, but one was taken and another seen July 2, 1921, on the shore of Minnewaukan Bay. Seven of this species were on the beach at Creel Bay, May 7, two on May 16, and three on May 18, 1921. 82. Ereunetes pusillus (Linneus). Semipalmated Sandpiper. Cours, 1878, p. 639; abundant species noted at various points during the month of August, one taken at Mouse River, August 10, 1873. HoFFMAN, 1882, p. 403; not common near Fort Berthold, McLean County. Jupp, 1917, p. 13; equals the least sandpiper in numbers. ‘The species was very common after May 9, 1921, at Devils Lake. As late as May 31, a hundred or more were still there. ‘Two were seen on the return migration, July 18, 1921. On July 21, 1920, large flocks were seen at a mud pond on the Rock Island Military Reservation. 83. Ereunetes mauri Cabanis. Western Sandpiper. Jupp, 1917, p. 13; found with the semipalmated sandpipers, which they closely resemble. On May 31, 1921, a single western sandpiper was taken from a large flock of sandpipers feeding on the beach at Devils Lake. While this was the only one identified, the species no doubt occurs rarely during the migrations. , 84. Calidris leucophea (Pallas). Sanderling. Jupp, 1917, p. 13; not common, first date May 28, The sanderling seems to be a rather common species in the Devils and Stump lakes region, especially during the fall migration. In 1920 the first were seen at Stump Lake July 24, but it soon became abundant on all the lake beaches and by August 24 1t was one of the most common waders at Devils Lake. The first record for 1921 is May 26, at Devils Lake, when ten were noted. On May 31 over two hundred, and on June ro thirty were seen there. 85. Limosa fedoa (Linneus). Marbled Godwit. Barrp, 1858. p. 741; records the specimen mentioned below, No. 5450. as collected by Dr. Hayden, but does not give the date. However, it is listed in the Catalogue of the Trowbridge Collection and has the following data: “Limosa fedoa, Ord. Mar- bled Godwit. Male. Ft. Union, Nebraska (Dakota), July. Lt. Warren; Dr. Hayden. (5450.)” (The specimen is in the Museum of Zoology Collection.) Hayprn, 1863, p. 175; not common; took one at Fort Union. Cours, 1878, p. 641; a female was taken June 20, 1873, at Pembina, where they were probably breeding. Cours, 1897, p. 176; August 2, 1843, Bell saw several godwits near Fort Union. Rep, 1904, p. 119; a set of four eggs taken June Io, 1895, at Devils Lake by W. F. Hill. Benv, 1907, p. 160; saw about twenty-six godwits in Steele County, June 13, 1901. At Birp Lire oF Nort DaKkotTa arr Stump Lake, Nelson County, Dr. Louis Bishop noted a flock of about a hundred July 28, 1902, and on July 26, 1905, saw flocks of about forty adults and young. Jupp, IQI7, p. 13; a few pair nested in Towner County. Mr. Reid reports that a few breed in sloughs south of McKenzie, Bur: leigh County, each year. Noted June 9, 1921. In 1920 godwits were first seen at a mud pond in the Rock Island Mili- tary Reservation on July 21.. They were seen at Stump Lake, July 24, and flocks of eight to twelve became common in August at Devils Lake. They fed on the lake beaches as late as August 24, when all were in winter plu- mage. Two were seen on the shore of Creel Bay, Devils Lake, on May 11, 1921. It was not noted again until June 3, when a pair were found in a wet marsh near Cando, where they were breeding. 86. Limosa hemastica (Linneus). Hudsonian Godwit. Jupp, 1917, p. 13; a fairly common migrant in 1895, first seen April 24. There is a specimen in breeding plumage in the Williams Collection, taken at Grafton, May 7, 1911. An adult female, taken May 15, 1913, at Stump Lake, Nelson County, by Henry V. Williams, is now in the Museum of Zoology Collection. The species seems to be rare in the state. Coues did not see it in 1873, and I saw none during 1920 and 1921. 87. Totanus melanoleucus (Gmelin). Greater Yellow-legs. BairpD, 1858, p. 732; one taken at Fort Berthold, McLean County, in 1856. ALLEN, 1875, p. 66; seen a few times on the Heart River in September, 1873. Cougs, 1878, p. 642; not observed until the last week in July; very abundant in August and Sep- tember throughout the region. Specimens were taken on Mouse River, August 19, 23, 24, 25, 1873. Jupp, 1917, p. 13; fairly common migrant, first arrivals third week in April. FREEMAN, I919, p. 12; fairly common, May 5. Only one was seen. This was feeding on a mud flat near the Missouri River, at Williston, July 16, 1921. There is a mounted specimen in the Williams Collection taken at Grafton, April 30, 1909, and another, taken there April 19, 1914. 88. Totanus flavipes (Gmelin). Yellow-legs. Cours, 1878, p. 643; very abundant in the Turtle Mountains and Mouse River region, August 5, I0, 22, 25, 30. JUDD, I9I7, p. 13; common migrant, arrives in flocks about April 20. This was one of the most common waders present on May 3, 1921, at Devils Lake, where it lingered until May 17. In 1920 it was first seen July 13 at Devils Lake, where flocks of two hundred soon became common. At Lake Upsilon, Turtle Mountains, sixty were counted in a flock August 4, 1920. 89. Helodromas solitarius solitarius (Wilson). Solitary Sandpiper. Barrp, 1858, p. 734; one taken at Fort Union, July, 1856, by Dr. Hayden. Cougs, 1878, p. 643; occurs in abundance on all pools and water courses of the region during the autumnal migration, and there is reason to believe that some may breed in this latitude. Specimens taken at the Turtle Mountains, August 5, 1873, and Mouse River, August 22, 23, 1873. 32 NorMANn A. Woop Adults and young were seen at Devils Lake on July 30, 1920. The spe- cies was present and seemingly at home at Lake Upsilon, Turtle Mountains, on June 5, 1921. It is evident that it breeds in the lake and mountain regions at least. 90. Catoptrophorus semipalmatus inornatus (Brewster). Western Willet. Cours, 1878, pp. 641-642; Red River to Rocky Mountains. Breeds in this region. HoFrrMAN, 1882, p. 404; frequent along the sand-bars of the Missouri. Jupp, 1917. p. 13; common summer resident, arrives first week in May. Batey, 1918, pp. 113-114; family of nearly grown young near Sweetwater Lake, Ramsey County. There is a male skin in the Museum of Zoology Collection taken in Ram- sey County, June 5, 1900, by Charles L. Cass. Mr. Alfred Eastgate found a nest with eggs in a wet meadow near Bot- tineau in June, 1920. On July 23, 1920, several adults with young were seen at Odessa Narrows. The young were mostly well grown. Fifteen birds were seen on July 27, 1920, at Devils Lake, and it was common there until August 20. In 1921 the species was first seen on May 12 at Devils. Lake, and a few individuals fed on the beaches all of May and June. A pair were seen in a wet marsh near Cando, June 3, 1921. gt. Bartramia longicauda (Bechstein). Upland Plover. BairD, 1854, p. 738; taken at Fort Union, July, 1855. Aten, 1875, p. 67; very common on the prairies east of the Yellowstone, where it outnumbers all the other Grallae together. Cours, 1878, p. 644; extremely abundant over all the prairie of the Red River region, where they breed in June. ‘Twenty-six specimens taken at Pembina, June 4 to June 28, 1873, one at the Turtle Mountains, July 23, and one at. Mouse River, August 19, 1873. Assotr, 1880, p. 985; says it is one of the most common birds of the prairie. HorrMan, 1882, p. 404; rather common during first three weeks of September. Jupp, 1893, p. 382; shot in numbers. BENT, IQOI, p. 320; scattered about in pairs nesting in short grass. REED, 1904, p. 123; set of eggs col- lected June 10, 1897, at Stump Lake, Nelson County, by Alfred Eastgate. Jupp, 1917, p. 13; abundant summer resident formerly, but in 1910 the species was uncom- mon, and now only pairs are seen, and they are growing less in numbers very speedily. Baimnry, 1918, p. 112; a few on & strip of prairie near Sweetwater Lake. A mounted specimen is in the Williams Collection, taken June 6, 1909,. at Grafton, and there is a pair in the Museum of Zoology Collection taken. at Grafton, June 10, 1914, by Henry Williams, Mr. Williams tells me that on the night of September 12, 1921, he heard a flock passing south overhead. I was surprised at the scarcity of this species. In 1920 I saw but four in all the region worked, and in 1921 but three, on June 4, near St. Johns. It is rapidly becoming more rare. 92. Tryngites subruficollis (Vieillot). Buff-breasted Sandpiper. Jupp, 1917, p. 13; a few single birds and one large flock were seen the last week in May, 1895. FREEMAN, IQIQ, p. 12; one record, September 21. 93. Actitis macularia (Linneus). Spotted Sandpiper. Cours, 1878, p._644; summers throughout the region, specimens taken at Pem- bina, June 4, and Mouse River, August 10, 1873. JUDD, 1917, p. 15; a rare bird om the prairies, more common near the Turtle Mountains, where it breeds. Birp Lirk oF Norta DaAKota 33 A few were found at. Lake Upsilon, Turtle Mountains, from July 30 to August 7, 1920. At Devils Lake, in 1921, the first were seen on May 13, and they were observed almost daily on Creel Bay, where they no doubt nested. They occurred at all other places worked about the state. 94. Numenius americanus Bechstein. Long-billed Curlew. ALLEN, 1875, p. 67; a few pairs met with at quite distant intervals from the Mis- souri to Montana. Covues, 1878, p. 645; breeds in moderate numbers about Pembina, the only locality where it was observed during the first season. Cours, 1893, p. 276; April 17, 1805, Lewis and Clark report seeing a curlew, Numenius longirosiris, below Williston. Cours, 1897, p. 176; Bell shot two curlews near Fort Union in August, 1843. THWAITES, 1905, p. 369; on June 22, 1833, Maximilian records seeing the great curlew near the mouth of the Little Missouri River, and October 11, flocks of curlew near Fort Union. The long-billed curlew is apparently rare in the state. There is one in the Williams Collection taken at Mandan, June 14, 1903. There are mounted specimens in the Agricultural College Museum, taken at Williston, July 26, 1913, and at Mandan, June 14, 1903. The species is reported to breed across the Missouri River, about ten miles south of Williston. Mr. Henry Wil- liams reports that they breed at Ardoch, twenty miles south of Grafton. Mr. J. D. Allen states that in 1881 curlews nested on the prairie north of Mandan, but a few years later they became scarce and he has not seen one for several years. Senator Crawford, of Sentinel Butte, writes that he saw but two pairs during the summer of 1921 and that the last of June they were nesting in the vicinity. No live birds were seen during the time in the field. 95. Numenius borealis (J. R. Forster). Eskimo Curlew. Bairp, 1858, p. 744; records two specimens taken in 1856 by Dr. Hayden. Hay- DEN, 1863, p. 175; not uncommon, high up toward the sources of the Missouri. Speci- mens were taken near Fort Union. 96. Squatarola squatarola (Linneus). Black-bellied Plover. McCuHeEsney, 1878, p. 86; observed it in Dakota Territory only during the fall migration about October 25; quite common. JupD, 1917, p. 15; tolerably common migrant the last week in May, usually in very small flocks or singly, and sometimes_ seen with the golden plover. A mounted specimen is in the Williams Collection at Grafton, taken Sep- tember 21, 1906, and a mounted female at the Agricultural College from Upper des Lacs Lake, May 18, 1913, collected by H. EK. Peck. The Museum of Zoology Collection contains a male taken September I1, 1913, at Grafton, and another collected November 6, 1921, near Grafton, by H. V. Williams. This species was not noted in 1920, but in 1921 four were seen on Creel Bay, May 26, in company with Baird’s and other sandpipers. They fed there until May 31, after which date none were seen. 97. Charadrius dominicus dominicus (Muller). Golden Plover. Bairp, 1858, p. 691; one taken at Fort Berthold, Mclean County, September 16, 1856, by Dr. Hayden. Coues, 1878, p. 633; no golden plovers are seen in summer in any portion of the region explored. They pass through in large numbers in May 34 Norman A. Woop and return the latter part of September, being abundant at this time in the Mouse River region, where many specimens were shot September 27, 1873. Jupp, 1893, pb. 382; they pass in small flocks, feeding on burned. fields or prairies. Jupp, 1917, p. 15; common migrant, usually arrives second week in May. A number of flocks were seen moving south early in July. 1895. FREEMAN, I919, p. 12; observed frequently in the fall, September 21. A male and female of this species, taken at Stump Lake, August 21, 1891, are in the Museum of Zoology Collection, the gift of Alfred Eastgate. There are several in the Agricultural College Collection, two of them labelled Kenmare, Ward County, May 13, 1914, H. E. Peck. The Williams Collec- tion contains one taken at Grafton, September 7, 1907. The species is apparently becoming much less abundant than formerly. Only one small flock was seen, at Devils Lake, May 24, 1921. Mr. Alfred Eastgate saw a flock of about five hundred at Bottineau in May, 1920. 98. Oxyechus vociferus (Linneus). Killdeer. Cours, 1878, p. 634; abundant throughout the summer in all suitable places. Specimen taken June 11, 1873, at Pembina. HorrMan, 1882, p. 403; rather frequent on sand-bars above the village (Fort Berthold) and along the pools of water farther inland. Jupp, 1917, p. 15; common summer resident; first arrivals, March 30. This species is quite common in all the parts of the state worked. In 1921 it was at Devils Lake, May 2, when we arrived, and a pair with downy young was seen there July 13, 1921. 99. Aigialitis semipalmata (Bonaparte). Semipalmated Plover. Jupp, 1917, p. 15; rare, one seen by Dr. Bishop at Rock Lake, May, 1895. Free- MAN, 1921, p. 12; rare, May 17. In 1920 a semipalmated plover was seen at Devils Lake on July 14 and several were noted in August at Lake Upsilon, Turtle Mountains. One was taken from a flock of least sandpipers on August 4. May 17, 1921, is the earliest record I have, when only two or three were seen. Others were noted on May 19 and 25 and June 1, and one was taken on July 19 at Devils Lake. 100. Azgialitis meloda (Ord). Piping Plover. REED, 1904, p. 130; nest from Devils Lake, Benson County, May 26, 1901. Jupp, 1917, p. 15; Devils Lake near Graham’s Island. In 1920 adul*s with half-grown young were seen July 14; a juvenile bird was taken from a brood of four at East Bay, Devils Lake, July 25; several were seen on August 21. On July 24, 1921, several adults with broods of young, some of which were only a few days old, were noted on the Devils Lake beaches; July 19 a brood of nearly grown young were found at Min- newaukan Bay. ‘The species is not rare as a summer resident in the Devils Lake region. tol. Podasocys montanus (J. K. Townsend). Mountain Plover. ALLEN, 1875, p. 66; two or three small flocks were met with in September, but generally they were seen only in single pairs at intervals of several days. Birp Lire oF NortH DAKOTA 35 102. Arenaria interpres ntorinella (Linneus). Ruddy Turnstone. Jupp, 1917, p. 15; tolerably common around Rock Lake in 1895. This species was not seen in July and August, 1920, but at Devils Lake in 1921 they were very common after May 23, when about fifty were seen, until June 11, when they left the lake. From May 25 to June 5 hundreds were seen each day feeding or resting along the rocky shores of Creel Bay and Fort Totten Bay. These were all in breeding plumage and quite tame. 103. Bonasa umbellus umbelloides (Douglas). Gray Ruffed Grouse. KEENEY, 1875, p. 220; found ruffed grouse along the Red River in scrub timber near Fargo, October, 1875. JUDD, 1917, p. 15; common in the Turtle Mountains. The species was found to be very common in the timber of the Turtle Mountains in July and August, 1920. There were twelve mounted specimens in the museum at the State Fish Hatchery at Lake Upsilon, which had been taken in the vicinity. Most of these birds were light gray with black ruffs and tail bands, but a few were a dark reddish chestnut with red ruffs and copper tail bands. Mr. Alfred Eastgate, who has collected many of these birds, says that both the red and gray forms occur in the same covey. My own experience agrees with this and suggests a dichromatic phase. In the Museum of Zoology Collection are seventeen specimens from the Turtle Mountains, ten with gray-colored tails and seven with brown- or chestnut-colored tails ; all but two have black ruffs and tail bands; the excep- tions have chestnut-colored ruffs and copper-colored tail bands. ‘This seems to be almost the extreme of the red phase, for I saw but one other in North Dakota that had more of the red color. 104. Lagopus lagopus lagopus (Linneus). Willow Ptarmigan. Mr. Russell Reid reports that a poorly mounted specimen taken October, 1909, in the Kildeer Mountains, Dunn County, was sent to J. D. Allen, at Mandan. The species is a straggler or accidental winter visitant in the state. Sen- ator Lewis F. Crawford, of Sentinel Butte, writes that he has seen one specimen, killed a few years ago in the western part of the state near Kildeer. This is probably the specimen referred to above. 105. Tympanuchus americanus americanus (Reichenbach). Prairie Chicken. THOMPSON, 1890, pp. 514-515; in 1883 it began to be common at Pembina. Jupp, IQI7. p. 15; common simmer resident. The species seems to gradually follow the settlement of the country and to now have a quite general distribution over the state in the prairie regions. There were thirty or more, August 9, 1921, in one flock near Bottineau, fifteen miles south of the United States boundary. A few were seen near Grafton in the Red River Valley, about forty miles south of Pembina, where Dr. Coues did not find it in 1873. A few were found at Medora and at Williston residents said it was common, 36 NorMAN A. Woop 106. Pedioecetes phasianellus campestris Ridgway. Prairie Sharp-tailed Grouse. Bairp, 1858, p. 627; one taken at Fort Union by Dr. Hayden; one taken at Fort Union, 1843, by J. J. Audubon. Cougs, 1878, p. 630; commen from Pembina to the Mouse River; secured a number of specimens from June 4 to August 24, 1873. HorrMan, 1882, p. 403; several specimens secured by the Indians in timber four miles west of the Agency at Fort Berthold. Cours, 1893, p. 274; April 15, 1805, on the high plains of McLean County, Lewis and Clark record many grouse which appear to be mating. Cours, 1897, p. 26; Audubon says on June 11 he saw a fine flock of sharp-tailed grouse near the mouth of the Knife River, Mercer County: p. 87; on July 13, 1843, Harris saw six sharp-tailed grouse near Fort Union, and on July 29 Harris and Bell shot an old and young of this species. THWAITES, 1906, p. 199; October 11 to 30, 1833, Maximilian records seeing flocks of prairie hens near Fort Union: p. 208; October 31, numerous flights of prairie hens were crossing the river in flocks of thirty or forty. FREEMAN, 1921, p. I2; winter visitant, uncommon. There are three immature specimens in the Museum of Zoology Collec- tion, taken in Ramsey County, August 2, 1900, by Charles L. Cass. The species is common in the Devils and Stump lakes region, especially on brushy areas near timber. It also occurs in summer on the open prairie, and is found throughout the state in suitable habitats. It was found to be very common in the Turtle Mountain region in the clearings and open fields. In winter this species migrates southward and enters all the timber tracts in numbers. Mr. Williams, of Grafton, tells me that the Columbian sharp- tail is found in the northern parts of the state at that time. 107. Centrocercus wrophasianus (Bonaparte). Sage Hen. ALLEN, 1875, p. 65; mone seen east of the Little Missouri. Horrman, 1882, p. 403; not common at the village, Fort Berthold, McLean County, but reported more abundant further west. RoosrvEir, 1885, p. 149; the sage grouse is preéminently a grouse of the plains and is never found near trees; he shot them on the high plains west of his ranch. Bernopirk, 1892, pp. 106-107; range of sage grouse in the state is western North Dakota. The sage hen was not seen alive, but several people reported its occur- rence. Carl Olsen told me that he had shot some in recent years about thirty miles southwest of Medora. Victor File, of Medora, said that in 1915 he found them common near Marmarth, Slope County, where he shot sev- eral. Hon. Lewis Crawford, of. Sentinel Butte, Billings County, in a letter to the writer dated August 23, 1921, says the sage grouse is plentiful about thirty miles south of there. At one place on the Cannonball River he has seen thousands of cocks on an old prairie dog town, bulling. In the breeding season all the males come together for this show-off, and at this time are very gentle and can be approached up to within fifty feet. The females are on their nests at this time and very rarely seen at the gathering. 108. Ectopistes migratorius (Linneus). Passenger Pigeon. Extinct. Coves, 1874, p. 388; along the Missouri often saw small flocks and noticed quite a number of nests in small trees between Fort Pierre and Fort Berthold. Cours, 1878, p. 628; countless flocks seen during the voyage down the Red River and at Pembina, where specimens were collected from June 4 to 14, 1873. Took one nest containing a single egg, June 13, and one bird in the Turtle Mountains in July. Apsorr, 1880o,. Birp Lire or Noryu DaKora hehe p. 985; one small flock met with in July, 1879, near Pembina. Cours, 1897, p. 156; August 23, 1843; Audubon says near mouth of Little Missouri they were passed by passenger pigeons. THWAITES, 1906, p. 32; on July 10, 1833, Maximilian records see- ing many wild pigeons near Fort Union. Jupp, 1917, p. 29; J. F. Rickbell, of Cavalier, Pembina County, saw a number of flocks of fifty to a hundred in 1880 and 1881, and shot several. 109. Zenaidura macroura carolinensis (Linneus). Mourning Dove. Cougs, 1878, p. 628; common at Pembina in June, 1873. HorrmMan, 1882, pp. 402- 403; several specimens observed only during the first half of September. THwaires, 1905, p. 385; at Fort Union, Maximilian records seeing Caroline doves in July. Jupp, 1917, p. 15; tolerably common summer resident; nests on the prairies and in ~ the Turtle Mountains. This was found to be a very common species throughout the state. It is one of the most beneficial birds, for it feeds largely on seeds of injurious weeds. 110. Cathartes aura septentrionalis Wied., Turkey Vulture. Cours, 1878, p. 627; frequently seen in the Red River region. Horrman, 1882, p. 402; apparently not rare at Fort Berthold, McLean County. S. C. C., 1887, p. 414; seen occasionally near Bismarck. THWAITES, 1905, p. 337; on June 13, 1833, Maxi- milian saw a high tree entirely covered with turkey buzzards, in Emmons County along the Missouri River. Jupp, 1917, p. 17; straggler in prairie region, tolerably common around Devils Lake. BatLty, 1920, p. 69; three turkey vultures seen at Sweetwater Lake. The Williams Collection contains a specimen taken at. Ardoch, Walsh County, October 4, 1914, by Mr. Williams. He has other records for Graf- ton of April 20, 1902, and May 22, 1921. Mr. Russell Reid informs me that it is not common at Bismarck; a specimen was taken there July 19, 1921. A specimen was seen near Devils Lake on May 9, 1921, and another on May 18. 111. Catharista urubu (‘Vieillot). Black Vulture. BoarDMAN, 1880, p. 106; saw a flock of black vultures a few miles north of San- born, Barnes County, and shot one to make sure of the identity. This is the only Dakota record known to the writer. 112. Elanoides forficatus (Linneus). Swallow-tailed Kite. Coves, 1878, p. 147; “I am informed by my valued correspondent, Dr. C. E. McChesney, U. S. A., of the occurrence of Elanoides forficatus at Fort Sisseton, Dakota, during nearly the whole of last winter. [This fort was in Roberts County, not far from the North Dakota line.] This account tallies with Trippe’s Minnesota record, north of Mille Lac, lat. 47°, while at Pembina, Dakota, lat. 49°, I was assured by an officer of the occasional appearance of the bird there.” 113. Circus hudsonius (Linnzeus). Marsh Hawk. ALLEN, 1875, p. 64; rare in the breeding season, more common in August and September. Cours, 1878, p. 619; common throughout the region. Specimens were taken at Pembina, June 3 and 11, 1873; at Turtle Mountains, July 28, and Mouse River, August 10-19. Horrman, 1882, p. 402; not common at Fort Berthold. Juop, IQI7, p. 15; very common summer resident. 38 NorMAN A. Woop This is a common species in all parts of the state. On June 29 a nest containing five small young was found in a dry marsh in tall, coarse grass, This is one of the most beneficial hawks, as it feeds principally on injurious mammals, all kinds of mice and ground squirrels. 114. Accipiter velox (Wilson). Sharp-shinned Hawk. Cours, 1878. p. 620; recognized on one occasion when a female was taken at Mouse River, September 3, 1873. WARREN, I890, p. 348; a specimen from Fort Buford, Williams County, taken September, 1887. Jupp, 1917, p. 17; rare, two speci- mens secured in 1805. No live birds were seen, but there is one from the Turtle Mountains mounted in the Biological Station Museum at Devils Lake. The Williams Collection contains one taken at Grafton, September 8, 1912, by Mr. Williams. 115. Accipiter coopert (Bonaparte). Cooper’s Hawk. BENDIRE, 1892, p. 193; Cooper’s hawk in the vicinity of Grand Forks, North Dakota. occasionally nests on the ground according to George G. Cantwell. The Williams Collection contains a female Cooper’s hawk taken at Grafton, May 24, 1914, by Henry Williams. The species was not fully identified in the field, although a hawk seen at a distance was recorded as Cooper’s. 116. Astur atricapillus atricapillus (Wilson). Goshawk. JUDD, 1917, p. 17; have seen head and feathers of two shot in north end of Towner County. A male in the Museum of Zoology Collection was taken January 21, 1917, at Grafton, by H. V. Williams; the Williams Collection has one taken at Grafton, April 9, 1907; and there is a mounted bird without data in the State Biological Station Museum. The species enters North Dakota in the fall, and some winter there, according to Mr. Williams. It is one of the most injurious of all the hawks, as it lives principally on game birds and mammals. 117. Buteo borealis kridert Hoopes. Krider’s Hawk. Jupp, 1917, p. 17; common migrant and also breeds in the same region as the red-tail, and probably types of both will be found intermingling and breeding. The Williams Collection contains a fine specimen taken at Grafton, Sep- tember 6, 1909, by Mr. Williams. On August 23, 1920, I saw at short range a hawk which I was able to identify as Krider’s. ‘This, like the red-tail, is mostly beneficial in its food habits, feeding largely on injurious mammals. 118. Buteo borealis calurus Cassin. Western Red-tail. Cours, 1878, p. 624; frequently observed in the west, but the only individual taken was at Mouse River, September 14, 1873. JUDD, 1917, p. 17; common migrant, breeds in Turtle Mountains and in timber around Devils and Sweetwater lakes. The above note is given for the eastern form, and both that and the western doubtless occur in the state. Several were shot in August, 1920, Birp Life oF NortH DAKOTA 39 and a number of others seen which were hard to identify as either species. There are a number of mounted birds of both species in the Williams Col- lection. A fine specimen was also seen there, a partly albino male. 119. Buteo borealis harlani (Audubon). MHarlan’s Hawk. ScuMiptT, 1904; central section as a migrant. CAMERON, 1907, p. 262; one from North Dakota, spring of 1890, now in British Museum. SAUNDERS, I92I, p. 173; this subspecies has been taken in North Dakota. Mr. Alfred Eastgate is responsible for two records for this hawk; one taken at Stump Lake, October 2, 1902, and another at Dry Lake, October 14, 1912. The only specimen seen was in the Williams Collection, a large female collected at Grafton, May 1, 1916, by Henry Williams. It is now in the Museum of Zoology Collection. 120. Buteo lineatus lineatus (Gmelin). Red-shouldered Hawk. ScHMIpD?T, 1920, p. 298; summer resident. BENpDIRE, 1892, p. 219; probably occurs in small numbers in the more heavily timbered parts of North Dakota. The species was not seen in the field. 121. Buteo swainsom Bonaparte. Swainson’s Hawk. BarrpD, 1858, pp. 21, 22; female taken at Heart River, North Dakota, September 21, 1856, and two females from the Little Missouri River, taken September 12, 1856, by Dr. Hayden. These were published as Buteo bairdii Hay, which name was later reduced to the synonymy of the present species. ALLEN, 1875, p. 64; more or less common; obtained an adult male on Heart River, June 25, 1873. Cours, 1878, p. 624; very abundant in North Dakota, where many specimens were taken from July 15 to September 9, 1873. BrENpIRE, 1892, p. 239; Capt. B. F. Goss says he found this speceis breeding in North Dakota in the high timber along the streams from forty to fifty feet up and in low brush patches on the prairie in the lake region, where its nests were but two to four feet from the ground. REED, 1904, p. 162; set of, eggs collected May 21, 1897, in Stark County, by Roy Dodd. Jupp, 1917, p. 17; an abundant sum- mer resident, nesting in small bushes and also in tall trees. In July they appear in large numbers to feed on grasshoppers, gophers, mice, toads, etc. Writing in IgI0o, one would say they were rare birds and not found nesting at all, nor is there a sum- mer flight of this species. A male in the Museum of Zoology Collection was taken at Stump Lake, May, 1897, by Alfred Eastgate. In the collection of Mr. Williams at Grafton are a number of mounted birds, all in different plumages, some very dark and others quite light. This species must have greatly decreased in num- bers or has changed its habits. At present it is rarely found nesting and is not seen in many localities where it was formerly abundant. It was first seen at Cando, June 3, 1921, when I secured one. In July and August I saw a few others in the prairie regions, but they seemed no more common than the red-tails. This is doubtless the most beneficial of the hawks in its food habits and should not be destroyed. 122. Buteo platypterus (Vieillot). Broad-winged Hawk. Jupp, 1917, p. 17; fairly common straggler, no evidence of its breeding here has been noticed. An adult female was taken August 23, 1900, at Devils Lake by Alfred Eastgate, and is now in the Biological Station Museum. One was seen and 40 NorMAN A. Woop taken at Cando, in a small grove of trees at the edge of town, on June 7. I saw one in 1920 at Lake Upsilon, Turtle Mountains, on August 2, and took one at Devils Lake, August 23. Another was seen there August 24, 1920. 123. Archibuteo lagopus sancti-johannis (Gmelin). Rough-legged Hawk. Jupp, 1917, p. 17; very early spring and late fall migrant, first seen the last of March, tolerably common. Mr. Reid reports that it is sometimes seen during migration and that a pair wintered in a small coulee north of Bismarck during 1920-21. This bird was not seen in life, but there is a mounted specimen in the Williams Collection, taken November 1, 1908, at Grafton, where the species is a winter visitor. 124. Archibuteo ferrugineus (Lichtenstein). Ferruginous Rough-leg. Bairp, 1858, p. 35; a female taken October 12, 1855, on the Little Missouri River by Lieut. E. H. Warren. ALLEN. 1875, p. 65; one of the most common species, sev- eral nests found containing young. The nest is often a large, bulky structure, three or four feet in diameter, built of large sticks, mixed with the ribs of antelope and bison, and is placed on the ground or rocks usually near the summit of buttes. CouEs, 1878, p. 626; found breeding in the Pembina Mountains by Lieut. F. V. Greene’s party. Horrman, 1882, p. 402; found in the timber of the river bottoms, though not very common. BENpIRE, 1892, p. 260; Capt. Goss wrote that the ferru- gineus rough-leg was not uncommon in the high, broken prairie and lake regions of northwest North Dakota. In May, 1880, he took four sets, the nests all on the ground on rocky hillsides and generally near large boulders. REED, 1904, p. 165; a set taken April 29, 1900, in Stark County, by Roy Dodd. Bwgnv, 1907, pp. 213-214; corrects the name of a rough-legged hawk, supposed to have been the American, found breeding in Nelson County, June 4, 1901, and published in Auk, Vol. XVIII, p. 303, and says: “This nest, I am satisfied, belonged to a pair of ferrugineous rough-legs.”” On June 17, 1902, Dr. L. B. Bishop found a nest near Lake Washington, North Dakota. All these developed the melanistic plumage. Jupp, 1917, p. 17; very common summer resident, breeding in considerable numbers in the hills in the northern part of Towner County. First seen early in April. In the collection of the Museum of Zoology is a male in melanistic plu- mage, collected October 3, 1920, at Grafton, Walsh County, by H. V. Wil- liams. ‘The museum at the Biological Station, Devils Lake, contains a mounted bird, and there is another in the Williams Collection, taken at Grafton, October 1, 1920, by Henry Williams. The only nest known to the writer was found at Stump Lake, July 24, 1920. A pair had nested for years in a thick ash forest at the Wishart Ranch named ‘Hawks’ Rest” by Mrs. F. M. Bailey, who made this place her headquarters while studying the birds of the Stump Lake Region and who wrote of this particular nest (Condor, 1918, Vol. XX, No. 4, p. 136). This species must be greatly reduced from its former numbers, as but two live birds were seen, one . August 9, 1920, and another July 11, 1921, both near Bottineau. It is almost wholly beneficial, living on mice, ground squirrels and gophers. 125. Aquila chrysaetos (Linneus). Golden Eagle. ALLEN, 1875, p. 65; occasional: young one captured) on Heart River. Cougs, 1897, p. 107; Audubon says on July 21, 1843, near Fort Union, he saw young golden eagles. This is a rare visitant or straggler in North Dakota. Only three are in Brrp LiFe oF NortH DAKoTA 4I the state collections. One in the Agricultural College Collection is from sismarck; there is a mounted one in the Rough Riders’ Hotel at Medora, taken near there, and there is an immature bird in the Williams Collection, taken April 5, 1921, at Olga, Cavalier County. We were told by Mr. Hal- liday, a federal trapper, of a nest on an isolated butte near Medora. ‘This species is considered quite destructive to game birds, and also kills hares, rabbits, and even young lambs, pigs and poultry. 126. Haliwetus leucocephalus leucocephalus (Linneus). Bald Eagle. Covers, 1878, p. 627; frequent last week in. May, 1873, along the Red River from Moorehead to Pembina. Several nests were noticed upon the tops of tall, isolated trees; upon one of the nests the parent was observed sitting. HorrmMan, 1882, p. 402; occasional visitor in the vicinity of Fort Berthold. Cours, 1893, p. 265; April 10, 1805, Lewis and Clark shot bald eagles and saw many nests in tall cottonwoods near Fort Berthold, and on April 12 shot one near the mouth of the Little Missouri. CouEs, 1897, p. 176; August 2, Owen climbed a tree to a white-headed eagle’s nest and drove a young one out, which was brought to Fort Union alive. It is remark- able that the young should be in the nest at this late season. JuppD, 1917, p. 18; not common, breeds; a young bird was taken from the nest in the Turtle Mountains in 1890; two were seen in 1895, one of which was taken near Rock Lake. No live birds were seen, but I have examined a mounted bird, taken at Medora, November, 1920. Captain Thomas Lonnevik took an immature male, October 15, 1921, at Devils Lake. This is in the Biological Station Museum. ‘There is an immature male in the collection of the Museum of Zoology, taken by Henry Williams at Northwood, Nov. 25, 1913. The bald eagle lives principally on fish, either dead ones cast on shore or upon those captured by other fish-eating birds. It is reported to feed upon water fowl also. 127. Falco rusticolus rusticolus Linneus. Gray Gyrfalcon. Mr. Reid reports two specimens at Mandan, taken at Hazen, Mercer County, March, 19109. There is one adult in the Williams Collection taken at Grafton, October 7, 1908. ‘This species must be considered as an accidental visitant or strag- gler in North Dakota. 128. Falco mexicanus Schlegel. Prairie Falcon. HayneEn, 1863, p. 152; along the Missouri, though not abundant. JUDD, IQI7, p. 18; common, especially during August and September. The only species of hawk seen taking poultry. Have seen them swoop down in the main street of the town after their prey. The species may have been common some years ago, but I failed to find it so during my work in the state. Mr. Alfred Eastgate tells me that it is a common migrant and breeds along the Mauvaise Coulee in Towner County. On August 9, 1920, I noted a fine bird, the only one seen, on a fence post near a small alkali lake, about five miles southwest of Bottineau. 129. Falco peregrinus anatum Bonaparte. Duck Hawk. CouEs, 1897, p. 156; Audubon says he saw many peregrine falcons near Fort Clark, Oliver County: p. 176; on August 2, Bell and Owen saw a peregrine falcon feeding 42 Norman A, Woop its young near Fort Union. Reep, 1904, p. 169; a set of eggs from Stark County, taken May 4, 1901, by Edw. Dodd. Jupp, 1917, p. 18; not common, several seen dur- ing the season. One female taken in June, 1895, by Dr. Louis Bishop. Barry, 1920, p. 69; a duck hawk flew swiftly in, darted down and seized a coot at Island Lake, Ramsey County. Mr. Alfred Eastgate, who lived for several years (1894-1915) on the shore of Stump Lake, Nelson County, told the writer that duck hawks were common there from October 10 to November 10 each year. Mr. Williams mounted one taken in October, 1904, at Glasston, by C. Storey. On May 19, 1921, I saw a bird perched on a stone duck blind on the point west of Creel Bay, Devils Lake. This was the only specimen seen alive, but Mr. D. R. Ducke, of the State Fish and Game Commission, gave me a skin of an immature bird taken by him in Eddy County, April 10, 1g1o. 130. Falco columbarius columbarius Linneus. Pigeon Hawk. ALLEN, 1875, p. 64; seen at distant intervals on the Heart River in September. Cougs, 1897, p. 162; on September 5, on the upper Missouri, Audubon saw a pigeon hawk chase a spotted sandpiper, which dove under the water and escaped. Jupp, 1017, p. 18; hawks seen flying at a distance were assigned to this species. Rare, if it occurs at all. Mr. Alfred Eastgate says it is a common migrant. The skin of a male taken September 16, 1914, at Grafton, Walsh County, by H. V. Williams, is in the Museum of Zoology Collection. No living birds were seen, but there is a mounted one without data in the Biological Station Museum at Devils Lake, and another in the Williams Collection, taken at Grafton, June 18, 1912, by Mr. Williams, who says it nests there rarely. There is also a mounted male in the Agricultural College Museum from Kenmare, Ward County, taken May 2, 1913, by H. E. Peck. 131. Falco columbarius richardsont Ridgway. Richardson’s Pigeon Hawk. Covers, 1874, p. 349; young male taken at Fort Rice by General Alfred Sully. Cougs, 1878, p. 623; one specimen, the only individual of this species observed, was taken on the headwaters of the Mouse River, September 8, 1873. BENpIRE, 1802, p. 304; U. S. National Museum contains some from North Dakota. The species was not seen in the field, and so far as known there are no specimens in the state collections. 132. Falco sparverius sparverius Linneus. Sparrow Hawk. Cours, 1878, p. 623; very abundant throughout the region surveyed. Many speci- mens were taken from Pembina to Mouse River, June 19 to August 30, 1873. Horr- MAN, 1882, p. 402; frequently seen in pursuit of small birds near the village. Cougs, 1897, p. 24; on June 10 a sparrow hawk was killed above the Little Missouri River. This is a common summer resident of all the wooded areas; it was pres- ent on my arrival at Devils Lake, May 3, 1921, and one was found nesting the last of that month in an oak tree at the side of the road on the Rock Island Military Reservation. Of four specimens taken, three are, no doubt, of this eastern species, while one is referred to the western or desert form. The sparrow hawk is a very beneficial species, living princpally upon insects and injurious mammals. Birp LiFE oF NortH DAKOTA 43 133. Falco sparverius phalaena (Lesson). Desert Sparrow Hawk. There is a specimen in the Williams Collection, taken at Grafton, May 12, 1907. One of my specimens, an adult female taken at Devils Lake, May 10, 1921, is referred to this form. 134. Pandion haliaétus carolinensis (Gmelin). Osprey. Jupp, 1917, p. 18; rare, one seen at Snyder Lake, Towner County, April 24, and again on May 28, 1895. This species was not seen in the field, but Mr. Williams told me that he saw one which was shot at Niagara, Grand Forks County, on September 25, 1920, and another at Grafton, April, 1921. 135. Aluco pratincola (Bonaparte). Barn Owl. This seems to be a very rare species in North Dakota. It was not observed in the field. Mr. Alfred Eastgate has a record for Grand Forks, November, 1891. ‘There is a mounted specimen in the Williams Collection taken at Oakes, Dickey County, September 1, 1912. Mr. J. D. Allen, of Mandan. has mounted a bird collected in Stutsman County, September 7, 1921, by Eddie Schutt, and two from Hebron, Morton County, taken Sep- tember 6, 1921, by Joe Wagenhopper. 136. Asio wilsonianus (Lesson). Long-eared Owl. ALLEN, 1875, p. 64; occasional, two specimens obtained. Jupp, 1917, p. 18; not common; one nest found May 25, 1895, in bushes along Big Coulee in the northern part of the county. Bame£y, 1910, p. 227; family of three-quarter grown specimens on Sweetwater Lake on August 17, 1917. A nest of this species was found at Hay Creek, near Bismarck, on May 22, 1921, by Russell Reid (see photograph). A mounted specimen without data is in the Biological Station Museum. It was probably taken near Devils Lake. I saw, in a deep draw near Medora, a female with three young and an unhatched egg in an old magpie’s nest built in a thicket of trees. On June 19 the oldest of these young was one- - half grown, the next was much smaller, while the youngest was only about half as large as the oldest. This is a very beneficial species, feeding almost entirely on mice and small mammals injurious to the farm crops. 137. Asio flammeus (Pontoppidan). Short-eared Owl. ALLEN, 1875, p. 64; apparently the most common of the owls. LARson, 1913, p. 94; occasionally seen; could be found almost anywhere in McKenzie County in the winter of 1912-13. BatLEy, 1915, p. 174; saw a number flying about in the daytime, July 8 and 9, 1912, between Stump and Devils lakes. Jupp, 1917, p. 18; common summer resident, usually arrives third week in April; full set of eggs was found May 18. This owl is, no doubt, the most common one in North Dakota. Mounted specimens are in all the state collections. I found one, May 26, 1921, in a marsh at the north end of Creel Bay, and at Bottineau, July 11, saw a num- ber in the road after dark. ‘This species destroys immense numbers of small mammals injurious to the farm crops and, like the two preceding species, should always be protected. 44 NorMAN A, Woop 138. Strix varia varia Barton. Barred Owl. Cours, 1874, p. 309; Dr. Hayden’s Missouri River specimen remains the western- most on record for the species. ALLEN, 1875, p. 64; met with on the Missouri at Fort Rice, 1873. HorrMan, 1882, p. 401; said to be of frequent occurrence, though only a single specimen was found at Fort Berthold. My only record is of a mounted specimen in the Williams Collection, taken at Grafton, October 3, 1904. This species is also a beneficial one, but as it is rare in North Dakota is not of much economic importance. 139. Scotiaptex nebulosa nebulosa (J. R. Forster). Great Gray Owl. RIDGWAY, 1914, p. 637; one from Mandan, North Dakota. Mr. Alfred Eastgate writes me that he has a specimen taken at Grand Forks in January, 1892. This species, like the last, was not noted by Coues or Judd, nor did I see any in life. A mounted specimen in the Williams Collection was taken near Pembina, February 3, 1900. It only occurs as a rare winter visitor. It is not beneficial, since it feeds mostly upon game birds and mammals. 140. Cryptoglaux funera richardsoni (Bonaparte). Richardson’s Owl. Jupp, 1917, p. 18; rare winter resident and migrant. The Williams Collection has a specimen taken at Grafton, April 8, 1904. The Agricultural College Museum contains a male taken at Grafton, Decem- ber 16, 1910, by Mr. Williams, who reports it as common only during the winter months. Mr. Alfred Eastgate records one from Stump Lake, March 27, 1904. ‘This, like the great gray owl, is only a rare winter straggler in North Dakota. 141. Cryptoglaux acadica acadica (Gmelin). Saw-whet Owl. Jupp, 1917; rare, taken in winter only. The Agricultural College Museum contains a mounted bird taken on the College Campus, October 10, 1920. I saw specimens in the Williams Col- lection, taken November 10, 1909. Mr. Williams told me it nested at Graf- ton. ‘This and the preceding species are of little economic importance, as they occur but rarely. In general, they are beneficial, feeding mostly upon mice. 142. Otus asio asio (Linneus). Screech Owl. FREEMAN, 1921, p. 12; fairly common, both the fuscous and gray phases found. A pair have been observed for several years in Island Park. Mr. Reid records one in the red phase found west of Mandan, November 13; 1921. I saw a mounted specimen in the collection of William Shunk, of Anselm, that had been taken there recently. One in the Williams Collection was taken January 9, 1908. One was killed at Stump Lake, 1906, and mounted by Alfred Eastgate. 143. Bubo virginianus virginianus (Gmelin). Great Horned Owl. Barrp, 1858, p. 51; female taken at Fort Union: Coves, 1878, p. 618; a pair with two young observed at Pembina early in June, 1873. HoFFMAN, 1882, p. 401; was Biro Lire oF NortaH DAKoTa 45 informed that the species is not rare. A wing was found in possession of an Indian who used it as a fan. Cougs, 1893, p. 272; April 14, Lewis and Clark record the killing of a large hooting owl. Cours, 1897, p. 97; Audubon says he shot a great horned owl near Fort Union. The Williams Collection contains three forms of horned owls; one of these, which I have referred to this species, was dark, marked with yellow and rufous, like our Michigan birds. It was taken at Grafton, November 10, 1903. Another typical specimen was also collected at Grafton, Novem- ber 1, 1903. Mr. Eastgate records one from Larimore, January, 1906. 144. Bubo virginianus pallescens Stone. Western Horned Owl. JUDD, 1917, p. 19; common resident of Turtle Mountains, where it breeds. Nests were found with eggs in February and March. This is the most common form found in North Dakota. I saw several mounted specimens in the state collections. On May 28, 1921, I found a nest in a big willow tree at the edge of Fort Totten Lake in Sully Hill National Park near Fort Totten. It contained three downy young nearly half grown. These were brought to me at the Biological Station on June 11 by the keeper of the park, Mr. Triska, who told me that the parents were catching the tame ducks in the park pond, so he shot the female, but it escaped. The largest one of the young is in the North Dakota Biological Station Museum. It was well grown, with tail and wing feathers about half length. The two smaller ones are in the Museum of Zoology Collection. The species probably destroys not only a considerable amount of poultry and game birds but also numbers of small mammals. 145. Bubo virginianus subarcticus Hoy. Arctic Horned Owl. RIDGWAY, IQI4, p. 751; one from Devils Lake. North Dakota. This form is supposed to be only a winter visitant in North Dakota, but Mr. Williams, in whose collection are some fine specimens, says that it breeds at Grafton. A fall record was December 3, 1903. 146. Nyctea nyctea (Linneus). Snowy Owl. DEANE, 1907, pp. 218-219; Mr. J. D. Allen, of Mandan, writes on March II, 1907: “The snowy owls have not been as abundant this winter as last, when I had thirty sent to me. Some years ago I secured over 500 specimens and the following year over 350.” LARSON, 1913, p. 94; only one seen in McKenzie County in winter of 1912- IG13. JUDD, 1917, p. 19; common winter resident and migrant. Seen here from Octo- ber to May. One seen at Rock Lake, July, 1gro. This species seems to be common in collections. Mr. Williams had a dozen or more, two of which were practically pure white. At the Williams shop seventy-five of these birds were mounted in one winter. A pair in the flesh was received at the Museum of Zoology, December 24, 1921, sent from Grafton by H. V. Williams. On account of their great numbers and destructive food habits they are, with the horned owl, a great menace to the game birds and mammals of the state. However, at some seasons of the year they destroy mice, gophers and other injurious mammals. 46 NorMAN A. Woop 147. Surnia ulula caparoch (Miller). Hawk Owl. JUDD, 1917, p. 19; rare, one specimen taken in the fall of 1902. This species was seen only in collections. The Agricultural College Museum has a mounted male taken at Joliette, Pembina County, October 24, 1912, by Mr. Williams. At Grafton, I saw one taken there December 10, 1908. Mr. Williams reports that he had seen them only during that winter, when they were quite common. 148. Speotyto cunicularia hypogaea (Bonaparte). Burrowing Owl. ALLEN, 1875, p. 64; not numerous, met with at intervals in the prairie dog towns from the Little Missouri westward. HorrMan, 1882, p. 402; rather common ten miles east of the village in the prairie dog town. WArREN, 1890, p. 365; a specimen from Fort Buford, taken September 29, 1887. BaiLey, 1918, p. 176; was told that a bur- rowing owl, here at the extreme eastern limit of its range, had nested in an old badger hole at Sweetwater Lake, Ramsey County. The species ranges farther east than Bailey records. On July 24, 1921, I saw two families of five and six young each in an old pasture north of Grafton, Walsh County. They were living in old badger or ground squirrel burrows. At that time they were able to fly well. On July 11, 1921, two families with five and six young each, also able to fly, were observed about fourteen miles west of Bottineau. None were noted near Medora, though the habitat seemed very favorable for them. ‘This owl is beneficial in habits and should be protected. 149. Conuropsis carolinensis (Linneus). Carolina Paroquet. Tuwaltes, 1906, p. 250; Maximilian mentions this species in a list of birds at Fort Clark, 1833. In 1843 Audubon saw them in South Dakota near Fort Pierre. 150. Coccyzus americanus americanus (Linneus). Yellow-billed Cuckoo. ALLEN, 1874, p. 63; several times observed along the Heart River. Jupp, 1917, p. 19; E. S. Bryant claims to have taken this species at Freshwater Lake. FREEMAN, 1919, p. 13; rare, occurs the latter part of May or early June. This species was not seen, but it no doubt occurs rarely in the south and eastern parts of the state, as it is noted as a common breeder in South Dakota by Visher.° On October 10, 1921, Henry Williams, of Grafton, received a bird in the flesh from Professor Miller, of Fargo. 151. Coccysus erythrophthalmus (Wilson). Black-billed Cuckoo. Cours, 1878, p. 615; a nest in Pembina Mountains was.found July 12, 1873, which contained a single young bird. Jupp, 1017, p. 19; tolerably common summer resident in Turtle Mountains. A common summer resident about Devils Lake in 1920 and 1921, and observed in the Turtle Mountains on August 7, 1900, and on June 4, 1921. It is a very beneficial bird, eating worms, caterpillars and other injurious forms of insect life. 'An Annotated List of the Birds of Sanborn County, South Dakota. Auk, Vol. XXX, No. 4, October, 1913. Brirp Lire oF NortoH Daxora 47 152. Ceryle alcyon (Linnzus). Belted Kingfisher. Cours, 1878, p. 615; general distribution along the Red and Mouse rivers. One was taken at Pembina, June 9, 1873. HorrMan, 1882, p. 401; not common at the settlement, but seen above and below it. Tuwaires, 1906, p. 186; near Fort Union, September 28, 1833, Maximilian saw many kingfishers. JUDD, 1917, p. 19; tolerably en migrant. Rarely found nesting. FREEMAN, 1921, p. 13; a few nest here, pril 2. This species was not observed at Devils Lake, but two were seen at Lake Upsilon, Turtle Mountains, August 5, 1920, and at Lake Winnebegosis in the western end of the mountains on July 11, 1921. It was also seen at Cando, June 10, 1921, by Elmer T. Judd, and I found it at Fargo, July 27, 1921, along the Red River. 153. Dryobates villosus villosus (Linnzeus). Hairy Woodpecker. HoFFMAN, 1882, p. 401; observed among the groves near the bad lands east of the village. RipGEWAyY, 1914, p. 202; two adult males from North Dakota. Scumunv, 1904; lists this species from North Dakota. ScuMupr, 1920, p. 302; lists this species as a common summer resident. FREEMAN, 1921, p. 13; permanent resident, seen every month in the year. I did not see this species, but it no doubt occurs in the Red River region. 154. Dryobates villosus leucomelas (Boddaert). Northern Hairy Wood- pecker. CouEs, 1878, p. 615; observed in heavy timber in the Turtle Mountains. One specimen taken there, July 20, 1873. (The above was given under the previous species name, villosus, but the bird was doubtless this form.) ScHMIDT, 1904; nests. Jupp, 1017, p. 19; tolerably common resident in the Turtle Mountains; breeds. On July 20, 1920, a typical adult female was taken at Devils Lake. I did not see it during my stay in the Turtle Mountains. Two specimens are in the Museum of Zoology, a male and female, taken at Grafton, February 28, 1922, by H. V. Williams. The species feeds upon wood-borers and injurious insects. 155. Dryobates pubescens medianus (Swainson). Downy Woodpecker. Jupp, 1917, p. 19; common summer resident in the Turtle Mountains; first seen in the middle of May. A common resident of all the wooded areas worked. It was seen in the Turtle Mountains, August 1 to 7, 1920, and at Devils Lake, May 4, 1921. This, as well as the hairy woodpecker, is very beneficial in food habits, rid- ding the trees of wood-borers and other injurious insects. 156. Picoides arcticus (Swainson). Arctic Three-toed Woodpecker. Jupp, 1917, p. 19; rare, Bryant collected one in the spring of 1894. A mounted specimen is in the Williams Collection at Grafton, taken there January 23, 1912, by Mr. Williams. This was the only one seen. It occurs only rarely. 157. Sphyrapicus varius varius (Linneus). Yellow-bellied Sapsucker. Cours, 1878, p. 616; plentiful at Pembina, where it was breeding in June, 1873. 48 Norman A. Woop Specimens were taken June 5 to July 8 at Pembina, and September 16 at Mouse River. Jupp, 1917, p. 19; summer resident of Turtle Mountains. FREEMAN, I9Q1Q, p- 13; common summer resident, April 4. Mr. Russell Reid reports that the species is sometimes seen during migration. Specimens were taken on April 24 and May 5, 1920, near Bismarck. Many adults and young were observed in the Turtle Mountains from July 30 to August, 7, 1920, and on June 5, 1921. I saw a nest in a big cot- tonwood tree on the Fargo College Campus in July, 1921. This species is considered injurious to trees, especially fruit trees, which it often girdles with its punctures. 158. Phlwotomus pileatus abieticola (Bangs). Northern Pileated Wood- pecker. The only specimens from North Dakota known to me are one in the Williams Collection taken at Grafton, May 30, 1905, and another in the Fargo College Museum, labelled Red River, October 16, 1915, Henry Wil- liams. This is a beneficial species, eating ants and all kinds of wood-boring beetles and larvae. 159. Mélanerpes erythrocephalus (Linueus). Red-headed Woodpecker. Cours, 1878, p. 617; common along the Red and Upper Missouri rivers. A specimen was taken at Pembina, June 9, 1873. Jupp, 1917, p. 19; rare; a straggler from the Red River flight. This species was not seen at Devils Lake, but it was common at Medora in June, 1921, and was noted at Fargo, June 13, and at Grand Forks, July at, 192i: 160. Asyndesmus lewisi Riley. Lewis’s Woodpecker. ; p ScHMIDT, 1904; western and central North Dakota, rare. A mounted specimen is in the Williams Collection, taken at Neche, Pem- bina County, October 13, 1916, by Peter D. Heilly. 161. Centurus carolinus (Linnzus). Red-bellied Woodpecker. I am unable to find a published record from North Dakota. A skin of a male taken in September, 1897, at Stump Lake, Nelson County, by Alfred Eastgate, is in the Museum of Zoology Collection. It must be recorded as a rare straggler in the state. 162. Colaptes auratus luteus Bangs. Northern Flicker. CovuEs, 1878, p. 617; common along the Red and Mouse rivers and seen at Pem- bina and Turtle Mountains. Specimens were taken from June 6 to September 3, 1873. Horrman, 1882, p. 401; saw no living or fresh specimen, but in the Great Lodge of the Arickaras was a true auratus poorly mounted. Ripcway, 1914, p. 18: east of Rocky Mountains except North Dakota and Minnesota. On page 20 he refers the North Dakota form to C. a. borealis. Jupp, 1917, p. 19; common summer resident, Turtle Mountains. This was found to he a very common species at Devils Lake from May 3, 1921, nesting in the ash forest and feeding on ants along the beaches of Birp Lire oF NortH DAKOTA 49 the lake. It was also seen at all of the localities worked. This is a very beneficial species, eating many injurious insects, seeds and wild fruits. 163. Colaptes cafer collaris Vigors. Red-shafted Flicker. HoFFMAN, 1882, p. 401; during the whole month of September at Fort Berthold, McLean County. BEnpire, 1895, p. 134; ranges east to North Dakota. Covers, 1897, p. 24; June 12, Harris saw several red-shafted woodpeckers near Fort Union: p. 41; June 19, Harris and Pell brought a new bird like our common golden-winged wood- pecker, but with a red mark instead\ of a black one along the lower mandible. In a foot-note Dr. Coues says, here is the original nidification of the curious flicker of the Upper Missouri region, which Audubon named Picus ayresti. It is the Colaptes hybridus of Baird, in which the specific characters of the golden-winged and red- shafted flickers are mixed in every conceivable degree and which later puzzled Audubon; p. 71; Audubon says: “On July 1, Harris and Sprague brought in the most curious set of five birds that I ever saw and which I think will puzzle all the naturalists in the world. These all belonged to one nest and were all differently marked.” Mr. Russell Reid reports that the species is frequent about Bismarck and that a pair nested on the High School grounds at Bismarck last May. He had seen hybrids of this and luteus. But one specimen was seen in life, at Williston, near the Missouri River, June 7, 1921. At Grafton I saw a mounted specimen in the Williams Col- lection taken by Henry Williams on December 3, 1919. It was seen in the locality some time before it was taken. The form is no doubt very rare in that part of the state. 164. oY 88 Figure I. Figure 2. Figure 3. NorMan A. Woop PLATE II Swampy meadow and prairie near Devils Lake. Missouri River near Williston. Little Missouri River and grove of cottonwood near Medora. NortuH Dakota Birps Prats IL FIGURE J FIGURE 2 ’ FIcuRE 3 fase ha ers Ny, ete Reg aeieg eey go Figure I. Figure 2. Norman A. Woop PLATE III Nest of ferrugineous roughleg hawk, near top of butte, north of Esmond. Photograph by P. B. Peabody. Saskatchewan horned lark on nest, Bismarck, June 6, 1921. Photograph by Russell Reid. PLATE IIE NortH Dakota Brirps TL a ee FIGURE I FIGURE 2 oy ee Rate : I) ay - - hood 7,7 r » » 4 : ‘f ; as i A aa rr ih She tg Ee ee ‘ay H i ne 4 y i Po tt y ort t » yu re i nt en a a ; : ; g2 Figure I. Figure 2. Figure 3. Norman A. Woop PLATE IV Long eared owl on nest. Hay Creek near Bismarck, May 22, 1921. Photograph by Russell Reid. Nest of chestnut-collared longspur near Esmond. P. B. Peabody. Nest of yellow rail near Esmond, June 7, 1912. P. B. Peabody. Photograph by Photograph by PLATE IV NortH Dakota Brirps FIGURE 2 FIGURE I 94 Figure I. Figure 2. NorMAN A. Woop PLATE V Nests of double crested cormorants. Photograph by Crystal Thompson. Lark bunting on nese Bismarck, July 6, Russell Reid. Bird Island, Devils Lake, 1920. 1921. Photograph by PEATE “Vv NortH Dakota BrIrRDs I y FIcuR FIGURE 2 96 Figure Figure Figure Figure il i Norman A. Woop PEATE, Vi Nest and eggs of LeConte’s sparrow. North Dakota. Photograph by P. B. Peabody. Western grasshopper sparrow. Photograph by P. B. Peabody. Nest of Baird’s sparrow near Esmond, June 5, 1919. Photograph by P. B. Peabody. Nest of ring-neck duck. Devils Lake Region, July 20, 1919. Photograph by P. B. Peabody. Norrie Dakota Brirps PLATE VI FIGURE I ape SO SNR BES hh DAE Gl h S a ie J ra » 2 A SATS “ARS EY ty ARA\ Zi On / Fan. CFP 6 fs TAY aT Rite. pe rare ae Le. ee jt FN ¥ f vi 5 Bae i] > SNR Vie es {A aut a wt FIGURE 4 UNIVERSITY OF MICHIGAN MUSEUM OF ZOOLOGY Miscellaneous Publications No. 11 Notes on the Genus Erythemis With a Description of a New Species (Odonata) BY FE. B. WILLIAMSON The Phylogeny and the Distribution of the Genus Erythemis (Odonata) BY CLARENCE HAMILTON KENNEDY ANN ARBOR, MICHIGAN PUBLISHE DEY THE UNIVERSITY JULY 14, 1923 ADVERTISEMENT The publications of the Museum of Zoology, University of Michigan, consist of two series—the Occasional Papers and the Miscellaneous Publi- cations. Both series were founded by Dr, Bryant Walker, Mr, Bradshaw H. Swales and Dr. W. W. Newcomb. ‘ The Occasional Papers, publication of which was begun in 1913, serve as a medium for the publication of brief original papers based principally upon the collections in the Museum. The papers are issued separately to libraries and specialists, and, when a sufficient number of pazes have been printed to make a volume, a title page, index, and table of contents are sup- plied to libraries and individuals on the mailing list for the entire series. The Miscellaneous Publications include papers on field and museum technique, monographic studies and other papers not within the scope of the Occasional Papers. The papers are published separately, and, as it is not intended that they shall be grouped into volumes, each number has a title page and, when necessary, a table of contents. ALEXANDER G. RUTHVEN, Director of the Museum of Zoology, University of Michigan. NOTES ON THE GENUS ERYTHEMIS WITH A DESCRIPTION OF A NEW SPECIES (ODONATA) By FE. B. WILLIAMSON This paper discusses in a general way the distribution of the species of the genus and some possible factors involved, recording specimens taken in the Canal Zone, Panama, Colombia, and Venezuela; the genus is defined, following earlier authors; a key to the species, based on some characters heretofore used and on some new characters, and therefore supplementary to keys by Calvert and Ris, is given; and a new species, the first to be rec- ognized as such in this genus for many years, is described. The paper deals only with imagos. A GENERAL CONSIDERATION OF THE DISTRIBUTION OF THE SpEcIES OF ERYTHEMIS A reference to the literature, summarized by Ris in Coll. Zool. de Selys, will reveal at once the wide distribution of the species of this genus. For example, seven of the eight known species occur in the West Indies, yet none is peculiar there, and of the seven species, five have each a range at least from Mexico to Paraguay. These dragonflies have been generally successful in passing over or around whatever obstacles the varied topography of trop- ical America could offer to their dispersal. Dragonflies as larvae and adults are active and more or less independent of any one particular source of food. The larvae are aquatic and are hence dependent on some water supply. The adults are on the wing when they seize their food, and hence can sustain themselves only under certain weather conditions which are equally important for mating and ovipositing. In con- formity with the complexities of multiform bodies of water under diverse climatic conditions, we find equally complex adjustments of dragonflies to meet varying conditions. Regions with pronounced seasons, involving changes in precipitation or temperature or both, and dry or frigid regions, have relatively fewer species of dragonflies than tropical regions with a more uniform climate and an abundant water supply throughout the year. In eastern North America the broad belt of dragonfly fauna bounded on the south by the northernmost extension of such genera as Dythemis, Orthemis, Brachymesia, and Neoneura, for example, and on the north by Cordulia, Agrion, and the bulk of Somatochlora and Aeshna, for example, is probably confined to its area and protected from encroachments from the south and north largely by the temperature factor in the climate. Directly associated with the adjustments of dragonflies to seasonal changes is their dependence on the orderly sequence of these seasons. Hence, in any 1See first paragraph, Wisc. Publ., Mus. of Zoology, University of Michigan, No. 9. 4 EK. B. WiL1AMSON region of marked seasonal changes there are more or less frequent changes in the dragonfly fauna of various habitats due to unusual or abnormal sea- sons. It is obvious that the premature drying up of a pond might result in the death of the larvae of several species. An unusually high water stage of a river might submerge all the resting places for emerging larvae at the time when they are due to emerge, or might coat larvae with a deposit of stiff clay, resulting in the wholesale destruction of many species. Late frosts in early summer or a premature freeze in the fall might cause the death of imagoes and thus at least temporarily eliminate certain species from certain habitats. Unusual seasons may also operate to the advantage of certain extralimital species. For example, an early and continuously warm spring may permit the extension northward of a southern species, and it is probably this factor, associated possibly also with favorable air currents, that some- times carries the southern Pantala hymenaea as far north as Ontario. In the low-lying American tropics temperature ceases to be an important factor in itself in determining dragonfly distribution. Drought takes its place. Drought makes itself felt even in the well-watered and forested equatorial belt, though, due to permanent rivers and lakes, its effect is not universal over a region as low temperature is. It is obvious, with low temperature absent as a limiting factor in the seasonal distribution of dragonflies, at least as imagoes, that more numerous adjustments in other factors are possible in the life histories of these insects. For example, in the north the rigors of winter can be bridged by a dragonfly only in the egg or larval stage, but in the tropics it is not impossible that the period of drought might be passed as an egg, a larva, or animago. ‘The existence in the tropics of both temporary and permanent bodies of water have resulted in many diverse adjustments in the life histories of dragonflies, both as larvae and imagoes, to these periods of drought. For example, warmth and abundant food obviously might permit the shortening of larval life, while other factors might permit the indefinite extension of imaginal life. When, as is usually the case, temporary and permanent bodies of water occur in close proximity in a region, the result is a rich and varied dragonfly fauna. It is interesting to note that in the tropics many of the most successful and specialized dragonflies are those which depend on, or are able to endure, a temporary water supply (wet weather ponds or streams) for their larval life, and that in the north our highly specialized species are pond and lake (not stream) dwellers. Too little is known of the detailed life histories of neotropical dragon- flies to allow us to cite particular examples of the various adjustments indi- cated as possible in the preceding paragraphs. ‘This is true of the species of Erythemis, and yet enough is known to permit of a few positive statements and some warranted conclusions. The larvae of the species of Erythemis, so far as known, are pond or lake dwellers, always so, possibly, in the case of some species, and frequently or usually in the case of all. The adults have a long seasonal range, occurring on the wing apparently throughout the year, and are apparently as numerous, if not more numerous, during the dry season. Since the larvae of some, if not all, may live in temporary as well NOTES ON THE GENUS ERYTHEMIS 5 as permanent ponds, and since during the dry season the imagoes are numer- ous, it is very probable that the larval life is brief, and it is not improbable that the dry season is bridged by imagoes rather than by drought resistant eggs or larvae. We know that old and worn females, which might well have lived throughout the dry period, are still able to oviposit; and we know that when dry depressions become pools with the coming of the rainy season ovi- positing individuals of species of several genera, including Erythemis, imme- diately put in an appearance. It is not improbable that species of dragonflies whose larval life is spent in temporary pools and whose imaginal life is continued during the dry sea- son are less bound to the vicinity of the location of these pools than are spe- cies whose larval life is spent in permanent bodies of water and which would therefore develop a stronger fixed habitat tendency through numberless gen- erations of larvae and adults attached to a particular habitat. A longer larval life might also strengthen this tendency. It may be safely concluded that dragonflies emerging from temporary pools after a short larval life and with a long imaginal life, under favorable conditions for extended flight (the dry season in the tropics) before them, will be wide wanderers, and, other con- ditions permitting, species of wide distribution. So far as species of Erythemis are concerned, temperature is not a limit- ing factor over the large area included in tropical America. Swamps, ponds, pools, and lakes of varied characters, permanent and temporary, are abun- dant and widely scattered over this region, forming, with the exception of the southwest coast of South America, a practically continuous habitat. Spe- cies of Erythemis as imagoes are active and conspicuous, apparently living happy, care-free lives so far as predaceous enemies are concerned. They are voracious feeders on a varied diet of other insects, including species which are not dependent on bodies of water during their own lives, such as Lepidoptera. Hence, they can find a living anywhere and are not held to the immediate vicinity of their birthplace, as are many species whose imaginal food consists largely, if not entirely, of aquatic insects, and whose adaptability to a varied diet may be limited. Because of their larval life in both tempo- rary and permanent pools, individuals are numerous as compared with other simliar species whose larval life is confined to permanent pools, and, at the time of ovipositing, when active competition takes place, the males of one species protecting their own females against attacks and at the same time hustling away from the water the females of other species, Erythemis, because of its numbers and activity, is at no disadvantage. The eggs of Erythemis are dashed in the water and oviposition is as safe against parasitism of the eggs as it is in the other large number of species of similar habit. There is no reason to think the larvae of Erythemis are exposed to any particular danger. Their use of temporary pools would relieve them of competition or attack of many forms of aquatic life, while their successful use of perma- nent pools indicates they can meet the conditions there. From the above considerations, we can understand the wide distribution of species of Erythemis which have been able to cross the great river valleys of South America, which valleys, marking frequently the broad areas which 6 E. B. WILLIAMSON separated earlier distinct continental land masses, have proved efficient bar- riers for many species. ‘These valleys may be mechanical barriers or they may merely mark separate regions in each of which, through long adjustment, conditions have resulted which prevent the occupation of one region by many species from another region. Barriers, in any case, exist only in connection with specialized habits of dragonflies. For a high stream species a long mountain range might offer a continuous highway, which would be an insur- mountable barrier for a low pond species. A long, broad valley, forming a great stretch of suitable habitat for the pond species, might prove an impas- sable barrier for the high stream species. In either case, if the river valleys are barriers, Erythemis has occupied and passed them, and if the higher regions themselves are exclusive to species from other regions, Erythemis, because of certain characteristics discussed and doubtless others, has success- fully forced its way, at least to some extent, into the new environment. As a result, the present distribution of species offers, I believe, an unsolvable problem for the student who would determine the place of origin and. trace the routes of distribution of the species of this genus. For example, Funda- cidn, Colombia, lies in the relatively isolated Santa Marta region, and at this one station five species of Erythemis occur, and these five occur in Central America, the West Indies, and to the south to Paraguay. At Tucacas, Vene- zuela, four species have been recorded, and I have no doubt more careful col- lecting there would add at least two species. At El Banco, in the lower Magdalena, six species have been taken. When we speak of the distribution of dragonflies or anything else, we imply, of course, the known distribution. The relation of the known distri- bution to the real distribution depends on the thoroughness with which col- lections are made. And this thoroughness depends on the number and distri- bution of collectors and the degree of success of each collector in obtaining everything in his specialty in each locality he works. And the collector’s degree of success, other things being equal, depends on the relative conspicu- ousness of the things he collects. The imagoes of species of Erythemis are conspicuous insects, and the records of their capture are correspondingly numerous. Though collectors and localities visited in the past have been few in numbers with what we may expect in the future, probably our knowledge of the distribution of species of Erythemis corresponds at least in a broad way with the real distribution of these insects. Compare the status of our knowldge in this genus with that of some of the gomphines, for example. ‘lo cite one case, we know that the genus Erpetogomphus occurs in South America because of the capture, which might be described as acci- dental, of two specimens in Colombia and three in Venezuela—a total of five specimens for the continent. DEFINITION OF THE GENUS The following definition of Erythemis is compiled almost entirely from other authors. Head moderate or small; eyes joined for only a short dis- tance; frons moderately prominent, rounded, without anterior ridge; vertex moderately high, apex slightly emarginate. Hind lobe of prothorax erect, NOTES ON THE GENUS ERYTHEMIS 7 as wide or wider than the other lobes, slightly or scarcely indentate in the’ middle posteriorly, the posterior and lateral margins with long hairs. Thorax‘ moderately robust. Abdomen variable, stout to Slendes) inflated or not at the base, never distinctly dilated apically, segments 2 and 3, but not 4, each with a distinct median transverse carina. Legs stout, third femur of the male with an antero-ventral row of 15-30 small, regular teeth, and a postero-ven- tral row of similar but less numerous teeth, which rows reach from: the base to about two-thirds the length of the femur, the apical third with three stout equal spines in the antero-ventral row and two of the same size in the postero-ventral row; second femur similarly formed with the basal rows of teeth shorter, reaching only about half the length of the femur, the number of teeth reduced and the apical spines variable in number, three. or four in each row; third femur of female similar to second femur: of male, ten or fewer teeth in the basal rows which reach from the base. to, or not as far as, the middle of the femur, each row followed by 3 or 4 spines of unequal size; second femur of female similar to the third, but the apical spines reduced in number; spines on third tibia in both sexes stout, about 8 in each row, slightly weaker and less numerous on the other tibiae; claws of tarsi toothed before the tip. Accessory genitalia of male small, hamule two-branched. Abdominal appendages of male simple, superiors: slightly constricted postbasally, then enlarging to beyond mid-length and then more rapidly reduced to a superior acute apex, a ventral row of small black teeth; inferior appendage triangular, shorter than the superiors, apex blunt or truncate, not deeply emarginate. Lateral margins of segments 8 of female not expanded. Vulvar lamina a more or less erect plate or trough. Wings long, moderately broad, the triangles at about the same level in front and hind wings, triangle of front wing narrow, the anterior side much shorter than the proximal, once crossed, triangle of hind wing and supertriangle of | both wings free, subtriangle of front wing three-celled, sectors of arculus joined in a long stalk, arculus between first and second antenodal, base of triangle of fc wing at the arculus, 11-15 antenodals in the front wing, the. last not continuous, one bridge cross-vein, one cubito-anal cross-vein in each wing, normally two cross-veins under the stigma, space between the last post- nodal and stigma normal, not distinctly longer than adjacent spaces between postnodals, basal and apical parts of stigma about the same color, not dis- ' tinctly bicolored, Cu, and Cu, in hind wing separate at origin, Rs and Rspl separated by one row of cells, M, with a single slight curve, not distinctly waved, Cu, in front wing strongly curved, the area between it and M, near their terminations in the wing margin wide, in front wing two (rarely three for one row’s length) rows of cells between A and the hind margin of wing at level of arculus, anal field of hind wing broad, usually, three (sometimes four and always two in credula) rows of cells between A, (A, of Laidlaw) and hind margin at level of base of triangle, two postanal cells in front wing, A, (Cuspl of Laidlaw) in hind wing slightly angled, the angle at about the level of distal angle of triangle, less rather than more distal, in hind wing three to five cells between Cu, and Cu, from triangle to distal angle of anal loop, one cell only between posterior angle of triangle and A, (Cuspl of 8 E. B. WILLIAMSON Laidlaw) in hind wing, triangle of front wing followed by three cells (rarely two in credula) followed by three rows (usually but not always by two rows in part of the field in credula), increasing to more at or distal to the level of the distal cell between M, and Mspl, hind wing with two postriangular rows of cells increasing to three at or distal to the level of forking of M,., and M,. The striking difference in the form of the abdomen of haematogastra or plebeja, for example, on one hand and muthroides or peruviana on the other, more than any other one character, led authors, prior to Calvert in 1906, to distribute the species of Erythemis in three genera. But between the extremes named above the remaining five species may be arranged from more slender to less slender in approximately the following order: carmelita, credula, sim- plicicollis, attala, and collocata. Another striking difference within the genus is the color of adults, espe- cially of males, some having more or less of the body, but especially the abdo- men, brilliant red, while in others no red or reddish appears in the colora- tion. This latter group includes some which are largely greenish or yel- lowish as tenerals and become dark bluish pruinose with age, while other darker species become deep black with age. This not-red group includes both slender and stout species, and also among the reds we have the extreme forms of abdomen. In the same way, among tenerals there are species with a distinctly patterned thorax and species which lack a definite pattern, the thorax being practically concolorous. And in each group are slender and stout species, and red and not-red species. Thus, the species within the genus are well marked and easily recognized, but efforts to define more than one generic group seem futile. In addition to the characters mentioned above, the accessory genitalia of the male and the vulvar lamina of the female are specifically distinct. In the case of the males, however, those individuals in which the penis is extruded usually, but not invariably, have the internal branch of the hamule bent out- ward and backward, so that the normal relation between the internal and external branches is lost. Key To THE SPECIES OF ERYTHEMIS ie Lateral and ventral carinas on abdominal segment 3 at the apex, measured along the apical carina, separated by .£8 mm. or less; the distance between the same carinas, opposite the meeting point of the lateral and median transverse carinas, equal to more than one and one-half times the distance between them at apex; lateral and ventral carinas on segment 4 separated by a distance less than one-sixth the length of the lateral carina of segment 4 (possibly not always true of credula). Ventral teeth on the superior appendages of the male extending posteriorly beyond the level of the apex of the inferior appen- AGI Tas Bp av deg tae ciebnie ide ms epm ete eaves Bs byw eterauthels 62 Ae aena einen Pee Relea 2. Tes Lateral and ventral carinas on abdominal segment 3 at the apex, measured along the apical carina, separated by I mm. or more; the distance between the same carinas, opposite the meeting point of the lateral and median transverse carinas, equal to one and one-half times or less the distance between them at apex; lateral and ventral carinas on segment 4 separated by a distance greater than one-sixth the length of the lateral carina of segment 4 (possibly not always true of simplicicollis). Legs of adults largely black or dark brownen see ea Si Zh): 3 (2). 4 (3). Ss (i): NOTES ON THE GENUS ERYTHEMIS 9 Abdomen less than 30 mm. long, shorter than the hind wings. Dorsum of thorax distinctly patterned, paler above, bordered on either side with ante- humeral black or the thorax black. Yellowish or greenish to black insects. Male with the external branch of the hamule triangular in shape, the postero- ventral edge longer than the distance between the apex of the internal branch and the antero-ventral angle of the external branch. Female lamina short (not SECLIM DVAMIINC:) Mais 5, o\c/tc chat at Meare RRO ett nicl > Ne ecuatemeparevnratc a wakes, arecie as credula. Abdomen 30 mm. long or longer. Male with the external branch of the hamule rounded, the postero-ventral edge equal to or shorter than the distance between the apex of the internal branch and the antero-ventral angle of the external |BSEZW AXE) oY SS ANREP RGIS ccs ciclo 2 6, 6 C1 0 CORREO ESE ec Part iar a en cao OO 3: Tibiae, and to a lesser extent the femora, largely dark brown or black. Abdo- men and hind wings about equal in length. Male with the external branch of the hamule rounded, the postero-ventral edge equal to or slightly shorter than the distance between the apex of the internal branch and the antero-ventral angle of the external branch. Female lamina trough-shaped, projecting ven- trad, the apex not distinctly bent and not directed caudad................... 4. Legs largely pale. Hind wings longer than the abdomen. Thorax not dis- tinctly patterned paler and darker. Yellowish or greenish to red insects. Male with the external branch of the hamule rounded, the postero-ventral edge about equal to, or slightly more than, half the distance between the apex of the internal branch and the antero-ventral angle of the external branch. Female lamina distinctly bent so the apical lobe is directed caudad, and, in lateral view, there is a deep posterior emargination between the basal lobes and the apex; from apex to base of basal lobe of lamina about .g mm.; apex of sternum of segment.o with a patch of sparse bristles. ..........-..8c00000%: carmelita n. sp. Dorsum of abdominal segments 5-10 predominantly dark or black; dorsum of thorax distinctly patterned, paler above, bordered on either side with a black antehumeral band, or the thorax black. Yellowish or greenish to black insects. Male with external branch of the hamule directed ventrad, in posterior view erect. Female lamina from base of basal lobe to apex I.5 mm.; apex of sternum of segment 9 with a patch of sparse bristles (80%) or with a single EEAMISMEESE “LOW, COON Oem neem te sec inte aati Sie a. cicns clave. ate wala cepai'n ly dae plebeja. Dorsum of abdominal segments 5-10 predominantly pale; dorsum of thorax not distinctly patterned paler and darker. Green or yellowish-green to red insects. Male with the external branch of the hamule directed more caudad than ven- trad, not erect in posterior view. Female lamina from base of basal lobe to apex .75 mm.; apex of sternum of segment 9 with a single row of bristles (80%) or with a patch of sparse bristles (207¢)....,-...20.-s0.: haematogastra. Labrum, face and frons pale colored, green and yellow. Wing bases unmarked or, at the most, hind wings basally tinged with yellow. Thorax not definitely patterned; green or yellowish green to dark pruinose. Male with ventral teeth on the superior appendages extending posteriorly beyond the level of the apex of the inferior appendages; external branch of the hamules directed more caudad than ventrad, genital lobes reaching ventrad far beyond the level of the hamules. Female lamina from base to apex about 1.25 mm., in antero- ventral view rounded triangular in shape, in lateral view the posterior basal lobe scarcely evident; apex of sternum of segment 9 with a single, or at most a double row, of bristles (true also of peruviana).......... ee eee rereeesace 6. Labrum, face and frons not as above. Base of hind wings distinctly dark or at least (some females) tinged with yellow. Thorax yellowish or greenish yellow to red or black. Male with ventral teeth on the superior appendages extending posteriorly to about the same level as the apex of the inferior appen- 7 ae) 8(7’). FE. B. WILLiAMSONn dage or not that far; external branch of the hamules directed ventro-caudad, extending ventrad about as far as or farther than the genital lobes. Female lamina in antero-ventral view, if triangular, swollen at base on either side, in lateral view the posterior basal lobe distinct; apex of sternum of segment 9 with a patch of bristles (except peruviana) yy, Lateral and ventral carinas on segment 4 separated by a distance about one- nfth the leneth of the lateral: carina, 9..- aeeecces usec (oo eee simplicicollis. Lateral and ventral carinas on segment 4 separated by a distance about one- thied the Jenothset the latetal carina. 6c. sob eee tae Se eee ae collocata. Dark basal area on hind wings reduced, not reaching the cubito-anal crossvein. Thorax distinctly patterned with a median rectangular pale area above, bor- dered on either side with black, to entirely black in old males which have abdominal segments 4-10 bright red. Lateral and ventral carinas on segment 4 separated by a distance about equal to one-third the length of the lateral carina. Male with the external branch of the hamules and the genital lobes reaching ventrad to about the same level, or the genital lobes very slightly longer; apical ventral teeth on the superior appendages about the level of the distal end of the inferior appendage. Female lamina in antero-ventral view semicircular, length from base of basal lobe to apex about .65 mm.; apex of sternum of segment 9 with a single row of bristles.................. peruviana. Dark basal area on hind wings reaching beyond the cubito-anal crossvein. Thorax not patterned with a median rectangular pale area bordered on either side with black. Lateral and ventral carinas on segment 4 separated by a dis- tance less than one-third the length of the lateral carina. Male with the exter- nal branch of the hamules reaching ventrad distinctly beyond the level of the genital lobes. Female lamina in antero-ventral view not semicircular, length from base of basal lobe to apex about 1.2 mm.; apex of sternum of 9 with a Patchy of (bristleszis;. ta whee Hove he cote ES s Gene eee eee Seen eee 8. Thorax largely dark or black, intricately patterned with paler, to completely black in old males. Abdomen black in old males, in others largely dark with conspicuous large basal or subbasal pale (yellow) areas, only narrowly sepa- rated by black in the median line, on segments 4 and 7. Dark colored to black insects. Lateral and ventral carinas on segment 4 separated by a distance equal to or less than one-fourth the length of the lateral carina. Male with the apex of the inferior appendage reaching beyond the level of the ventral teeth on the superior appendage. Female lamina in antero-ventral view nearly as long as wide, rounded triangular, slightly swollen at base on either side........ attala, Thorax pale yellowish or greenish to dark red, without dark markings. Abdo- men largely pale, yellowish to bright red. Pale colored to red insects. Lateral and ventral carinas on segment 4 separated by a distance equal to between one-third and one-fourth the length of the lateral carina. Male with the ven- tral teeth on the superior appendages extending posteriorly to about the same level as the apex of the inferior appendage. Female lamina in antero-ventral view trilobed, the width about one-half greater than the length..... mithroides. Erythemis carmelita, new species Plate I, figs. 1-6 Abdomen, male 31-31.5, female 32-33; hind wing, male 34.5-35, female 35; stigma, male 3-3.2, female 3.4-3.6. Male.—Rear of head brownish red or yellow, labium yellow, darkening to red anteriorly and mesially, labrum orange, clypeus, frons, vesicle and occiput red, the last and the anteclypeus duller. NOTES ON THE GENUS ERYTHEMIS Il ' Prothorax and thorax yellowish or greenish red, possibly bright red in life in old males, unmarked. Legs yellowish (ochraceous-tawny of Ridg- way), the antero-dorsal surface of the first femur and tibia black; second femur with the same surface largely black, and a narrow line of black on the same surface of the second tibia; third femur with the same surface with an ill-defined, restricted, subapical black area equal in length to about one-fourth the length of the femur; a narrow interrupted line on the same face of the third tibia; all tarsi darker, obscure to black; all spines and teeth on the legs, black, of usual form. Front wings at base with a trace of yel- lowish red. Hind wings with a yellowish red basal spot which anteriorly reaches the first antenodal and posteriorly is slightly more extensive, a few of the included cells sometimes with a dark reddish brown spot at the center. Stigma brown, narrowly shaded darker along the anterior edge. Abdomen, including accessory genitalia and apical appendages red, the genitalia and appendages paler, yellowish; all unmarked, except for restricted black edgings on some carinas, especially on the apical carina; segments 4-9 in dorsal view shading apically into darker red. Abdomen inflated at base, 3 at base 5 high, 4, 1.6 high; 4-6 in dorsal view measure about as follows: 4, 1.6 wide at mid-length, and 4.8 long; 5, 1.9 wide and 4.6 long; 6, 2.1 wide and 4.4 long; 7-9 are about as wide as 6, 9 slightly narrower, and 10 is about 1.8 wide; in length, 7-10 are, respectively, 3.6, 2.4, 2, and 1. In profile at mid-length, 4 is about 1.1 high; from this point the abdomen gradually thickens posteriorly to the apex of 8, which is about 1.8 high. Lateral and ventral carinas on segment 3 separated at the apex by about .8, at the level of the point where the lateral meets the median transverse by about 1.6; on 4 the lateral carina is about 4.7 long and the lateral and ventral carinas are separated by a distance of .5 to .7. Female—Rear of head pale dull yellowish, a reddish brown spot on either side above and below against the eye, head pale as in the male but dull pale greenish brown without trace of red, the labium and anteclypeus yellowish. Thorax dull green (about Saccardo’s olive of Ridgway in dried speci- mens), unmarked. Legs slightly duller than in the male, similarly marked. Wings hyaline without basal markings or with yellowish red at the base to a varying extent, from a mere trace to, at the other extreme. in the front wings about one-third the distance to the first antenodal and in the hind wings slightly beyond the first antenodal- Stigma brown, narrowly shaded darker along the anterior edge. Abdomen dull green, about the same color as the thorax, with 4-10 apic- ally dark; 1-3 with the carinas slightly darker or more yellowish than the body of the segments; 4-7 shading to black apically, the apical third to half of each segment dark; 8-10 largely or entirely dark to black; the appendages and the lamina supra-analis lying between them yellowish to black, the latter about .67 long and 1 wide at base, triangular, the apex truncated, about .3 wide, and each posterior angle a rounded knob or tubercle so the apex is broadly emarginate or divided. Abdomen inflated at base, 3 at base about 4.5 high, 4 about 2.1 high; 4-6 in dorsal view measure about as follows: 4, 12 E. B. WILLIAMSON 1.7 wide at mid-length and 5.6 long; 5, 1.9-2 wide and 5-5.2 long; 6, 2.1-2.4 wide and 4.8-5 long; 7-9 are about as wide as 6, 9 slightly narrower, and 10 is about 1.6 wide; in length, 7-10 are, respectively, 3.8-4, 2-2.4, 1.6-1.9, and .8-.9. In profile, at mid-length 4 is about 1.8 high; from this point poste- riorly the abdomen is about the same height to near the apex of 8, the apex of which segment is 2.6-2.8 high. Lateral and ventral carinas on segment 3 separated at the apex by about .8, at the level of the point where the lateral meets the median transverse by about 1.2-1.4; on 4 the lateral carina is about 5-2 long and the lateral and ventral carinas are separated by a distance of .6-.7. Venation, male and female—Antenodals, front wing 13-15 (15 in 14%, 13 in 21%, 14 in 64%), hind wing 9-11 (11 in 14%, 9 in 28%, I0 in 57%); postnodals, front wing 9-10 (Io in 35%, 9 in 64%), hind wing 10-11 (11 in 14%, 10 in 85%). In front wing usually two rows of cells between A and the hind margin at level of arculus, but in four male wings the distal row of cells (the row against the subtriangle and the postanal cells) consists of three. In the hind wing three or four rows of cells between A, (A, of Laidlaw) and the hind margin. In hind wing 4+ cells between Cu, and Cu, from triangle to distal angle of anal loop, except in one female wing, where there are 5 cells. In anal loop 9-11 cells between A, and A, (Cuspl and A, of Laidlaw) (11 in one female wing, 9 in three male wings, 10 in all others). Described from four males and one female, E1 Banco, Colombia, January 23-25, 1917; one male, Rio Neuvo, between Magangué and El Banco, Colom- bia, January 22, 1917 (both localities in the lower Magdalena River Valley) ; one female, Tucacas, Venezuela, March 23, 1920 (all above material in coll. EK. B. W.); and one female, Boa Vista do Jaquiri, near the mouth of the Rio Teffé, Brazil, September 5, 1920, W. T. M. Forbes, coll. Cornell Uni- versity. Type, the male from Rio Neuvo, allotype female, El Banco, January 25, 1917. Named in honor of Mrs. M. A. Carriker, Jr. (née Carmelita Flye), whose courage and industry as a member of Mr. Carriker’s expedi- tions in the American tropics merit recognition. Specimens of this species have been sent to Drs. Calvert and Ris, and both pronounce it new. Unfortunately, our notes give no particular data as to habits of the species. The male at Rio Neuvo was taken in a bit of dry (at that season) pastured woods. At El Banco we collected a variety of habitats, but most of our material was taken in dry, (at that season) woods and brush adjoining the river, and nearly dried up adjacent pools. At Tucacas on March 23 we collected at a large lagoon, possibly an old river bed, lying on the north side of the railroad and between four and five kilo- meters out of town. ‘This lagoon has large areas of deep open water bor- dered at places by more extensive low areas filled with rosette-leaved plants, sedges, grasses, and several species of bushes. At places large areas of this sedgy, brushy marsh were floating on the deeper waters of the lagoon, and about such places, crocodiles were very numerous. The more solid ground about the lagoon is covered with pastured grass, dense thickets of small spiny trees or shrubs, and forests. Strong flying libellulines are numerous about the margins of the deeper water in the lagoon." NorTEs ON THE GENUS ERYTHEMIS 13 Carmelita is separated from all other species by its paler legs. It is sep- arated from the less densely veined, very distinct and smaller credula by venational characters indicated in the definition of the genus. With the exception of simplicicollis and collocata, the species of Erythemis are very distinct and it is difficult to distinguish relationships within the genus. How- ever, carmelita is probably more closely related to haematogastra than to any other species. Characters given in the key readily separate them. A List OF THE SPECIES RECORDED IN THIS PAPER, WITH PLACES AND Dates oF CAPTURE? The species are arranged alphabetically and numbered consecutively. Following each locality under each species are numbers in parentheses indi- cating the other species taken at the same place. For example, E. plebeja Burmeister. Venezuela: Encontrados, (1, 5, 6), indicates that attala, mith- roides, and peruviana, in addition to plebeja, were taken at Encontrados. 1. Erythemis attala Selys 5. Erythemis mithroides Brauer 2 = carmelita Williamson 6. es peruviana Rambur zB: a credula Hagen Ts ss plebeja Burmeister 4. os haematogastra Burmeister 1. Erythemuis attala Selys. Colombia: Fundacion, (4, 5, 6, 7), January 13, 1917; El Banco, (2, 4, 5, 6, 7), January 23 and 25, 1917. Venezuela: Palma Sola, (5, 6, 7), March 8, 1920; Encontrados, (5, 6, 7), April 25, 1920. Attala is a widely distributed and therefore apparently a common species, but I have never found it in great numbers at any particular habitat. In fact, the total Colombia-Venezuela catch numbers only four males and three females. However, Bates, as recorded by Ris, notes: “Dry woods at San- tarem, very common, much more abundant than plebeja.’ As opposed to my general statement that tropical species of the genus are probably on the wing throughout the year is the fact that Hine and I failed to get attala at Amatillan, Gualan, and Los Amates, in Guatemala, in January, 1905, while I took it at the three localities in June, 1909. But available records show attala on the wing somewhere every month in the year, except September, November, and December, and there is no doubt it flies these months, too; and our failure to find it in January in Guatemala was probably due to its relative scarcity and to its having deserted the permanent bodies of water (where we did most of our collecting) to range about in the dry brush. 1For notes on localities, see: A Collecting Trip to Colombia, South America, by E B. Williamson, Miscellaneous Publications, Univ. of Mich., Mus. of Zool., 'No. 3, February 22, 1918; and Notes on the Habitats of Some Tropical Species of Hetaerina, by E. B. Williamson, Occ. Papers, Univ. of Mich., Mus. of Zool., No. 130. Localities not discussed in these papers or in the description of &. carmelita are briefly described at the end of this paper. The student may be interested to note that at the following locations, described in Occasional Papers, No. 130, no species of Erythemis have been recorded: Aroa, Boqueron, Cincinnati, Cristalina, La Fria, Maraquita, Nirgua, San Esteban, and Tachira. 14 E. B. WILLIAMSON \ The elevation of the three Guatemalan localities mentioned are, moreover, above the apparently optimum habitats of species of Erythemis (due prob- ably rather to the relative absence of such pools as occur at lower levels than to elevation itself), and it may be noted that on February 5, 1905, we found the species at sea level at San José, Guatemala. Attala is an active species in a genus of generally active species, and at times individuals are seen whose incessant activity and wariness effectually prevent their capture. It is a matter of common observation among strong- flying species of dragonflies that in an assemblage of only a few individuals of a species such individuals are much more wary of the collector than indi- viduals in a numerous assemblage when their attention is largely given to each other, and the collector finds their capture less difficult. This will often explain the extreme wariness of attala which, as stated above, has not been noted as very numerous about any particular habitat in Colombia or Venezuela. 2. Erythemis carmelita Williamson, Colombia: Rio Neuvo, (5, 6), Jan- uary 22, 1917; El Banco, (1, 4, 5, 6, 7), January 23-25, 1917. Venezuela: Tucacas, (5, 6, 7), March 23, 1920. See description of this species in this paper for further notes. 3. Erythemis credula Hagen. Panama: Gamboa, Canal Zone, (6), August 5, 1920, J. H. Williamson. Credula is apparently, next to carmelita, the rarest species of the genus and J. H. W.’s record for Panama is the first for Central America. I have seen the species alive at only one locality, Pitch Lake, Trinidad, where the males were flying about the small pools of water in the pitch. 4. Erythemis haematogastra Burmeister. Colombia: Rio Frio, (5, 6, 7), January 8, 1917; Sevilla, December 15, 1916; Fundacion, (1, 5, 6, 7), Jan- Uatyt2-and 13,1917; Hl (Banco, (1, 2,.5;. 6,17) hanuaty, 24 aadeas signa Puerto Berrio, (5, 6), January 8 and February 21, 1917. Veneguela: El Guayabo, (5, 6, 7), April 20, 1920. Bates, as recorded by Ris, notes: ‘Rather common species at Santarem ; it frequents only the shade of the woods and settles frequently on dried twigs. Banks of brooks, Para, not uncommon.” So far as I have observed, this species is rare, being found usually only as a solitary individual sunning itself on some twig or broad leaf near or far from water. In fact, the total Colombian-Venezuelan catch numbers only fifteen specimens, nine males and six females; the number of specimens taken at each locality varies from one to three, except at El Banco, where five specimens were collected. 3. Erythemis mithroides Brauer. Colombia: Santa Marta, (7), Decem- ber 19, 1916; Rio Frio, (4, 6, 7), January 7, 1917; Fundacion, (1, 4, 6, 7), January 13 and 14, 1917; Bolivar, (6, 7), December 20, 23, and 26, 1910; Rio Neuvo, (2, 6), January 22, 1917; El Banco, (1, 2, 4, 6, 7), January 23-25, 1917; Puerto Berrio, (4-6), January 30, 1917. Venezuela: Tucacas, NovrEs ON THE GENUS ERYTHEMIS 15 (2, 6, 7), March 23 and 24, 1920; Palma Sola, (1, 6, 7), March 4, 6, and 10, 1920; Encontrados, (1, 6, 7), April 23 and 25, 1920; El Guayabo, (4, 6, 7), April 20 and 22, 1920. In life mithroides is a rich-colored and active species, usually with lower flight than the slender species haematogastra and plebeja, which are possibly more wary and active than mithroides. About the ponds where mithroides often occurs in considerable numbers it is often seen resting on the leaves of coarse grasses and sedges, its color and activity making it conspicuous. We have found it in numbers on several occasions: the pond on the upper side of the railroad out from Santa Marta and just beyond the bridge over the Manzanares; the large swamp south of Mamatoca near Bolivar; and the little ponds along the railroad and beyond the Cana Fraile out from El Guayabo. One of two males taken at Rio Neuvo has the triangle of both front wings free. One of the large males taken in Guatemala and described by Dr. Ris (Coll. Selys, p. 1173) was referred to Dr. Kennedy, who kindly examined the penis of the specimen and found it agreeing with penes of smaller specimens of the species. 6. Erythemis peruviana Rambur. Panama: Gamboa, Canal Zone, (3), August 5, 1920, J. H. Williamson. Colombia: Rio Frio, (4, 5, 7), January 4 and 6-8, 1917; Fundacion, (1, 4, 5, 7), January Io and 12, 1917; Bolivar, (5, 7), December 19, 23, and 26, 1916; Rio Neuvo, (2, 5), January 22, 1917; Magangué, January 21, 1917; El Banco, (1, 2, 4, 5, 7), January 24, 1917; Las Delicias, January 28, 1917; Puerto Berrio, (4, 5), January 31 and Feb- ruary 8, 1917. Venezuela: Tucacas, (2, 5, 7), March 23 and 24, 1920; Palma Sola, (1, 5, 7), March 4, 6, 8, and 9, 1920; Bejuma, (7), February 16, 1920; Encontrados, (1, 5, 7), April 23, 25, and 26, 1920; El Guayabo, (4, 5, 7), April 20, 1920. Bates, as recorded by Ris, states: “Banks of brooks near Para, Obydos very plentiful.” In fact, this species may occur about some favored habitat in possibly larger numbers than we have ever seen any other species of the suborder Anisoptera. Under such circumstances, usually about a pond or muddy, sluggish, and sun-exposed creek or ditch, most, if not all, of the individuals will be teneral or lacking the colors of the fully adult. However, a few mature ones in full flight, by their restless activity and general imper- tinence toward other species, may make collecting of other Anisoptera at the same habitat difficult or well-nigh impossible. And when at some swamp or pond the collector finds peruviana represented by only one or two adult males, he will usually find the capture of these a difficult or even impossible feat if he depends solely on his insect net. As in other species observed, the females of peruviana are more voracious feeders than the males. A female taken at Encontrados, Venezuela, was feeding on a butterfly, Thymelicus athenion Huebner, as determined by Mr. William Schaus. In fact, Lepidoptera seem to be a favorite food for species of Erythemis, and I have records of simplicicollis in Indiana feeding on Thecla liparops Bd. and Lec., and Haploa confusa Lyman, both determina- tions by Dr. Dyar. 16 E. B. WILLIAMSON In marked contrast to the dull, common-looking tenerals and females, the fully adult males of peruviana are among the most brilliant and handsome of dragonflies. In flight they are swift and agile, and suggest simplicicollis, but I have never observed them indulging in the complicated evolutions of the latter species. 7. Erythemis plebeja Burmeister. Colombia: Puerto Colombia, Decem- ber 10 and 11, 1916; Santa Marta, (5), December 13, 16, and 18, 1916, and January 3, 1917; Don Jaca, December 17, 1916; Rio Frio, (4, 5, 6), Jan- uary 4-8, 1917; Fundacion, (1, 4, 5, 6), January 13, 1917; Bolivar, (5, 6), December 20, 21, 23, and 26, 1916, and January 2, 1917; Zambrano, January 20, 1917; El Banco, (1, 2, 4, 5, 6), January 23, 1917; Gamarra, January 27, 1917. Venezuela: Tucacas, (2, 5, 6), March 23, 25, 1920; Palma Sola, (1, 5,6), March 8 and 10, 1920; Bejuma, (6), February 13, 18, and 24, 1920; Encontrados, (1, 5,6), April 25, 1920; El Guayabo, (4, 5,6), April 20, 1920. A male of this species, collected in Jamaica and received from Carpenter, is labelled haematogastra, but I do not think this mistake is likely to be fre- quent as, if any confusion occurs as regards plebeja, it is likely to be with attala; and as Calvert has pointed out, the adult male of haematogastra super- ficially resembles the adult male, not of another Erythemis, but a species of another genus, Rhodopygia hollandi, just as another Erythemis, peruviana, superficially resembles Planiplax sanguiniventris rather than some other spe- cies of Erythemis. Bates, as recorded by Ris, states of plebeja: “Banks of the river Tapajos near Santarem, settling on bushes in the woods.” It is an agile species and is found widely scattered in practically all the environments of any locality where it occurs. CONCLUSIONS Finally, I should lke to call attention. to the present status of Erythemis and a very different genus of American dragonflies, Hetaerina, and to certain facts in their life histories, leaving to other students any con- clusions as to the significance of these facts in explaining present conditions. ‘There are nine known species of Erythemis as against about fifty species of Hetaerina. And yet, in any locality (an area three to ten miles in diameter) the largest known list of Hetaerinas is four species, or eight per cent of the total, while in Erythemis the largest local list is six species, or sixty-six and two-thirds per cent of the total. Moreover, in any locality the species of Erythemis there tend to more fully occupy the various environments of the locality than do Hetaerinas, or one environment in the locality may harbor all the Erythemis found there, while these conditions are usually reversed in Hetaerinas, each species often showing a marked preference for a certain ecological habitat. The species of Erythemis, as larvae, live in stagnant water, and Hetae- rinas live in running water. Streams offer a greater variety of ecological conditions than fresh-water ponds offer. Hetaerinas as imagoes are closely bound to the streams which make their larval home. Their dispersal to new habitats is doubtless rare and accidental. NOTEs ON THE GENUS ERYTHEMIS 17 It is entirely conceivable that a single fertile female in a certain instance might furnish the entire original Hetaerina invasion of the headwaters of . some small mountain creek. Here the species, by direct descent from this single female, might maintain itself without any admixture of new blood for many generations. After a period of time, the infrequent and accidental invasion by new blood would be quickly swamped in the mass of old stock. Individuals of Erythemis, on the other hand, are wanderers, and the Ery- themis fauna of any pond may be partly or largely descended from new immi- grants of the preceding season, their descendents in turn scattering with the dry season and many of them visiting other ponds when the period of ovi- positing comes. Specimens of any species of Erythemis are remarkably uniform through- out the range of the species. I know of only one exception, that of the Guatemalan specimens of £. mithroides, which are distinctly larger than South American specimens of the same species. On the other hand, a num- ber of cases of differences in a species of Hetaerina in its range can be cited. The most striking case I recall is the remarkably large, dark specimens of cruentata occurring at a high elevation in the isolated Santa Marta Moun- tains of Colombia. Almost equally striking is the large, dark form of macro- pus at an elevation of 2,230 feet in Peru. On a larger scale, but less dis- tinctly marked, are the pale forms of caja and macropus east of the Andes in western Venezuela, as compared with the dark forms west of these mountains. LocaLities Not HERETOFORE DESCRIBED IN THIS OR OTHER PAPERS 1. Bejuma, Department of Caraboba, Venezuela. In describing the coun- try about Bejuma in Occasional Papers, No. 130, several swamps or shallow ponds which lie in the valley were not mentioned. One of these is about one mile east of Bejuma. The area is about an acre, surrounded by thorny mimosas inside which grew several species of sedges, water hyacinth, another aquatic with three yellow petals, orange at the base, another with small pur- ple flowers, and a small yellow Utricularia. This pond was in the last stages of drying up when we saw it on February 16, 1920, with very little open water, which had a maximum depth, with the mire in which one sank, of about three feet. 2. Encontrados, Department of Zulia, Venezuela. Encontrados is on the right bank of the Catatumbo River and is the lower terminus of the Gran Ferrocarril del Tachira. Its elevation is about 138 feet, and it lies in a low, flat country of brushy forest, large areas of which near town and along the railroad are now in grass, bananas, cocoa, and some sugar cane. Just above town is a long lagoon filled with aquatics and bordered with brush and pas- tures where odonate life, especially strong-flying libellulines, was very abun- dant. Along the river below town are many flats, covered with water during the rainy season, and during the dry season consisting of stretches of mud, grass, and bushes, with shallow pools. Three or four kilometers below town on the left river bank is a cafia eight to twelve feet wide, mostly in the sun, 18 E. B. WILLIAMSON muddy and with many logs, the banks high and grass-covered, and the adja- cent country in corn, bananas, and abandoned fields, with occasional patches ‘of forest with heavy brush and frequent dense Heliconia growths. These bits of forest are often bordered and sometimes completely surrounded by an impenetrable jungle of spiny shrubs overgrown with razor-edged sawgrass. 3. Las Delicias, Department of Magdalena, Colombia. A woodpile on the banks of the Magdalena about 120 leagues above Barranquilla and two or three leagues below Boca de Rosario. The surrounding country is the usual low, level-lying river plain of the lower Magdalena. 4. Zambrano, Colombia. A village on the Magdalena between Calamar and Magangué in the usual level, pastured land of the lower river valley. THE PHYLOGENY AND THE GEOGRAPHICAL DISTRIBUTION OF THE GENUS ERYTHEMIS (ODONATA) By CLARENCE HAMILTON KENNEDY The following paper has been prepared at the suggestion of Mr. E. B. Williamson, who, in the accompanying paper, has presented the systematics of this group. The material has been arranged to parallel the arrangement in the writer’s article’ on the genus Libellula, and is one of a series of papers which it is planned to present on the relationships of the American Odonata. Time is not available and material is not at hand for a definition of the relationships of this genus to the other genera in the Libellulinae. For this reason, the writer has followed the definition of groups presented by Ris,? who followed Calvert.*. The penes of Lepthemis and of all the species of Erythemis recognized by these authors have been examined. Ris includes in this group only one other genus, Rhodothemis of the East Indies. As no material of this species is in Mr. Williamson’s or the writer’s collections, it has not been studied. However, Ris figures its wings and gives a detailed description of its peculiarities. From this account it is evidently the least specialized of the three genera in the armature of the legs, which have the spines of the femur less developed; and in the venation, which is ‘‘reduced” as it is in credula only of the American genera. The location of Rhodo- themis, the least specialized member of this series, in the East Indies sug- gests an Oriental origin for the group as a whole and that the stock of the American genera at some time in the past has spread into its present habitat. As the majority of the species are tropical, the spreading into America may have taken place when the Asia-Alaska land-bridge had a climate suitable for such species, which must have been in Pre-Miocene times. The other possi- bility 1s that they came across from Africa, but so far we have not recog- nized any African relatives of the group. Credula, the most primitive of the American species, has not been found north of Panama and St. Thomas, while the two species that are found farthest north, haemdatogastra*t and sim- plicicollis, are two of the most specialized, which distribution seems to con- firm a southern or tropical entrance into America. The homologies of the lobes of the penis in the genera Libellula, Sym- petrum, Orthetrum, and Orthemis were worked out by the writer in a pre- vious article.* All parts that were defined for these appear also in the series 1The Phylogeny and the Geographical Distribution of the Genus Libellula (Odo- nata). Ent. News, XXXIII, pp. 65-71, 105-111, 1922. 2“TLibellulinen,” Fasc. XIII, pp. 591-608 in Coll’s Zool. du Selys, 1911. 3 Nomenclature of Certain North American Odonata, Ent. News, XVII, pp. 30-31, 1900. oe Williamson doubts the Georgia records of this species. 5 The Morphology of the Penis in the Genus Libellula. Ent. News, XXXIII, pp. 33-40, 1921. 20 CLARENCE HAMILTON KENNEDY of Erythemis penes. In comparison with the penes discussed in the above article, the first peculiarity in the Erythemis penes that strikes one is the ‘development of a small, chitinized, two-lobed hood over the medial lobes. Apparently, it is a specialization of Erythemis and its near relatives. It appears in all the species except credula, in which species the penis is most like that organ in the less specialized members of Libellula, Sympetrum, etc. As this species has the generalized characters of venation found in Rhodo- themis, we have considered it the most primitive of the American species of the Erythemis series. Even though it lacks this hood, there is no doubt that credula belongs in or next to Erythemis, as it has in a pair of tubercles the homologue of the hood in the other species (see fig. 8). The American species, by penis characters, divide into four groups which are exactly equivalent with the groups as outlined by Calvert and Ris on other characters. ‘To these, Rhodothenus rufa may be added as an extra group. In the key J have placed it first as Group I because of its generalized characters. ‘The fifth and largest group is composed of species with penes so much alike that no good characters appear on which to divide them fur- ther, though they seem to fall into two sub-groups on the shape of the abdo- men. The writer would be inclined to call the whole series of five or six groups Lepthemis and would consider the individual groups as subgenera, which could be given generic rank by those who wish to use smaller genera. Lepthemis, for the whole series from Rhodothemis rufa to Erythemis haema- togastra, is no more comprehensive, in the writer’s opinion, than Libellula for the series of species included in the latter by Ris and later by the writer. The comprehensive genus would have to be Lepthemis because of the pri- ority in the use of that name by Hagen.® The groups may be outlined as follows: ite Oxiental “CRhodothetis)p4 o-edejen: «eo sees sarees es 2 ae Ree Group I—rufa American. (Lepthemss ‘and Erythemis)).\00%s 5.052) o-0s ees eet ee ee eee 2 2 (1’). Hood over media! lobes undeveloped (represented by two tubercles).......... SiG aan devils Ney SHAS teeta, AA MRI Ca bats 1S yee RR BHR RSH SIO Group IIl—credula 2h Hood over medial lobes developed into an arched two-lobed affair.......... 3. 3 (2'). Posterior lobe of penis lacking, the three cornua fused into a single conspicu- ous, terminal. Lobe. Vase is taed fete ete ne so gee Reet Pett ae Be eae 4. ’ 3% Pasterion Wobelpresentis ci ci laere: cries <5 ats) Sete hla ane what ove. eee ota tee ok ened eto 5s 4 (3). Lateral lobes and cornual lobe parallel to the axis of the penis (Lepthemis)... SRAM EE ATARGA. BPRS PRUE RU OMSL Resa ite rade tage sors ses fovceenay roveramenene Group IlI—vesiculosa 4. Lateral lobes and cornual lobe bent across the axis of the penis (Mesothemis) .. abe Mis Selle Bete Pd eh ere Mee tare hee ee one Group 1V—collocata simplicicollis 5 (3'). Abdomen not especially slender ((Eirythemis) 0-2 nosnetse eee eee eerie Group 'V A—peruviana mithroides attala ne Abdomen slender (Erythemis)...GroupV B—plebeja carmelita haematogastra © Synopsis of the Neuroptera of North America, p. 160, 1861. DISTRIBUTION OF THE GENUS ERYTHEMIS ~ 21 This series of eleven species, small as it is and with several of its species occupying nearly identical habitats, is so non-committal as to its origin and routes of migration that these will have to be worked out by some less direct method than that of a study of the degrees of specialization of the individual species. ‘The latter method gives results where the group is large and many connecting links remain. It is possible that vesiculosa belongs near Group V B and that the loss of the posterior lobe has occurred independently in Groups II] and IV. A careful check of the other characters should decide this. We have also the problems of the generic rank of Mesothemis and the value of its two forms, collocata and simplicicollis. By penis characters alone, Mesothemis is as valid a genus as Lepthemis. In fact, the specialization of the fusion of the lobes of the cornua and the loss of the posterior lobe has gone farther from the Erythemis type that is found in the majority of the Erythemis species than it has gone in the Lepthemis penis. The forms, col- locata and simplicicollis, as viewed in the light of recent work by geneticists on other insects, are probably good species. Just how nearly homozygous each is under all circumstances will have to wait until a careful study can be made of the two forms where their habitats overlap. 22 CLARENCE HAMILTON KENNEDY EXPLANATION OF PLATE I Figs. 1-6, by E. B. Williamson. Figs. 7-16, by C. H. Kennedy. Figs. 1-6, Erythemis carmelita n. sp. Figs. 1-4, type male; figs. 5-6, allotype female. Fig. 1, accessory genitalia of segment 2 in lateral view; fig. 2, same in latero- ventral view; fig. 3, abdominal appendages in lateral view; fig. 4, inferior appendage in ventral view; fig. 5, vulvar lamina in lateral view; fig. 6, the same in antero-ventral view. Figs. 7-16, penes of species of Lepthemis and Erythemis. Each pair of figures represents, in the upper member, a lateral view of the tip or third segment of the penis and, in the lower member, the same view with the lateral lobe removed to show the cornual and medial lobes. Where not otherwise indicated, the specimen is in the collection of Mr. E. B. Williamson. The lettering of the figures is that used in “The Morphology of the Penis in the Genus Libellula,” Ent. News, XX XIII, pp. 33-40, as follows: H=hood, M=medial lobes, C=cornua, I,=—lateral lobes, P = posterior lobes. Fig. 7, Lepthemis vesiculosa, Dutch Guiana, O. S. U. coll.; fig. 8, Erythemis cre- dula, Pitch Lake, Trinidad; fig. 9, Erythemis peruviana, Tucacas, Falcon, Venezuela; fig. 10, Erythemis mithroides, El Banco, Colombia; fig. 11, Erythemis attala, Funda- cion, Colombia; fig. 12, Erythemis collocata, Fresno Beach, California; fig. 13, Ery- themis simplicicollis, Columbus, Ohio; fig. 14, Erythemis plebeja, Fundacion, Colom- bia; fig. 15, Erythemis haematogastra, Fundacion, Colombia; fig. 16, Erythemis carme- lita, El Banco, Colombia. LEPTHEMIS AND ERYTHEMIS PLATE I BEG E. carmelita n. sp.; figs. 1-4, type male; figs. 5-6, allotype female. 16/— carmelita Penes of the species of Lepthemis and Erythemis, figs. 7-16. ‘ if _ j sa = = ay VY, Pater) ee eee UNIVERSITY OF MICHIGAN MUSEUM OF ZOOLOGY Miscellaneous Publications No. 12 The Genus Gyrotoma BY CALVIN GOODRICH Published in Co-operation with the Geological Survey of Alabama ANN ARBOR, MICHIGAN PUBLISHED BY THE UNIVERSITY FEBRUARY 23, 1924 ADVERTISEMENT The publications of the Museum of Zoology, University of Michigan, consist of two series—the Occasional Papers and the Miscellaneous Publi- cations. Both series were founded by Dr. Bryant Walker, Mr. Bradshaw H. Swales and Dr. W. W. Newcomb. The Occasional Papers, publication of which was begun in 1913, serve as a medium for the publication of brief original papers based principally upon the collections in the Museum. The papers are issued separately to libraries and specialists, and, when a sufficient number of pages have been printed to make a volume, a title page, index, and table of contents are sup- plied to libraries and individuals on the mailing list for the entire series. The Miscellaneous Publications include papers on field and museum technique, monographic studies and other papers not within the scope of the Occasional Papers. The papers are published separately, and, as it is not intended that they shall be grouped into volumes, each number has a title page and, when necessary, a table of contents. ALEXANDER G. RUTHVEN, Director of the Museum of Zoology, University ci Michigan. . THE GENUS GYROTOMA By CALVIN GoopRICH The genus Gyrotoma has received no systematic treatment in fifty years. Since 1873, when Tryon’s monograph was published, three collectors, T. H. Aldrich, A. A. Hinkley and H. H. Smith have taken many hundred times more specimens than were available for study by the naturalists who de- scribed the species. For the present inquiry I have examined material from several sources. ‘The collection made by Mr. Smith and now in the Ala- bama Museum of Natural History is unrivaled, and, because of the water- power development in the Coosa River, flooding the shoals in which Gyro- tomae live, it can probably never be duplicated. Acknowledgments are due to the Alabama Museum, the U. S. National Museum, Dr. Bryant Walker and Miss Mina Winslow. Without their aid this work could not even have been attempted. THe RECORD _ Dr. Isaac Lea was the first descriptive naturalist to receive specimens of Gyrotoma. These he described as Melania excisa and Anculosa incisa, sug- gesting at the same time the generic term Schizostoma. When later he found the name already in use for a fossil he changed it to Schizochilus. In the meanwhile Shuttleworth had defined his genus. The fossil Schizostoma as well as the Schizostoma of Lea finally were forced into the synonymy. The good doctor, one of whose picturesque qualities was tenacity of pur- pose coupled. with a certain fervid plausibility, sought to dig his original name out of its grave. The canons of nomenclature do not allow for cases of resurrection, and Gyrotoma has had to stand as the true generic desig- nation. Lea’s specimens were sent to him by Dr. B. W. Budd. Because these shells were few in numbers, the condition of many of them that of drift material, their place of origin confused, and because Dr. Budd was himself interested in conchology and hardly likely to take ten or twenty shells where he might have gathered thousands, it is improbable that he was the collector. About the same period, Gyrotomae came to Mighels and apparently from the same correspondent. This was 1841, possibly 1842, to 1845. All the forms were of the Wetumpka aspect. In July, 1845, Shuttleworth described two species. His specimens had been collected by Rugel, who apparently visited the Coosa River at Greensport or thereabouts and may have gone as far down as Childersburg. It was not until 1904, when Mr. H. H. Smith began to explore the Coosa systematically, that Shuttleworth’s one “good” species was found by an American collector. This history is an illustra- tion of the casual, one might say accidental, collecting of early-day zoology in America. 4 CALVIN GOODRICH About fifteen years after Dr. Budd’s sendings, Gyrotomae, from the reaches between the Yellowleaf Creek of Shelby County and Fort William Shoals, Talladega County, were taken by Dr. Schowalter of Uniontown, Alabama. It would seem that Dr. William Spillman of Columbus, Missis- sippi, visited the Coosa River in the same places and at about the same time, and it is possibly fair to infer that the two were together. Lea and An- thony erected eighteen species from this new material and Reeve described still others under names supplied by Anthony. Reeve’s monograph of Melatoma appeared in 1860-61. Swainson had established Melatoma as a subgenus of Melanopsis Lam. Reeve believea it to be the same as Gyrotoma. Tryon says that Melatoma “is actually an exotic marine shell.” ‘The connection of Anthony’s name with Melatoma was an error. Reeve’s illustrations are excellent and in most instances the species can be recognized from these pictures. But the descriptions, written by Reeve, convey little information of value. In two of the descrip- tions no mention is made of the fissure, the distinctive character of Gyrotoma. In the case of one species the slit is described as “small, almost obsolete”; in one as “slightly channelled”; in two, “moderate”; in two, “broad’’; in ten, “deep”; in two, “rather deep’; in one, “very deep.” The reader of Reeve is reminded of the childhood question, “How deep is a well and how long is a string?” The Anthony types described in this monograph are in the Museum of Comparative Zoology at Cambridge, Massachusetts. Reeve’s own types are in the British Museum and these were examined by Dr. Bryant Walker and compared with material taken by Mr. Smith. In 1862 Brot issued a “Catalogue Systematique des Especes qui Com- posent la Famille des Melaniens.” He spoke about the genus Gyrotoma being little known in Europe and of the sparse distribution of examples among the collections. He indicated a suspicion that these shells might be variations or mutations of “certain species of the group of M. Haysiana, which live in the same locality.” A very just criticism was made of authors who pictured their species in the conventional manner and failed thereby to show the “characteristic notch.” Lacking a sufficient number of speci- mens and illustrations more decisive Brot declined to pass upon the valid- ity of the genus. It is this catalogue I have mentioned later in the text as the Brot list. In 1869, Dr. James Lewis (4) made a list of the Gyrotomae as part of a catalogue of mollusks of the Coosa River. Tryon’s (10) study of the genus appeared in 1873. For half a century it has been a guide for the collector of Pleuroceridae. It erred in the synonymy. The classification was artificial. ‘Too many species were recognized as valid. But when it is remembered that thirteen species of Gyrotoma had been erected from just thirteen specimens and that comparatively few additional shells were available for Tryon’s study, that Shuttleworth’s species were unknown in this country and Mighels’ types had been destroyed by fire, any sharp criticism of Tryon’s work with the genus would be unfair. The next writing upon the subject was by Mr. A. A. Hinkley, whose paper (2) appeared in 1904. He had collected the year before in the THE GENUS GyROTOMA 5 Coosa at Fort William Shoals, Wilsonville and Wetumpka. He distributed a great number of specimens, many of which I have had the opportunity to examine. Hannibal (2, p. 179) recognized Gyrotoma as a genus and set down Goniobasis as a subgenus. Under the name Gyrotoma olivula excisa (Lea) he threw together all the forms that are mentioned in Tryon as being deep- fissured, eighteen species in all. ‘Twenty-two, described as having short or shallow fissures, suffered telescoping under the designation Gyrotoma laeta incisa (Lea). Mr. Hannibal was bold, but not without a few misgivings. Preparatory to a prodigious scrambling of the Pleuroceridae he wrote, “It is not anticipated that this treatment will meet with general approval in every instance.” The Alabama Museum collection of Gyrotomae is, of course, unparalleled. It contains thousands of specimens carefully labelled and so painstakingly cleaned that color and sculpture can be made out quickly and with certainty. Notes by Mr. Smith upon his labels and observations he made in his cor- respondence with Dr. Walker have been invaluable in this study. THE ENVIRONMENT The genus is confined to the Coosa River. It has not been found even within the mouths of creeks flowing into the river. The nature of the habitat supplies the explanation. Mr. Smith repeatedly wrote of Gyrotomae as living in rapid heavy currents. Speaking of the genus at Wetumpka he said that it was “almost confined to swift water, and those specimens in pools do not look healthy and are apt to be deformed.” A permanent habitat of this kind is rarer among Pleuroceridae than is generally supposed. Io and Anculosa do occupy the swifter parts of rivers, the old as well as the young, but-most species of Pleurocera are to be found in quiet water. Pleurocera unciale (Hald.), a characteristic east Tennessee species, is in the main part of streams only while partly grown. The adults seek the pools and banks. I have noticed the same thing of Goniobasis livescens (Menke) in Ohio, Michigan and Indiana. ‘Colonies of livescens are found upon wave-beaten rocks in Lake Erie, but far larger colonies inhabit places much more protected. The ordinary bulbous form of livescens lives plentifully in rapids of the upper Wabash River. A depauperate form of the species was collected by Hinkley under stones in the lower part of the river. Such a site, corresponding to that in which Gyrotoma flourishes, was an unhealthy one for a species probably the most adaptive of all Goniobases. DISTRIBUTION Gyrotoma first appears in the Coosa at Lock 2, St. Clair County. The only species there and about as far down as the mouth of Yellowleaf Creek of Shelby County is pyramidatum. In the short distance between this point and the foot of Peckerwood Shoals, Talladega County, occur five species which are confined to these reaches, namely, amplum, cariniferum, spillmanii, lewisii and hendersonit. The second and the last of these five are known from Fort William Shoals only. Alabamensis is first met with sparingly 6 CALVIN GOODRICH on Peckerwood Shoals and disappears below Duncan’s Riffle, Chilton Coun- ty. Walkerit has a somewhat shorter range, Weduska Shoals, Shelby Coun- ty, to Butting Ram Shoals, Coosa County. Five species occur at Wetumpka. Their up-river range from this point is: Exctsum, to Three-Island Shoals, Talladega County ; Jacimatum, to Fort William Shoals ; incitsum and pumilum to Weduska Shoals; pagoda, to The Bar, Chilton County. The localities for Gyrotoma mentioned in this paper are here set down in their order, up-stream to down-stream, the county designations being given as upon Mr. Smith’s labels, though the river in most of these places is the border between counties: Lock 2, St. Clair County. Ten-Island Shoals, St. Clair County. Lock 4, St. Clair County. Riverside, St. Clair County. Truss Shoals, St. Clair County. Clarence Shoals, St. Clair County. Hall’s Island, Talladega County. Yellowleaf Creek (upper), Shelby County. Three-Island Shoals, Talladega County. Fort William Shoals, Talladega County. Peckerwood Shoals, Talladega County. Weduska Shoals, Shelby County. Cedar Island, Chilton County. Adam’s Ferry, Chilton County. Yellowleaf Creek (lower), Chilton County. Butting Ram Shoals, Coosa County. . Higgin’s Ferry, Chilton County. Wetumpka, Elmore County. RELATIONSHIPS AND GENERAL CHARACTERS The affinities of Gyrotoma are with certain Goniobases which should be separated from that genus. These mollusks are characterized by large shells and large opercula. They have the same wide aperture of Gyrotoma and the same microscopic sculpture. The group has not been carefully studied, but these species unquestionably belong to it: Goniobasis wmpressa Lea, laeta Jay, showalterit Lea (1860), lewisti Lea, bellula Lea and ovalis Lea. Probably others are gratiosa Lea and lachryma Anthony, which Tryon (10, pp. 342, 343) assigned to Eurycaelon. Occasionally all these species develop incipient fissures. Mr. Smith collected several specimens with fis- sures nearly as large as in pyramidatum and incisum and yet, in other re- gards, retaining their usual Goniobasic features. Every one has a deposit of callus at the top of the columella. This is found in typical Gyrotoma. One specimen of Goniobasis showalterit, in addition to having a very small fissure, has a girdle for about 5 mm. upon the body whorl. After a little study it becomes apparent that Gyrotoma is not a compact genus whose members can be traced with certainty to a common origin. It differentiates into five natural groups. Pyramidatum and spillmaniu form one of these groups. A second consists of pagoda, pumilum, alabamensis and cariniferum. These two groups would appear to have developed from Goniobasis laeta or some other species very like it. Lewisit and hender- THE GENUS GYROTOMA 7 soni constitute a third group whose relations with Goniobasis impressa are quite plain. Incisum, walkeri and amplum, though not always easily to be distinguished from one another, manifestly differ from the other species of Gyrotoma. The fifth group is made up of excisum and lacimatum. Their well-fixed characters and their long range in the river justify the belief that they are the oldest species of the genus. [ have been unable to find in Goniobasis any species definitely ancestral to the excisum and incisum groups. The specific value of some of the races of Gyrotoma, I recognize, is open to question. It requires considerable familiarity with them, and perhaps the habit of mind which goes with studying a family of extreme complexity, to be altogether confident as to the identifications of certain forms, icisum and amplum for example. Characters overlap. Characters sometimes prominent in one locality fade nearly to invisibility in another. It might possibly be doubted whether variations in the depth of the fissure, being largely a matter of size, warrant recognition beyond subspecific rank. But difficulties arise in attempting a wholesale lumping of the Gyroto- mae. One would have to ignore distinct differences in many of the young and these differences point to ancestral distinctions. The derivation of one group is clearly not the derivation of two other groups. A fourth and a fifth group cannot be traced to any known form. Lumping would consolidate forms independently developed, adding confusion instead of achieving sim- plicity. A possibly important question in connection with Gyrotoma is whether it is a genus in course of developing or disappearing. J. C. Willis (11, p. 166) recites three explanations for endemism which he chooses to consider tivals: “(1) That endemics are very specialized species (and genera) suited only to the areas upon which they have been found; (2) that they are old species (and genera) which have been driven into quiet nooks or left in odd ‘corners by the competition of better adapted species; and (3) * * * that in general they are young beginners, descended from the ‘wides’.”’ Willis’ argument is largely devoted to the advocacy of the third explanation as the one for the majority of cases. I see no reason why all three hypotheses may not apply with almost equal force. The Vitrinizonites of the southern Ap- palachians may be examples of highly specialized forms narrowly confined to particular Iccalities. An illustration of one form of a relict race is given by Ortmann (7, p. 3-6). Margaritana margariifera (L.) occurs in the headwaters of the Schuylkill River, and in no other place in Pennsylvania. The shells are confined to cold trout streams and in the parts of these thet are So to 1200 feet above sea level. ‘The Pennsylvanian area of this species is not only the most southern extensiou of its range in eastern North A-nerica, but it also has the peculiarity of being the only one to the south of the Terminal Moraine. ‘Thus it may be regarded as a part of the Glacial Preserve (refugium) of this species.” The great numbers in which Gyrctomae occur in the Coosa River, their variability which is so character- istic of a genus that has not yet become “fixed” and their failure, for all their numbers and apparent vigor, to get beyond a restricted habitat, convince 9°99 me that they represent “young beginners descended from the ‘wides’. 8 CALVIN GOODRICH The observations of skilled zoologists with a great deal of field ex- perience have been warnings against giving hastily the emphasis of specific names to the thirteen species herein recognized. Ortmann (6, p. 512) says, “Tsolation is, in my opinion, a necessary factor in the differentiation of spe- cies and I do not think that a case ever will be discovered where two closely allied species possess precisely the same distribution.” Jordan’s law (3, p. 547) lays it down that, “Given any species in any region the nearest related species is not likely to be found in the same region nor in a remote region, but in a neighboring district separated from the first by a barrier of some sort.” More briefly Ruthven (9, p. 192) declares that “di- rectly related forms on any line of descent occupy neighboring environ- ments.” Several of the closely allied species of Gyrotoma do appear to have ex- actly the same distribution to all intents and purposes. Unfortunately I have not all the details of Mr. Smith’s collecting experiences. But the im- pression is obtained from his letters and from the notes accompanying the labels that a given reef of a series of shoals is not occupied by a heterogeneous assemblage of mollusks, but by some particular form of the genus, each such colony being cut off from other colonies of the genus by deep water, quiet water, rolling gravel, by mud or by rocks that in dry seasons are exposed to the air. Thus within a small area, too small to be especially observed by the collector not upon the lookout for it, there may be a distinct differen- tiation of locality, small barriers with as powerful an influence upon Gyro- tomae as barriers quite plain to the eye—all the effects of isolation. THE OPERCULA The opercula of all Gyrotomae are large, leathery, dark red to dark brown. The growth lines are crowded. The surface in instances is more or less marked with microscopic wrinkles which tend to form irregular lines or bands at right angles to the growth lines. The nucleus is indented. Opercula wear down quickly and become irregular in shape so that in the cases of adults the shape is frequently quite different from that in the young. In old specimens the nucleus or area of the inner whorls is often completely eroded away. In this study the opercula were first examined in connection with the shells. Later they were more carefully re-examined and compared with one another, species by species. There are recognizable differences, but I cannot feel that they are of great importance. The opercula of pyramidatum, spillmanit, lewisit and hendersoni are quite broad and with the inner whorls well within the operculum. Pagoda has a narrower, darker operculum, the apex more blunt. In pumilum, alabamensis and carmiferum the opercula are mostly of the pagoda form, but have become slightly broader and no- ticeably larger. The opercula of exciswm and laciniatum are practically identical, being smaller than in pyramidatum, darker, thinner and broader than in pagoda. The inner whorls are tightly coiled rather than loosely. The opercula of incisum are thin, ragged, variable in shape. In walkeri and amplum the opercula are much the same as those of incisum, but in amplum THE GENUS GYROTOMA 9 they appear to run decidedly larger and this is not to be explained by the size of the aperture. The nucleus of all except pyramidatum is quite close to the left margin and near the base. A few opercula have been illustrated, but without any intention of ac- centing differences. Many of the differences are possibly due to variations in the erosion to which this feature is subjected in strong currents. The opercula of quite young Gyrotomae are very nearly uniform. GENUS GYROTOMA SHUTTLEWORTH Schizostoma Lea, (non Bronn, 1835), Proc. Amer. Phil. Soc., II, 1843, p. 242; Trans. Amer. Phil. Soc., IX, 1844, p. 26; X, 1848. pp.. 67, 03; I. n: s., 1852, p. 205; Proc. Acad. Nat. Sci., Phila., XII, 1860, p. 186; Journ. Acad. Nat. Sci., Phila., V, n. s., 1862, pp. 218, 243, 245. 246; VI, 1866, p. 152; Obs. Gen. Unio, IV, 1848, pp. 26, 67, 93; V, 1852, p. 51; LX, 1863, pp. 40, 65, 67, 68; XI, 1867, p. 108; Tryon, Amer. Journ. Con- chol., I, 1865. pp. 114-116, 131; II, 1866, p. 134; Lewis, Amer. Journ. Conchol., V, 1869- 70, p. 168; Tryon, Monog. Strepom., 1873, pp. XxXili-xxXVii, XXXili, XXXV, Xxxix, xlvi, 351; Hinkley, Nautilus, XVIII, 19004, p. 40. 41. Gyrotoma Shuttleworth, Mittheil. der Nat.-forsch. Gesellsch. in Bern., July 22, 1845, p. 88; H. & A. Adams, Genera, I, 1854, p. 305; Gray, Guide td Mollusca, I, 1857, p. 103; Chenu, Man. de Conchyl., I, 1850, p. 293; Anthony, Proc. Acad. Nat. Sci., Phila., XII, 1860, p. 53; Binney, Check List, 1860; Brot, List, 1862, p. 27; Tryon, Monog. Strepom., 1873, pp. xv, xviii, xxiii, xxiv-xxvii; Paetel, Cat., 1888, p. 361; Pilsbry, Proc. Acad. Nat. Sci. Phila., 1806, p. 497; Hannibal, Proc. Malacol. Soc., X, 1912, p. 179. Schizochilus Lea, (non Gray, 1850), Trans. Amer. Phil. Soc., X, n. s., 1852, p. 295; Obs. Gen. Unio, V, 1852, p. 51; IX, 1863, p. 67; Proc. Acad. Nat. Sci., Phila., XII, 1860, p. 186; Journ. Acad. Nat. Sci. Phila., V, n. s., 1862, p. 245; Tryon, Monog. Strepom., 1873, pp. xxiv, xxxiii. Melatoma Anthony, Gray, Proc. Zool. Soc. of London, 1847, p. 153; Woodward, Manual, 1851, p. 131; Reeve, Conch. Icon., 1860. Apella Mighels, MS., 1860. “Shell turreted ; columella incurved, thickened above by an intermarginal callus ; aperture oval, somewhat spread out (effuse) at the base; lip simple, sharp, with a narrow fissure, deeply notched. “Animal. —————; operculum corneous, spiral.” —Shuttleworth. Group oF Gyrotoma pyramidatum Rather heavy, pyramidal especially in the young, sometimes nodose, smooth to striate; fissure shallow, oblique, not exceeding 8 mm. in depth; girdle not always conspicuous; aperture ovate; sinus not pronounced ; bands usually four and of equal width. Gyrotoma pyramidatum Shuttleworth Pi iefigsi i 2and.3 Gyrotoma pyramidata Shuttleworth, Mittheilungen der Naturforschenden Gesell- shaft in Bern, No. 50, July 22, 1845, p. 88; H. & A. Adams, Genera, I, 1858, p. 305; Binney, Check List, 1860, No. 329; Brot, List, 1862, p. 27; Paetel, Cat., 1888, p. 361. Gyrotoma conica Shutt. (?), Troschel, Gebiss der Schnecken, I, 1857, p. 111, pl. 8, fig. 12; Brot, List, 1862, p. 27; Tryon, Monog. Strepom., 1873, p. 380. 10 CALVIN GoopDRICH Gyrotoma laeta incisa (lea), Hannibal, Proc. Malacol. Soc., X, 1912, p. 182. Schizostoma pyramidatum Shutt., Tryon, Monog. Strepom., 1873, pp. xv, 365; Lewis, Fauna of Ala., 1876, p. 23. Description: Shell pyramidal, heavy, olivaceous or blackish, concentrically sul- cate-costate, more often with rough knobs (nodules); banded with brown, apex eroded; whorls 5 or 6; fissure short, broad, columella thickened tuberculously above. Length, .9; breadth of the ultimate whorl, .412-.5 of an inch. Length of aper- ture, .312; length of fissure, .1 of an inch. Obs.—Somewhat resembles Melania canaliculata; varies as the preceding (G. ovoidea). A few specimens . . . occur among others, with the upper whorls acutely carinated, ridges and furrows fewer and almost obliterated; epidermis paler and apex little eroded.— Shuttleworth. This is the first Gyrotoma appearing in the river. Mr. Smith’s highest station for it was Lock 2, which is at the head of Ten-Island Shoals, between St. Clair and Etowah counties. Shuttleworth in his preface states that Rugel collected in the Coosa River, near Wetumpka. But there is sound reason for believing that this is an error. The species was never taken by persons who collected from Wetumpka as far up the stream as Fort William Shoals. The nodulous character of the shells sent to Europe is a peculiarity almost entirely confined to Pleuroceridae upon and near Ten- Island Shoals. Not only are the Gyrotomae of this locality so marked and the so-called Eurycaelon, but also Anculosae and Goniobases. In a letter to Dr. Walker, Mr. Smith said that even certain Pleurocerae are likewise affected. ‘The oddity is not in mollusks collected farther up the river. It tends to disappear below these shallows. The fissure? is shallow, oblique, and is angled rather than curved or sinuous as in imcisum, also having a short fissure. At Lock 2, the fissure has a maximum depth of 3 1-2 mm. At Hall’s Island, it is as deep as 5 mm. and is not so oblique as in material from the higher locality. Oblique- ness is more pronounced in young specimens than in old; in juveniles the fissure is a mere notch. The hem is only faintly marked, being of much the same color as the rest of the shell; it is to be observed in hardly more than half the shells. The pyramidal aspect is that mostly of individuals one-half to three- fourths grown. The base is usually rounded in adults. Knobs, when pres- ent, give the shell a squared, four or five-sided appearance, and are not spaced regularly. Color? ranges from honey-yellow to raw-umber. ‘The prevailing banding formula is four bands of equal width. The aperture is ovate. The white or bluish-white columella is smoothly rounded, slightly *In the literature of Pleurotomariidae the fissure is spoken of as the “anal sinus” and the girdle or hem as the “sinus band”. I have preferred to use the terms of the early writers upon Gyrotoma even if the others be more descriptive. Doubtless as in the marine mollusk the fissure of Gyrotoma serves a purpose for the discharge of excreta. *The colors assigned in this study are according to “Color Standards and Color Nomenclature” of Ridgway, 1912. While a fairly large number of names have been civen in order to be as exact as possible it will be remarked that the color rangé in this genus is rather limited. Tue GENUS GYROTOMA II produced at the base and has a node at the fissure. It attains full develop- ment earlier in this species than in any other of the genus. The outer lip is sinuous, The largest pyramidatum measured, which probably would have eight whorls if uneroded, is 21 x II 1-2 mm. A young specimen from Lock.2, having five whorls, is conic, carinated at the periphery of the whorls, sharply angulate of base. The whorls are flat and without folds. The columella is undeveloped. ‘There is scarcely more than an indication of fissure. Bands are three in number. Mr. Smith’s lowest locality for pyramidatuwm is Hall’s Island, Talladega County. There are two specimens in the Schowalter Collection which I judge to be this species. The locality is mouth of Yellowleaf Creek (Shelby County). This is a few miles below Hall’s Island. Dr. Walker compared specimens taken by Mr. Smith with types of pyramidatum and conica in Geneva and found them to be the same. Conica does not appear to have been formally described. Dr. Walker, who came upon material in the Geneva Museum so labeled, suggests that the re- marks at the end of the description of pyramidatwm might have been in- tended to refer to this form. Gyrotoma spillmanii (Lea) Pl. I, figs. 4 and 5 Schizostoma Spillmanti ea, Proc. Acad. Nat. Sci., Phila., XIII, 1861, p. 54; Journ. Acad, Nat. Sci. Phila, V, n. s., 1862, p. 250, pl. 35, fig. 55; Obs. Gen. Unio, IX, p. 72, pl. 35, fig. 55. Schizostoma babylonicum Lea, Tryon, Monog. Strepom., 1873, p. 360. Gyrotoma Spillmanii Lea, Paetel, Cat., 1888, p. 362. Gyrotoma laeta incisa, (Lea), Hannibal, Proc. Malacol. Soc., X, 1912, p.- 182. Schizostoma Wheatleyi Lea, Proc. Acad. Nat. Sci., Phila., XX, 1868, p. 153; Journ. “cad. Nat. Sci. Phila., VI, 1868, p. 342, pl. 54, fig. 27; Obs. Gen. Unio, XII, 1868, p. 342, pl. 54, fig. 27; Lewis, Amer. Journ. Conchol., V, 1869-70, p. 169; Lewis, Fauna of Ala., 1876, p. 23. Description: Shell striate, subcylindrical, rather thick, yellowish-brown, imper- forate; spire obtuse, conoidal; sutures impressed; whorls six, very much banded, flat- tened, the last large; fissure oblique and rather short; aperture large, ovate and banded within, obtusely angular at the base; columella white, thickened above; outer lip sharp and sinuous. Operculum ovate, spiral, rather large, dark brown with the polar point near to the left edge, about one-fifth above the basal margin. Habitat: Coosa River, Alabama; Dr. F. R. Showalter. Diameter, .48; length, .92 of an inch.—Lea. The fissure of spillmanii is like that of pyramidatum from Hall’s Island. The young, though larger, whorl for whofl, are hardly to be separated from juvenile pyramidatum. ‘The types carry the resemblance because they have unusually shallow fissures for adult spillmani. But this species in general is a larger, coarser race, more boldly sculptured, pyramidal only in the adoles- cent stage and so stepped at the suture as to suggest to Tryon that it was the same as babylonicum Lea. I think it can stand as a species, related closely to pyramidatum or springing at no great distance in time from the same ancestral form. 12 CALVIN GOODRICH The fissure is 3 to 8 mm. in depth, always oblique. Of 20 specimens in one lot, 11 have a wide fissure, in 9 it is narrow. Most of these latter in- dividuals show signs of injury or deformation at the top of the whorl. The girdle is cord-like, irregular, more or less conspicuous by reason of differing in color from the body of the shell. ; Spillmanti is heavy, nodulous, though sometimes smooth, frequently constricted, slightly plicate upon the early whorls. Fourteen shells in the lot of 20 have stout, revolving folds, 5 have folds nearly obsolete and 1 is smooth. Growth lines are fine to coarse, the scars of rest periods being quite prominent. An adult with three remaining whorls measures 25 x 13 1-2 mm, Color varies from sulphine-yellow to buffy-citrine. Bands are four, all of the same width. This is the prevailing formula. Variations are modifications of this formula. The aperture is wide, ovate, the sinus at the base scarcely noticeable. The columella is white, thick, broadly rounded and has a heavy callous at the top. The outer lip is slightly sinuous, A young shell of five whorls is pyramidal, obscurely plicate, subangulate at the base. The columella is small, undeveloped ; the fissure a mere curving indentation. A full grown shell would have eight whorls probably if un- eroded. The more nearly typical shells examined in the Alabama collections were taken by Dr. Schowalter and the locality given is “Coosa River, Shelby County, Ala.” A smooth form, less robust and with a shallow fissure cor- responding with the types, was taken by Mr. Smith in Peckerwood Shoals. No other localities are known. The type of wheatleyi, save in point of size, looks very much like spill- mani in form, sculpture, bands and fissure. This seems to be the only shell of the kind that was collected by Schowalter. Mr. Smith found two small shells which I feel convinced are dwarfed spillmanii, but curiously they differ greatly from Lea’s wheatleyt. In his formal description, Lea mentions Dr. Schowalter as the collector from whom he received spillmanii. Later, under “observations,” he says, “I have a number of specimens, chiefly young, from Dr. Spillman and a fine suite of different ages from Dr. Schowalter.” All these shells are in the United States National Museum. Certain ones, with a label (1) pasted upon them, I have ventured to assume to be the types. They correspond to the Schowalter shells in the Alabama Museum collection. With them are two specimens of G. pagoda Lea and a third Gyrotoma broken and indeterminate. These probably are the sendings from Dr. Spillman. Three other lots under the name of Spillmanii agree with the Schowalter mollusks and a fifth consists again of pagoda. Group oF Gyrotoma pagoda Conical to globose, smooth to striate, the young carinate and more or less obscurely plicate; fissure oblique to straight, not exceeding 10 mm. in depth; girdle of varying size and color; aperture more frequently elliptical than ovate; sinus usually distinct; bands, ordinarily four and equidistant. THE GENUS GYROTOMA 13 Gyrotoma pagoda (Lea) Pipi hess 6 7. S$ and 9 Schizostoma pagoda Lea, Proc. Amer. Phil. Soc., IV, 1845, p. 167; Trans. Amer. Phil. Soc., X, 1848, p. 93, pl. ix, fig. 52; Obs. Gen. Unio, IV, 1848, p. 93, pl. ix, fig. 52; Journ. Acad. Nat. Sci. Phila. V, n. s., 1862, p. 247; Obs. Gen. Unio, IX, 1863, p. 60; Lewis, Amer. Journ. Conchol., V, 1869-70, p. 169; Tryon, Monog. Strepom., 1873, pp. xv, 365; Lewis, Fauna of Ala., 1876, p. 23. Gyrotoma pagoda Lea, Chenu, Manuel, I, 1859, fig. 2,020; H. & A. Adams, Genera, I, 1858, p. 305; Binney, Check List, 1860, No. 327; Brot, List, 1862, p. 27; Paetel, Cat., 1888, p. 362. Melatoma pagoda Lea, Reeve, Monog. Melatoma, 1860, sp. 1. Schizostoma Wetumpkaensis Lea, Proc. Acad. Nat. Sci., Phila., 1860, p. 187; Journ. Acad. Nat. Sci., Phila., V, n. s., 1862, p. 251, pl. 35, fig. 56; Obs. Gen. Unio, IX, 1863, p. 73, pl. xxxv, fig. 56; Lewis, Amer. Journ. Conchol., V, 1869-70, p. 169; Tryon, Monog. Strepom., 1873, p. 366; Lewis, Fauna of Ala., 1876, p. 23. Melatoma Wetumpkaense Lea, Reeve, Monog. Melatoma, 1861, sp. 17. Melatoma ornata Anth., MSS., Reeve, Monog. Melatoma, 1861, sp. 11. Gyrotoma ornata Anth., Paetel, Cat., 1888, p. 362. Melatoma Anthonyi Reeve, Monog. Melatoma, 1861, sp. 12. Gyrotoma Anthonyi Reeve, Brot, List, 1862, p. 27; Paetel, Cat., 1888, p. 361. Schizostoma Anthonyi Reeve, Lewis, Amer. Journ. Conchol., V, 1869-70, p. 168; Tryon, Monog. Strepom., 1873, p. 368; Lewis, Fauna of Ala., 1876, p. 22. Melatoma funiculatum Lea, Reeve, Monog. Melatoma, 1860, sp. 5. Schizostoma pyramidatum Shutt., Hinkley, Nautilus, X’VIII, 1904, p. 41. Gyrotoma lacta incisa (Lea), Hannibal, Proc. Malacol. Soc., X, 1912, p. 182. Description: Shell carinate, conical, rather thick, dark horn-color; spire rather short; sutures very much impressed; whorls six; fissure small; aperture elliptical, within whitish; columella smooth. Habiiat: Tuscaloosa, Alabama. Diameter, .35; length, .75 of an inch—Lea. The pagoda-like form of this species is especially striking in the juve- nile specimens and to such an extent, indeed, that these shells resemble Goniobasis catenaria Say. The carinae are very wide and turned upward at the edges. The suture is deeply channelled, the fissure a mere indentation. Embryo shells are about 1 1-4 to 2 whorls in size, unsculptured. The first two carinated whorls are scalloped by strong and regularly spaced plicae. As the animal grows larger, the carinae tend to disappear so that in many instances the shell is ventricose and smooth. The Wetumpka form is char- acterized by revolving folds, these disappearing in up-river specimens. M1i- croscopical sculpture typically gyrotomoid. Color varies from Ridgway’s “old gold” to warm sepia. The proportion of banded to unbanded shells is about five to four. The prevailing banding system is four equidistant bands of the same width. The columella is rather narrow, rounded, with a slight nodulous thickening above. The outer lip is sharp-edged, slightly sinuous. A Wetumpka specimen measured 20 x 12 I-2 mm., the largest from up-river shoals, 22 I-2 x 14 mm. The fissure is short, straight and bluntly rounded at the opening in old shells. It is seldom more than 3 mm. in depth. Compared with the fissure in pumilum that of pagoda is shorter and more flaring; apparently it is always straight. 14 CALVIN GOODRICH The known range for the species is from The Bar, Chilton County, to Wetumpka, Elmore County. Lea speaks of wetumpkaensis as umbilicate. This is merely an effect of erosion. Sometimes the animal leaves a chink close to the border of the colu- mella and the opening is widened irregularly by corrosion. Three lots of wetumpkaensis in the Lea collection agree very well with pagoda, though less carinated and having color bands. Ornata is a young shell. The locality given, North Carolina, is apparently, as Tryon points out, a confusion with Anculosa ornata Anth. from that state. Dr. Walker examined the type of anthonyi in the British Museum and found it to be identical with pagoda. Lewis (5, p. 23) suggested the possibility of wetumpkaensis being a variety or the young of pagoda. Gyrotoma pumilum (Lea) Pl. I, figs. 10 and I1 Schizostoma pumila Lea, Proc. Acad. Nat. Sci., Phila., XII, 1860, p. 187; Journ. Acad. Nat. Sci., V, n. s., 1862, p. 252, pl. 35, fig. 57; Obs. Gen. Unio, IX, 1863, p. 74, pl. xxxv, fig. 57; Lewis, Amer. Journ. Conchol., V, 1869-70, p. 169; Tryon, Monog. Strepom., 1873, p. 357; Lewis, Fauna of Ala., 1876, p. 23. Gyrotoma pumila Lea, Binney, Check List, 1860, No. 328; Brot, List, 1862, p. 27. Gyrotoma olivula excisa (Lea), Hannibal, Proc. Malacol. Soc., X, 1912, p. 182. Melatoma Alabamense Lea, Reeve, Monog. Melatoma, 1861, sp. 20. Schizostoma globosa Vea, Proc. Acad. Nat. Sci., Phila., XII, 1860, p. 186; Journ. Acad. Nat. Sci., Phila., V, n. s., 1862, p. 252, pl. 35, fig. 58; Obs. Gen.. Unio, IX, 1863, p. 74, pl. xxxv, fig. 58; Lewis, Amer. Journ. Conchol., V, 1860-70, p. 168. Schizostoma excisum Lea, Hinckley, Nautilus, XVIII, 1904, p. 41. Gyrotoma globosa Lea, Binney, Check List, 1860, No. 321; Brot, List, 1862, p. 27; Paetel, Cat., 1888, p. 362. Melatoma globosum Lea, Reeve, Monog. Melatoma, 1861, sp. 18. Description: Shell striate, top-shaped; rather thin, pale horn-color, imperforate; spire very obtuse; sutures much impressed; whorls six, ventricose, the last very large; fissure straight and rather short; aperture rather small, ovate, white within, angular at the base and somewhat canaliculate; columella white, twisted and thickened below; outer lip acute and sinuous. Habitat: Alabama; B. W. Budd, M. D. Diameter, .40; length, .63 of an inch.—Lea. The fissure of pumilum is “rather short” only as compared with the character in such species as excisum. It may be considered as moderately deep for the genus. It is narrow, usually straight, rarely slightly oblique. In Wetumpka specimens examined, this slit has a maximum depth of 7 I-2 mm., in a specimen from Higgin’s Ferry, it is 9 1-4 mm. deep. The girdle is more marked in some specimens than in others, varying also in color and size. Cross sections showed the girdle to be hollow for about 3 or 4 mm, behind the fissure. Behind this hollow space it was filled with fine crystalline shell material. The adult is ovate, with whorls slightly convex, the base rounded; often very heavy. The spire in partly grown shells is obtuse, with faint plicae, often missing. The largest Wetumpka specimen measures 18% x 11% mm., the largest up-river shell, 22 x 12mm. ‘Ten or twelve revolving folds appear on the body whorl in Wetumpka lots, tending to disappear in colonies THE GENUS GYROTOMA : I5 farther up the Coosa. For example, at Butting Ram Shoals, half the pu- milum are smooth. Color varies from sulphine-yellow to dark olive. About one shell in four lacks color bands. The prevailing formula is four bands, the third from the top being smaller than the others. The aperture is ovate to elliptical, white to yellowish-white. In Wetumpka shells the folds show through as broad white bands in the aperture. The columella is rounded, not wide, white to bluish-white. It is not nearly as stout as might be expect- ed in a shell of this size. At the mouth of the fissure is a spot of white callus. The outer lip is curved outward at the top and is slightly crenulate in shells with folds. In juveniles the spire is smooth, conic, the whorls flat, the suture not much impressed, the periphery angular. The fissure begins with the fifth whorl. Embryo shells are of about 114 whorls, closely coiled and without sculpture. The species occurs from Weduska Shoals to Wetumpka. The Weduska Shoals specimens are a dwarf form, occurring in very small numbers. The largest is 19x 9 mm. The shell is delicate, shining, the upper whorls having low plicae, the base bearing obscure folds. The fissure in these specimens is straight to slightly oblique, 3 to 4 mm. in depth. The temptation has been to allow them a distinctive name, but the form appears to correspond with dwarf pumilum collected by Mr. Smith at Higgin’s Ferry, The Bar and Duncan’s Riffle. : A note upon one of Mr. Smith’s labels for this species reads, “On rocks, in very swift water, 1-3 feet deep.” Hinkley (2, p. 41) makes pumilum a synonym of excisum. An examina- tion of the types of the two species compels me to disagree with him. Gyrotoma alabamensis (Lea) PIM ites: 12 and 13 Schizostoma Alabamensis Lea, Proc. Acad. Nat. Sci., Phila., XII, 1860, p. 187; Journ. Acad. Nat. Sci., Phila., V, n. s., 1862, p. 250, pl. 35, fig. 54; Obs. Gen. Union, IX, 1863, p. 72, pl. xxxv, fig. 54; Lewis, Amer. Journ. Conchol., V, 1869-70, p. 168; Tryon, Monog. Strepom., 1873, p. 367; Lewis, Fauna of Ala., 1876, p. 22. Gyrotoma Alabamensis Lea, Binney, Check List, 1860, No. 305; Brot, List, 1862, p. 27; Paetel, Cat., 1888, p. 361. _Gyrotoma olivula excisa (Lea), Hannibal, Proc. Malacol. Soc oe Ones pa dere Melatoma babylonicum Lea, Reeve, Monog. Melatoma, 1860, sp. 6. Description: Shell striate, elliptical, stout, yellowish-olive, imperforate; spire obtusely conical; sutures very much impressed; whorls six, banded, rather inflated, the last very large; fissure oblique and rather short; aperture rather large, ovate, banded within and obtusely angular at the base; columella white, somewhat thickened above and below; outer lip sharp and sinuate. Habitat: Alabama; B. W. Budd, M.D., and Dr. E. R. Showalter. Diameter, .50; length, .9o of an inch.—Lea. The description, “fissure oblique and rather short” so misled Mr. Smith that when he found alabamensis he was compelled to believe that he had come upon a new species. Examination of the type disclosed, however, that the shells of ea and Smith were the same. ‘The fissure varies from straight to oblique. Of 37 shells from The Bar, Chilton County, 16 had 16 : CALVIN GooDRICH straight fissures. ‘The difference did not seem to be a matter of age. The depth of fissure is from 6 mm. in half-grown shells to 10 mm. in old. The girdle is yellow, rather wide, and is well-marked in nearly all specimens. The species is probably the most robust of the genus. The largest specimen in the Alabama collection measures 2642 x 13% mm. ‘The base is sub-angular up to three-fourths of the growth. In adults, it is rounded. The folds are low, not always regularly-spaced, and are stoutest in young shells. Occasional smooth shells have been taken. A distinct tendency to- ward plication is noticeable upon the upper whorls. The nearly miscroscopic sculpture consists of fine growth lines, crossed by discontinuous, wavy trans- verse lines. ‘The color is buckthorn-brown to deep olive. Two bands of equal width, at top and base respectively, and a third band, much wider, at the periphery constitute the commonest banding formula, In many speci- mens coloring matter is deposited in the base of the folds and this shows in the aperture as bands or lines. The aperture is elliptical rather than ovate. The columella is white, not very wide, thinning above the center into a bluish-white wash with a porcelainous node at the mouth of the fissure. The sinus is distinct. The outer lip is sharp-edged, crenulate, slightly sinuous. Young shells are ovate, sub-angulated, plicate upon the post-embryo whorls; the fissure straight, not wide; the columella narrow, yellowish- white; the sinus having a shallow channel. Seven or eight whorls are indi- cated for the adult specimen. The range of alabamensis is from Peckerwood Shoals, Talladega County, to Duncan’s Riffle, Chilton County. Peckerwood Shoals shells are wider in proportion to height than speci- mens from other known localities and they show, more than these others, the relationship of alabamensis and carimferum Anth. A varietal name does not seem to be warranted in the case of this race unless other varieties, and there are several, are also so baptized. In nearly all the lots are shells close to pumilum in general appearance. Pagoda may be considered a third or fourth cousin to the southward and cariniferum a first or second cousin to the north. Gyrotoma carimferum Anthony Pl. I, figs. 14 and 15 Gyrotoma carinifera Anth., Proc. Acad. Nat. Sci., Phila., XII, Feb. 1860, p. 66; Binney, Check List, 1860, No. 310; Brot, List, 1862, p. 27; Paetel, Cat., 1888, p. 361. Melatoma cariniferum Anth., Reeve, Monog. Melatoma, 1861, sp. 13. Schizostoma cariniferum Anth., Lewis, Amer. Journ. Conchol., V, 1869-70, p. 168; Tryon, Monog. Strepom., 1873, p. 353; Lewis, Fauna of Ala., 1876, p. 22. Schizochilus Showalterii Lea, Proc. Acad. Nat. Sci. Phila.. XII, March, 1860, p. 9 ios) Schizostoma Showalteriti Lea, Journ. Acad. Nat. Sci., Phila., V, n. s., 1862, p. 246, pl. 35, fig. 49; Obs. Gen. Unio, IX, 1863, p. 68, pl. xxxv, fig. 40. Gyrotoma Showalterii Lea, Binney, Check List, 1860, No. 334; Brot, List, 1862, p. 28. Gyrotoma olivula excisa (Lea), Hannibal, Proc. Malacol. Soc., X, 1912, p. 181. THE GENUS GYROTOMA 17 Description: Shell conic, thick, dark brown; ‘spire obtusely elevated, truncate, though not abruptly so, six whorls remaining, one or two having apparently been lost by truncation; carinations elevated, subacute and found on all the whorls, two on each of the spiral ones and three or four on the body whorl; fissure direct, broad and moderately deep, extending about one-fifth around the shell; sutures irregular, much modified by the carinae, and often concealed in part by them; aperture ovate and band- ed within; columella much rounded, callous at the lower part only; outer lip irreg- ularly waved, its outline modified by the carinae on the body-whorl; no sinus. Habitat: Coosa River, Alabama. Length of shell, 7-8; breadth of shell, 1-2 of an inch. Length of aperture, 514-16 of an inch; breadth of aperture, 1-4 of an inch.—Lea. The opportunity has been given to examine the type of cariniferum, which is in the Museum of Comparative Zoology. ‘The shell is about three- fourths grown. ‘The fissure is rather narrow as compared with the same character in several other species, is slanting and 814 mm. deep. In twenty- five other specimens the fissure of seven is straight, very slightly oblique in one and noticeably oblique in seventeen. The girdle is narrow and of much the same color as the body of the shell. Full grown specimens are not so stoutly ribbed as younger ones, and in instances the folds become almost obsolete. A few greatly resemble G. alabamensis Lea. The largest shell measured 24 x 15 mm. It should be said that in the cut published by Tryon (10, p. 353) the ribs are made to seem more twisted than they actually are in the type. ‘The color is brown- ish olive. The banding formula, common to the group, is four bands, the third from the top being smaller than the rest. Variations occur, but they are modifications of this formula. The aperture is more often elliptical than ovate. In old specimens the columella is white, wide, continuous and with the usual node of porcelain-like callus at the top. Young shells are angulated at the base. They indicate that an uneroded adult would have eight whorls. The only locality known for this species is Fort William Shoals. In 1914, shortly after a visit to this site, Mr. Smith wrote to Dr. Walker, “On a reef * * * J struck G. cariniferum; the only ones I had seen before were a broken specimen in the Schowalter collection and about ten I had collected last year in a lower part of Fort William Shoals. In this new place we gathered about 200 in three hours, and I think we scraped the place nearly clean. The species occurred only in about fifteen yards of the reef and always in the swiftest water where other Gyrotoma cannot stick. There was one little pool under a fall which must have yielded over 100 fine ones.” Hinkley (2, p. 41) mentions showalterti Lea. It is not clear that he re- fers to the species which has had to give way to cariniferwm by reason of the rule of priority orto the one given the same name by Lea in 1864. Group oF Gyrotoma lewisit Elongate to ventricose, the whorls almost flat, sometimes ligulate, the shell rather heavy and resembling Goniobasis impressa Lea in sculpture; fissure oblique, with a maximum depth of 6 mm.; girdle not pronounced ; aperture narrow; the sinus distinct; bands consisting usually of many fine lines. 18 ; CALVIN GooDRICH Gyrotoma lewisii (Lea) Pl. I, figs. 16, 17, and 18 Schizostoma Lewisti Lea, Proc. Acad. Nat. Sci., Phila.. XXI, 1860, p. 62; Journ. Acad, Nat. Sci. Phila., VIII, 1874, p. 61, pl. 21, fig. 16; Obs. Gen. Unio, XIII, 1874, p. 65, pl. xxi, fig. 16; Lewis, Amer. Journ, Conchol., V, 1869-70, p. 168; Lewis, Fauna of Ala., 1876, p. 23; Hinkley, Nautilus, XVIII, 1904, p. 41. Description: Shell closely striate, subcylindrical, rather thin, yellowish-brown, imperforate; spire conical, folded; sutures very much impressed; whorls about seven, the last one large; fissure oblique and short; aperture large, rhomboidal, banded within; outer lip crenulate and sinuous; columella white, thickened and twisted. Habitat: Coosa River, Alabama, Dr. E. R. Schowalter. Diameter, .5; length, .o inch.—Lea. The fissure is oblique, shallow, 234 to 334 mm. in depth. There is no distinct girdle. A broadening or-shouldering of the whorl immediately below the suture appears to serve the purpose of covering the fissure as growth proceeds. In young shells this shoulder does not occur though the mollusk has a well-developed fissure at this stage. Lewisii is stout, ventricose to elongate and is covered with fine, sharply cut folds or riblets, closely set and varying slightly in size. Growth lines are fine, being crossed by minute discontinuous revolving lines not always distinct beneath a hand lens. A few specimens are constricted upon the body whorl though none taken by Mr. Smith is so much constricted as in the figure of Lea’s type. Color usually citrine. All the ribs seem to con- tain more or less coloring matter. In young shells this shows through into the aperture, giving an effect of numerous bands. As shell material is deposited the finer bands or lines tend to disappear. The usually narrow and elliptical aperture has a distinct sinus at its base. ‘The columella is broad and, in immature specimens, merely a wash of white above the center of the aperture. ‘The node at the top is not prominent. One specimen has a reddish columella. The largest shell is 2214 x 12% mm. Half-grown specimens are conic, subangulate, slightly carinate and have a pronounced sinus. The fissure is very shallow, but well-marked. Mr. Smith’s shells, and presumably those taken by Dr. Schowalter, were collected on Fort William Shoals. Hinkley (2, p. 41) gives Wilsonville as another locality. This would seem to be on or within touch of Three- Island Shoals. Dr. Lea remarked upon the resemblance of his species’ to Goniobasis impressa Lea. Save in regard to the fissure of Jewisti, the two shells are exceedingly alike. The one, I believe, can be considered an offshoot of the other. Gyrotoma hendersoni H. H. Smith, new species Pi li ness 10) 2onand 21 Shell: Subcylindrical, rather heavy, whorls almost flat, the last broadly rounded at the base. Apex eroded, only two whorls remaining. The entire shell is covered with low folds of varying size which tend to wave or break where new shell material has been deposited after rest periods. Between the folds of the penultimate whorl the fine THE GENUS GyROTOMA 19 vertical growth lines are crossed by transverse lines, very delicate and usually dis- continuous. Fissure oblique, 6 mm. in depth. The suture is covered by a narrow, cord-like, irregular girdle somewhat lighter than the general body color. In a broken place in this “hem” the suture appears as deeply impressed or channelled. Color of shell brownish-olive. The bands are deposits of coloring matter in the base of the folds, sometimes coalescing so that for the eleven folds of the body whorl six bands show in the aperture. Columella narrow, rounded from center to base, flattened above the center and having a heavy callus at the fissure; white from center to base, dull purple above. Aperture elliptical, with a distinct sinus below. Outer lip a little crenulate, firm, broadly curving outward at the top and slightly inward at the base. Operculum: Leaf-like, leathery, very dark red. Apex acute, base rounded; right margin much more curved than the left. Whorls three, the first two tightly coiled and close to the left margin and about 1 mm. from the base. The region of the inner whorls is excavated or crater-like. Radiating lines are of varying strength. Altitude & 1-2 mm.; diameter 4 I-2 mm. Measurements: Altitude, 21 mm.; diameter, 12 mm. Aperture: Lowest part of fissure to base, 10 I-2 mm.; diameter, 4 I-4 mm. Type Locality: Coosa River, Fort William Shoals, Talladega County, Ala. Type in Museum of Zoology, University of Michigan; paratypes in Alabama Mu- seum of Natural History. The affinities of this species are with G. lewisii (Lea). It differs in hav- ing fewer and stouter folds, a more rounded body whorl, a deeper and less oblique fissure and less shouldered aspect. While the question may rise as to whether lewisii is Goniobasis impressa Lea with a gyrotomoid — outer lip the position of hendersoni as a true Gyrotoma is not to be ques- tioned. The fissure varies from 3'%4inm. in depth in young shells to 6 mm. in adults. As depth increases, the fissure becomes less oblique, or to put it another way, more narrow. There is little variation in form, sculpture and color. The bands are from five to ten, following no formula as in most other species of the genus. A young specimen of hendersoni is thin, suban- -gulate. The fissure is shallow, the suture rather indistinct, the bands many and linear. Eight whorls are indicated for a full-grown specimen. The only locality known for this species is Fort William Shoals. Writing from this place on July 17, 1914, Mr. Smith said: “On a reef just above we got the regularly striate form which I have been callng G. lewisii, though I am not at all sure. With it another which I take to be G. babylonicum, and a third form undescribed which I propose to name G. hendersoni, it is striate much like Jewisii, but has a square and moderately deep slit.” In September of the same year, Mr. Smith wrote: “The species which I propose to name hendersoni is one of the most beautiful of the genus and is apparently very constant.” Group oF Gyrotoma excisum Quadrate, bulbous, conic or elongate; the whorls smooth or with folds, often faintly plicate; fissure ordinarily straight and as deep as Io mm.; girdle well-marked, though sometimes absent; aperture elliptical; sinus distinct ; bands: three in excisum, four in lacinatum. 20 CALVIN GOODRICH Gyrotoma excisum (Lea) Pl. II, figs. 22 and 23 Melania excisa Lea, Proc Amer. Phil. Soc., II, 1843, p. 242; Trans. Amer. Phil. Soc., IX, 1844, pp. 25, 29; Trans. Amer. Phil. Soc., X, 1848, p. 93; Obs. Gen. Unio, IV, 1848, pp. 25, 20, 93; Jay, Cat., 4th ed., 1852, p. 273. Schizostoma excisa Lea, Wheatley, Cat. Shells U. S., 1845, p. 28; Lea, Journ. Acad. Nat. Sci. Phila., V, n. s., 1862, p. 250; Obs. Gen. Unio, IX, 1863, p. 72; Lewis, Amer. Journ. Conchol., V, 1869-70, p. 168; ‘Tryon, Monog. Strepom., 1873, pp. xv, xxiii, xxv, 356; Lewis, Fauna of Ala., 1876, p. 23. Gyrotoma excisa Lea, H. & A. Adams, Genera, I, 1858, p. 305; Binney, Check List, 1860, No. 317; Brot, List, 1862, p. 27; Paetel, Cat., 1888, p. 362. Gyrotoma olivula excisa (Lea), Hannibal, Proc. Malacol, Soc., X, 1912, p. 181. Melatoma excisum Lea, Reeve, Monog. Melatoma, 1860, sp. 2. Melatoma Showalterii Lea (prob. excisum), Reeve, Monog. Melatoma, 1861, sp. 23. Gyrotoma ovoideum Shutt., Mittheil, Bern. Nat. Gesell., No. 50, July 22, 1845, p. 88; H. & A. Adams, Genera, I, 1858, p. 305, pl. 32, fig. 4; Paetel, Cat., 1888, p. 362. Schizostoma ovoideum Shutt., Tryon, Monog. Strepom., 1873, pp. xv, 356; Lewis, Fauna of Ala., 1876, p. 23; Hinkley, Nautilus, XVIII, 1904, p. 41. Gyrotoma bulbose Anth., Proc. Acad. Nat. Sci. Phila., Feb. 1860, p. 65; Binney, Check List, 1860, No. 300; Brot, List, 1862, p. 27; Paetel, Cat., 1888, p. 361. Melatoma bulbosum Anth., Reeve, Monog. Melatoma, 1861, sp. 22. Schizostoma bulbosum Anth., Lewis, Amer. Jour. Conchol., 1869-70, p. 168; Tryon, Monog. Strepom., 1873, p. 361; Lewis, Fauna of Ala., 1876, p. 22. Gyrotoma ovalis Anth., Proc. Acad. Nat. Sci., Phila., Feb., 1860, p. 65; Binney, Check List, 1860, No. 325; Brot, List, 1862, p. 27; Paetel, Cat., 1888, p. 362. , Schizostoma glans Lea, Proc. Acad. Nat Sci. Phila., 1860, p. 186; Journ, Acad. Nat. Sci., Phila., V, n. s., 1862, p. 248, pl. 35, fig. 52; Obs. Gen. Unio, IX, 1863, pp. 70, 71, 74, pl. xxxv, fig. 52; Lewis, Amer. Journ. Conchol., V, 1869-70, p. 168; Tryon, Monog. Strepom., 1873, p. 363; Lewis, Fauna of Ala., 1876, p. 23; Hinkley, Nautilus, XVIII, 1904, p. 40. . Gyrotoma glans Lea, Binney, Check List, 1860, No. 320; Brot, List, 1862, p. 27. Melatoma ellipticum Anth., MSS., Reeve, Monog. Melatoma, 1861, sp. 21. Gyrotoma ellipticum Anth., Brot, List, 1862, p. 27; Paetel, Cat., 1888, p. 362. Schizostoma ellipticum Anth., Lewis, Amer. Journ. Conchol., V, 1869-70, p. 168; Tryon, Monog. Strepom., 1873, p. 359; Lewis, Fauna of Ala., 1876, p. 23; Hinkley, Nautilus, XVIII, 1904, p. 40. Description: Shell striate, subfusiform, rather thick, yellowish; spire ovately conical; sutures impressed; whorls flattened; aperture cut out above, small, elliptical, white. : Habitat: Alabama. Diameter, .40; length, .64 of an inch—Lea. As variation goes in Gyrotoma this species is remarkably wanting in variability. At Wetumpka it bears folds in common with many other Pleu- roceridae there and at Three-Island Shoals, the other extreme of its occur- rence, it is ordinarily lighter in color and smooth. Otherwise there is little difference between the members of the two colonies. This may argue that excisum is the oldest of the genus, time having brought its usual uniformity, and this hypothesis is supported by the fact that the species has the longest range of all Gyrotomae. The fissure is deep, up to 10 mm., in the largest adults, very narrow in the young and widening with age; ordinarily straight, but in a few shells Tue GENUS GYROTOMA 21 waving upon the lower edge. The girdle is thread-like to rounded and thick, and of slightly lighter color than the body of the shell. Forms are quadrate, bulbous, ovate, elongate, any one colony having specimens of these shapes. Where folds occur they are low, not very con- spicuous. Of 38 specimens in one lot taken at The Bar, 29 are obscurely plicate upon the upper whorls, 9 are without plicae. This ratio of sculp- tured shells to those lacking plicae is about the same wherever the species occurs. In general the shell is of a soft silky appearance. Color is aniline- yellow to mummy-brown, The commonest banding formula is three bands, the one at the periphery being wider than the others. The columella in the oldest specimens is heavy, broad, complete to the top and thickened at the fissure. Color is usually white, but at times yellowish-white, pink or purple. The sinus is distinct, pointed. The outer lip is slightly sinuous. Mr. Smith’s largest exciswm measured 23 X 12 mm. The embryo is smooth, tightly coiled. Shells developed to the fifth whorl are smooth with a slight carina at the periphery; the sinus being well marked; the fissure straight, tight and about 3 mm. in depth. Partly grown individuals are not so angular at the base as in most other species. The whorls indicated for an adult are seven or eight. Shuttleworth’s shells from the Coosa River were sent to him by Rugel and this collector took pyramidatum not far from the northernmost known locality for excisum. It seems probable that the shell to which Shuttle- worth gave the name ovoideum came from this point or very close to it. G. ellipticum Anth. in the Philadelphia Academy, received from Anthony himself, is the same as his bulbosum except that it is elongate. Bulbosum is excisum in its old age, ovalis in its graceful youth. ‘The tuberculation of glans, cited by Tryon as if a distinguishing character, occurs rarely and seems to be without significance. Gyrotoma laciniatum (Lea) Pl. II, figs. 24 and 25 Schizostoma laciniatum Lea, Proc. Amer. Phil. Soc., IV, 1845, p. 167; Trans. Amer. Phil. Soc., X, 1848, p. 95, pl. ix, fig. 57; Obs. Gen. Unio, IV, 1848, p. 95, pl. ix, fig. 57; Journ. Acad. Nat. Sci., Phila., V, n. s., 1862, p. 253; Obs. Gen. Unio, IX, 1863, p. 75; Lewis, Amer. Journ. Conchol., V, 1869-70, p. 168; Tryon, Monog. Strepom., 1873, pp. xv, 359; Lewis, Fauna of Ala., 1876, p. 23. Gyrotoma laciniata Lea, H. & A. Adams, Genera, I, 1858, p. 305; Binney, Check List, 1860, No. 324; Brot, List, 1862, p. 27; Paetel, Cat., 1888, p. 362. Schizostoma castanea Lea, Proc. Acad. Nat. Sci. Phila. XII, 1860, p. 186; Journ. Acad. Nat. Sci., Phila. -V, n. s., 1862, p. 247, pl. 35, fig. 50; Obs. Gen. Unio, TX, 1863, p. 69, pl. xxxv, fig. 50; Lewis, Amer. Journ. Conchol., V, 1869-70, p. 168; Tryon, Monog. Strepom., 1873, p. 355; Lewis, Fauna of Ala., 1876, p. 22; Hinkley, Nautilus, XVIII, 1904, p. 40. Gyrotoma castanea Lea, Binney, Check List, 1860, No. 311; Brot, List, 1862, p. 27; Paetel, Cat., 1888, p. 361. Gyrotoma olivula excisa (Lea), Hannibal, Proc. Malacol. Soc., X, 1912, p. 181. 22 CALVIN GOODRICH Description: Shell smooth, obtusely conical, rather thick, banded, yellowish horn-color; spire obtuse; sutures excavated; whorls convex; fissure deep; aperture elliptical, whitish within; columella smooth, thickened above. Habitat: ‘Tuscaloosa, Alabama. Diameter, .25; length, .45 of an inch.—Lea. This is the smallest species of the genus. ‘The largest shell examined measures 16 x 10 mm. ‘I‘he average size of fourteen specimens is 14.9 x 8.5 mm. The species resembles excisum Lea in sculpture, fissure and banding system. It occurs from Fort William Shoals, Talladega County, to Wetumpka, Elmore County. A note by Mr. Smith with lacimiatum from Fort William Shoals reads, “Found in a few places along the eastern or Talladega County side and quite close to the banks where the water is swift; on or clinging to the lower side of stones. In one place, partly under overhanging bushes, we collected over 100 within a few yards, but generally the species is rare.” In laciniatum the fissure is very deep in porportion to the shell; it is narrow and hugs the body whorl. Its depth in adults is 7 to 7% mm. The girdle is small and thin. Of 48 specimens the girdle is absent in 13. The shell is conic, subangulated, with whorls flat to slightly convex. Up- per whorls are smooth to carinate, showing a tendency to form a few ob- scure plicae. Commonly there are four or five low folds upon the base, seldom very distinct. Beneath the fissure are low vertical puckerings, made conspicuous by spots of coloring matter between them. Growth lines are fine, irregular, crossed on parts of the shell by revolving lines which in few instances go clear around the whorl. Color is buckthorn-brown to clove-brown. ‘The prevailing banding formula is four bands of equal width, the two in the center approximate. These latter bands sometimes coalesce into the formula ordinarily occurring in excisum. ‘The aperture is elliptical, nearly pointed at the top, more rounded at the base and having a slight sinus. The columella is small, regularly rounded, white and having a node at the fissure. In only one specimen observed was: the columella developed fully above the center. The outer lip is sinuous, more angular at the top than rounded. The embryo shell is very small, smooth, tightly coiled. Some juveniles are strongly carinate, others smooth. Irregular plicae occur. ‘There are three bands, that at the center dividing into two. In shells of five whorls the fissure is quite deep. The adult laciniatum has seven or eight whorls. A comparison of the types of laciniatum and castaneum proved them to be identical. Lea’s first specimens were apparently from Wetumpka. It was not until about fifteen years later that he saw the up-river forms. Group oF Gyrotoma incisum Cylindrical to globose; whorls shouldered, often constricted, sometimes nodulous; smooth to striate; fissure usually shallow and oblique though in amplum as deep as 9 mm. and in instances straight; girdle stout and ir- regular when present; aperture ovate; sinus more noticeable in young than in adult shells; bands three. THE GENUS GYROTOMA 23 Gyrotoma incisum (Lea) Biliehios. 20,027) 20. atid. 2o Anculosa incisa Lea, Proc. Amer. Phil. Soc., II, 1843, p. 243; Trans. Amer. Phil. Soc., IX, 1844, p. 28. Schizostoma incisa Lea, Wheatley, Cat. Shells U. S., 1845, p. 28; Lea, Trans. Amer. Phil. Soc., X, 1848, p. 93; Lea, Obs. Gen. Unio, IV, 1848, pp. 28, 93; Lea, Proc. Acad. Nat. Sci., Phila., XII, 1860, p. 186; Hanley, Conchol. Misc., Melania, 1854-58, pl. 5, figs. 44, 45; Lewis, Amer. Journ. Conchol., V, 1869-70, p. 168; ‘Tryon, Monog. Strepom., 1873, pp. xxiii, xxv, 378; Lewis, Fauna of Ala., 1876, p. 23; Hinkley, Nautilus, XVIII, 1904, pp. 40, 41. Melania incisa Lea, Jay, Cat., 4th ed., 1852, p. 274. Leptoxis incisa Lea, Haldeman, in Chenu’s Illustrations Conchyliog., 1843-53, liv. 73, p. 2, pl. 1, figs. 24-26; Binney, Check List, 1860, No. 363. Gyrotoma incisa Lea, H. & A. Adams, Genera, I, 1858, p. 305; Binney, Check List, 1860, No. 323; Brot, List, 1862, p. 27; Paetel, Cat., 1888, p. 362; Goodrich, Anculosae Ala. Drain., 1922, p. 56. Gyrotoma laeta incisa (Lea), Hannibal, Proc. Malacol. Soc., X, 1912, p. 182. Melatoma incisum Lea, Reeve, Monog. Melatoma, 1860, sp. 4. Schizostoma curta Mighels, Bost. Proc., I, Oct. 1844, p. 189; Tryon, Monog. Strepom., 1873, pp. xv, 363; Lewis, Fauna of Ala., 1876, p. 22. Gyrotoma curta Migh., H. & A. Adams, Genera, I, 1858, p. 305; Binney, Check List, 1860, No. 314. Gyrotoma curta Gould, Brot, List, 1862, p. 27; Paetel, Cat., 1888, p. 361. Gyrotoma olivula excisa (Lea), Hannibal, Proc. Malacol. Soc., X, 1912, p. 181. Schizostoma Buddii Lea, Proc. Amer. Phil. Soc., IV, 1845, p. 167; Trans. Amer. Phil. Soc., X, 1848, p. 94, pl. ix, fig. 53; Obs. Gen. Unio, IV, 1848, p. 94, pl. ix, fig. 53; Obs. Gen. Unio, IX, 1863, p. 74; Lewis, Amer. Journ. Conchol., V, 1869-70, p. 168; Tryon, Monog. Strepom., 1873, pp. xv, 370; Lewis, Fauna of Ala., 1876, p. 22. Gyrotoma Buddii Lea, H. & A. Adams, Genera, I, 1858, p. 305; Binney, Check List, 1860, No. 308; Brot, List, 1862, p. 27; Paetel, Cat., 1888, p. 36r. Schizostoma constrictum Lea, Proc. Amer. Phil. Soc., IV, 1845, Dp.) 167 > Trans: Amer. Phil. Soc., X, 1848, p. 94, pl. ix, fig. 55; Obs. Gen. Unio, IV, 1848, p. 94, pl. ix, fig. 55; Proc. Acad. Nat. Sci., Phila. XII, 1860, p. 186; Journ, Acad. Nat. Sci. Phila., V, n. s., 1862, p. 246; Obs. Gen. Unio, IX, 1863, p. 68; Journ. Acad. Nat. Sci., Phila., ‘VI, 1866, p. 149; Obs. Gen. Unio, XI, 1867, p. 105; Lewis, Amer. Journ. Conchol., V, 1869-70, p. 168; Tryon, Monog. Strepom., 1873, pp. xv, 373, 374; Lewis, Fauna of Ala., 1876, p. 22. Gyrotoma constricta Lea, H. & A, Adams, Genera, I, 1858, p. 305; Binney, Check List, 1860, No. 302; Brot, List, 1862, p. 27; Paetel, Cat., 1888, p. 36r. Schizostoma funiculatum Lea, Proc. Amer. Phil. Soc., IV, 1845, p. 167; Trans. Amer. Phil. Soc., X, 1848, p. 95, pl. ix, fig. 56; Obs. Gen. Unio, IV, 1848, p. 95, pl. ix, fig. 56; Tryon, Monog. Strepom., 1873, pp. xv, 371. Gyrotoma funiculata Lea, H. & A. Adams, Genera, I, 1858, p. 305; Binney, Check List, 1860, No. 318; Brot, List, 1862, p. 27; Paetel, Cat., 1888, p. 362. Gyrotoma quadrata Anthony, Proc. Acad. Nat. Sci., Phila., XII, Feb. 1860, p. 65; Binney, Check List, 1860, No. 330; Paetel, Cat., 1888, p. 362. Melatoma quadratum Anth., Reeve, Monog. Melatoma, 1860, sp. 7. Schizostoma virens Lea, Proc. Acad. Nat. Sci., Phila., XII, 1860, p. 187; Journ. Acad. Nat. Sci., Phila., V, n. s., 1862, p. 253, pl. 35, fig. 59; Obs. Gen. Unio, IX, 1863, ‘Diba pl. xxv, se. 50. : Gyrotoma virens Lea, Binney, Check List, 1860, No. 335; Brot, List, 1862, p. 28; Paetel, Cat., 1888, p. 362. Gyrotoma obliqua Anth., MSS., Tryon, Monog. Strepom., 1873, p. 378; Paetel, Cat., 1888, p. 362. 24 CALVIN GOODRICH Description: Shell smooth, ovately gibbous, thick, yellowish-brown; spire short; whorls four, flattened; columella thickened above; aperture large, ovate, white. Habitat: Alabama. Diameter, .44; length, .64 of an inch.—Lea. In incisum the fissure is shallow and oblique. It is as deep as 7% mm. in specimens from Weduska Shoals. It is most shallow in specimens from Butting Ram Shoals down to Wetumpka, not exceeding 234 mm. in depth. The girdle, which represents the material filling the fissure as growth pro- ceeds, is usually well-marked when it is noticeable at all, particularly to- ward the end of growth of adult individuals. The shell is heavy, broadly rounded at the shoulder, frequently with a broad constriction around the center of the whorl. Hinkley (2, p. 40) speaks of incisum as never nodulous. Occasional specimens do have nodes at the shoulder, but these markings appear to be without significance. ‘The largest shell observed measures 22 x 14% mm. The growth lines are usual- ly fine and regular and these are often crossed by discontinuous transverse lines, plainest at the top of the whorl. Of 78 shells from one lot from Wetumpka, 11 showed revolving folds of more or less prominence. The folds are seldom present in imciswm coming from farther up the river. Color varies from olive-ocher to cinnamon-brown. Bands occur much more com- monly than not. Of 77 banded shells from Wetumpka, 74 had a band at the top of the whorl, one at the periphery and a third at the base. This is the prevailing banding formula in the species from other localities. The aper- ture is ovate, varying slightly. The columella in adults is white, rather wide, broadly rounded and regularly curved vertically, being thickened into a small node at the mouth of the fissure. The outer lip is sharp-edged, sinuous. Juveniles are conic, angulated at the base, having the whorls flat and without carinae. A channelled sinus, present in all the young examined, is much more pronounced than in the case of adults. The fissure begins with the third whorl and is easily distinguished in the fourth. Six or seven whorls for the fully-grown shell are indicated. The range of incisum is from Weduska Shoals, Shelby County, to We- tumpka, Elmore County. Lea’s type is a narrow, somewhat constricted form of the Wetumpka aspect. His virens is a young shell. Curta Mighels appears to me to be readily recognizable from the description. [ examined the type of buddi Lea, the locality for which is given in error as Tuscaloosa, but felt uncertain about the identification. Mr. W. B. Marshall of the National Museum kindly furnished me later with an excellent photograph and I can say with confidence that the shell is the same as incisum. The cut in Tryon (6, p. 371) could not have been made from the type. The constriction in constrict- um is fairly common to the species. Funiculatum Lea is a deformed mol- lusk. At the mouth of the Yellowleaf Creek of Chilton County, Mr. T. H. Aldrich took numbers of a small form of Gyrotoma which, though suggestive of walkeri because of its delicacy, I assign to imciswm on account of the deeper fissure. These shells have a kind of family resemblance to Anculosa aldrichi H. H. Smith, taken at the same locality. This Anculosa also is THE GENUS GYROTOMA 25 dwarfed. Larger Gyrotomae and larger Anculosae have been collected in the same place. The student is led to wonder whether these dwarfed shells do not dwell in some particular spot on these shoals where the con- ditions of food supply, current, temperature or some other cause tend to stunt molluscan growth as sometimes in very cold springs shell life is at once small and plainly of mature development. Gyrotoma walkeri H. H. Smith, new species Pl; ihies 308 31, ande32 Shell: Cylindrical, shouldered, solid; body whorl slightly constricted, apex erod- ed, base rounded and with only a suggestion of a sinus. Growth lines very fine, crossed by delicate, waving lines to be seen on all parts of the exterior except where it is worn smooth. Rest period scars rather rough, forming a few irregularly spaced nodules. Fissure very shallow, very oblique, 2 mm. deep at the suture. There is no girdle. Suture impressed, irregular. Color buckthorn-brown. Bands three, one ending at lower edge of fissure, one at periphery and one at the base. Aperture ovate, yellowish- white within. Columella white, broadly-rounded, having a heavy deposit of its material at the fissure. Outer lip firm, sharp-edged, forming a sinuous line with the fissure. Operculum: Ovate, thick, dark reddish-brown, measuring 3 I-2 mm. diameter by 5 I-2 mm. altitude. The left margin is thickened, smooth-edged; right margin broadly curved, broken and worn; basal margin rounded, eroded close to the polar point. Whorls three, the first two regularly and tightly coiled. Growth lines fine to coarse. Measurements: Altitude, 15 mm.; diameter, 10 mm. Aperture, base to suture, 10 I-2 mm.; base to lowest point of fissure, 6 mm.; diameter 4 I-4 mm. Type Locality: Coosa River, Cedar Island, Chilton County, Alabama. Type in Museum of Zoology, University of Michigan; paratypes in Alabama Museum of Natural History. This species appears to be a derivative of G. imcisum Lea which has developed fixed, uniform specific characteristics. It is small for the genus and varies much more in the size of mature specimens than in form. The largest shell measures 17 x 11 mm. It is from Weduska Shoals. No folds appear upon any of the material. Growth lines are usually fine, but some- times coarse and roughened near the shoulder. The waving transverse lines are strikingly beautiful in a few specimens which have not suffered a great amount of wear. The fissure varies slightly in depth, but in none is it deeper than 3% mm. It is frequently only a sinuous indentation. ‘The girdle is absent in all the specimens examined. Color is sometimes as dark as burnt-umber. Of 65 shells, 51 have bands, the formula in all being exactly that of the type. Very young shells are smooth, lacking carinz, plice and all other sculp- ture except microscopic growth lines. The base is obtusely angulate. The sinus is well-marked, slightly oblique. The fissure, about 1%4 mm. deep, begins with the fourth whorl and is hardly more than a sinuous, inturned part of the outer lip. Eight whorls are indicated for full-grown individuals. The known localities for walkeri are Weduska Shoals, Shelby County ; The Bar, Cedar Island, Higgin’s Ferry and Duncan’s Riffle, Chilton County ; Butting Ram Shoals, Coosa County. One of Mr. Smith’s labels carries the note, “swift water.” 26 CALVIN GoopDRICH Gyrotoma amplum Anthony Pl. II, figs. 33, 34, and 35 Gyrotoma ampla Anth., Proc. Acad. Nat. Sci. Phila., XII, Feb., 1860, p. 66; Binney, Check List, 1860, No. 306; Brot, List, 1862, p. 27; Paetel, Cat., 1888, p. 361. Melatoma amplum Anth., Reeve, Monog. Melatoma, 1861, sp. 16. Schizostoma amplum Anth., Lewis, Amer. Journ. Conchol., V, 1869-70, p. 168; ‘Tryon, Monog. Strepom., 1873, p. 360; Lewis, Fauna of Ala., 1876, p. 22. Gyrotoma recta Anth., Proc. Acad. Nat. Sci., Phila, XII, Feb., 1860, p. 64; Linney, Check List, 1860, No. 331; Brot, List, 1862, p. 27; Paetel, Cat., 1888, p. 362. Melatoma rectum Anth., Reeve, Monog. Melatoma, 1861, sp. Io. Gyrotoma salebrosa Anth., Proc. Acad. Nat. Sci. Phila., XII, Feb., 1860, p. 66; Binney, Check List, 1860, No. 333; Paetel, Cat., 1888, p. 362. Melatoma salebrosum Anth., Reeve, Monog. Melatoma, 1861, sp. 15. Schizostoma salebrosum Anth., Lewis, Amer. Journ. Conchol., V, 1869-70, p. 169; ‘Tryon, Monog. Strepom., 1873, p. 376; Lewis, Fauna of Ala. 1876, p. 23; Hinkley, Nautilus, XVIII, 1904, p. 40. Gyrotoma robusta Anth., Proc. Acad. Nat. Sci., Phila., XII, Feb., 1860, p. 67; Pinney, Check List, 1860, No. 332; Brot, List, 1862, p. 28; Paetel, Cat., 1888, p. 362. Melatoma robustum Anth., Reeve, Monog. Melatoma, 1861, sp. 14. Gyrotoma demissa Anth., Proc. Acad. Nat. Sci. Phila., XII, Feb., 1860, p. 64; Binney, Check List, 1860, No. 316; Brot, List, 1862, p. 27; Paetel, Cat., 1888, p. 362. Melatoma demissum Anth., Reeve, Monog. Melatoma, 1861, sp. 9. Schizostoma demissum Anth., Lewis, Amer. Journ. Conchol., V, 1869-70, p. 168; Tryon, Monog. Strepom., 1873, p. 372; Lewis, Fauna of Ala., 1876, p. 23. Schizostoma glandula Lea, Proc. Acad. Nat. Sci, Phila., XII, May, 1860, p. 187; Journ. Acad. Nat. Sci., Phila, V, n. s., 1862, p. 249, pl. 35, fig. 53; Obs. Gen. Unio, IX, 1863. p. 71. pl. 35, fig. 53; Lewis, Amer. Journ, Conchol., V. 1869-70, p. 168; Tryon, Monog. Strepom., 1873, p. 377; Lewis, Fauna of Ala., 1876, p. 23. Gyrotoma glandula Lea, Binney, Check List, 1860, No. 319; Brot, List, 1862, p. 27; Paetel, Cat., 1888, p. 362. Schizostoma Hartmanii Lea, Proc. Acad. Nat. Sci, Phila. XII, May, 1860, p. 187; Journ. Acad. Nat. Sci., Phila., V, n. s., p. 247, pl. 35, fig. 51; Obs. Gen. Unio, IX, 1863, p. 60, pl. 35, fig. 51. Gyrotoma Hartmanii Lea, Binney, Check List, 1860, No. 322; Brot, List, 1862, p. 27; Paetel, Cat., 1888, p. 362. Melatoma nucula Anth., MSS., Reeve, Monog. Melatoma, 1861, sp. 19. Gyrotoma nucula Anth., Brot, List, 1862, p. 27; Paetel, Cat., 1888, p. 362. Schizostoma nuculum Anth., Lewis. Amer. Journ. Conchol., V, 1869-70, p. 169; Tryon, Monog. Strepom., 1873, p. 361; Lewis, Fauna of Ala., 1876, p. 23. Schizostoma Showalterii Lea, (non Lea, 1860), Proc. Acad. Nat. Sci., Phila., XVI, 1864, p. 112; Journ. Acad. Nat. Sci, Phila. VI, 1866, p. 149, pl. 23, fig. 56; Obs. Gen. Unio, XI, 1867, p. 105; pl. 23, fig. 56. Schizostoma \Showalteriana “ea”, Tryon, Amer. Journ. Conchol., I, Oct., 1865, p. 335; Tryon, Monog. Strepom., 1873, p. 375; Lewis. Fauna of Ala., 1876, p. 23. Schizostoma constrictum Lea, Hinkley, Nautilus, XVIII, 1904, p. 4o. Gyrotoma olivula excisa (Lea), Hannibal, Proc. Malacol. Soc., X, 1912, p. 18r. Gyrotoma laeta incisa (Lea), Hannibal, Proc. Malacol. Soc., 1912, p. 182. Description: | Shell smooth, ovate, rather thick, olivaceous; spire not elevated, but acute; whorls 6-7, subconvex; sutures well defined; fissure broad, rather deep and waved; aperture moderate, elliptical, flesh-colored and banded within; columella . THE GENUS GyROTOMA 27 smooth, or slightly thickened only at the fissure; body-whorl striate and banded; whorls of the spire not banded, but having a thickened, cord-like line near the suture. Habitat: Coosa River, Alabama. Length, eleven-sixteenths; breadth, seven-sixteenths of an inch. Length of aper- ture, seven-sixteenths; breadth of aperture, four-sixteenths of an inch—Anthony. This species is close to the form of inciswm which has a comparatively deep fissure, as at Weduska Shoals. But the young of that form have the typically shallow fissure. It is not so in the case of amplum. Moreover, the fissure in amplum is sometimes straight, never in incisum. Sometimes the lower edge of the fissure of amplum is waved, but this has not been observed in incisum. ‘The depth of the slit in the species under consider- ation varies in adults from 6% to 9 mm. The girdle is stout, often ir- regular. In form the shell is somewhat cylindrical, noticeably shouldered and in instances slightly constricted though not so frequently or so deeply as in mcisum, Ordinarily the mollusk is without revolving folds, but occas- ionally they do occur. Commonly the body whorl bears low nodules just below the suture. Growth lines are fine, often prominent and are crossed in places by waving transverse lines usually discontinuous. The largest speci- men taken by Mr. Smith measures 23% x 144% mm. Color varies from olive-yellow to Mars-brown. The prevailing banding formula is three bands equidistant and of the same width. Bands in this species show a marked tendency to become obsolete. The columella is white to bluish-white, hav- ing a porcelainous node at the fissure and ending at the base in a distinct sinus. Anthony erected several species at the same time from shells which seem to have come from Dr. Schowalter just after that collector’s first visit to Fort William Shoals. Their novelty led to the placing of undue emphasis upon characters of no great significance. Thus recta was made a species principally upon the form of the shell and the absence of bands. The ac- centuation of rest scars was the justification of salebrosa. Robusta was differentiated from salebrosa by reason of its smoothness. Demissa, a sin- gle specimen, appears to have been one of the not very common adults of the genus which have a fully developed columella. Dr. Lea published his descriptions of hartmanii and glandula hardly three months after Anthony’s names had been printed. His hartmani is an old, worn and broken speci- men of amplum. It is somewhat difficult to distinguish glandula from in- cisum, but I was enabled to do this with care by means of a photograph sent to me by Mr. W. B. Marshall of the United States National Museum. Showalteriana is a pathological individual. The synonymy of amplum is an illustration of a curious competition in species-naming which obtained among American naturalists in the first half of the nineteenth century. Hinkley (2, p. 40), who found this to be the most plentiful Gyrotoma on Fort William Shoals, identified it as constrictum, apparently following Try- on’s decision in the case of recta. I have seen both the type of constrictum and some of Hinkley’s shells and can say confidently they are not the same. The range of the species is Three-Island Shoals to Peckerwood Shoals. 28 CALVIN GooDRICH UNIDENTIFIABLE SPECIES Gyrotoma babylonicum (Lea) Schizostoma Babylonicum Lea, Proc. Amer. Phil. Soc., IV, 1845, p. 167; Trans. Amer. Phil. Soc., X. 1848, p. 04, pl. ix, fig. 54; Obs. Gen. Unio, IV, 1848, p. 94, pl. ix, fig. 54; Journ. Acad. Nat. Sci. Phila. V, n. s., 1862, p. 248; Obs. Gen. Unio, IX, 1863, p. 70; Lewis, Amer. Journ. Conchol:, V, 1869-70, p. 166; Tryon, Monog. Strepom., 1873, pp. xv, 369; Lewis, Fauna of Ala., 1876, p. 22. Gyrotoma Babylonicum Lea, Chenu, Man. de Conchyl., I, 1859, f. 2, o2t; H. & A. Adams, Genera, I, 1858, p. 305; Binney, Check List, 1860, No. 307; Brot, List, 1862, p. 27; Paetel, Cat., 1888, p. 361. b Gyrotoma lacta incisa (Lea), Hannibal, Proc. Malacol. Soc., X, 1912, p. 182. Description: Shell striate, somewhat fusiform, rather thick, chestnut-color; spire obtusely conical; sutures impressed; whorls flattened; fissure small; aperture large, elliptical, somewhat flesh-colored within; columella smooth, angular at the base, thick- ened above.—Lea. This species was described from one specimen and as from Tuscaloosa. ‘Like Lea’s funiculatum and his showalterti of 1864 it is a deformity and broken. ‘The shell is now in the National Museum, being No. 119, 408. It is very much shouldered and is lacking in color bands. The fissure is slightly more oblique than in specimens of pagoda having unusually oblique fissures. It is much more oblique than in the case of pumilum which the shell of babylonicum resembles. While the mollusk does not suggest inciswm in form, the fissure is that of many specimens of this species. I feel unable to decide as to which of these species babylonicum belongs. The chief char- acteristic upon which its specific position was established does not obtain among thousands of Gyrotoma examined. It is unquestionably an abnor- mal shell. If it should be decided that babylonicum is the same as pumulurit the latter species, described in 1860, will go into the synonymy. Hinkley (2, p. 41) made this shell a synonym of zetwmpkaensis and the latter the same as pyramidatum. Whatever babylonicum may be it belongs to the members of the genus occurring at Wetumpka, while pyramidatusm is the first species appearing in the river and does not go down to Elmore County. Gyrotoma cylindraceum (Mighels) Schizostoma cylindracea Mighels, Bost. Proc., I, Oct. 1844, p. 189; Tryon, Monog Strepom., 1873, pp. xv. 361; Lewis, Fauna of Ala., 1876, p. 23; Hinkley, Nautilus, XVIII, 1904, p. 40. Gyrotoma cylindracea Migh., H. & A. Adams, Genera, I, 1858, p. 305. Gyrotoma cylindracea Mull., Binney, Check List, 1860, No. 315. Gyrotoma cylindracea Gould, Brot, List, 1862, p. 27; Paetel, Cat., 1888, p. 362. Gyrotoma olivula excisa (Lea), Hannibal, Proc. Malacol. Soc., X, 1912, p. 181. Description: Shell nearly smooth, cylindrical, thick, with slight, revolving un- dulations; epidermis olivaceous; spire ovate-conic, eroded; whorls three or four, flat- tened, shouldered; suture distinct, aperture oval; fissure deep and wide——Mighels. The author gave Warrior River, Alabama, as the locality for his species. Mr. C. W. Johnson informs me that the Mighels collection was destroyed in the fire which burned the quarters of the Portland Society of Natural History in 1854, another fire in 1866 destroying a second collection, The THE GENUS GYROTOMA 29 type of cylindraceum must be considered lost and the species cannot be recog- nized with any exactness from the description. Hinkley (2, p. 40) thought the shell might be a form of ellipticum Anth. SPURIOUS SPECIES Melatoma sphaerica Anth. is apparently an Anculosa belonging to a described species of the group A. picta Conrad. Habitat, Coosa River, Ala- bama. References: Melatoma sphaerica Anth., MSS., Reeve, Monog. Melatoma, sp. 8, April, 186r. Schizostoma sphaericum Anth., Lewis, Amer. Journ. Conchol., V, 1869-70, p. 169; Tryon, Monog. Strepom., 1873, p. 364; Lewis, Fauna of Ala., 1876, p. 23. Anculosa sp., Goodrich, Anculosae of Ala. River Drain., 1922, p. 56. H. & A. Adams list Gyrotoma altilis Anth,, comca Say, curvata Say and foremanti Lea. The.altilis possibly refers to Gillia altilis Lea. Shuttle- worth’s G. conica may be intended or Melania conica Say, a Pleurocera. Tryon (10, p. 132) appears to think that curvata refers to Plewrocera cur- vatum Lea. Two Melanians were named foremanu by Lea, but neither belongs to Gyrotoma. Paetel repeats these errors and adds aequalis Lea, circumlineata Tryon, Coosaensis Lea, ellipsoides Lea, rubicunda Lea and vauxiana Lea, all belonging to other genera than Gyrotoma, besides angu- lata Lea, having reference probably to Melania angulata Anth., a Gonio- basis. REFERENCES 1. HANNIBAL, HaArotp, “A Synopsis of the Recent & Tertiary Freshwater Mol- jusca of the Californian Province”, Proc. Malacol. Soc., X, 1912. 2. Hinxiry, A. A. “List of Alabama Shells Collected in October and November, 1893,” Nautilus, XVIII, 1904, pp. 40, 41. 3. Jorpan, D. S., “The Origin of Species Through Isolation”. Science, n. s., XXII, 1905. 4. Leéwis, JAMEs, “Catalogue of the Shells of the Coosa River, Alabama.” Amer. Journ. Conchol., V, 1869-70, p. 168. 5. Lewis, JAMES, “Fauna of Alabama; I, Fresh Water and Land Shells. Geolog. Survey of Ala., preface dated Oct. 24, 1876. 6. ORTMANN, A. E., “The Crawfishes of the State of Pennsylvania.” Memoirs Carnegie Museum, II, 1906. : 7. Ortmann, A. E., “A Monograph of the Naiades of Pennsylvania.” Part III, Memoirs Carnegie Museum, VIII, Dec., 1910. 8. Reeve, L. A., “Conchologia Iconica.” Melatoma, London, 1860-61. 9. Ruruven, A. G., “Variations and Genetic Relations of the Garter-Snakes.” U. S. Nat. Mus., Bulletin 61, 1908. 10. Tryon, Gro. W., Jr., “Land & Freshwater Shells of North America, Part 4, Strepomatidae.” Smithsonian Misc. Coll., No. 253, 1873. 11. Wits, J. C., “Age and Area.” Cambridge, 1922. 30 _. CAtvin GoopricH PLAC GyROTOMA All figures of shells are 3/2 times natural size Figures 1, 2, and 3. G. Pyramidal un Shuttleworth. Coosa River, Ten-Island Shoals, St. Clair County. (Cat. No. 27425.) Figures 4 and 5. G. spillmanti (Lea). ‘Coosa River, Shelby County. Schowalter Collection. (Cat. No. 27426.) Figures 6, 7, and 8. G. pagoda (Lea). Coosa River, Higgins Ferry, Chilton County. (Cat. No. 27427.) Figure 9. G. pagoda (Lea). Coosa River, Higgins Ferry, Chilton County. Young shell. (Cat. No. 27428.) Figures 10 and 11. G. pumilum (Lea). Coosa River, Wetumpka, Elmore County. (Cat. No. 27420.) Figures 12 and 13. G. alabamensis (Lea). Coosa River, Weduska Shoals, Shelby County. (Cat. No. 27430.) Figures 14 and 15. G. cariniferum Anthony. Coosa River, Ft. William Shoals, Talladega County. (Cat. No. 27431.) Figures 16, 17, and 18. G. lewis (Lea). Ft. William Shoals, Talladega County. (Cat. No. 27432.) GYROTOMA rice 1) Masi \ : We Ns vit * “ean Pe * i SS X PLATE [ ps ato Patt, es OE Ee. SMUT EAR ae? /8 Ad "9. hl bak o Beara Az is' al 7 UJ 7 j -* * v aR SD 3 ens 32 CALVIN GoopRICH PLAT tt GyroToMA All figures of shells are 3/2 times natural size Figures 19, 20, and 21. G. hendersoni H. H. Smith. Type. Coosa River, Ft. William Shoals, Talladega County. (Cat. No. 27433.) Figures 22 and 23. G. exciswm (Lea). Coosa River, Higgins Ferry, Chilton County. (Cat. No. 27434.) Figures 24 and 25. G. laciniatum (Lea). 'Coosa River, Weduska Shoals, Shelby County. (Cat. No. 27435.) - Figures 26 and 27. G. incisum (Lea). Coosa River, Wetumpka, Elmore County. (Cat. No. 27436.) Figures 28 and 29. G. incisum (Lea). Coosa River, Weduska Shoals, Shelby County. (Cat. No. 27437.) Figures 30, 31, and 32. G. walkeri H. H. Smith. Type. Coosa River, Cedar Island, Chilton County. (Cat. No. 27438.) Figures 33, 34, and 35. G. amplum Anthony. Coosa River, Ft. William Shoals, Talladega County. (Cat. No. 27439.) GYROTOMA PLATE II UNIVERSITY OF MICHIGAN MUSEUM OF ZOOLOGY Miscellaneous Publications No. 13 ( Studies of the Fishes of the Order Cyprinodontes BY CARL L. HUBBS ANN ARBOR, MICHIGAN PUBLISHED BY THE UNIVERSITY January 18, 1924 ADVERTISEMENT The publications of the Museum of Zoology, University of Michigan, consist of two series—the Occasional Papers and the Miscellaneous Publi- cations. Both series were founded by Dr. Bryant Walker, Mr. Bradshaw H. Swales:and Dr. W. W. Newcomb. The Occasional Papers, publication of which was begun in 1913, serve as a medium for the publication of brief original papers based principally upon the collections in the Museum. ‘The papers are issued separately to libraries and specialists, and, when a sufficient number of pages have been printed to make a volume, a title page, index, and table of contents are sup- plied to libraries and iridividuals on the mailing list for the entire series. The Miscellaneous Publications include papers on field and museum technique, monographic studies and other papers not within~the scope of the Occasional Papers. The papers are published separately, and, as it is not intended that they shall be grouped into volumes, each number has a title page and, when necessary, a table of contents. ALEXANDER G. RUTHVEN, Director of the Museum of Zoology, University cf Michigan. . STUDIES OF THE FISHES OF THE ORDER CYPRINODONTES By Cart L. Hupss I. A CLASSIFICATION OF THE FISHES OF THE ORDER II. An ANALYSIS OF THE GENERA OF THE POECILIIDAE III. THe Species oF PROFUNDULUS, A NEW GENUS FROM CENTRAL AMERICA IV. Tue Susspecies oF PSEUDOXIPHOPHORUS BIMACULATUS AND OF PRIA- PICHTHYS ANNECTENS I. A CLASSIFICATION OF THE FISHES OF THE ORDER I The order Cyprinodontes Agassiz (equivalent to Microcyprini Regan) comprises certain fishes which have in the past been more or less confused, first with the Cyprinidae, then later with the Haplomi (E'sox, Umbra, Dallia). Toward each of these groups the Cyprinodontes show, indeed, a superficial resemblance, but this resemblance is apparently not indicative of genetic relationship. It has long been clear that these fishes have no close affinity with the cyprinids, but not until 1911 was their close relationship with the Haplomi critically questioned. It appears probable, in reference to the main line of evolution of fishes, that the Cyprinodontes are of distinctly more advanced organization than the Haplomi. They differ from the Haplomi and most other primitive orders of fishes, and resemble the acanthopterygian fishes in the following respects: I. The premaxillaries alone form the margin of the upper jaw. 2. The pectoral fin is elevated in position, and the base of the fin is lat- eral and vertical, rather than ventral and transverse (in Dallia alone among the Haplomi the pectoral fin resembles that of the Cyprinodontes, but in that genus the basal bones are highly modified: the resemblance is not indicative of relationship). 3. The pelvic fins are more anteriorly inserted. 4. The pelvic rays are not more than 6 in number. 5. The vertebrae are typically reduced in number. 6. The larval adhesive organ, which the writer has found retained in Esox lucius of the eae is apparently never developed in any of the Cyprinodontes. 1 Regan, Ann. Mag. Nat. Hist., 7, 1911, p. 323. 4 Cart L. Husss 7. Finally, a fact perhaps most conclusively indicative of a more advanced organization, the branchiostegal rays of the Cyprinodontes are of the acan- thopterygian type, whereas those of the Haplomi are of a distinctly more, generalized type.” II The writer agrees with Jordan® that the old family Poeciliidae or Cyprino- dontidae should be dismembered. While as a whole Jordan’s division of the group seems sound, in certain details it apparently requires modification. The subfamily Fundulinae is equivalent to the Haplochilinae (properly the Aplocheilinae) of Garman, a name of earlier date. The Orestiidae might better be regarded also as a subfamily (Orestiinae) of the Cyprinodontidae. The Characodontidae and Goodeidae should not be separated, for to do so would destroy the extreme naturalness of the combined group. The name Characodontinae is synonymous with Goodeinae, which is the older; the family should therefore be named Goodeidae. This family may best be divided into two groups (quite unlike those recognized by Jordan), namely, the Zoogoneticinae (new subfamily name), containing Zoogoneticus and Girardinichthys, and characterized by the simple teeth and short intestine, and the Goodeinae proper, consisting of Characodon, Chapalichthys, Goodea and Skiffea, and characterized by the bicuspid teeth and more or less elongate intestine. The Poeciliidae (misprinted Poecilidae) may be accepted as outlined by Jordan, except that four generic names listed under this family should be transferred to the Cyprinodontidae. ‘These names follow: Gambusinus, Cynodonichthys, Ilyodon (relationship doubtful) and Cynopoecilus. The Fitzroyidae and Anablepidae should probably be reduced to subfam- ily rank, for the latter group was very probably derived from the former. The name Anablepidae holds for the combined group, being the older. The name Fitzroyinae must be altered to Jenynsiinae, according to the interna- tional rules as at present interpreted (Fitzroyia and Jenynsia were proposed in the same work by Giinther; Fitzroyia has page priority, but the name Jenynsia was accepted by the earlier writers, who regarded the two genera aS synonymous). Ill The families of Cyprinodontes may therefore stand as follows: Cyprinodontidae Poeciliidae Adrianichthyidae Phallostethidae Goodeidae Amblyopsidae Anablepidae . 2 See Hubbs, Jour. Morph., 33, 1919, p. 67. 8 A classification of fishes including families and genera so far as known. Stan- ford Univ. Publ. (Biol. Sci.), 3, 1923, PP. 77-243, 1-x. FIsHES OF THE ORDER CyPRINODONTES GONOPODIA OF POECILIIDAE Ficure 4 : M Tih AAR yi = aS 29 PyatsE III oY SL YSooe= = a ereas Ce Sakae — WU earal \) — D 7 Sy IN — asa = “—_N -——. ED == ase meme ee FIGURE 5 30 Cart L. Husrs PIA aL Figure 6. Distal portion fe the gonopodium of Allopoecilia caucaia, from a specimen 26 mm. long to caudal fin, from Rio Camarones at Arrcyo de Arena, Colom- bia. Figure 7. Distal portion of the gonopodium of Mollienisia latipinna, from a specimen 51 mm. long, from Louisiana. In figures 6 and 7 the membranous hook is labelled + h, the retrorse segment at tip of ray 5, r s. 31 FISHES OF THE ORDER CyPRINODONTES PLATE IV GONOPODIA OF POECILIIDAE | doth A Spades Se ee ee Sa SSS S092, SoS oo Ss ? SS CU Sec Seis rs) ite) ¢ , oe amar rym oe FIGURE 7 FIGURE 6 Pb bee “Tu i} J = i (a Laer) ae v/ : eye Pa Ay 208 a yi i- ‘PR = ee ee FISHES OF THE ORDER CYPRINODONTES 5 II. An ANALYSIS OF THE GENERA OF THE POECILIIDAE I Since 1913, when Regan published A Revision of the Cyprinodont Fishes of the Subfamily Poeciliinae,——an epoch-making work in the study of this group of tropical American fishes——a number of new types have been dis- covered, and new light has been thrown on the characters and relationships of forms previously described. It seems advisable, therefore, to bring this revision by Regan up to date and to include at the same time some original observations and deductions. As Regan’s review demonstrated, by far the most valuable of all the char- acters which have been used in the classification of these little fishes are those taken from the structure of the gonopodium? (this is the intromittent organ of the male, formed by the elongation and singular elaboration of rays 3, 4. and 5 of the anal fin; see figures). With the aid of gonopodial charac- ters, Regan and other writers have been able to demonstrate repeatedly that species which previously had been closely associated together in reality should be distributed to distant parts of the series, and that species which had been widely separated in the older classifications were, nevertheless, very closely related. In the older systems the characters chiefly used involved the length of the intestine and the shape and arrangement of the teeth, features corre- lated closely with food, and now known -to have been repeatedly and inde- pendently altered on the different lines of evolution within the family (these characters still appear to be of high value in the classification of the Poeci- liidae, however, even now that it has been shown that they cannot safely be employed in the primary division of the group). Unfortunately, males are often not taken, even when a considerable series of these fishes are collected, or, if taken, the males may be either young or non-breeding individuals, in either case not showing the characteristic struc- ture of the gonopodium. Unless a species is known and recognized, it can- not be certainly classified.in the absence of males with perfect gonopodia; usually even its approximate position in the series cannot be determined. In fact, species of one cyprinodont family known only from females may be, as they have been at various times, placed in genera belonging to another family of the group. Gonopodia may be quickly mounted on a slide, using preferably the glycerine sodium silicate medium recently described by Creaser and Clench,’ and thus readily studied and drawn with the aid of a microscope, or, better, of a projection apparatus. In routine identification most of the gonopodial features may be made out with the aid of a 12X or 18X hand lens. 1 Regan, Proc. Zool. Soc. London, 1913, pp. 977-1018; figs. 168-173, and pls. 99-IOI. The Poeciliinae of Regan are equivalent to the Poeciliidae of the present paper. 2 Gonopodia were figured and used as systematic characters as early as 1848 (Heckel), 1855 (Poey), and 1907 (Eigenmann), but Regan was the first to use gono- podial characters in a comprehensive revision of the entire group. 8 Trans. Am. Micr. Soc., 42, 1923, pp. 69-71. 6 | Cart L. Hupss II The gonopodium is the chief distinctive feature of the Poeciliidae (as here delimited). In the other groups of viviparous cyprinodonts the anal fin, to be sure, is also modified as an intromittent organ, but in so different a fashion as to indicate an entirely independent origin as compared with the gonopodium. It is highly probable that the Goodeidae, Poeciliidae and Ana- blepidae, as well as the remarkable Phallostethidae of the Malay Peninsula, originated independently from the oviparous Cyprinodontidae.* It is, there- fore, probable that viviparity has been independently acquired four times among the cyprinodont fishes. If this view is correct, we have in the present case a most striking example of the independent attainment, within one group of animals, of one end through diverse adaptation. ‘The oviparous Cyprinodontidae, it would further follow, have had and probably still possess the potentiality of developing viviparity. In fact, many of the species of this family show an approach toward the viviparous condition in the develop- ment of clasping structures and habits.° In the following key to the genera of the Poeciliidae the family is divided into four subfamilies—the Gambusiinae, Poeciliopsinae, Poeciliinae and Tomeurinae. These groups are of unequal size, but are recognized as coor- dinate because each appears to have evolved from the basal stock of the whole family, each being characterized by certain primitive as well as special- ized features. Of the four subfamilies, the Gambusiinae are without ques- tion the least specialized, while Tomeurus is the most highly modified poeciliid. Within these subfamilies certain clusters of genera may be recognized, by reason of the remarkable codrdination which they display between geo- graphical distribution and gonopodial structure, as distinct groups of inde- pendent origin. As it is planned to return later to a fuller consideration of these phenomena, it seems desirable to assign definite names to these groups. For this purpose the term tribe (with the ending -ini) may be borrowed from entomology. III In the key one new subfamily, nine new tribes, nine new genera, and one new species are diagnosed. A list of these new names follows: New subfamily name: Poeciliopsinae. New tribe names: Gambusiini, Xiphophorini, Heterandriini, Poeciliini, Girardinini, Alfarini, Cnesterodontini, Tomeurini. Pamphoriini, : New genera: Panamichthys, Xenophallus, Allogambusia, Neopoecilia, Darienichthys, Parapoecilia, Alloheterandria, Allopoecilia. Phallichthys, New species: Neopoecilia holacanthus. 4The major classification of the cyprinodont fishes is considered in the preceding paper of this series. 5 See Newman, Biol. Bull., 12, 1907, pp. 314-348, 3 pls. and 22 figs. FISHES OF THE ORDER CyPRINODONTES 7 IV An ANALYSIS OF THE GENERA OF THE POECILIIDAE A1.—Lower edge of caudal peduncle rounded, with a median series of normal scales. Pelvic fins well developed in both sexes, and inserted behind the base of the pectorals. B1.—Pelvic fins alike in the two sexes. Gonopodium elongate; the two halves of posterior branch of ray 5 not separated. C1.—Rays of the gonopodium not imbricated so as to form an enclosed tube, all lying strictly in the same plane (see figures I, 2, and 4-7). Jaws firmly United (Except ..im, Giranc 1ittsy eerie ais ou. 4 chee ore so SPaieiars csale ators cre Gambusiinae. D1.—Terminal segment of ray 3 (of anal fin in male — first prolonged ray of gonopodium) not forming a specialized transverse process. E1.—Ray 5 and posterior branch of ray 4 each with a small, usually retrorse hook at its extremity; anterior branch of ray 4 with a thickened node, which is sharply produced, except in Belonesox, to form the “elbow.” (Tribe Gambusiini).*: F1.—Jaws normally formed. Thickened node of ray 4 shorter but more sharply produced, forming a well-defined “elbow” (see figure 1). G1.—Ray 3 without erect spinules; serrae of ray 4 less modified, and wholly apicad to elbow; apical processes of rays 4 and 5 scarcely hooked. Dorsal fin more posteriorly inserted (about twice as distant from eye as from base of caudal in female)......... Heterophallus.? G?.—Ray 3 with erect spinules more or less sharply developed; serrae of ray 4 better developed, in part at least and usually wholly proximad to elbow; apical processes of rays 4 and 5 strongly hooked, except in one subgenus (undescribed). Dorsal fin less posteriorly inserted.. FRE tren hors Roi toate C SURE COC DOC OCA TOE ote Gambusia. F2.—Jaws produced into a short beak. Thickened node of ray 4 longer and not sharply produced to form a well-defined “elbow”; ray 3 without erect spinules; serrae of ray 4 sharply developed and wholly proximad to thickened node; apical processes of rays 4 and 5 scarcely hooked.. eared na cdtece pe arale wae Meee cat's ads st eee ek othe? Betenesene E?.—Ray 5 and posterior branch of ay 4 each without a trace of a retrorse hook at or near the tip, the posterior branch of ray 4 not entering into the distal modification of the gonopodium; the anterior branch of ray 4 not provided with a thickened node, except in Allogambusia’ (see figure 2). H1.—Ray 3 without a pair of curved, horn-like appendages ; ray 5 always smooth on posterior edge. (Tribe Heterandriini).’ I1._Ray 3 without developed serrae. Teeth conical (slightly compressed in Heterandria). Ji1—Ray 3 without definite processes of any kind; ray 4 with no trace of an enlarged node resembling the “elbow” of the Gambusiina. K1.—Rays 4 and 5 contiguous distally, or nearly so, never separated by a notch in the membrane; rays 3 and 4 not together support- ing a small knob. L1.—Rays 4 and 5 both entering into the formation of tip of gono- podium. 6 Gambusiini, new name (as a tribe name). 7 Regan, Ann. Mag. Nat. Hist., (8) 14, 1914, p. 66, fig. A. Haplotype, Heterophallus rachovii Regan, from Vera Cruz, Mexico. 8 The thickened node of ray 4 in Allogambusia is probably of independent origin, as the genus -does not otherwise especially resemble the Gambusiini, and as the node differs in certain details from the “elbow” of the group. ® Heterandriini Hubbs, new tribe name. 8 Cart L. Hupss M1.—Gonopodium rather short; ray 4 curved more or less back- Ward at atipenecies cc cewtisecfiees reece saan. Brachyrhaphis. M2.—Gonopodium rather longer and more slender; ray 4 curved forward at tip. N1.—Tip of gonopodium gently recurved; ray 5 evenly and gently curved; serrated segments of ray 4 numerous..... separ sa rapetetetels fetes dicho nacbuiseeareereusen eee MADIC HUN ee N?2.—Tip of ysiecnaluaten abruptly curved forward to form a hook-like appendage; ray 5 arched backward behind the HEWe'SELEAC OL LAY. Alyse ctersiecaecare clots iereloyeretels Panamichthys.11 L?.—Anterior branch of ray 4 alone entering into the formation of the specialized forward-curved tip of the long gonopodium (see figure 2). O1.—Terminal segment of ray 4 greatly elongate; anterior edge of ray 4 roughened, the posterior edge with numerous serrae. Teeth strictly conical, and mouth with a developed lateral gape. Dorsal fin much longer and inserted farther forward than anal. Size of fish moderate. Sexual dimorphism slight EME erode vere ishcreve RIES Ove aioe atberseieverce ereuaee re Pseudoxiphophorus. O*.—Terminal segment of ray 4 much less notably enlarged; ante- rior edge of ray 4 smooth, the posterior edge with few serrae (see figure 2). Teeth somewhat compressed, the mouth more strictly transverse. Dorsal fin about equal in size and oppo- site in insertion to anal. Size minute. Sexual dimorphism SIAM acavcpevsiateeictecs oats ees acesc i pere stare Oke aasneredeeonets ners Heterandria. K.—Rays 4 and 5 not contiguous distally, separated by a notch in the membrane; the finely segmented tip of the anterior branch of ray 4 and the elongated terminal segment of ray 3 together supporting a small knob at tip of fin’. ..0.. oo. > Priapella. J?.—Ray 3 with a short, suberect, curved spur near tip; anterior branch of ray 4 with a much enlarged node somewhat resembling the “elbow. ‘OL, the GambuSiii... «-cacer sommes eeae Allogambusia.12 I?.—Ray 3 with developed serrae along its anterior edge; ray 3 without trace of any processes other than the serrae; anterior branch of ray 4 curved forward distally (straight in Alloheterandria), and alone extended to the extreme tip of the gonopodium. ‘Teeth conical or compressed. P1.—Ray 3 abruptly hooked forward and upward at tip; anterior branch of ray 4 slender throughout, and without an enlarged terminal segment;. the posterior branch of ray 4 with the serrae weak, few, and widely separated from tip of ray. ‘Teeth conical............. Seeac ra ratateer er netata Grane eve aovetekeihaite, ramalaresitets Mea ot ey einre eee Darienichthys.** P?.—Ray 3 scarcely curved forward at tip. 10 Priapichthys Regan. Type (as designated by Jordan), Gambusia annectens, the only species the gonopodium of which was known to Regan. The other species’ which have been referred to Priapichthys are not strictly congeneric with Priapichthys annectens, 11 Panamichthys Hubbs, new genus. ‘Type species, Priapichthys panamensis Meek and Hildebrand (Pub. Field Mus., Zool., 10, 1916, p. 322). One species, P. panamensis. 12 Allogambusia Hubbs, new genus. Type species, Gambusia tridentiger Garman, Mem. Mus. Comp. Zool., 19, 1895, p. 80, pl. 4, fig. 10 (Priapichthys tridentiger Meek and Hildebrand, Publ. Field Mus., Zool., 10, 1916, p. 320, fig. 6). Two species, A. tridentiger Garman and A. cana Meek and Hildebrand. 13 Darienichthys Hubbs, new genus. Type species, Gambusia dariensis Meek and Hildebrand, Publ. Field Mus., Zool., 10, 1913, p. 88 (Priapichthys dariensis, Meek and Hildebrand, ibidem, 10, 1916, p. 321, fig. 7). One species known, D. dariensis. FISHES OF THE ORDER CyPRINODONTES 9 Q1.—Anterior branch of ray 4 with an enlarged terminal segment; posterior branch of ray 4 with the serrae few but strong, extend- ing nearly to tip of ray; serrae of ray 3 short. R1.—Terminal segment of gonopodium not curved, but with a back- ward directed spinelet; segments of anterior branch of ray 4 everywhere slender. Mouth with a lateral cleft; teeth conical. Size of fish small....... Sur Shed oh a ated are Alloheterandria.* R*.—Terminal segment of gonopodium strongly curved forward, and without spinelets; two subterminal segments of anterior branch of ray 4 considerably thickened. Mouth vertical; teeth some- what expanded at tip. Size minute......... Neoheterandria.15 Q?.—Anterior branch of ray 4 with the terminal segment not enlarged; posterior branch of ray 4 with the serrae more numerous, but not nearly extended to tip of ray; serrae of ray 3 long. Teeth incisor-like; mouth without developed lateral cleft............. ER he bone Sb.- 6 Cb aae OO cho RP CROPS ET eae Pseudopoecilia. H?.—Ray 3 with a pair of curved horn-like appendages near tip; ray 5 with the posterior edge serrate. (Tribe Girardinini.)* S1—Bones of the lower jaw firmly united; the teeth fixed. T1—Teeth of the outer row spear-shaped and widely spaced....Toxus. T?—Teeth of outer row chisel-shaped and close-set...... Glardichthys. S?.—Bones of lower jaw loosely connected; teeth of outer row movable, Spatulatien ere crate era ee eens ss cthanslans Gen oleeises Girardinus. D?.—Terminal segment of ray 3 forming a more or less highly specialized proc- ess. (Tribe Cnesterodontini.)“ U1.—Process not trowel-like. V1—Process not forked, nor antler-like. W1.—Process short, with a simple and slight curve; ray 3 with serrae on both edges. “Teeth or Otiter row comie-+. sas. 5 «.e: Diphyacantha.18 W?.—Process inordinately elongate, with a strong, double curve; ray 3 without developed serrae. Teeth of outer row chisel-shaped, fol- lowed by a single series of conical teeth............... ‘Cnesterodon. V?—Process forked, and each fork antler-like. Teeth of outer row oar- shaped, slender, followed by a band of villiform teeth..... Phalloceros. U?.—Process trowel-shaped; a most remarkable structure, formed like a rela- tively immense scoop, with a compressed horn on each side, and attached to the fin by a narrow pedicel. Teeth oar-shaped, in a single series..... BOR BOC SY COO SOS SOP os De Re ee Phallotorynus.19 C?.—Rays of the gonopodium imbricated so as to form an enclosed tube on SANE GKGKEY apt Seto, (AS weKeTIAE Ae hou obmo meee lon bese o aces > coue Poeciliopsinae. X1.—Gonopodium without horn-like appendages; ray 4 not forming a hook at its tip. Y1—All well developed serrae of ray 4 directed backward. 14 Alloheterandria Hubbs, new genus. ‘Type species, Gambusia nigroventralis Eigenmann and Henn, Jndiana Univ. Studies, No. 16, 1912, p. 26 (Priapichthys nigro- ventralis Henn, Ann. Carn. Mus., 10, 1916, p. 115, fig. 3). In addition to A. nigroven- tralis, Gambusia caliensis Eigenmann and Henn, in Henn, J. c¢., p. 113, may prove to belong to this genus. 15 Neoheterandria Henn, Ann. Carn. Mus., 10, 1916, p. 117, fig. 5. Haplotype, Neoheterandria elegans Henn, from Colombia. 16 Girardinini Hubbs, new tribe name. 17 Cnesterodontini Hubbs, new tribe name. 18 Diphyacantha Henn, Ann. Carn. Mus., 10, 1916, p. 113, fig. 2. Haplotype, Diphy- acantha chocoensis Henn. 19 Phallotorynus Henn, Ann. Carn. Mus., 10, 1916, p. 126, figs. 8-13. Haplotype, - Phallotorynus fasciolatus Henn, from Brazil. 10 Cart L. Hupss Z1.—Mouth with a lateral cleft; bones of lower jaw firmly conriected; teeth fixed jiandeconicaly ists cee wove or tai Otte, ee eenerarere sirens Leptorhaphis. 2—Mouth with the cleft transverse and vertical; bones of the lower jaw loosely connected; teeth movable and somewhat compressed. al.—Body deep and the form not distorted. Dorsal inserted farther for- ward than anal. Gonopodium longer, reaching nearly or quite to caudal base when depressed; posterior branch of ray 4 not nearly FEACHIAE AAD TOE CTI. ha cite wiciaie bleu Sipahelare oss emistialtte s sitlshe Phallichthys.2+ a?.—Body slender, and distorted in form, as usual in Gambusia. Dorsal inserted farther back than anal. Gonopodium of moderate length, not reaching caudal base when depressed; posterior branch of ray AlSileebelnigandteetclauhokes (ohonOueimilyands¢onandgechsooo5p6o0c- Poeciliopsis.22 Y2,—Well developed serrae of ray 4 directed forward and outward in part. Other characters as here given for Poeciliopsis............ Phalloptychus. X2,.—Gonopodium with two prominent horn-like appendages,—one, like the excrescence on a pelican’s bill, on the front edge of ray 3; the other a long curved horn attached at base to posterior edge of ray 4. Other characters as here piven for Leptorhaphisys .,..c\..25 vines Lose aneeen Xenophallus.2? B?.—Pelvic fins enlarged and modified in the male, the first ray with a soft swollen tip, the second ray thickened and usually much elongated. Ray 5 of anal fin in male (the last ray of the gonopodium) forming a trough posteriorly, the two sides of the posterior branch of the ray being widely separated, and more or less twisted to one side; ray 3 with a membranous swelling along anterior margin (characters not known for the Pamphortini) i. am eveer ace Poeciliinae. b*—Gonopodium elongate, about one-third as long as the body. Bones of the lower jaw more firmly united. Inner ventral rays abruptly shorter than the second ray. (Tribe Pamphoriini.)* : c!.—Teeth of outer series slender, pointed; bones of lower jaw firmly united... sopeteroncig arateheieehe Risks. 4 eye rensaaya © Brey s]5/s ee hoe eRe ee lain p hontchtins: c?.—Teeth of outer series broad incisors; bones of lower jaw somewhat less Hriiliy = wmated i. ce spake ooress Geeks sic Ge roelo o eae a cinete eee tenet are Pamphoria. b?.—Gonopodium short. Bones of lower jaw loosely united. Teeth of outer series rather slender, spatulate. d1.—Membranous swelling on front of gonopodium not forming a definite pre- puce-like hood; ray 3 ending in a very large and strong hook, about the end of which the tip of the anterior branch of ray 4 is curved; posterior branch of ray 4 with serrae both distal and proximal to tip of ray 5; ray 5 arched backward toward tip (see figure 4). Inner rays of ventral of male not abruptly shorter than the second ray. (Tribe Xiphophorini.)* el,—Ray 5 not ending in a hook. Caudal alike in the two sexes; dorsal fin shorter, swithovtommr rays )meria ceteeioticiee Aa aa ie ert Clee Platypoecilus. e2.—Ray 5 ending in a backward-directed hook. Lower rays of caudal fin in the male ending in a long, sword-like appendage; dorsal fin tonger, with Tt OES LAYS sic We wre sttveesseuats ic louwea et om aimelatalien se asere terete tome Retort Xiphophorus. 20 Poeciliopsinae Hubbs, new subfamily name. 21Phallichthys Hubbs, new genus. Type species, Poeciliopsis isthmensis Regan. Poeciliopsis isthmensis from Panama and Pocecilia pittieri Meek from Costa Rica are probably identical with Poeciliopsis amates Miller, from Guatemala and Honduras (a specimen from Tela, Honduras, is at hand). 22Poeciliopsis Regan. ‘Type, Poecilia presidionis Jordan and Culver (as designated by Henn, 1916; not Poeciliopsis. isthmensis as designated by Jordan, 1920). 23Xenophallus Hubbs, new genus. Type species, Gambusia umbratilis Meek, Publ. Field Mus., Zool., 10, 1912, p. 70 (type material, from Costa Rica, re-examined). 24 Pamphoriini Hubbs, new tribe name. ‘This tribe perhaps belongs in the Gam- busiinae, the distinctive features of its two genera not being well known. 25 Xiphophorini Hubbs, new tribe name. FISHES OF THE ORDER CyPpRINODONTES I! d?.—Membranous swelling along anterior margin of gonopodium modified into a prepuce-like hood; ray 3 with at most a very small hook; ray 4 straight or nearly so, with serrae only proximal to end of ray 5; ray 5 not arched backward (see figures 5-7). Inner ventral rays abruptly shorter than the second ray. (Tribe Poeciliini.)* f1.—Ray 5 with the last segment of the posterior branch not elongate, nor sharply retrorse (see figure 5). g!—Tips of all rays very slender; ray 3 with either serrae or a terminal hook. h1.—Ray 3 serrate along anterior margin, not abruptly nor extensively slender at tip, and without a trace of a terminal hook. i1.—Anterior branch of ray 4 strictly entire on both margins; anterior edge of ray 5 more or less roughened, but hardly spinous. . Poecilia. i?—Both branches of ray 4 serrate on posterior margin; anterior edge of ray 5 with long spines (see figure 5) ........... Neopoecilia.27 h?.—Ray 3 without developed serrae, and abruptly narrowed distally to form a long, slender tip, which bears a terminal hook......... Limia. g°.—Tip of ray 3 and of posterior branch of ray 4 markedly clavate; ray 3 lacking both serrae and terminal hook................ Parapoecilia.” f{°.—Ray 5 with the last segment of the posterior branch elongate and abruptly retrorse, entering the posterior profile of the gonopodium (see figures 6 and 7). j1—Ray 3 without processes on posterior margin; terminal hook of same ray weakly or not at all developed. Fishes of minute size. k1.—Tips of all the ray$S of the gonopodium slender; anterior as well as posterior branch of ray 4 serrate on posterior margin; serrae of ray 3 strong; terminal hook of ray 3 wholly undeveloped. Sexual dimor- phism in color and size strongly marked...................- Lebistes. k*.—Tips of ray 3 and of posterior branch of ray 4 clavate (but less so than in Parapoecilia) ; posterior branch only of ray 4 serrate; serrae of ray 3 weak; terminal hook of ray 3 often weakly developed (see figure 6). Sexual dimorphism slight.............5.4 Allopoecilia.” j°.—Ray 3 with long, more or less spinous processes on posterior margin; terminal hook of same ray always strongly developed; tips of all the rays slender; posterior branch only of ray 4 serrate; serrae of ray 3, strong (see figure 7). Larger fishes, with sexual dimorphism slightly marked (except in the development of the dorsal fin)...... Mollienisia. A?.—Lower edge of the caudal peduncle sharp, without a median series of scales. Pelvic fins better developed in the male sex. Bones of lower jaw firmly united; Lectin COliCalll wate A sett mectne uN cee ratte Pea ae ot 3 oN fe sees nee eee, 2 ci hee: Tomeurinae. I1.—Pelvic fins well developed in both sexes, located behind the base of the pectoral fins. Gonopodium very simple in structure, without appendages or serrae; ray 3 with knob-like enlargement. Body rather deep. (Tribe Alfarini)2°....Alfaro ’—Pelvic fins absent in female, minute and below gill opening in male. Gonopo- dium with long appendages. Body elongate (tribe Tomeurini).*1...Tomeurus. 26 Poeciliini Hubbs, new name (as a tribe name). 27 Neopoecilia Hubbs, new genus. Orthotype, Neopoecilia holacanthus Hubbs, new species (based on the material recorded as Poecilia vivipara by Evermann and Marsh, in Bull. U. S. Fish Comm., 1900 (1902), p. 97). 28 Parapoecilia Hubbs, new genus. Type species, Limia hollandi Henn, from Brazil (Henn, Ann. Carn. Mus., 10, 1916, p. 138, fig. 17). *9 Allopoecilia Hubbs, new genus. ‘Type species, Girardinus caucanus Steindachner (Mollienisia caucana Henn, Ann. Carn. Mus., 10, 1916, p. 136, fig. 16). 80 Alfarini Hubbs, new tribe name. 31 Tomeurini Hubbs, new name as a tribe name. 12 Cart L. Huss III. Tur Species oF PRoFUNDULUS, A NEW GENUS FROM CENTRAL AMERICA I Profundulus, new genus Orthotype, Hundulus punctatus Gunther Several closely related species from Central America, hitherto referred to Fundulus, Adinia, or Zoogoneticus, comprise a natural group, differing from Fundulus as here restricted in several important respects. The lateral rims of the genital aperture of the adult female are scarcely produced, surrounding not more than the first anal ray: thus not forming the genital pouch diagnostic of Fundulus. The anal fin in the adult male is lower, instead of higher, than in either the young or the adult female. The dorsal and anal fins are relatively long and low, the anal especially being composed of more rays (13-17) than usual in Fundulus. The two fins are inserted far posteriorly, and near the same vertical, somewhat as in a pike. ‘The interorbital is more convex, the head .being formed more like that of an Umbra than that of a Fundulus. In the absence of a genital pouch, at least, the species of Profundulus resemble those’ comprising the African group Nothobranchius? (walkeri, orthonotus, melanospilus, neumaniu, guenthert, taemopygus, and possibly others, such as sjoestedti and microlepis). All of these African types, how- ever, have a much narrower preorbital. Another African species, commonly referred to Fundulus, namely, capen- sis, has the dorsal fin more advanced in position than in Profundulus, while the remainder? have higher and sharper vertical fins, and mouths of different structure. ‘ II As the more generalized members of Fundulus, Cynolebias and some other American genera of the Cyprinodontidae, as well as the less specialized Goodeidae, resemble the species of Profundulus in habitus, it seems not improbable that Profundulus, of all American genera, diverges least from a general ancestral cyprinodont type. The geographical distribution of the groups in question harmonizes with such a view. 1 For descriptions, figures and references, see Boulenger, Cat. Fresh-water Fishes of Africa, 3, 1915, pp. 23-29. 2 Recognized by Regan (Ann. Mag. Nat. Hist., 7, 1911, p. 324) as distinct from Fundulus. 3 Fundulus nisorius Cope, described as from West Africa, is excluded from present consideration, as it was probably based on American material erroneously labelled. FISHES OF THE ORDER CyPRINODONTES 13 III Four or more species of Profundulus may be distinguished. ‘Two are fromm the Atlantic, the others from the Pacific drainage of Central America. 1. Profundulus punctatus Gunther Fundulus punctatus Gunther, Cat. Fishes Brit. Mus., 6, 1866, p. 320 (original description; page priority over guatemalensis) ; Trans. Zool. Soc. London, 6, 1868, p. 482, pl. 84, fig. 5 (redescription of types, with figure); Jordan and Evermann, Bull. U. S. Nat. Mus., 47, pt. 1, 1806, p. 637 (after Giinther); Meek, Publ. Field Mus. (Zool.), 5, 1904, p. 104; tbid., 7, 1907, p. 136; Regan, Biol. 'Centr. Am., Pisces, 1907, Dp. 2 (excepting oaxacae in synonymy); Fowler, Proc. Acad. Nat. Sci. Phila., 68, 1916, p. 47. Fundulus guatemalensis Giinther, Cat. Fishes Brit. Mus., 6, 1886, p. 321 (original description) ; Trans. Zool. Soc. London, 6, 1868, p. 482, pl. 84, figs. 3, 4 (redescription of types, with figures) ; Meek, Publ. Field Mus. (Zool.), 5, 1904, p. 103; ibid., 7, 1907, pp. 136, 184. ?Fundulus guatemalensis Eigenmann, Proc. U. S. Nat. Mus., 13, 18901, p. 64 (Western Ecuador; locality probably erroneous). Adinia guatemalensis Jordan and Evermann, Bull. U.S. Nat. Mus., 47, pt. 1, 1806, p. 660 (punctatus was here retained in Fundulus!). Zoogoneticus guatemalensis Meek, Publ. Tield Mus. (Zool.), 3, 1903, p. 94 (name only; the species erroneously and uncritically placed in the unrelated new genus Zoogoneticus). Fundulus pachycephalus Ginther, Cat. Fishes Brit. Mus., 6, 1866, p. 321 (a doubtful synonym of punctatus, which has page priority); Trans. Zool. Soc. London, 6, 1868, p. 483, pl. 84, fig. 6 (type figure and redescription of type) ; Regan, Biol. Centr. Am., Pisces, 1907, p. 77. Fundulus parvipinnis Garman, Mem. Mus. Comp. Zool., 19, 1895, p. 100 (in part: Profundulus punctatus confused with Fundulus parvipinnis!). Specimens collected by George P. Engelhardt in the Rio de Moka, Guate- mala, were identified with this species by Dr. S. FE. Meek, but the record was never published. All of the material published on by Dr. Meek (see syn- onymy) has been re-examined. I provisionally follow Regan in uniting the nominal forms guatemalensis and punctatus under one specific name. The specimens at hand from (the outlet of ) Lake Amatitlan, regarded by Meek as typical of guatemalensis, are, it is true, of stouter build and paler colora- tion than stream specimens (typical punctatus). But some otf the specimens from the outlet of the lake show evident, though indistinct, spots on the dorsal and caudal fins, and closely resemble the type of punctatus in the form and general coloration of the body. The degree of spotting is also very vari- able in stream specimens. Hence it is provisionally assumed that the name guatemalensis was based on well-nourished, pale-colored lake specimens of punctatus, and that the form is at most only racially distinct. Cart L. Hupss 14 -pilu ay} Burpnypour Cor +S Cor Cer Se JSipar aes Ze gt oe Sic aa cee og re a ov ev Iv ce ve Ge Sie. SLe Sg'e S2e Ce aes $9 vS Qh tI e1 VI ol II II ZI II ZI ve ee ce UD/JYDUL GY ayD’T £0 43]1NO Cer Ee cae, ae Iv coe c6'~ 1'¢ Si eI rane ZI ve br I ve At ¢° I ts oe coe i [bn Oz Ve €Z TI rat ZI ve ee (it Cal ant gt ace gt oe Sz Cie ee eee re ev ZV JESS VAe Le SOs pyar. = AVETe oz Ve gt Gre 09 ZO LS SI VI SI ZI ZI ZI ZI ZI ZI te ee ee D]DWAIIDN+) ‘obupuajven ff ze Ze aes Y SEE cQ°z Se; c-1S VI II ZI ve ‘Salas [BSIOp ce I (Ay pee en OS ODEN OI a0 coe wece peoy ‘UIPIM Gs ¢ gt 2 6)ne, 6 6) a) oe), 86. 6) eRe [e}iqioo1d UIPIM ALA PG © ROTO OTC eueded selene) site [e1g10.193U1 ‘UIPIM vs? gts: enis ny 0.10076, © 010, 0, Sue late evel) 6\9:/0,10 . ynous ‘y}sud’T I'v V lie) s (pial 6 elie ValimAniep es els “or 6) ©) eve'7e.m . aAa ‘yy sua] Soe LS Sie olerta ue) ¢ ecedeelceelay sy 86/61. 6nb) 8)o.n peoy ‘yy sUo’T Sgz gz —eppee JO aseq 0} [eSiop JO UISIIO ee gt cmaneietelie) eelia ls) {e\s) ace jelteis|-0)emale oe Apoq ‘yidoq cg LS a evette Jota © alee) ute) ee eufaner ave; a)'s) 6/0 “WU ‘yy su’T VI PI Aetiecec ia Cran ake) s).0x@"uye sehen a le.8 Jeue ‘sAVI-ULT ee CT ilenalfe Teles py eme,elonente gan = acho ao jes1op ‘SABI-ULT 25 ZIocrresttetteeeeesees sss ssueay 1 S3]B9G Pe PE @n 1006) 6.0, 6 ae @" 0) 0) 0, 68:5) 0.66) 20) “BUOT S SITBIS DI DUL AION) ‘olop] ap 412 0°T aso ups : snyojound snjnpunfor1g f0 Sjuamadnsvayy pup syunoy FISHES OF THE ORDER CyPRINODONTES 15 2. Profundulus oaxacae Meek Fundulus oaxacae Meek, Publ. Field Mus. (Zool.), 3, 1902, p. 90 (original descrip- tion) ; ibid., 5, 1904, p. 104, fig. 28 (redescription and figure of type material). Fundulus punctatus Regan, Biol. Centr. Am., Pisces, 1907, p. 78 (in part). As also in P. labialis and P. punctatus, the anal fin is higher in the female than in the male; it is also high in the young, as in punctatus. Re-examination of Dr. Meek’s types from Oaxaca seems to confirm the validity of his species. It is a smaller fish than punctatus, and the eye is smaller than in specimens of the Guatemalan species of comparable size (see measurements of each). ‘The scales average fewer in the longitudinal series, but more in the transverse series. The dorsal and anal fins are nearly unicolor, whereas in punctatus these fins have a light marginal band, just within which the dark pigment is intensified. The two species, however, are very closely related, and may prove identical (as Regan has held after exam- ining more abundant material). Counts and Measurements of Profundulus oaxacaé Locality, Oaxaca, Mexico Beales Loe. Loi P ia vc dale. «ogra teres St 30 31 2 33 31 EEE 28 Dic Ces a en 13 13 13 12 12 13 Paretys. Gorsalt 20s ib.0 sa ieee wate 128 II es II II II 1 TDS SE ROS. hill i eee 8 15 13 14 13 13 13 ean AMUN eGo, 608 os Sis wre. b-bd SAE 48 47 44 43 43 39 EMEC OCY. oi saints fuss s)= its sss one 3.05 3.4 ae 3.6 3.8 arg Origin of dorsal to base of caudal.. 2.85 2.9 3.0 3.0 2s 2.85 Wenethmheade wn ceieec cence. aaa BA) 2.5 3.4 3.4 35 2.4 WRPPREOS CV El).« oic4 Gites Pesan edad vere 4.7 4.7 4.6 4.5 4.5 4.0 Meeti@ti SHOUE jb - 5 cic 2 aise. cals ss B35 3.2 3.4 BiG 3.6 3.4 Width: anterorbital .. 2s )../04/.. 0 2.2 2.3 2.4 2.35 2.4 2.3 Width enreorpital, gicccac 4. oe 3.5 3.6 BG) 3.4 B57) 3.6 MCLE EAC aalets-csc e cracit's Sucre anes 1.45 TS 1.5 1.4 1.6 15 3. Profundulus candalarius, new species Holotype, a male specimen 73 mm. long to base of caudal; collected with a female paratype of 68 mm. in a limestone spring at Candalaria, near the Mexican boundary of Guatemala, 6 leagues distant from Nenton, on June * 1, 1906. This is in the basin of the Rio Chiapas of the Atlantic drainage. The specimens were collected by Mr. O. F’. Cook, and both are deposited) in the Field Museum of Natural History. P. candalarius in general appearance is similar to P. labialis Giinther, the only species hitherto known in the Atlantic drainage, it having been taken from Lake Yzabal and from the basin of the Rio Ouke tributary thereto. 5 In five others: 30, 30, 31, 32, 32. 6 Holotype. 7 In five others: 12, 12, 12, 13, 13. 8 Miscounted 14 in original description. 16 Cart L. Husss From labialis, candalarius differs in the number of fin-rays and scales, in which characters it agrees with the Pacific species (punctatus and oaxacae). From these two species candalarius differs in form and proportions, as noted in the course of the following description. In its form Profundulus candalarius suggests a pike. The back is heavy at the shoulders, but flat in profile back to the posteriorly inserted dorsal fin. The top of the head is flatter than in related species, both transversely and longitudinally ; the anterior profile is a little incurved. The snout is rather more produced and pointed than in the Pacific species. Depth of body, 3.4 (3.5) in total length, minus the caudal; least depth of caudal peduncle, 2.3 (2.2) in head, 1.35 (1.6) in its length from anal fin. Length of head, 3.25 (3.6); eye, 4.5 (4.5), larger than in the species of the Pacific drainage; upper jaw, 3.2 (3.0), not reaching to below eye; least preorbital width, 2. (2.8) in postorbital length of head, decidedly wider than in punctatus and oaxacae. ‘The body and the head are much more compressed than in the species just named: greatest width of head, 1.65 (1.7) in its length; least interorbital width, 2.5 (2.5). Branchial aperture free to upper opercular angle. Gill-rakers only about as long as the pupil, 15 (16) on the lower limb of the outer arch. ‘Teeth in rather broad bands in the jaws, the outer series a little enlarged. Scales larger than in P. labialis, in 33 (35) rows from branchial aperture to caudal base, 12 (13) longitudinal rows from pelvic fin to and including the mediodorsal series. Fin rays: dorsal, 12 (12); anal, 14 (15); pelvic, 6 (developed rays), Dorsal fin set farther backward than in punctatus and oaxacae, the interval between its origin and the caudal base being contained 3.2 (3.2) times in the total length without caudal, rather than from 2.65 to 3.1 times. In the male holotype the dorsal fin is just as high as long, the anal a little longer than high; in the female specimen the dorsal is a little higher than long, while the two dimensions of the anal are about the same. ‘The broad caudal is truncate behind ; the pectoral is broadly rounded; the pelvic short, not nearly reaching to anus. ‘The intestine is little coiled. The coloration is not very distinctly preserved. The scales of the male type are light where exposed, but marked by dark crescents at their bases; dark streaks are evident between each of the longitudinal scale.rows. ‘There is a dark axillar blotch. The fins are not distinctly marked; the anal prob- ably had a light margin. ‘The female specimen is rather plainly colored, the scales marked by dark reticulations; an indistinct dark lateral streak was apparently developed; coloration of fins as in the male. FIsHES OF THE ORDER CyPRINODONTES 17 IV. Tuer SupsprecieEs oF PSEUDOXIPHOPHORUS BIMACULATUS AND OF PRIAPICHTHYS ANNECTENS I Two species of cyprinodont fishes from Middle America exhibit geograph- ical variation sufficiently well defined and extensive as to require division into subspecies. "These species are Pseudoxiphophorus bimaculatus of south- ern Mexico and Guatemala, and Priapichthys annectens of Costa Rica. II Pseudoxiphophorus bimaculatus Heckel This species, which inhabits the Atlantic drainage from central Mexico to Guatemala, comprises a complex of races distinguished only by average, intergrading characters. Provisionally it may be divided into three subspe- cies, two of which have been regarded as distinct species. Of these sub- species, the typical bimaculatius is intermediate, the two extreme subspecies being P. b. taeniatus Regan and P. b. jonesii Ginther. ‘These subspecies form a series grading from jonesii of the higher mountain streams and lakes of east-central Mexico through the intermediate bimaculatus of the lower and somewhat warmer waters back of Vera Cruz into the more southern form. The characters which distinguish these subspecies are those of form and proportions, color, and, particularly, the number of dorsal rays. From extreme fdeniatus to jonesu, that is, toward the north and toward higher ele- vations, the body becomes deeper, the head longer, the color paler, the dorsal rays reduced in number. The difference in the number of dorsal rays is summarized in the following table: Frequency table—Number of dorsal rays in the three subspecies of Pseudoxipho- phorus bimaculatus: . Number of dorsal rays ist 12 I3 T4 TS5 16 D7 ee Dee TOMES din sip. cina ont eS soars 15 51 Oy) 6 3 ZED DUMUACULOUUS 2). le wpeeiaree << oe I 29 21 3 aX - Pe LG ENACIUS® <0: a.Bje erase'oy a cc oes 2s a bho yk ats vi 13 4 ais The correlations of these variational phenomena will be considered by the writer in a subsequent paper. Pseudoxiphophorus bimaculatus jonesii Gunther Mollienisia jonesii Giinther, Ann. Mag. Nat. Hist., (4) 14, 1874, p. 371; Garman, Mem. Mus. Comp. Zool., 19, 1895, p. 52; Jordan and Evermann, Bull. U. S. Nat. Mus., 47, pt. 1, 1896, p. 698. Gambusia jonesii Regan, Ann. Mag. Nat. Hist., (7) 19, 1907, p. 260; Biol. Centr. Am., Pisces, 1907, p. 97, pl. 12, fig. 8. Pseudoxiphophorus jonesii Regan, Proc. Zool. Soc. London, 1913, p. 993. 1 Miller, in recording specimens from Guatemala, gave the dorsal rays as 10 to 17, but no one else has found fewer than 14 rays in the dorsal fin of any specimen of this species found so far south. Perhaps Miller confused some other species with the present one. 18 Cari L. Hupss Pseudoxiphophorus bimaculatus Woolman, Bull. U. S. Fish. Comm., 14, 1804, p. 65; Jordan and Evermann, Bull. U. S. Nat. Mus., 47, pt. 1, 1896, p. 678; Meek, Publ. Field Mus., Zool., 5, 1904, p. 127 (in part). Pseudoxiphophorus pauciradiatus Regan, Ann. Mag. Nat. Hist., (7) 13, 1904, p. 256; (7) 16, 1905, p. 362. The pertinence of Mollienisia jonesu to the present genus and its approx- imate identity with his own pauciradiatus has already been indicated by Regan. The material of this form examined is part of that recorded by Woolman (1894), as Pseudoxiphophorus bimaculatus, from the Rio Blanco at Orizaba. - ‘This is the type locality of pauciradiatus. , Pseudoxiphophorus bimaculatus bimaculatus Heckel Xiphophorus bimaculatus Heckel, Sitz. Akad. Wiss. Wien, 1, 1848, p. 206, pl. 9, figs. I, 2. Pseudoxiphophorus bimaculatus Bleeker, Ichth. Arch. Ind. Prodr., Cypr., 2, 1860, p. 485; Atl. Ichth., 3, 1863, p. 140; Troschel, in Muller’s Reise Mexico, 3, App., 1865, p. 104; Giinther, Cat. Fishes Brit. Mus., 6, 1866, p. 332; Garman, Mem. Mus. Comp. Zool., 19, 1805, p. 81, pl. 3, fig. 6, and pl. 8, fig. 9; Bean, Proc. U. S. Nat. Mus., 21, 1898, p. 541; Meek, Publ. Field Mus., Zool., 3, 1902, p. 98; Regan, Ann. Mag. Nat. Hist., (7) 13, 1904, p. 256; Meek, Publ. Field Mus., Zool., 5, 1904, p. 127 (in part) ; Regan, Proc. Zool. Soc. London, 1913, p. 993, fig. 170C (in part); Cockerell, Proc. Biol. Soc. Wash., 28, 1915, pp. 153, 156. Gambusia (Pseudoxiphophorus) bimaculata Regan, Ann. Mag. Nat. Hist., (7) 19, 1907, p. 260. Gambusia bimaculata Regan, Biol. Centr. Am., Pisces, 1907, p. 98 (in part). (?) Poecilioides bimaculatus Steindachner, Sitzb. Akad. Wiss. Wien, 48, 1863, p. 176, pl. 4, fig. 2; Troschel, in Miiller’s Reise Mexico, 3, App. 1865, p. 104 (not Xiphophorus bimaculatus Heckel). Pseudoxiphophorus reticulatus 'Troschel, in Miller’s Reise Mexico, 3, App., 1865, p. 104; Gtinther, Cat. Fishes Brit. Mus., 6, 1866, p. 333. The typical subspecies of Pseudoxiphophorus bimaculatus inhabits the streams back of Vera Cruz, chiefly at moderate elevations. In material collected by Meek the dorsal rays vary as indicated in the following table: | Frequency table-—Dorsal rays in Pseudoxiphophorus bimaculatus bimaculatus: Dorsal rays 12 13 I4 I5 San Francisco River basin at Xico and Jalapa........... I 24 9 2 Blanrcomnavectesbasineate GCOLnd opal eer rie tae ei Me I 10 I Papaloapan basin at Obispo, Perez and El Hule......... is 4 ee Pseudoxiphophorus bimaculatus taeniatus Regan > Pseudoxiphophorus bimaculatus Meek, Publ. Field Mus., Zool., 5, 1904, p. 127 (in part); Regan, Ann. Mag. Nat. Hist., (7) 16, 1905, p. 363; Miller, Bull, Am. Mus. Nat. Hist., 23, 1907, p. 105 (in part?) ; Regan, Proc. Zool. Soc. London, 1913, p. 993, fig. 170C (in part). Gambusia bimaculata Regan, Biol. Centr. Am., Pisces, 1907, p. 98 (in part). Pseudoxiphophorus bimaculatus var. taeniatus Regan, Ann. Mag. Nat. Hist., (7) _ 16, 1905, p. 363; Biol. Centr. Am., Pisces, 1907, p. 98, pl. 14, fig. 4. FISHES OF THE ORDER CyPRINODONTES 19 The name taeniatus was proposed by Regan, as a “variety,” to distinguish a form from San Domingo de Guzman, southern Mexico, the distinctive fea- tures as pointed out being the development of a median longitudinal stripe and a higher number of dorsal rays. Since the color feature will probably prove to be only a phase without geographical significance, and since the high number of rays appears to be a general characteristic of the southern form of the species, it is here proposed to use the name taeniatus for that subspecies. The range of taeniatus may then be extended to cover the Atlantic drain- age of the Isthmus of Tehuantepec, Guatemala and British Honduras. Frequency table—Number of dorsal rays in Pseudoxiphophorus bimaculatus taeniatus : Dorsal rays Locality Authority I4 I5 16 vy Sanborn, Mexico FAUT Sls cyarctel Qeverccleroiers I 2 * a Rio Tonto, Mexico Regan (1905) o......; 5 7 3 =p San Domingo de Guzman, Mexico Regan (1905) ....... .. I : I San Domingo de Guzman, Mexico Regan (in lit.) ...... I 2 I Belize, British Honduras BUNS ee ahd 3 Se sdone sts wo e8 I III Priapichthys annectens Regan This species as here construed includes two subspecies: typical annectens, of the lowland streams on the Atlantic slope of Costa Rica, and the new sub- species lesperis, inhabiting the headwaters of the Rio Grande de Tarcoles, a Pacific stream. The form hesperis, when compared alone with typical an- nectens, appears to represent a distinct species, but the differences between the two forms are found to be bridged over when one determines the charac- ters of series from Rio Reventazon, which flows down the east.slope of Costa Rica opposite these headwaters of the Tarcoles. The connection between the two forms is not by an even areal intergradation, but rather by what may be termed mosaic intergradation. Excluding for the present the intermediate races of the Reventazon basin, the differences between the two subspecies of Priapichthys annectens may be expressed as follows: Character annectens Intermediate races hesperis Head in length, in adult female.... 3.5 to 3.75 3.6 to 4.2, 3.8 to 4.35 rarely to 4.5 err M CAGES: 2.0 arose c's, cfs Pale coerce 2 SE 3.0 to 3.7 2.9 to 3.75 3.6 to 4.3 Mites LENE: ers Sleteaetyois Fitba echiere Slee ates usually various usually wider narrower Pigment of margins of scale pockets. tending to various less con- form spots centrated Finally, the dorsal rays are reduced in number in the subspecies hesperis, as is indicated in the following tabulation: Frequency table—Number of dorsal rays in the two subspecies of Priapichthys annectens: ———————Dorsal rays 8 9 IO II Average IPL), ARO LS OS GO OC SOC DO ECT ae igs 22 6 10.21 PiowMIE Mahe TACES co hele div ashe nese neeens I 54 28 I 9.34 PEM ES OTIS 1h TAT T o claraih pices cies seeess 3 79 18 * 9.15 20 Cart L. Husss Comparison will show that the differences between the Tierra Caliente and the mountain forms of Priapichthys annectens are rather closely paral- leled by the differences which distinguish the races of Pseudoxiphophorus bimaculatus inhabiting respectively the same types of environment. Further emphasis will be given this point in a paper to follow the present one. Priapichthys annectens annectens Regan Gambusia annectens Regan, Ann. Mag. Nat. Hist., (7) 19, 1907, p. 259; Biol. Centr. Am., Pisces, 1907, p. 97, pl. 14, figs. 5-7. Priapichthys annectens Regan, Proc. Zool. Soc. London, 1913, p. 992, fig. 170B; Meek, Publ. Field Mus., Zool., 10, 1914, p. 114 (in part). A series of specimens from Parismina, Costa Rica, taken by Dr. Meek in the Rio Parismina, a lower tributary of the Reventazon, agrees closely with the type-description of Gambusia annectens Regan. Hence it is taken as typical of the subspecies Priapichthys annectens annectens. ‘The 23 females vary in length to caudal from 18 to 58 mm.; the three males, from 28 to 34 mm. ‘The 6 adult females more than 40 mm. long vary in proportions as follows: greatest depth of body, 3.0 to 3.3; length of head, 3.65 to 3.75; snout, 3.1 to 3.3, longer than eye (but no longer than eye in smaller females, “as long as or shorter than eye,” according to Regan) ; eye, 3.3 to 3.7; scales, 28 to 30; dorsal rays Io or 11 (ten in 19 specimens, eleven in 6) ; anal rays, 9, the fin not strongly falcate (except in some young). The coloration con- sists principally of conspicuous black borders around the scale pockets, the pigment on the ventral part of the scales becoming mostly concentrated to form a round black spot at apex of scale pocket ; the spots forming longitud- inal rows, somewhat as in Gambusia. Along the posterior half of the axial series of scales the black spots become transversely divided, one portion being located above, the other below, the middle of the scale. In some specimens the spots on the scales above and below the axial series have been, shifted from their normal median position to a location adjacent to that of the divided spot of the scales of the axial series, which then appear to be bounded above and below by a zig-zag line. On young specimens dark transverse bars extend across the posterior half of the body, passing through every second or third scale of the axial series. In both sexes the bars become faint with age and, in the females at least, completely disappear. Three half-grown females from Guapiles (spelled Guapilis by Dr. Meek), taken in the Rio Guapiles, an upper tributary of the Rio Negro which flows toward the mouth of the San Juan, agree with the females of the series from Parismina described above. Dorsal, 10; anal, 9. Priapichthys annectens: races intermediate between P. a. annectens and P. a. hesperts. Gambusia annectens Meek, Publ. Field Mus., Zool., 7, 1907, p. 145 (in part). Priapichthys annectens Meek, Publ. Field Mus., Zool., 10, 1914, p. 114 (in part). Under this head there are recorded several series of specimens, mostly from the valley of the Rio Reventazon, and all showing traits variously inter- FISHES OF THE ORDER CYPRINODONTES 21 mediate between those of P. a. annectens on the one side and those of P. a. hesperis on the other. 1. Chitaria; elevation, 340 meters. One female, 48 mm. long to caudal base: depth,. 3.3 ; head, 3.8; snout, 3.4; eye, 3.4; scales, 28; dorsal rays, 9; anal, 9, the fin not strongly falcate. This specimen has the large eye and the coloration of annectens, but only 9 dorsal rays, as in hesperis, and the head of intermediate length. 2. El Guayabo; elevation, 360 meters. Five females, 44 to 56 mm. long to caudal: depth, 3.0 to 3.4; head, 3.6 to 3.9; snout, 3.3 to 3.4; eye,-3.0 to 3.5; scales, 29 or 30; dorsal rays, 10 (four specimens) or 11 (one). ‘These speci- mens have the large head, large eye, and the number of dorsal rays of typical annectens, but the dark margins of the scale pockets are not concentrated into spots. Another series from Guayabo, collected by J. F. Ferry, contains: (1) young females with the typical coloration of annectens; (2) four males, 24 to 30 mm. long to caudal; (3) four females, resembling the lot described above, 42 to 50 mm. long. The adult females present these variations: head, 3.6 to 3.9; eye, 2.9 to 3.2; scales, 29 or 30; dorsal rays, 9 (three specimens), as in hesperis. 3. Turrialba; elevation, 600 meters. One female, 40 mm. to caudal: depth, 3.4; head, 3.65; snout, 3.0; eye, 3.2; scales, 29; dorsal rays, 9; anal, 9, strongly falcate. This specimen agrees with typical annectens in proportions and coloration, but with hesperis in the number of dorsal rays. 4. Tucurrique; elevation, 941 meters. Four females, 40 to 49 mm. long; head, 3.6 to 4.0; eye, 3.0 to 3.4; scales, 29 to 31; dorsal rays, 9 (all four specimens) ; anal strongly falcate; dark margins of scale pockets concentrated ventrally into spots, which are not divided along axis of fish. The fishes of this lot have the proportions of annectens, but only g dorsal rays, as in hesperis. 5. Quebrada de los Negros; elevation, (?). ‘Twenty-six females, varying in length to caudal base from 34 to 52 mm. The specimens of this lot more than 45 mm. long have the following characteristics: head, 3.8 to 4.2; eye, 3.2 to 3.6; scales, 29 to 31; dorsal rays, 9 (ten specimens) or 10 (16 speci- mens) ; anal more or less strongly falcate; coloration as in the Tucurrique lot. The specimens of this series have the large eye characteristic of typical annec- tens, but in the size of the head and in the number of dorsal rays they are intermediate between typical annectens and the subspecies hesperis. 6. “Quebradilla’ (this locality cannot be found on the available maps, but according to a note left by Dr. Meek with the specimens, Quebradilla is on the Atlantic slope of Costa Rica). A large series collected by Dr. A. Alfaro, including about fifty males 21 to 29 mm. long. Eight females, 51 to 59 mm. long to base of caudal, have the following proportions: head, 3.8 to 4.5; snout, 2.7 to 3.3; eye, 3.4 to 3.75; scales, 29 or 30; dorsal rays, 8 to Io (eight in one specimen, nine in 35, ten in 8); anal strongly falcate. ‘This series has the general appearance of hesperis, but the eye is not quite so small as in that subspecies. 22 Cart L. Husrs Priapichthys annectens hesperis, new subspecies Gambusia annectens Meek, Publ. Field Mus., Zool., 7, 1907, p. 145 (in part). Priapichthys annectens Meek, Publ. Field Mus., Zool., 10, 1914, p. 114 (in part). This new form is represented in the collections of the Field Museum of Natural History by a very large series of specimens from San José, Costa Rica, a city on the Rio Maria Aquilar, an upper tributary of the Rio Grande de Tarcoles (elevation, 1,165 meters) ; and from ‘Tobosi (1,200 meters) ; and San Isidro (1,260 meters), on other tributaries of the same stream. Lower in the same basin the species appears to be replaced by Brachyrhaphis olomina. The holotype is a female, 59 mm. long, from San José. Many female paratypes are nearly as large, but the largest male among about 125 is 49 mm. in length, and none of the other males are longer than 36 mm. Body rather heavy; dorsal contour rather strongly arched before the dor- sal fin, its greatest point of elevation being midway between tip of snout and base of caudal (in smaller specimens the dorsal contour is less Strongly arched, and its highest point is at the origin of the dorsal fin) ; the belly is rounded; greatest depth of body in length to caudal base, 2.8 (2.8 to 3.4); caudal peduncle deep, its length from anal fin to base of caudal, 2.8 (2.8 to 3.2) in length of body; least depth of caudal peduncle, 1.35 (to 1.45) in head. The head is short, broad, and deep; its length to end of opercle, 4.3 (3.8 to 4.35) in length of body; its greatest breadth, 1.3 in its length; its depth below occiput, equal to (or a little shorter than) its length. The flatness of the top~ of the head is continued backward with decreasing width to the dorsal fin; least interorbital width, 1.8 (1.7 to 1.9). Length of the broadly rounded snout, 3.2 (2.8 to 3.3) ; diameter of the eye, 3.7 (3.6 to 4.3), larger, of course, in smaller specimens. The mouth is oblique and broad, but with well-devel- oped lateral cleft; teeth conic, pointed, in bands, those of the outer series spaced and moderately enlarged. Scales in 31 (29 to. 31) series. Color in alcohol brownish, becoming lighter on the belly and on the flat- tened back, but without a definite dark streak along the mid-dorsal line, nor along the axial line of the body, nor along the edges of the caudal peduncle. The scale-pockets are broadly margined with blackish brown except on the under part of the body. In smaller females the scales on the lower part of the body frequently have their pigment concentrated into apical spots, but usually neither so definitely nor so extensively as in H. a. annectens; as in that form the pigment of the median three rows of scales tends to be con- centrated along the upper and lower margin of the scales of the axial series posteriorly. Small females have dark vertical bars in the posterior part of the body, separated by one or two scales. ‘The dorsal fin is dark, with a row of interradial black spots near base of fin; caudal dark to light dusky, the scales on its base with conspicuous dark borders; anal and paired fins dusky. Fins usually smaller than annectens proper, but variable. The dorsal is composed of 9, sometimes 10, rarely 8, rays, and its origin is nearly as close to the caudal base as to the head, usually but not always farther back than in annectens proper. The caudal is usually short and truncate, as long as (or a fifth longer than) the depth of the caudal peduncle. Anal composed FISHES OF THE ORDER CYPRINODONTES 23 of 9 rays, the third to the fifth being prolonged to form a short lobe, render- ing the edge of the fin strongly falcate; these are the same rays which in the male are much more produced to form the gonopodium; the origin of the anal is midway between the base of caudal and a point behind the eye equal to half or all its diameter. The paired fins are broadly rounded and short, the pectoral being about as long as the postrostral length of the head, the ventrals reaching only two-thirds the distance to the anus or to the origin of the anal fin. The male, as mentioned before, is much smaller than the female. Its body, while actually not deeper, appears to be so, as the back is more strongly arched ; the body is decidedly more compressed; the width of the head is contained about 1.6 times in its length. The proportions are like those of the young females, or like the adult females of annectens proper. ‘The coloration and color range agree with those of the female. ‘The dorsal fin is inserted midway between the base of the caudal and the anterior half of the postorbital length of the head; the origin of the anal midway between base of caudal and tip of mandible; the ventral fins extend to the origin of the anal. The gonopodium (intromittent organ) is elongate, reaching when depressed nearly to the lower procurrent rays of the caudal fin. Its finer structure is like that of annectens proper. The organ is developed from the third, fourth, and fifth anal rays when the fish has reached a length of about 20 mm. to caudal. Over’one hundred males were examined with the organ developed; in length these specimens vary from 21 to 49 mm., few being more than 30 mm. long, and only the largest being longer than 36 mm. In the specimens from 21 to 36 mm. long the serrae-bearing segments of the second produced ray vary in number from 8 to 12, averaging fewer in the smaller specimens than in the larger ones; in the largest male (49 mm. long) the number is increased to at least 18. The number of serrae doubtless increases with age, as might well be expected, for one serra is developed from each of the distal articulations of the ray, and the number of articula- tions, of course, increases with age. EXPLANATION OF FIGURES All of the figures are of gonopodia, the intromittent organ formed by the modi- fication of rays 3, 4 and 5 of the anal fin in the male. In each figure the rays are numbered, and if divided the anterior (a), median (m) and posterior (p) branches of each ray are lettered as indicated; further, the left (1) and right (r) halves ot certain branches are also shown, and so lettered. The gonopodia are all shown as semitransparent objects, and all, with the exception of that of Pocciliopsis lutsi (figure 3) are viewed from the left side. All the figures were drawn with the aid of a projection machine in order to secure accuracy of proportions; detailed correc- tions of the outlines of individual segments were made while the object was being examined under a high power of the microscope. 24 Cart L. Hupss PATE, i Figure 1. Distal portion of gonopodium of Gambusia puncticulata, from a speci- men 26 mm. long to caudal fin, from Havana, Cuba. The terminal hooks of rays 4 and 5 are indicated by the letters ¢ h. Figure 2. Distal portion of the gonopodium of Heterandria formosa, from a speci- men 12.5 mm. long to caudal fin, from De Leon Springs, Florida. 25 FISHES OF THE ORDER CYPRINODONTES Prats ik GoNOPODIA OF POECILIIDAE 26 Cart L. Husss PLATE II Figure 3. Distal portion of the gonopodium of Poeciliopsis lutzi, from a speci- men 26 mm. long to caudal fin, from San Juan, a village in Acayucan, Mexico. Figure A shows the right side of the gonopodium, while figures B to F digrammat- ically show in oblique views, the cross-cut ends of the gonopodium at various indi- cated levels; the latter are introduced to show the complex imbrication of the rays to form a tube, and the bilateral asymmetry of the two halves of certain rays The crescentic horn in the membranous tip of the gonopodium is indicated by the letters ch. FISHES OF THE ORDER CyPRINODONTES 27 GONOPODIA OF POECILIIDAE Pirate. bi FIGURE 3 28 Cart L. Huspss PEATE Tul Figure 4. Distal portion of the gonopodium of Platypoecilus maculatus, specimen 23 mm. long to caudal fin. The membranous hook is labelled m h. Figure 5. Distal portion of the gonopodium of Neopoecilia holacanthus, specimen 34.5 mm. long, from Arroyo, Porto Rico. The prepuce-like hood and the following two figures is indicated at p h. from a from 2 in this UNIVERSITY OF MICHIGAN MUSEUM OF ZOOLOGY Miscellaneous Publications No. 14 The Genus Perilestes (Odonata) BY E. B. WILLIAMSON AND J. H. WILLIAMSON ANN ARBOR, MICHIGAN PUBLISHED BY THE UNIVERSITY JULY 15, 1924 ADVERTISEMENT The publications of the Museum of Zoology, University of Michigan, consist of two series—the Occasional Papers and the Miscellaneous Publi- cations. Both series were founded by Dr. Bryant Walker, Mr. Bradshaw H. Swales and Dr. W. W. Newcomb. The Occasional Papers, publication of which was begun in 1913, serve as a medium for the publication of brief original papers based principally upon the collections in the Museum. The papers are issued separately to libraries and specialists, and, when a sufficient number of pages have been printed to make a volume, a title page, index, and table of contents are sup- plied to libraries and individuals on the mailing list for the entire series. The Miscellaneous Publications include papers on field and museum technique, monographie studies and other papers not within the scope of the Occasional Papers. The papers are published separately, and, as it is not intended that they shall be grouped into volumes, each number has a title page and, when necessary, a table of contents. ALEXANDER G. RUTHVEN, - Director of the Museum of Zoology, University of Michigan. THE GENUS PERILESTES (ODONATA) By E. B. WILLIAMSON AND J. H. WILLIAMSON The genus Perilestes contains three hitherto described species to which six are added in this paper. They are of similar general appearance and have long slender abdomens and relatively short wings. They bear a striking superficial resemblance to species of the protoneurine genus Protosticta of the Oriental fauna. The obvious superficial difference is in the shape of the apical segments of the female where Protosticta has the characteristic pro- toneurine apical enlargement beginning at the apex of segment 6, while in Perilestes 7 is only of normal size, and the great development in Perilestes of segment 9 into an almost globular form does not occur in Protosticta, though both have the segment relatively very large when compared with the abbreviated eighth segment. The genus is characterized by the following venational characters: the apical posterior angle of the quadrangle reaches the wing margin, M, arises at the first or between the first and second postnodals and Rs arises from the fifth to the ninth postnodal. Kennedy has erected for the genus the sub- family Perilestinae of the family Lestidae. It had been placed in the Legion Podagrion by De Selys (Megapodagrioninae of more recent authors). The genotype is fragilis Hagen as redescribed in this paper. GENERIC AND SPECIFIC CHARACTERS In addition to the venational characters mentioned above, which make the recognition of the genus an easy matter, the species have a number of other characters in common. The dorsum and rear of the head are black, largely shining and with greenish or bronzy reflections. The thorax is dark striped with pale, the sterna are black, and the abdomen is patterned in light and dark colors, the most striking of which is a bright pale ventral area on the apical segments of the male. The wings are hyaline with unicolored stigmas and largely black venation. The legs are moderate in length, rather weak and with a few relatively long spines. The two to five basal spines in the ventro-anterior row of the first femur are modified similarly to those form- ing the comb-like grouping of spines in the same row near the apex of the first tibia. The superior male appendages are forcipate, and the inferior appendages are rudimentary. In the female the valves of the ovipositor are smooth along the ventral edge except at the extreme apex where each valve terminates in two strong, nearly equal, ventrad directed teeth, these two teeth in tandem, one anterior to the other. Specific differentials are given in the synoptic key and in the descrip- tions which follow. They may be briefly indicated. In Group II, probably including the more primitive species of the genus, the venation is more reduced than in species of Group I, as indicated especially by the origin of M,,. In counting the cells behind Cu, we have used as the apical point 4 E. B. WimLuIAMSON AND J. H. WILLIAMSON the first cell basal to the first cell extending from Cu, to the wing margin. Usually there are scattered or more or less continuous double cells between Cu, and the wing margin distal to the first single cell behind Cu,, so the real termination of Cu, is indefinite, but it seemed that no more precise way of indicating this character could be employed. , There is a real difference in the development of Cu, in different species and the method of counting cells as we have done indicates this as accurately as any other method. In fact the only other method is to begin at the most distal double cell between Cu, and the wing margin and count the marginal cells from there to the origin of Cu,. Another venational character, which occurs occasionally and which gives a rather peculiar venational pattern but which offers no difficulty to precise description or tabulation, is the arising of M, and Rs at the same level. More rarely they are separated at their origin by two postnodal spaces. There are two colors of stigmas, black (brown in tenerals) or distinetly reddish brown. Both colors exist in each group. The posterior border of the hind lobe of the prothorax and the mesostigmal lamina offer specific characters. The color of the face, the leg color pattern, the abdominal color pattern, especially of segments 3-5, and in the female of segment 9, and the color of the metasternum differ in different species. In the male the anterior lamina with its posterior process, the second hamule and especially the superior abdominal appendages offer specific characters. In the male of remotus the anal plate is more conspicuous in dorsal view than in any other species. The longer anterior lamina of species of Group I, as compared with Group II, is associated with a relatively longer second segment in species of Group I. The length of the ovipositor is a specific character. HISTORICAL Hagen, in 1862 (in De Selys’ Syn. des Agrionines, Legion Podagrion), described the genus Perilestes and the species fragilis from two female specimens, one from Congonhas, Brazil, the other, a smaller female, from the Essequibo River, British Guiana. The specimens were in two Euro- pean museums and there is nothing to indicate that direct comparison of the two was ever made. At the present time there are several reasons for 'thinking the two specimens are not conspecific and no reason for thinking they are. In this paper the name fragilis is retained for a Brazilian species. In the Revision du Syn. des Agrionines, Premiere Partie, 1886, De Selys adds to the original generic description and describes a male and female (or females) from the Amazon, referring them doubtfully to fragilis which he knew only from Hagen’s inadequate description, but in the final paragraph proposing the name cornuta for the Amazonian speci- mens if they should later prove different from the two females described by Hagen. The name cornuta (cornutus) is retained in this paper for one of the five known Amazonian species and it is assumed that the two i) | THE GENUS PERILESTES or more specimens seen by De Selys are really conspecific, though females in this genus are not readily recognized. But it happens that cornutus belongs to one group in the genus while the remaining four Amazonian species belong to another group, and these groups are venationally dis- tinct. De Selys calls attention to this character in the male, and almost certainly recognized it in the female. Of the latter sex he probably had more than one specimen, as he gives two measurements for the female abdomen, but the text, excepting the localities, implies a single specimen of each sex. Immediately following the description of cornutus De Selys describes attenuata (attenuatus), apparently from a single female collected by Bates at Santarem. The description is so complete that we are able to almost certainly identify one of the four Amazonian species of the group to which attenwatus certainly belongs as that species. Calvert in the B. C. A. doubtfully refers a single immature male from Costa Rica to fragilis. This is a specimen of remotus described in this paper. In the Odonata of the Neotropical Region he discusses two males and two females from Chapada, Brazil, and summarizes in a tabulation the characters of these specimens, of cornutus from De Selys’ descrip- tion, and of the Costa Rican male, but because of the inadequacy of the earlier descriptions he was unable to certainly identify the Chapada speci- mens which were referred doubtfully to fragilis. We have examined these specimens, through the kindness of Mr. Kahl, and they are discussed below under the name solutus. RELATIONSHIPS WITHIN THE GENUS AND GEOGRAPHICAL DISTRIBUTION Species of Perilestes are found from Costa Rica in the north to Rio Janeiro, Brazil, in the south, approximately between parallels 10° north and 25° south. In South America at least, they are confined to the Atlantic drainage. The genus probably had its origin in the Brazilian highlands. Two well defined groups, as indicated in the synoptic key to groups and species, exist in the genus. Group I of this key is northern, Group II southern. The northern extension of Group I may indicate that this group originated during the time of and on the front ranks of the in- vasion of the present Amazonian valley. The northern Group I of four species has only one species in the Amazonian basin. Cornutus, while clearly a member of Group I, is very distinct from the remaining three species of the group. It has been taken on the upper Amazon and on the Madeira River and is thus widely sep- arated geographically from the nearest member of the same group, gwian- ensis, known certainly only from French Guiana and doubtfully from British Guiana. The two remaining species of the group, magdalenae and remotus, occur in the lower Magdalena Valley in Colombia, and remotus, like several other northern Colombian dragonflies, is found also in Central America, in this particular case in Panama and Costa Rica. It seems clear that Group I has penetrated to the north over two routes, one by 6 E. B. WILLIAMSON AND J. H. WILLIAMSON the Rio Negro and the Magdalena, the other probably by way of the Rio Braneo and Guianan rivers. At least the southern tributaries of the Orinoco should’ yield some species of Perilestes but at present none is known from Venezuela. The southern Group II has its headquarters in the Amazonian basin. It is probably older and is certainly not as homogeneous as Group I. One very distinct species, fragilis, which about equals in size the larger species, remotus and magdalenae of Group I, is, like these two species, the most widely separated from the Amazonian basin, being known only from the vicinity of Rio Janeiro. Its nearest rival in size in its group is kahli, prob- ably its nearest relative, also a very distinct species, and also one on the fringe of the geographical distribution of the group, being known only from the Madeira River in Matto Grosso, and from the Rio Japacani in eastern Bolivia. Another very distinct species of the group is attenwatus. In fact, as regards probable relationship with other species of the group, it is the most distinct of all. It has a wide distribution, occurring on the Rio Beni in Bolivia, on the Madeira at Porto Velho, Amazonas, on the Rio Negro at Mandos, and on the lower Amazon at Santarem. The two remaining species of the group, solutus and minor, are closely related. Solutus occurs on the Madeira River in Matto Grosso, and its range ex- tends southward into the Paraguay in the same state. It is also probably along the entire lower Amazon as it has been taken at Para. Munor is the smallest species of the genus and is known only from the Madeira River at Porto Velho, Amazonas. Hasits All Perilestes are forest stream dwellers. The small creeks where we have invariably found them may be briefly described. The three streams at Cristalina, Colombia, have been described in Mise. Publ. No. 30, Univ. of Mich., Mus. of Zool. To briefly summarize this description, the creeks flow in a rolling or hilly, heavily forested country at an elevation of about 320 meters. They vary in size from a few inches to ten or twelve feet in width. Generally the flow is gentle, pools alternating with short ripples. The beds are mixed mud and gravel. The flora is rich with over- hanging bushes and clumps or stretches of aquatics. When we collected there the streams were at low water stage and the water was clear. This description will do equally well for the two streams in the Canal Zone, Panama, where we found Perilestes. In Brazil, similar streams were found at Porto Velho (60 meters above sea level), Albuna, Villa Murtinho, Manaos and Para, each inhabited by from one to three species of Perilestes. The streams near Cashuelo Esperanza, Bolivia, contained a few outerop- pings of rock but otherwise fell within the general description given above. On Mt. Tijuca, in the vicinity of Alto Boa Vista, a suburb of Rio de Janeiro, there are several creeks about one foot deep, varying in width from, eight to twenty feet and having a rapid descent among boulders and outcroppings of solid rocks. Perilestes was found in limited num- bers along the wooded portion of these creeks. THE GENUS PERILESTES On the position at rest and in flight, we made ‘the following note on remotus at Cristalina. In so far as we did not in the field recognize mag- dalenae as distinct it is probable the note will apply equally well to both species. ‘‘Rest with abdomen hanging, wings halt spread, and with the last four abdominal segments often curled up over the back. Flight swift, abdomen ‘dragging’.’’ And at the same locality on February 14, we noted of both sexes: ‘‘Flies till at least 4:30 P. M., when the ravine is dark and gloomy. Always rests with the abdomen hanging but the flight is strong. Males frequently hover head on just as Heteragrion does. At rest they are easily caught in the fingers.’’ Of attenuatus, J. H. W. noted, ‘‘Hirst male seen (at Abunda) hanging with folded wings from tip of leaf eight feet above the ground; the second was flying in the woods a few inches above the ground.’’ On pairing habits we have the following note on remotus, made at Cris- talina: ‘‘Female hung from vine; male alighted on the apex of her abdo- men and walked, up to her head which he held by his feet while he fastened his appendages. He then filled his seminal vesicle, kicking the female vigorously about the head with his hind legs while ‘doing so. Copulation followed immediately, and then the seminal vesicle was filled again, with more kicking.’’ And at Villa Murtinho, J. H. W. noted of solutus: ‘‘Saw a female resting on an almost horizontal twig. A male came up from below and grasped her, she apparently aiding by using front legs. They then flew to a perpendicular twig about a foot away and the tip of the male’s abdomen was brought for an instant into contact with his genitalia. Female then brought the tip of her abdomen to genitalia. After a second or two the sun was clouded and I captured the pair by catching the male by the wings, in my fingers.’’ Sources oF MATERIAL AND METHODS OF STUDY We are indebted to Mr. Morton and Mr. Kahl for the privilege of study- ing certain specimens as indicated under each species in the material studied. The bulk of the material available to us, representing by far the larger number of all known specimens, has resulted from two of our collect- ing trips. Of first importance is the Brazilian collection made by J. H. Wil- liamson and John W. Strohm, in 1922. The other trip which resulted in considerable material in this genus was the University of Michigan- Williamson Expedition to Colombia. We failed to find Perilestes on our trips to British Guiana, Trinidad, Venezuela and Peru. This was espe- cially surprising to us in Venezuela, but the suitable streams (as we under- stand them) which we collected in Venezuela were on the north side of the mountains. Evidently the Magdalena species, doubtless derived from the Amazonian headquarters, have been able to work northward into Central America, but have not spread eastward across the Maracaibo region into the Venezuelan streams flowing into the Caribbean. In this paper, in the section Generic and Specific Characters, we have explained the method followed in counting the wing cells, posterior to Cu,. 8 E. B. WILLIAMSON AND J. H. WILLIAMSON Sometimes the first cross-vein normally crossing Cu, stops at Cu, and does not reach the hind margin. In such a ease the basal cell posterior to Cu, is the equivalent of two cells, and in counting cells posterior to Cu, this basal peculiarity reduces the number of cells by one, but has nothing to do with the length or distal termination of Cu,, thus introducing a misleading factor in the description of Cu,. All measurements in this paper are in millimeters. Length of abdo- men includes the appendages. The color descriptions of preserved speci- mens, because of indefinite patterns and obscure markings, are not very satisfactory and there is a possibility that they give a very imperfect picture of the living insects. However our brief color notes, made from fresh material, while certainly less complete than we now wish, indicate no particularly striking differences between the living and preserved individuals. Probably even in life, color patterns are obscure and in- definite, and detailed descriptions would be difficult, tedious and more or less incomprehensible. It is not impossible that some delicate and evanes- cent colors, overlooked by us in our hurried note making of freshly caught specimens, may be detected in the future by more leisurely and better equipped students. It must be remembered also that an obscure or in- definite color pattern may, by a slight increase of pigment, become sharp and definite and that such increase of pigment may take place nearly at the end of imaginal life and after sexual maturity has been attained. Abdo- minal appendages of both sexes are measured in dorsal view from apex of dorsum of segment 10 to apex of appendages. The anterior lamina is measured from the apex of the sternum of segment 1 to the apex of the posterior process. Possibly better fixed points might be selected, as a bending of the abdomen between segments 1 and 2, either dorsally or ventrally, moves the basal point. But in measuring the length of the expanded postero-ventral wing of the tergum of 2, and the distance from its apex to the base of the segment, while the antero-ventral angle of the tergum is theoretically an admirable fixed point, in actual practice it was found that due to bending of the abdomen this angle was often concealed. The valve is measured from the posterior angle of the basal plate of the ovipositor to the apex of the valve. The form of the hind border of the prothorax is described as seen in antero-dorsal view. And similarly the mesostigmal lamina is described as seen in postero-dorsal view. KEy TO GROUPS AND SPECIES OF PERILESTES i M,, arising usually at the level of one cell before the stigma or more proximad. Abdominal segments 3-6 each with a pale postbasal ring or area proximal to an extensive dark median area. Face similar in the two sexes, dark to black, not sharply patterned, the anteclypeus if paler only slightly so and not strikingly different from the labrum and postclypeus. Male superior appendage with a basal dilatation ending on the inner border in an acute spine. Blade of the second hamule slender, narrowly spoon-shaped or tapering, if flattened the apical blade with its flat surface in the line of the long axis of the abdomen; anterior lamina longer (1.5 or more), slenderer, 1: 2 (1). Pole 3 (2’). 3’. 4 (37). 4’, 8 (1’). SG THE GENUS PERILESTES 9 usually largely dark colored. Adult male with metasternum and femora largely or entirely black (except in gwianensis as shown by scanty material). Adult male with sterna of 9-10 yellowish or blue (except cornutus which has Sterna Of 8—10 yellow OF DUE) ..cccccccscrsssccsssssssenessssnseee Group 1. Males 2. Females 5. Mia arising usually at the level of the proximal end of stigma or more distad. Abdominal segments 3-6 each with a pale basal ring, the dark area following it continuous with the dark median area. Face dissimilar in the two sexes, in the male sharply patterned black and bright pale blue, in the female less distinctly patterned, no bright blue, duller, the black or dull labrum with pale basal areas or darker edged. Male superior appendage without a basal dilatation, or, if present, with it ending on the inner border in a tongue-like plate with a rounded apex. Blade of the second hamule wider, very broadly spoon-shaped or angled in cross-section, the flat or ex- ternal surface transverse to the long axis of the abdomen; anterior lamina shorter (less than 1.5), largely pale colored. Adult male with metasternum and femora not largely or entirely black. Adult male with sterna of 8-10 bright blue (or possibly bright yellow in fragilis) ....ccccsccccseeerene Group II. 8. Stigma reddish brown; spine of superior appendage small, placed slightly distad to midlength of the appendage and directed caudo-ventrad........... cornutus Stigma’ DIACK cee ee cee meen semeete Fae ees ts, ARNON BN Ak, A cee css Be Metasternum and femora largely pale; spine of superior appendage large, placed at about midlength of the appendage, directed ventrad and only peVlrea) g1F ger Cee OUD EZTO artes ont tees crore types tere eee ee en eee eer guianensis Metastermum ‘and: femorarmlame ely tole c ess acces ees saet create cet eee eee nee esersnctecetoe 4, Anal plate a small but conspicuous cordate tubercle projecting beyond the apex of segment 10 in dorsal view; spine on superior appendage basal to mid- length of the appendage and directed Caudadn.n.eesscecsssesneessecsssesesassneteete remotus Anal plate inconspicuous, a small rounded tuberele concealed within the apex of segment 10 in dorsal view; spine on superior appendage distal to midlength of appendage and directed more ventrad than caudad........... mag dalenae Hind lobe of prothorax bearing an erect Median SPiN... recesses: 6. Hind lobe of ‘prothoram Wot spined.: ‘sticrmis, Tae he cc cecceene ec ceeeetentpeemreeer nies Ue Stipmas pred Cushy, novi ee ee cece le ea Rta eect esiatenrdee te cornutus Stigma black; point of attachment of valvular styles under the abdominal apperndapes™ .. ie ccs. meee tee meer eee eee re che AEE 22 PARR ROME LAE BA Tires magdalenae Point of attachment of valvular styles under the abdominal appen- Gagesty .. SOCe PTET Ge Cee Leet he, 2 nee etre A EEY, AMS oh EM eee eee gwianensis. Point of attachment of valvular styles distal to the level of the apices of the abdominal append ames a cesses neers ne en ee eee Ra scone eed remotus Six or fewer cells behind Cu, in the hind wing; stigma black or brown (but not reddish brown). Male appendage without spine or plate in the basal half, distal to midlength a biparted spoon-like dilatation on the inner edge, about equal in length to the narrowed apex; male labrum and anteclypeus bright light blue, almost white, the former black-bordered in front, postelypeus shining greenish black. Female with point of attachment of valvular styles about the level of the apices of the abdominal appendages; hind border of the prothorax straightened, nearly horizontal, dropping off at either end with a nearly vertical lateral margin; mesostigmal lamina low, of uniform VS Id 1] ip MR ee ae Re Nite ot Cea Mes OO oh 8 Ae Oe nD en ee attenuatus More than six cells behind Cu, in the hind wing Males 9. Females 12 (female of minor not known). 10 E. B. WILLIAMSON AND J. H. WILLIAMSON 9 (8’). Male appendage with only an angle on the inner edge at about two-sevenths the length; apex of dorsum of segment 10 slightly elevated in the middorsal line and concave on either side of this and dorsal to the appendages; dorsum of segments 8-10 black, 8 with about the basal third pale (yellow or blue) ; stigma ted dish /browates fatness ene een mete kahli 9/, Male appendage with a flattened tongue-like blade or lobe on the inner edge at about one-third the length; apex of dorsum of segment 10 not elevated but low and flat; dorsum of segments 8-10 dark but not all distinctly black, the pale basal area on 8 reduced to less than one-third the segment... 10. 10 (9’). Stigma reddish brown; labrum very narrowly black edged; outer end of mesostigmal lamina ending in a laterally directed free, acute point................. solutus NO Stigma black; labrum with more black than in solutus; mesostigmal lamina DOL TONGUG iN p.6 Tres, jacabe, POnb sss 25 ace re per ile 11 (10’). Labrum with less than the anterior Half Pla eK ....cc.c.cccccsscscssssmenssscsmonesssssonessessen fragilis Ik iLabrum, Tatgely! black: stb 420:6tces) Sane pie ote els See ie ee, ee minor 12 (8’). Stigma reddish brown; outer end of mesostigmal lamina angular or a free DOUG AA A SER SRY nities Me es Ni sd 8 Ue 8 Dp eae ta ee See ee 13. 127% Stigma black; outer end of mesostigmal lamina scarcely or not elevated but low and rounded; valvular styles at their origin proximal to the level of the apex of 10; dorsum of 9 largely dull brown, shading imperceptibly into the palemateralabasal jarearion either, isid@rsc..c ace eee ee fragilis 13 (12). Dorsum 8-10 black; a large distinct pale lateral basal area on either side (0) ites pent e N-t a LEE UIST ite AOU OPUS GN ea teed Or 2 mi dS ele ecete kahli aye Dorsum of 8-10 brown; pale, lateral basal areas on 9 reduced, more widely separated by the dorsal brown, less sharply outlined... cccccceccecmeecmn solutus DESCRIPTIONS OF SPECIES 1. Perilestes cornutus Selys Abdomen male 46-50.5; hind wing male 22.5-25; stigma front wing male 1.20-1.47; stigma hind wing male 1.27-1.50; abdominal appendage male 1.40; male anterior lamina from apex of sternum of segment 1 to apex of the posterior process 2.1. Male.—Head similar to that of remotus; the labium with a brown area on the suture between the mentum and submentum. Prothorax similar to that of remotus, the proepimeron dark to black and this dark area extended broadly onto the proepisternum; hind border semicireular. Thorax similar to that of remotus, the dark colors possibly duller, dark brown instead of black over much of the dark areas of the mesothorax; the pale area of the metepisternum more nearly reaches the wing base above. . Mesostigmal lamina elevated from either extremity, in nearly straight edges, to a low but distinct obtuse angle placed nearer the lower extremity. Coxae similar to those of remotus. Femora black, more or less pale on the anterior face, especially just before the apex. ‘Tibiae pale to ob- secure, the first and second sometimes largely darker to black above, with less dark beneath, the third with little or no dark. Venation as in remotus; stigma reddish brown. THE GENUS PERILESTES alas Abdomen seen from above: 1 pale with obscure and indefinitely pat- terned median brown from base to apex, darkest at apex; 2 similar to that of remotus but the median longitudinal pale stripe duller, not as sharply patterned; 3 as in remotus but the postbasal pale area more or less re- duced; 4-6 as in remotus, but the anteapical pale area of 6 is discernible in cornutus, and the pale postbasal area of the same segment is less ex- tensive; 7 brown, slightly and indefinitely paler postbasally; 8 brown, the apical half or two-thirds indistinetly darker; 9 nearly black; 10 black. Seen from the side similar to that of remotus but on 3-6 the postbasal pale area is slightly reduced as compared with remotus and on 3 the ster- num, opposite the median dark area, is pale, not dark as it is in remotus; 7 with the postbasal pale area more extensive in cornutus than in remotus and with traces of the same area on 8 in cornutus; anterior lamina and posterior processes of 2 largely dark; lateral color pattern of 4-6 carried across the sterna; sternum of 7 largely dark, the basal fifth, or slightly more, pale; sterna of 8-10 bright pale colored, yellow (?) or more probably blue (?) in life. Appendages largely dark colored, black along the outer edge, shading out to brown on the inner edge and on the extreme base and apex; on the dorsal surface just basal to the spine is a smooth shal- low elliptical depression which is nearly or quite white (possibly bright blue in life). Length of the wing or lobe on the expanded postero-ventral margin of the tergum of 2 about .7, and the distance from the apex of this lobe to the antero-ventral angle of the tergum about 2.2. Venational characters based on four males Postnodals front wing, 13 (25%), 14 (25%), or 15 (50%) ; hind wing, 11 (12.5%), 12 (12.5%), 13 (62.5%), or 14 (12.5%). M, in front wing arising between the first and second postnodals (100%); in the hind wing at the first postnodal (50%), or between the first and second postnodals (50%). Rs in front wing arising at the seventh postnodal (37.5%), or at the eighth postnedal (62.5%); in the hind wing at the sixth postnodal (12.5%), at the seventh postnodal (62.5%), or at the eighth postnodal (25%). M, in front wing arising at or near the eighth postnodal (37.5%), or at or near the ninth postnodal (62.5%); in the hind wing at or near the seventh postnodal (37.5%), or at or near the eighth postnodal (62.5%). M,. in front wing arising three cells before the stigma (75%), or four cells before the stigma (25%); in the hind wing two cells before the stigma (12.5%), or three cells before the stigma (87.5%). Number of cells posterior to Cu, in front wing, 13 (37.5%), 14 (37.5%), 15 (12.5%), or 16 (12.5%) ; in hind wing, 9 (25%), 10 (12.5%), 12 (25%), 18 (12.5%), 14 (12.5%), or 15 (12.5%). Material examined: Brazil, Porto Velho, Amazonas, May 2 and 24, 1922, 5 males; Villa Murtinho, Matto Grosso, April 4, 1922, 2 males; all in coll. E. B. W. 12 E. B. WILLIAMSON AND J. H. WILLIAMSON This species was associated with solutus and kahli at Villa Murtinho and with attenuatus at Porto Velho. Unfortunately no females were taken and our knowledge of this sex is based on De Selys’ description. As to the correctness of our determination of the seven males before us as cornutus there can be little or no doubt. De Selys’ description of the male clearly indicates the group of the genus to which our specimens be- long (origin of M,, and spine on the superior appendage). No other species of the group is known in the Amazon basin, and the color of the stigma and detailed description of the superior appendages agree with our specimens. There is no reason to doubt that the female (or females) described by De Selys belonged to the same species as his male. 2. Perilestes guianensis, new species Abdomen male 44-46, female 39-42; hind wing male 21.5-23, female 24-25; stigma front wing male 1.13-1.20, female 1.24-1.35; stigma hind wing male 1.30-1.33, female 1.40-1.50; abdominal appendage male 1.30, female .40; male anterior lamina from apex of sternum of segment 1 to apex of the posterior process 1.6-1.7; female valve 2.3. Male—Labium as in remotus; labrum, clypeus and frons brown, the labrum sometimes narrowly darker along its base, anteclypeus largely darker with some greenish, postclypeus darker medianally and anteriorly, and frons darker at the extreme lateral margins. Prothorax obscure, sides pale, dorsum darker, especially on the hind lobe; the proepimeron largely and the proepisternum to a lesser extent, dark; hind border semicircular. Thorax similar to that of remotus but reddish brown largely or entirely replacing the black of the latter species, except in the second lateral suture, and with the following differences: the pale color on the metepisternum above and anteriorly nearly or quite reaching the wing base; the metepi- meron is entirely pale, the black stripe on the second lateral suture con- fined to the metepisternum and continued below across the upper edge of the metinfraepisternum, thus resembling magdalenae but more reduced even than in that species; no black on the latero-ventral carina and the merest trace of dark on the metepimeron. Beneath pale, the sterna black. Mesostigmal lamina low, very slightly elevated at the point where the angle occurs in that of cornutus. Coxae pale, unmarked. Femora pale, each slightly darker at apex and with a more or less distinct postbasal dark ring and a very slightly more extensive anteapical one, the two rings more or less saci along the anterior row of spines. Mibiae pale. Venation similar to that of remotus. Abdomen seen from above: 1 pale, brown ringed apically; 2-6 similar to those of remotus, the brown on 2 paler, duller and reduced; 7 with an ill or well defined, dull and pale postbasal area following the slightly darker and more reduced basal brown, the median brown area into which the postbasal pale area darkens is followed or not by an ill defined, scarcely THE GENUS PERILESTES aN} distinguishable anteapical pale area, the apex is darker brown to black; 8 shading from brown at base to nearly ‘black at the apex) ori largely brown throughout; 9 similar to 8 but darker; 10 black, or 9 and 10 brown. Seen from the side 1-6 similar to those of remotus but the basal inferior spot on 2 wanting or very small; 7-9 largely pale, each more or less darker at apex and along the ventral border; 10 black or brown. Sternum of 1 pale; anterior lamina pale, the posterior processes darkening to brown with the ventral edge of the plate black; 3-6 each with the lateral color pattern carried across the sternum; sternum of 7 largely dark to black; 8 black with a median pale area nearer base than apex; 9 and 10 pald (blue? in life), the latter discolored by postmortem changes. Appen- dages pale dull yellow to brownish horn color, slightly darker at base. Length of the wing or lobe on the expanded postero-ventral margin of the tergum of 2 about .6, and the distance from the apex of this lobe to the antero-ventral angle of the tergum about 1.8. Female.—Head similar to that of the male. Prothorax, thorax, mesostigmal lamina, coxae, legs, venation and stigma similar to those of the male. Abdominal segments 1-7 similar to those of the male; seen from above 8 and 10 are black or nearly so, and 9 is about as in remotus with the dark bar slightly narrower in the basal two-thirds; seen from the side 8 and 10 are brown, and 9 is largely blue, shading apically into the brown of the apically widened dorsal band, with the blue and brown shading below into black along the entire ventral border. Sternum of 1 and 2 pale, the latter with or without a trace of dark at midlength; 3-6 with the lateral color pattern of each carried across the sternum; sternum of 7 largely dark, pale for a short distance at the base; of 8 largely pale (blue? in life), an irregular dark median area beginning before the middle and extending to the apex. Appendages brown, shaped as in magdalenae. The valvular styles with points of attachment under the appendages. Venational characters based on three males and four females Postnodals front wing, 18 (12.5% female), 14 (50% males, 50% fe- males), or 15 (50% males, 37.5% females) ; hind wing, 11 (33.3% males), 12 (33.3% males, 62.5% females), or 13 (33.8% males, 37.5% females). M, in front wing arising between the first and second postnodals (100% males, 100% females); in the hind wing at the first postnodal (66.7% males, 62.5% females), or between the first and second postnodals (33.38% males, 37.5% females). Rs in front wing arising at the sixth postnodal (12.5% females), at the seventh postnodal (50% males, 50% females), or at the eighth post- nodal (50% males, 37.5% females) ; in the hind wing at the sixth postnodal (66.7% males, 62.5% females), or at the seventh postnodal (33.3% males, 37.5% females). M, in front wing arising at or near the seventh postnodal (12.5% fe- males), at or near the eighth postnodal (33.3% males, 37.5% females), or 14 E. B. WILLIAMSON AND J. H. WILLIAMSON at or near the ninth postnodal (66.7% males, 50% females) ; in the hind wing at or near the sixth postnodal (16.7% males), at or near the seventh postnodal (16.7% males, 75% females), or at or near the eighth postnodal (66.7% males, 25% females). M,. in front wing arising three cells before the stigma (83.3% males, 75% females), four cells before the stigma (16.7% males, 12.5% females), or five cells before the stigma (12.5% females); in the hind wing two cells before the stigma (50% males), three cells before the stigma (50% males, 62.5% females), or four cells before the stigma (37.5% females). Number of cells posterior to Cu, in front wing, 13 (16.7% males), 14 (33.3% males, 12.5% females), 15 (50% males, 75% females), or 16 (12.5% males); in hind wing, 11 (16.7% males), 12 (50% males, 12.5% females), 13 (16.7% males, 37.5% females), 14 (25% females), 15 (16.7% males), or 16 (25% females). Material examined: French Guiana, Pied Saut, Oyapok River, November, 1917, S. M. Klages, Carn. Mus. Ace. 6111, 1 male, 3 females, the male the type and one female the allotype; Tamanoir, Mana River, June, 1917, S. M. Klages, Carn. Mus., 1 male; St. Laurent du Maroni, coll. K. J. Mor- ton, 1 male, 1 female. None of the above material is perfect and all appear somewhat teneral or discolored. It is therefore probable that the color description will be found to be inaccurate for mature and well preserved or freshly killed specimens. At the same time the structures are well enough preserved in the material available for this paper and there should be no trouble in the future in recognizing the species. Hagen describes fragilis from two females, a larger from Congonhas, Brazil, and a smaller from the Essequibo, British Guiana. The larger Brazilian specimen evidently served largely, if not entirely, as the basis of both the generic and specific descriptions. His generic description will apply probably only to Group II, as defined in the synoptie key in this paper. Our material from the Guianas is very incomplete, but there is no reason to think Hagen’s two females are conspecific and it seems better to retain the name fragilis for the southern Brazilian species to which Hagen’s description better applies, and which is therefore the type of the genus, and to give a new name to the only species certainly known from the Guianas. This species, guzanensis,- belongs to our Group I, and its nearest relative is probably magdalenae from the lower Magdalena in Colombia. 3. Perilestes remotus, new species Abdomen male 50-54, female 41.5-43.5; hind wing male 23.5-25.5, fe- male 25-26; stigma front wing male 1.05-1.10, female 1.10-1.30; stigma hind wing male 1.20-1.30, female 1.27-1.40; abdominal appendage male 1.40, female .40-.46; male anterior lamina from apex of sternum of seg- ment 1 to apex of the posterior process 1.8-2.0; female valve 3.0. Male.—Labium pale, apices of median lobe, end hook and movable hook THE GENUS PERILESTES 15 shaded darker; labrum and postclypeus black, the anteclypeus paler, dull greenish or brown, the labrum sometimes with more or less brown. Prothorax brown above, narrowly on the front lobe, the sides pale, pro- epimeron largely brown; hind border shallowly trilobed. Thorax above black, middorsal carina very narrowly brown, a narrow pale greenish or yellowish antehumeral stripe, not reaching the antealar sinus above, and projected across the prothorax below, about .3 wide at the widest part; mesepimeron largely black, the same broad band of color carried across the mesinfraepisternum; metepisternum largely pale yel- lowish or flesh colored except above where a broad black band on the sec- ond lateral suture crosses it to join the black on the mesepimeron; metepi- meron pale colored like the metepisternum except where the pale area is encroached on by the broad black band on the second) lateral suture, a limited area of dark to black above and below, and a narrow edge of black on the latero-ventral carina posterior to its angle; the black stripe on the second lateral suture is carried across the metinfraepisternum leaving only its ‘lower angle pale. Metasternum black. Mesostigmal lamina elevated into a low lobe just external to the end of the fork of the middorsal carina; below (external to) this lobe the lamina is low and nearly symmetrically convex. Coxae pale, marked with dark to black; on the anterior face the first coxa has a mere trace of dark, the second nearly the inner half dark, and the third all dark but a narrow outer border; the posterior face of each coxa darker than the anterior face, and again the first is the palest and the third the darkest. Femora black, the last two and rarely the first each with an anteapical anterior pale spot. Tibiae black with the anterior surface pale. Venation black, main veins shading out, especially on the ventral sur- face, to nearly white at base; stigma black, brown (but not reddish brown) in tenerals. Abdomen seen from above: segment 1 pale, a more or less distinct nar- row longitudinal middorsal stripe and an apical ring, brown; 2 dark brown except at base and a longitudinal middorsal pale stripe, not reaching the apex; 3 with base pale, followed by a slightly darker area, followed by a slightly more extensive pale area, which darkens posteriorly into a dark area nearly or quite black at its middle, which occupies about one-third or slightly less of the segment, and which fades out posteriorly into a pale area about equal in extent to the pale area which bounds it anteriorly; this pale area darkens posteriorly into a more extensive black area which reaches to the apex; 4-6 similar to 3 but the pale basal area (a smooth area lacking minute transverse striae) is scarcely paler than the dark area which follows it, the large median dark area is relatively more ex- tensive and the anteapical pale area which separates it from the apical black is progressively reduced from 4 until it has quite disappeared on 6; the postbasal pale area is about the same extent on each of segments 3-6; 7-10 black. Seen from the side: 1 pale except at apex; 2 pale below except a basal spot at the inferior border; 3-6 as in dorsal view, except the post- 16 E. B. WILLIAMSON AND J. H. WILLIAMSON basal pale area on each is more extensive along the ventral border than in the middorsal line, and it is a conspicuous lateral spot on 7 on which segment it is not visible in dorsal view; 8-10! black, the apical ventral angle of 10 pale. Sternum of 1 largely or entirely dark; anterior lamina and posterior processes of 2 largely dark; lateral color pattern of 3-7 car- ried across the sterna; sternum of 8 largely dark, pale basally and apically; sterna of 9 and 10 bright pale colored, yellow (?) or more probably blue (?) in life. Appendages nearly white to pale reddish horn color. Length of the wing on lobe on the expanded postero-ventral margin of the tergum of 2 about .6, and the distance from the apex of this lobe to the antero-ventral angle of the tergum about 2.0. Female.—Head similar to that of the male. Prothorax dull pale color without definite or extensive markings, the dorsum of the two posterior lobes dusky, proepimeron largely dusky to brown; hind lobe as in the male. Thorax patterned similarly to that of the male, but the black is replaced by dark reddish brown except a narrow black border to the pale middorsal stripe, part of the extreme lower end of the broad stripe which covers the mesepimeron, and the broad stripe on and just anterior to the second lateral suture which remains distinctly and conspicuously black; there is a mere line of black above on the metepimeron and none at all on the latero-ventral carina. Metasternum pale, an indefinite trace of brown basally in the median line. Mesostigmal lamina similar to that of the male. Coxae pale, a small dark area on the anterior face of the third. Femora largely pale, obscurely darker above, black beneath and at the apex. Tibiae pale, a narrow dark streak beneath, and the first tibia narrowly dark streaked above. Venation and stigma similar to those of the male. Abdominal segments 1—5 colored about as in the male; in 6 the dark median area is not black in its darkest portion and it fades out posteriorly into a light reddish brown area which passes into darker apical brown, but not black; similarly 7 and 8 are reddish brown, instead of black as in the male, though there is an indefinite trace of black in the middorsal line of 8; 9 with a dorsal longitudinal ill defined dark bar about .8 wide at base, bounded on either side basally with blue for about two-thirds its length, widening apically and becoming paler below in its apical third till it fades out with the lateral basal blue spot in a general loss of color pattern along the ventral border, 10 black or nearly so. Sternum of 1 pale, of 2 largely dark, of 8 largely blue, apparently: the same color in life as the large lateral basal spot on 9. Appendages dark brown, black at apex, flattened cone-shaped, acute. Valves long, the valvular styles at their point of attachment distal to the apices of the appendages. Venational characters based on four males and four females Postnodals front wing, 13 (12.5% males), 14 (25% males, 25% females), THE GENUS PERILESTES 17 15 (50% males, 12.5% females), 16 (12.5% males, 50% females), or 17 (12.5% females); hind wing, 12 (25% males, 25% females), 13 (50% males, 12.5% females), or 14 (25% males, 62.5% females). M, in front wing arising at the first postnodal (12.5% males), or be- tween the first and second postnodals (87.5% males, 100% females) ; in the hind wing at the first postnodal (37.5% males) or between the first and second postnodals (62.5% males, 100% females). Rs in front wing arising at the seventh postnodal (50% males, 62.5% females), or at the eighth postnodal (50% males, 37.5% females) ; in the hind wing at the sixth postnodal (25% males, 12.5% females), at the seventh postnodal (62.5% males, 87.5% females), or at the eighth post- nodal (12.5% males). M, in front wing arising at or near the seventh postnodal, (12.5% males), at or near the eighth postnodal (50% males, 37.5% females), at or near the ninth postnodal (25% males, 62.5% females), or at or near the tenth postnodal (12.5% males) ; in the hind wing at or near the sev- enth postnodal (50% males, 25% females) or at or near the eighth post- nodal (50% males, 75% females). M,. in front wing arising three cells before the stigma (62.5% males, 37.5% females), four cells before the stigma (25% males, 62.5% females), or five cells before the stigma (12.5% males) ; in the hind wing two cells before the stigma (25% males), three cells before the stigma (62.5% males, 50% females), or four cells before the stigma (12.5% males, 50% females). Number of cells posterior to Cu, in front wing, 14 (25% females), 15 (12.5% males), 16 (50% males, 25% females), or 17 (37.5% males, 50% females); in hind wing, 12 (12.5% males, 12.5% females), 13 (50% females), 14 (75% males, 37.5% females), or 15 (12.5% males). The following brief color note was made on a male taken at Cristalina: eyes above dark green, lighter below, light thoracic markings very pale greenish yellow, almost white. Teneral specimens are entirely without black, being dull brown with obscure paler markings. Probably all de- grees of color between this dull brown and the clear bright black of the fully adult specimens are passed in the life period. Specimens otherwise apparently mature have been noted with traces of brown in the labrum and with the black of the mesepimeron duller above, shading out into brown. Material examined: Canal Zone, Panama, Rio Sardanilla, December 5, 1916, 1 male, 1 female, both tenerals; Rio Mazamba, December 6, 1916, four males, all tenerals; Colombia, Cristalina, Dept. Antioquia, Quebradas Cris- talina (73 specimens), Sabaleticus (44 specimens), and La Camelia (6 speci- mens), February 12-19, 1917, 116 males, 7 females, type male Quebrada Cristalina, February 12, allotype female Quebrada Sabaleticus, February 14, all in coll. E. B. W. This is the only species of Perilestes which has reached Central America. At Cristalina, Colombia, it was associated with the much rarer magdalenae. 4. Perilestes magdalenae, new species Abdomen male 49-52.5, female 46 ; hind wing male 23.5-25, female 26-27 ; 18 E. B. WILLIAMSON AND J. H. WILLIAMSON stigma front wing male 1.07—1.20, female 1.27-1.40; stigma hind wing male 1.27-1.40, female 1.47-1.50; abdominal appendage male 1.5-1.6, female .35-.40; male anterior lamina from apex of sternum of segment 1 to apex of the posterior process 1.55-1.60; female valve 2.14—2.26. Male.—Head similar to that of remotus. Prothorax similar to that of remotus, but slightly paler, the hind border straighter, the median and lateral lobes less evident. Thorax similar to that of remotus but duller; mesepimeron largely brown instead of black; the black stripe on the second lateral suture is slightly narrower and at the metinfraepisternum it is not joined across the metepi- meron with the ventral black as it is in remotus. Metasternum black as in remotus, but with a pale median longitudinal area of variable width from the level of the angle of the latero-ventral carinae to near the apex of the metasternum. Mesostigmal lamina low, nearly straight, a short shallow concavity just before its lower end. Coxae pale, a triangular dark area on the anterior face of the third, and the posterior face of the second and third largely dark. Femora black but with the anteapical pale spot of remotus enlarged to an anterior pale stripe on each femur from the coxa quite to or near the apex. Tibiae as in remotus. Venation and stigma as in remotus. Abdomen seen from above: 1 with a large squarish median basal brown spot reaching about two-thirds the length of the segment, narrowly separ- ated from an apical brown area which joins the apical ring of dark brown or black; 2-6 about like 2-6 in remotus, but the paler area on each seg- ment, except the postbasal, is slightly more extensive on 3-6 in magdalenae ; the postbasal pale areas on 7 are more nearly joined in the median line in magdalenae so that 7 is about the same pattern in magdalenae as 6 is in remotus; 8-10 black. Seen from the side: 1-7 as in remotus, but the basal ventral spot on 2 is reduced to an inconspicuous small brown area in mag- dalenae and the pale areas, except the postbasal pale areas, are more ex- tensive in magdalenae; 8-10 as in remotus. Sternum of 1 pale, brown at base, shading out into the pale color about midleneth of the sternum; anterior lamina as in remotus; lateral pattern of 3-7 carried across the sterna; 8-10 and color of appendages as in remotus. Length of the wing or lobe on the expanded postero-ventral margin of the tergum of 2 about .7 and the distance from the apex of this lobe to the antero-ventral angle of the tergum about 1.9. Female.—Head similar to that of the male. Prothorax largely pale, a longitudinal median pale brown stripe and proepimeron largely brown; dorsum of hind lobe, against the posterior border, elevated into an erect, acute, median spine about .5 long; lateral margin of hind border rather prominent, the supero-lateral angle about 90°. Thorax similar in color pattern to that of the male, but the black is replaced by dark reddish brown except a narrow black border to the pale middorsal stripe, and the stripe on the second lateral suture which re- THE GENUS PERILESTES 19 mains distinctly and conspicuously black; ventro-lateral carina and meta- sternum entirely pale except for the merest trace of a brown line on the median suture from the third coxae to the level of the angle of the latero- ventral carinae. Mesostigmal lamina low, the edge nearly uniformly curved. Coxae pale, a small dark aréa on the anterior face of the third. Femora largely pale, obscurely darker above, black beneath and at the apex. Tibiae pale, the first tibia narrowly dark streaked above and all dark streaked beneath. Venation and stigma similar to those of the male. Abdominal segments 1-6 patterned about as in the male but with the median dark area of 4-6 paler, darkest immediately following the post- basal pale area, on 6 searcely darker than the anteapical pale area; the postbasal pale area on 47 is slightly reduced in area in its dorsal but not in its lower ventral part as compared with the male and the apical dark area grows progressively paler posteriorly so it is brown instead of black on 6; 7 reddish brown, the postbasal pale areas widely separated in the median line; 8-10 black above, on 9 reduced to a well-defined median longitudinal bar about .8 wide basally which widens laterally, at about two-thirds the length of the segment, and passes into brown on the sides, terminating at the apex at about midheighth of the segment; this black bar is bounded basally on either side by a large blue sharply defined lateral spot; sides of 9 largely blue, the ventral border at midlength narrowly black, from which point the ventral black widens in both directions to base and apex giving a circular form to the lower edge of the lateral blue spot. Sternum of 1 largely pale, of 2 largely black, of 8 largely pale (blue? in life). Appen- dages pale brown, black at extreme apex, flattened cone-shaped, acute; as compared with those of remotus they are less tapering in the apical two- thirds or three-fourths and less acute,—evident differences when direct comparisons are made but difficult to describe. Valvular styles at their point of attachment about at the level of the apex of 10. Venational characters based on four males and two females Postnodals front wing, 13 (50% males, 25% females), 14 (50% males, 25% females), 15 (25% females), or 16: (25% females); hind wing, 10 (12.5% males), 11 (37.5% males), 12 (37.5% males), 13 (12.5% males, 75% females), or 14 (25% females). M, in both front and hind wings arising between the first and second post- nodals (100% males, 100% females). Rs in front wing arising at the seventh postnodal (75% males, 25% females), or at the eighth postnodal (25% males, 75% females) ; in the hind wing at the sixth postnodal (37.5% males), at the seventh postnodal (62.5% males, 75% females), or at the eighth postnodal (25% females). M, in front wing arising at or near the seventh postnodal (12.5% males), at or near the eighth postnodal (62.5% males), at or near the ninth post- nodal (25% males, 75% females), or at or near the tenth postnodal (25% females) ; in the hind wing at or near the seventh postnodal (75% males), 20 E. B. WILLIAMSON AND J. H. WILLIAMSON at or near the eighth postnodal (25% males, 75% females), or at or near the ninth postnodal (25% females). M,. in front wing arising at the proximal end of the stigma (12.5% males), two cells before the stigma (62.5% males, 100% females), or three cells before the stigma (25% males) ; in the hind wing one cell before the stigma (37.5% males), two cells before the stigma (62.5% males, 75% females), or three cells before the stigma (25% females). Number of cells posterior to Cu, in front wing, 11 (25% males), 13 (50% males), 14 (12.5% males, 50% females), or 15 (12.5% males, 50% females) ; in hind wing, 9 (12.5% males), 10 (75% males, 25% females), 11 (12.5% males, 50% females), or 12 (25% females). At the time of capture magdalenae and remotus were not recognized as two species, and all our notes are on envelopes containing the much com- moner species, remotus. Material examined: Colombia, Cristalina, Dept. Antioquia, Quebradas Cristalina (3 specimens), Sabaleticus (3 specimens) and La Camelia (4 specimens), February 12-18, 1917, 8 males, 2 females, type male Quebrada La Camelia, February 18, allotype female Quebrada Cristalina, February 12; all in coll. E. B. W. This species is known only from the streams about Cristalina, Colombia, in the Magdalena Valley, where it was associated with the commoner remotus. 5. Perilestes attenuatus Selys Abdomen male 43.5-48, female 37-37.5; hind wing male 20—21.5, female 20.5-22; stigma front wing male 1.05-1.10, female 1.10-1.20; stigma hind wing male 1.15-1.20, female 1.20-1.27 ; abdominal appendage male 1.40-1.60, female .30-.35; male anterior lamina from apex of sternum of segment 1 to apex of the posterior process 1.30-1.40; female valve 1.95-2.20. Male.—Labium pale, apices of parts very slightly darker, except the mov- able hook which has the apical half or more black as in kahli; labrum and elypeus in front light blue, almost white, the former in front with a straight black border, occupying scarcely one-fourth the length of the labrum, clypeus above black, the line between the black and pale areas smooth. Prothorax pale, slightly darker above, especially posteriorly where it is pale reddish or rusty brown; hind border straightened above, very slightly trilobed, the lateral extremities almost vertical. Thorax above like that of kahli except that the antehumeral pale stripe reaches the upper end of the mesepisternum; side of thorax similar to that of fragilis except that the dark stripe on the second lateral suture is slightly wider throughout and widens below on the metepimeron into a more or less bilobed expansion; metinfraepisternum dark reddish brown above, shading into black below, the lower angle pale. Metasternum pale, a black longi- tudinal median line from the coxae to about two-thirds the length of the metasternum where it ends in a fork or this black line may reach nearly the apex of the metasternum, widened into a circular area at about two-thirds THE GENUS PERILESTES 21 its length, and again widened or not at its apical end; on either side a longitudinal black bar just inside the latero-ventral carina, reaching pos- teriorly nearly to the apex of the metasternum,: anteriorly it reaches a greater length, measured from the level of the angle of the latero-ventral earina, than its length posterior to that point, sometimes with a greater or lesser interruption nearly reaching the anterior suture which bounds the metasternum. Mesostigmal lamina low and of nearly uniform heighth throughout. Coxae pale, the second and third each with the anterior face broadly black. Legs pale, of all the species the most distinctly marked with darker though the general pattern is similar to that of kahli; second and third femora dark at base and all femora dark at apex, between base and apex each femur with two encircling bands of brown to black, these dark areas more extensive than the pale areas adjoining them; tibiae paler and clearer than femora, each dark at base, a narrow line of black on the ventral sur- face, widest on the first tibia. Venation like that of solutus; stigma dark brown or black. Abdomen seen from above: similar to that of kahli, 7 brown rather than black, and the dark colors on the preceding segments are slightly paler in attenuatus ; 8-10 brown, darker to black on the base of 9 and the apex of 10, a pale ring occupying the basal fifth of 8. Seen from the side similar to that of kahli, except that the dark color, adjoining the pale ventral margin of each of segments 3-6 is paler and duller, so the ventral margin is less contrastingly colored but the dark color is more extensively extended from the dorsum to the ventral margin so the pale basal and pale anteapical areas of each segment are more widely separated in attenuatus than in kahli; on 8 only the extreme ventral margin is pale, this widened slightly and briefly just before it passes into the pale basal ring. Sternum of 1 pale; 3 pale at base, passing insensibly into very pale brown which fades quickly into pale at the anteapical pale area, then passing posteriorly into more or less distinct blue and then brown; 4-6 similar to 3, the dark (brown) progressively darker from the anterior to the posterior segments, and the faint blue, sometimes following the anteapical pale area on 3, growing fainter and disappearing on 5 or 6; 7 dark brown or black, restricted pale (blue?) at base and apex; 8-10 bright light blue. Appendages black. Length of tne wing or lobe on the expanded postero-ventral margin of the tergum of 2 about .8, and the distance from the apex of this lobe to the antero-ventral angle of the tergum about 1.7. Female.—Labium as in the male; labrum black, a large dull yellowish rounded basal spot on either side, these spots very narrowly or scarcely separated in the median line; clypeus patterned as in kahli but the pale color on the front is a dull leaden blue or obscure. Prothorax similar in color and shape to that of the male. Thorax above like that of the male except that the black adjoining the middorsal carina is only about half as wide as in the male; side of thorax similar to that of the male except that the dark colors are slightly paler, the dark stripe on the second lateral suture is not or very slightly widened 22 E. B. WILLIAMSON AND J. H. WILLIAMSON below, and the dark color (brown) of the metinfraepisternum does not shade into black below. Metasternum pale, a black longitudinal median line from between the coxae posteriorly to about the level of the angle of the latero-ventral carinae ; likewise the lateral dark area at the latero-ventral carina is less developed in the female than in the male, fading cut into brown anteriorly and posteriorly, this brown fading imperceptibly into the general pale color. Mesostigmal lamina as in the male. Coxae pale, the third coxa with or without a brown area on the anterior face. Legs similar to those of the male, possibly slightly duller. Venation and stigma similar to those of the male. Abdomen seen from above: 1-4 similar to those of the male; 5 is narrowly pale at base, then very briefly black which fades into brown, which is con- tinuous to just before the apical black where it fades into the restricted paler anteapical area (in the male the dark area which occupies the larger part of the segment darkens posteriorly from brown, just following the postbasal black to very dark brown or black just anterior to the anteapical pale area); 6 is largely pale or dark brown, the area adjoining the pale basal area being scarcely or not at all darker, the anteapical pale area poorly defined, and the apical area being only dark brown, not black; 7 brown, pale at base, scarcely if any darker at apex; 8 brown; 9 dark brown to black with a lateral basal blue spot on either side, patterned as in kahli but with the dark median longitudinal bar about .6 wide at base and 1.0 wide where it joins the dark apex; 10 dark brown, almost black. Seen from the side: 1—4 similar to those of the male; 5-7 paler than in the male, same as in dorsal view; 8-10 patterned as in kahli, 8 brown, shading to black at the extreme postero-ventral angle; dark colors on 9 black or nearly so, 10 black or nearly so. Sternum of 1 and 2 pale, 2 with a postbasal dark stretch occupying about one third the length of the segment; 3-6 similar to those of the male but without traces, or very faint ones, of anteapical blue; 7 dark to black, broadly pale at base and less definitely anteapically; 8 blue with a median black area as in kahla. Appendages dark brown to black, shaped like those of magdalenae. Valvular styles at their point of attach- ment at about the level of the middle of the appendages. Venational characters based on four males and three females Postnodals front wing, 11 (25% males), 12 (37.5% males, 33.38% fe- males), 13 (25% males, 66.7% females), or 14 (12.5% males) ; hind wing, 10 (37.5% males), 11 (50% males, 33.3% females), 12 (12.5% males, 50% females), or 13 (16.7% females). M, in front wing arising at the first postnodal (50% males, 16.7% fe- males), or between the first and second postnodals (50% males, 83.3% females) ; in the hind wing at the first postnodal (62.5% males, 16.7% females), or between the first and second postnodals (37.5% males, 83.38% females). Rs in front wing arising at the seventh postnodal (62.5% males, 16.7% females), or at the eighth postnodal (37.5% males, 83.3% females) ; in the THE GENUS PERILESTES 23 hind wing at the sixth postnodal (37.5% males), at the seventh postnodal (62.5% males, 83.3% females), or at the eighth postnodal (16.7% females). M, in front wing arising at or near the eighth postnodal (62.5% males, 16.7% females), or at or near the ninth postnodal (37.5% males, 83.3% females) ; in the hind wing at or near the seventh postnodal (37.5% males), at or near the eighth postnodal (50% males, 83.3% females), or at or near the ninth postnodal (12.5% males, 16.7% females). M,. in front wing arising at the distal end of the stigma (100% males, 83.3% females), or at the proximal end of the stigma (16.7% females) ; in the hind wing at the distal end of the stigma (100% males, 100% females). Number of cells posterior to Cu, in the front wing, 7 (12.5% males, 16.7% females), 8 (16.7% females), 9 (87.5% males), 10 (25% males), 11 (25% males, 16.7% females), or 12 (50% females) ; in hind wing, 3 (16.7% fe- males), 4 (37.5% males, 16.7% females), 5 (62.5% males, 33.3% females), or 6 (33.38% females). The following color notes were made from a male and female taken at Abuna, March 20. Male.—Eyes dull green above, slightly lighter below. Top of head metallic green with purplish reflections. Back of head metal- le grays and browns. Prothorax brown and transparent light color. Nar- row middorsal stripe light gray, bordered with black, in turn bordered with brown with metallic green reflections; a pearl colored antehumeral stripe; mesepimeron brown; posteriorly almost white with a brown stripe on the second lateral suture; very light below. | Legs white with black markings. Abdomen black, brown and white, 7 and 8 with areas of pale blue. Female.—Similar, thoracic colors the same but paler; blue areas on abdominal segments 7 and 8 paler. Material examined: Bolivia, Cashuela Esperanza, April 13, 1922, 1 male; Brazil, Abuna, Matto Grosso, March 16, 20 and 21, 1922, 4 males, 1 female; Porto Velho, Amazonas, May 2, 7, 18, 19, 23, and 24, 1922, 17 males, 2 females; Manaos, Amazonas, June 18, 1922, 2 males. ~ The reasons for assigning De Selys’ name to the above species are as follows: 1. His description agrees well, except that the female described by him has the abdomen longer than any of the few females we have seen. (It must not be overlooked that De Selys’ description is largely compara- tive with cornutus, though he speaks of it as fragilis.) 2. Our specimens are from widely scattered localities and there is no reason why the species should not be expected to occur at Santarem, the type locality. 3. More- over, attenuatus seems to be rather adaptable for a Perilestes as it occurs on sandy bottom creeks and on creeks with mud bottoms and banks, so there is this further reason for expecting to find it at Santarem. Attenuatus was the only species of Perilestes taken at Abuna, where it was found in woods along the creek at Km. 216. The first male seen was hanging with folded wings from the tip of a leaf eight feet above the ground; the second was flying in the woods a few inches from the ground. Likewise it was the only species taken at Cashuela Esperanza, where a single male was taken on a creek where two species of Heteragrion were common. At Porto Velho while cornutus was found only on the charcoal 24 E. B. WILLIAMSON AND J. H. WILLIAMSON burners’ creek, two and one-half miles east of town, attenuatus was found not only on the same ereek, but also on a creek crossed by the Rondon Telegraph Trail, and on a creek two feet deep and three to eight feet wide, with mud bottom and banks, flowing in a gully in the forest east of town. It was taken on this mud bed stream on May 19; on the same date minor was taken in the nearby jungle. At Mandaos attenuatus, the only species found there, was taken on a mixed gravel and clay bedded stream in heavy forest beyond Flores. 6. Perilestes kahli, new species Abdomen male 47-50, female 43 ; hind wing male 22-24, female 25; stigma front wing male 1.10-1.15, female 1.20; stigma hind wing male 1.20, female 1.30; abdominal appendage male 1.30-1.40, female .35; male anterior lamina from apex of sternum of segment 1 to apex of the posterior process 1.15- 1.20; female valve 1.90. Male.—Labium similar to that of solutus, the apical half or more of the movable hook black; face similar to that of fragilis, but the black on the apical part of the labrum is slightly produced basally in the median line and on the lateral margins. Prothorax pale, only slightly darker above; hind border flattened, above straightened, slightly rising from either side to a little higher at the center, the lateral extremities almost vertical. Thorax similar to that of fragilis with the black adjacent to the nar- rowly brown middorsal, carina from about as wide as the antehumeral yel- low stripe, (about two-fifths as wide as the reddish brown of the middle area of the mesepisternum) to fully occupying the area from the pale carina to the pale antehumeral stripe; mes- and metinfraepisternum extensively pale below and darker to black above to largely black with only the lower triangular area pale. Metasternum as in fragilis but the lateral dark areas may be the same or may be reduced and brown. Mesostigmal lamina low and straight above, the lower third elevated into a low symmetrical triangle. Coxae pale, the third with or without the anterior face largely dark to black. Femora pale, marked more extensively with dark than in fragilis to such an extent in the case of the Bolivian male that the color pattern of the first and second femora is obscured by the extensive black; first femur pale at base, the apical two-fifths to four-fifths of the dorsum brown or black with or without a trace of a pale anteapical spot on the antero-dorsal surface, black continued on the antero-ventral edge to near the base, dorsal surface with trace of a brown postbasal area; ventral surface pale, shading to brown apically or the entire surface black; second femur pale at base and black at apex with two brown areas between, each of which is more extensive than the pale areas adjacent to it and nearly or quite encircles the femur, the anteapical pale color, on the antero-dorsal surface, is pale blue or green; or the second femur is black with only a small pale postbasal and a small pale anteapical area on the antero-dorsal surface; third femur similar to the pale form of the second described above, but the basal pale THE GENUS PERILESTES 25 area is wanting, being occupied by the brown area which is postbasal on the second femur. Tibiae distinctly paler than the femora, very light yel- low, dark brown at base, marked as in fragilis. Venation and stigma as in solutus. Abdomen seen from above: 1 pale, only slightly darker above, the extreme apex brown, passing to black below; 2 black or brown, in the latter case pale at the base and passing into black at the extreme apex, in which case the brown is so pale basally that the median longitudinal pale area is ill defined in the dried specimen; 3—7 similar to those of fragilis, but slightly darker in the dark areas on the anterior segments and progressively more so on the posterior segments, so 7 is black with a pale basal ring; 8 black with the basal fourth or third pale (blue?), in the type specimen the black encroaching on the pale area in the median line with a triangular projec- tion; 9 and 10 black. Seen from the side: 1-7 similar to those of fragilis, but differing as follows: on 2 the dorsal dark does not extend quite so far on the side on kahli and the line of separation of the dark and pale is a straight line, not a dorsally convex line as in fragilis; segments 3-7 have the dark areas darker in color in kahli, on the posterior segments becoming black rather than brown, and they differ strikingly from fragilis in having the ventral margins more extensively pale, due to a complete or more nearly complete joining along the ventral margin of the pale basal area and the anteapical pale area, on 3 this joining is complete, on 4 there is a slight brief shading of brown just before the anteapical pale area, on 5 this brown is darker and more extensive, on 6 the dorsal black reaches the ventral margin along about the middle half of the segment and is thus nearly as extensive as in fragilis; 7 entirely black, excepting the lateral extension of the pale basal ring; 8 with the pale basal spot narrowing posteriorly but reaching the apex along the ventral margin; 9 all black with the merest streak of pale on the ventral margin; 10 with the ventral half pale. Ster- num of 1 pale; 3 dark, narrowed at the base, and interrupted at the ante- apical pale area; 4 dark to black, pale at base and at the anteapical pale area; 5 black, pale at base with a brown spot, and pale at the anteapical pale area; 6 similar to that of 5 but the basal brown spot larger and darker; 7 black, very briefly pale at base and apex; 8-10 bright light blue. Appen- dages black. ; Length of wing or lobe on the expanded postero-ventral margin of the tergum of 2 about .8, and the distance from the apex of this lobe to the antero-ventral angle of the tergum about 1.7. Female.—Labium as in the male; labrum black, a large dull yellow median geminate spot at the base; clypeus in front dull yellow, a vertical brown bar on either side equally distant from the median line and) the lateral margin; clypeus above black. Prothorax similar to that of the male but darker above, a round dark brown median spot on the front lobe, and the dorsum of the middle and hind lobes, excepting the elevated hind border of the latter, dark brown to black; hind border similar to that of the male but slightly more elevated. Thorax above with the dorsal black occupying one-half to two-thirds of 26 E. B. WILLIAMSON AND J. H. WILLIAMSON the mesepisternum, narrowly bordered along its outer edge with brown which fades into the pale but ill-defined antehumeral stripe; side of thorax as in the male, but the metinfraepisternum is entirely black except the lower nearly equilaterally triangular part. Metasternum as in the male but the lateral dark areas are larger and are black. Mesostigmal lamina as in the male. Coxae and legs similar to those of the male. Venation similar to that of the male, the stigma slightly duller, less reddish, brown. Abdomen seen from above: 1-5 similar to those of the male, but darker, the pale basal rings are as distinct as in the male but the anteapical pale area on each segment is less distinct, especially on the posterior segments, though the apical black, as distinguished from the color basal to it, is dis- tinct in all; 6 and 7 brown, pale ringed at base, and very slightly darker at apex; 8 and 10 black; 9 distinctly patterned, slightly more than the basal half pale (blue?) with a black median longitudinal bar from the base passing into the black which occupies the apical part of the segment; this black bar is about .4 wide at the base and about .7 wide where it passes into the apical black. Seen from the side 1-5 are similar to those of the male; 6 brown, pale basally and thence along the ventral margin for a short distance, and with ill defined traces of pale on the ventral margin at the region of the anteapical pale area; 7 brown, darker at the postero- ventral angle; 8 and 10 black, the ventral margin of 8 pale, widest at the base and tapering apically, not quite reaching the apex; 9 pale (blue?) with a triangular black area in the antero-ventral angle and with apical black, narrowing from the dorsum to the extreme postero-ventral angle, the black at base and the black at apex with their opposing lines parallel, that is, the pale area cuts diagonally across the segment. Sternum of 1 pale; 2—5 dark to black, pale at base and anteapically; 6 similar to the preced- ing segments but apparently lacking the anteapical interruption; 7 black, pale at base and extreme apex; 8 blue with a small median longitudinal dark area which reaches neither base nor apex, and is widest at mid-length. Appendages black, shaped like those of magdalenae. Valvular styles at their point of attachment slightly distal to the level of the apex of 10. Venational characters based on two males and one female Postnodals front wing, 12 (25% males), 13 (50% males), or 14 (25% males, 100% females) ; hind wing, 11 (50% males), 12 (25% males), or 13 (25% males, 100% females). M, in front wing arising between the first and second postnodals (100% males, 100% females) ; in the hind wing at the first postnodal (25% males, 100% females), or between the first and second postnodals (75% males). Rs in front wing arising at the seventh postnodal (50% males), at the eighth postnodal (25% males, 100% females), or at the ninth postnodal (25% males); in the hind wing at the seventh postnodal (100% males, 100% females). THE GENUS PERILESTES 27 M, in front wing arising at or near the eighth postnodal (25% males), at or near the ninth postnodal (50% males, 100% females), or at or near the tenth postnodal (25% males) ; in the hind wing at or near the eighth postnodal (100% males, 100% females). M,. in front wing arising at the distal end of the stigma (100% males, 100% females) ; in the hind wing at the distal end of the stigma (100% males, 100% females). Number of cells posterior to Cu, in front wing, 9 (25% males), 11 (25% males), 12 (50% males, 50% females), or 13 (50% females) ; in hind wing, 7 (50% males, 50% females), 8 (50% males), or 9 (50% females). Material examined: Brazil, Villa Murtinho, Matto Grosso, April 1, 1922, 1 male; Bolivia, Rio Japacani, East Bolivia, J. Steinbach, Carnegie Museum, Ace. 5574, 1 male, 1 female. The Brazilian male, coll. E. B. W., the type, and the Bolivian female, coll. Carn. Mus., the allotype. Named for Mr. Hugo Kahl of the Carnegie Museum, to whom we are indebted for many kindnesses, the last of which was the loan of all the specimens of Perilestes in the collections in his charge. Moreover, before this material was sent us, Mr. Kahl had studied the two Bolivian specimens and was unable to refer them to any described species. The single Brazilian specimen was taken at the creek at Km. 315 south of town. Associated with it on the same creek were the rare cornutus and the relatively abundant solutus. 7. Perilestes fragilis Hagen Abdomen male 50-52, female 50; hind wing male 25-26, female 27.5; stigma front wing male 1.35-1.60, female 1.50; stigma hind wing male 1.40-1.60, female 1.60; abdominal appendage male 1.50, female .40; male anterior lamina from apex of sternum of segment 1 to apex of the posterior process 1.30-1.40; female valve 1.85. Male.—Labium similar to that of solutus, apical parts not quite as dark; labrum light blue, almost white, a straight black bar on the anterior border, occupying about one-third) the labrum; anterior surface of the clypeus same color as basal two-thirds of labrum, dorsal surface black, the line separating the two colors smooth. Prothorax similar to that of solutus in color and shape. Thorax brown above with greenish reflections, shading to black along the middorsal carina which is very narrowly light brown; a pale yellow antehumeral stripe about .2 wide not reaching the upper end of the mesepi- sternum; mesepimeron reddish brown; metepisternum yellow or yellowish, with a dark brown stripe along the second lateral suture which widens above to extend across the sclerite while below it is carried across the metinfraepisternum by a broad area which includes all but the inferior angle of that sclerite; metepimeron yellow or yellowish, the same color as the metepisternum. Metasternum pale, with a median black line as in solutus and, in addition, with a black area on either side in the angle of the latero-ventral carina. 28 E. B. WILLIAMSON AND J. H. WILLIAMSON Mesostigmal lamina above low, gradually rising to near the lower end where it is rounded off so the lower third is symmetrically convex. Coxae pale, the third pair with a small black area on the anterior face. Femora pale yellowish, slightly tinged brownish at the base, and dis- tinctly black at the apex with two intermediate brownish rings or areas, about equally spaced, and each about the same extent as each of the three pale areas thus formed, the basal of these two darker areas the paler and less distinct; ventral surface black in the median line except at the base of the first femur and the apices of all. Tibiae pale like the femora, the first tibia with a narrow black line beneath, widening and more distinct at the comb, this line a mere trace on the second and third tibiae; a basal streak of brown on the dorso-anterior surface of all the tibiae, most distinct on the first, scarcely evident on the third. Venation like that of solutus; stigma black. Abdomen seen from above: 1 and 2 similar to those of solutus, but the median longitudinal pale bar on 2 shorter, only about two-thirds the length of the segment; 3-7 each with an obscure pale basal ring, darkest and least evident on 3, followed by brown which darkens posteriorly, becoming black on 4-6; 3-6 with this dark on each segment fading posteriorly into pale which in turn darkens into an apical black area about twice as extensive as the pale area which precedes it; 3-8 each with or without a median longitudinal thread line of pale which disappears before the apex of 8; 7 uniformly dark brown after the obscure pale basal ring, tending toward black along the middorsum; 8-10 dark brown, 8 the palest of the three, and 10 nearly or quite black. Seen from the side: 1 and 2 are similar to ‘those of solutus; 3-6 each pale at base and subapically along the ventral margin, these pale areas extending toward each other along the extreme ventral margin, and obscurely joined on 3, but progressively: more widely separated from 4 to 6 by the lateral extension of the large median dark area; 7 pale at base, then brown, the extreme apex usually darker; 8 with the entire side fading out to light bluish along the extreme ventral margin ; 9-10 black, 9 with the merest line of pale on the ventral border. Sternum of 1 pale; of 3 pale with a longitudinal streak, narrow and darkest at the base, then widening and fading out before the middle of the segment, darkening again to black in the apical black of the tergum; 4-6 each with the sternum pale at base, then darkening through brown to black which fades out to pale at or before the subapical pale area of the tergum, the apex as in segment 3; in life the pale areas on the sterna of 3-6 doubtless show more or less blue; 7 black, extreme base and apex apparently blue; 8-10 heavily pigmented, bright blue (or possibly yellow) in life. Appen- dages white to brown, the external edge brown to black. Length of wing or lobe on the expanded postero-ventral margin of the tergum of 2 about .8 and the distance from the apex of this lobe to tha antero-ventral angle of the tergum about 1.7. Female.—Labium as in the male; labrum yellowish brown, the apical half and the lateral margins black; clypeus in front greenish, a large brown area in the middle of each side, above dark brown to black. Prothorax as in the male. THE GENUS PERILESTES 29 Thorax similar to that of the male, the antehumeral yellow stripe very slightly wider. Mesostigmal lamina similar to that of the male. Legs similar to those of the male, the basal streak of brown on the dorso- anterior surface of the tibiae scarcely evident in the female. Venation similar to that of the male. Abdomen seen from above: 1 and 2 similar to those of the male but 1 paler at base; 3 black following the pale basal ring, fading out through brown beyond the midlength of the segment to the anteapical pale, the apical 1.5 m.m. black as in the male; 4 and 5 similar to 3; 6 brown rather than black postbasally, and the segment nearly uniformly. colored, the apical black and anteapical paler areas scarcely distinguishable; 7 and 8 similar to 6, slightly darker, the pale basal ring reduced on 7 and wanting on 8; 9 yellowish brown, narrowly and indefinitely black in the midline and on the apical edge, dark brown to black on either side apically, these lateral dark areas widely separated by the paler dorsal color; 10 black. Seen from the side 1 and 2 are similar to those of the male; on 3 and 4 the basal and subapical pale areas are joined along the ventral border but on 5 the median dark of the tergum separates them and on 6 the subapical pale area is scarcely or not distinguishable and the postbasal pale area is con- tinued very briefly along the ventral margin; 7 and 8 brown, slightly paler toward the ventral margin and basally on 7 and a small ventro-posterior brown area on 8; 9 pale yellowish brown, lightest along the ventral margin, apically dark brown or black for a width of about one-third m.m., this black narrowing ventrally, and with a small ventro-anterior brown area; 10 black. Sternum of 1 pale; 2 pale with a postbasal and an apical black or brown area; 3-6 similar, with the postbasal dark area on each extensive, occupying half the length of the sternum or more; 7 pale at base, then dark, apparently without an anteapical pale region; 8 pale (probably blue in life) with a narrow longitudinal median black line scarcely reaching base or apex and widest at about midlength of the sternum. Appendages black, shaped like those of magdalenae. Valvular styles at their point of attach- ment distal to the level of the apex of 10. Venational characters based on three males and one female Postnodals front wing, 13 (33.3% males), or 14 (66.7% males, 100% females) ; hind wing, 10 (16.7% males), 12 (66.7% males, 50% females), 13 (16.7% males), or 14 (50% females). M, in front wing arising at the first postnodal (50% males, 100% fe- males), or between the first and second postnodals (50% males); in the hind wing at the first postnodal (100% males, 100% females). Rs in front wing arising at the eighth postnodal (50% males, 100%! females), or at the ninth postnodal (50% males) ; in the hind wing at the sixth postnodal (33.3% males), or at the seventh postnodal (66.7% males, 100% females). M, in front wing arising at or near the ninth postnodal (33.3% males, 100% females), or at or near the tenth postnodal (66.7% males) ; in the 30 E. B. WILLIAMSON AND J. H. WILLIAMSON hind wing at or near the seventh postnodal (16.7% males), or at or near the eighth postnodal (83.3% males, 100% females). M,. in front wing arising at the proximal end of the stigma (83.3% males), or one cell proximal to the stigma (16.7% males, 100% females), in the hind wing at the distal end of the stigma (33.3% males), or at the proximal end of the stigma (66.7% males, 100% females). Number of cells, posterior to Cu, in front wing, 10 (50% females), 11 (16.7% males), 12 (83.3% males), 14 (50% females), 15 (33.3% males), or 16 (16.7% males) ; in hind wing, 6 (16.7% males), 8 (16.7% males, 100% females), 9 (16.7% males), 10 (16.7% males), 11 (16.7% males), or 12 (16.7% males). Material examined: Brazil, Rio de Janeiro, Alto Boa Vista Creek, below Alto Boa Vista, September 6 and 8, 1922, 3 males; Cascatinha Falls Creek, above Alto Boa Vista, September 14, 1922, 1 female. The larger of Hagen’s two female types of fragilis was taken at Con- gonhas, Brazil, a town about 250 miles north and a little west of Rio de Janeiro, and on the headwaters of the Rio San Francisco. This larger female type rather than the smaller female from the Essequibo in British Guiana, is the basis for Hagen’s generic and specific description. The description agrees well with the Rio female, and unless later collecting should reveal more than one species in the locality, the determination of the Rio specimens as fragilis may be considered certainly correct. Because of the wide separation of the locality from the Amazonian basin, it is im- probable that more than one species will be found near Rio in the future. 8. Perilestes solutus, new species Abdomen male 42-46, female 38-40; hind wing male 20-21, female 21-24; stigma front wing male 1.13-1.20, female 1.27-1.33; stigma hind wing male 1.12-1.20, female 1.27-1.80; abdominal appendage male 1.20-1.35, female .40; male anterior lamina from apex of sternum of segment 1 to apex of the posterior process 1.35-1.45; female valve 1.85-2.0. Male.—Labium pale, apices of median lobe, end hook and movable hook shaded darker; labrum, anteclypeus and the postelypeus narrowly in front and with or without a deep median rectangular indentation, bright blue, the labrum narrowly black margined in front. Prothorax pale, darker shaded above especially on the middle and hind lobes, the proepimeron largely light brown; hind border semicircular. Thorax black above with greenish reflections, the dark occupying about half or more of the mesepisternum, the middorsal carina narrowly brown; outer half or less of mesepisternum, except the extreme upper part which is black, pale brown or dark flesh with traces of bluish or yellowish; this pale color may fade out to a maximum along the humeral suture or there may be a well-defined narrow pale antehumeral stripe; mesepimeron red- dish brown, a long black spot above on the humeral suture and another shorter one above and posterior to it, with a pale (yellow?) area between them; the broad band of color of the mesepimeron carried across the mesin- THE GENUS PERILESTES 31 fraepisternum ; metepisternum pale, flesh colored below and yellowish above and behind where there is a black stripe along the second lateral suture; at midheighth this black stripe is about .1 wide, above it leaves the suture and; angles obliquely forward and upward about half across the upper end of the metepisternum, below it ends against the metinfraepisternum ; metepimeron pale flesh, almost white, bright rust colored below, this color continued across the upper part of the metinfraepisternum where it passes, at the extreme antero-ventral angle, into black. Metasternum pale like the metepimeron, a black longitudinal median line from the hind coxae posteriorly to nearly the level of the angle in the latero-ventral carina. Mesostigmal lamina very slightly concave above throughout its length, at its outer end produced into a free triangular outwardly directed point. Coxae pale, the third pair shaded a darker rust color on the anterior face. Femora pale, tinged yellowish or reddish, black at the apex, this black carried basally a third to a half the length of the femur on the anterior face of the first and second femora, the first femora the darker. Tibiae pale reddish like the femora, the anterior face paler and yellowish rather than reddish. Venation black, veins shading out to light brown at the wing base; stigma reddish brown (Sanford’s brown, Ridgway). Abdomen seen from above: largely dark to black with restricted pale markings; 1 with a narrow median longitudinal black bar, joined at the apex with a narrow black transverse ring; 2 with a corresponding but wide black bar which laterally extends far below the level of the black on 1 and to the level of the black on 3, a narrow median longitudinal pale bar from base to about four-fifths its length ; 3-7 black, each with a narrow transverse pale blue or yellowish basal ring and a pale median dorsal stripe extending posteriorly from it, nearly to the apex on 3, and successively narrower and shorter posteriorly till on 7 it is scarcely evident or may occupy as a maxi- mum one-third the length of the segment; 8-10 unmarked, with restricted traces of paler color on the base of 8. Seen from the side: 1 pale except the dark apical ring; 2 with about the ventral half pale, the ventral pale and dorsal black areas meeting in a nearly straight line; 3-7 each with the ventral margin narrowly pale from the base, where it joins the pale basal ring, to near the apex; 8 has a pale ventral border from base to apex but in this case the color is distinctly blue, not merely pale or yellowish as on 3-7, though in life 7 at least may show the same blue color; 9 has the ex- treme ventral edge blue, 10 is blue below, the dorsum black. Sternum of 1 pale; of 3 pale with a longitudinal black streak, widening into a bar, from the base to near the apex of the lateral pale color of the tergum where it fades out to again reappear in the apical black of the tergum; or the dark color may extend continuously from base to apex without any pale inter- ruption ; 4-6 similar, but there may or may not be a short postbasal inter- ruption of the black streak of about the same extent as the anteapical in- terruption; 7 blue with a median streak corresponding to that on 6, but lacking in the apical part; 8-10 blue, densely pigmented, usually faded to 32 E. B. WILLIAMSON AND J. H. WILLIAMSON leaden grey in dried specimens. Appendages black, at thin places dark reddish horn showing through. Length of the wing or lobe on the expanded postero-ventral margin of the tergum of 2 about .8, and the distance from the apex of this lobe to the antero-ventral angle of the tergum about 1.8. Female.—Labium as in the male; labrum and elypeus brown, the labrum narrowly black margined in front and the anteclypeus with a median spot and the lateral margin pale (bluish?). Prothorax as in the male. Thorax similar to that of the male with the following differences: mid- dorsal black reduced to a median bar about .5 wide, bordered on either side with broader brown, fading out into a pale bar along the suture, the latter probably blue in life; mesepimeron lacking the dorsal posterior black spot of the male; the bright rust less evident on the metepimeron. Meso- stigmal lamina similar to that of the male. Coxae similar to those of the male, the rust color scarcely evident on the third pair. Femora similar to those of the male, slightly paler and less reddish. Tibiae pale yellowish, the first pair dark beneath along the comb. Venation similar to that of the male; stigma slightly duller. Abdomen seen from above similar to that of the male with the following differences: dorsal bar on 1 fading out and disappearing at the base; on 3-5 the dark dorsal area on each segment is black only at its base against the basal blue ring, shading at once posteriorly into brown, which brown darkens anteapically into black which is more extensive than the basal black; 6 is brown throughout in the dark area, without black, but the apex darker; 7-10 brown, 9 darker, each side with a large lateral blue spot, the brown between them at base about .8 wide; in dorsal view the blue spot disappears at about two-thirds the length of the segment. In side view 1-5 patterned about as in the male but duller; 6 brown, except the pale base which is carried along the ventral border a short and indefinite dis- tance; 7, 8 and 10 brown; 9 largely blue, the antero-ventral area brown, and apically the brown dorsum is carried onto the side in an obscure ill- defined pattern, but does not reach the ventral margin which is pale throughout except at the base. Sternum of 1 pale; 2-5 each similar to 3 of the male; 6 and 8 each similar to 5 but no anteapical fading out of the black; 7 is like 6 or has the black fading out at the apex. Appendages black, flattened cone-shaped, acute, similar in shape to those of magdalenae. Valvular styles at their point of attachment at about the level of the apex of 10 or very slightly more proximal. Venational characters based on four males and four females Postnodals front wing, 11 (100% males), 12 (25% females), 13 (62.5% females), or 14 (12.5% females) ; hind wing, 10 (25% males), 11 (75% males, 12.5% females), or 12 (87.5% females). M, in front wing arising at the first postnodal (75% males, 50% females), or between the first and second postnodals (25% males, 50% THE GENUS PERILESTES 33 females) ; in the hind wing at the first postnodal (62.5%! males, 87.5% females), or between the first and second postnodals (37.5% males, 12.5% females). Rs in front wing arising at the sixth postnodal (87.5% males, 25% females), at the seventh postnodal (12.5% males, 62.5% females), or at the eighth postnodal (12.5% females) ; in the hind wing at the fifth postnodal (12.5% males), at the sixth postnodal (87.5% males, 75% females), or at the seventh postnodal (25% females). M, in front wing arising at or near the seventh postnodal (100% males, 25% females), or at or near the eighth postnodal (75% females) ; in the hind wing at or near the sixth postnodal (25% males, 12.5% females), at or near the seventh postnodal (75% males, 75% females), or at or near the eighth postnodal (12.5% females). M,. in front wing arising one cell distal to the stigma (25% males), at the distal end of the stigma (75% males, 50% females), at the proximal end of the stigma (37.5% females), or one cell proximal to the stigma (12.5% females) ; in the hind wing one cell distal to the stigma (37.5% males), at the distal end of the stigma (62.5% males, 87.5% females), or at the proximal end of the stigma (12.5% females). Number of cells posterior to Cu, in front wing, 9 (12.5% males, 12.5% females), 10 (12.5% males), 11 (87.5% males, 25% females), 12 (37.5% males, 25% females), 138 (12.5% females), or 14 (25% females) ; in hind wing, 7 (12.5% males), 8 (75% males, 75% females), or 9 (12.5% males, 25% females). The following living color notes were made from a male taken at Villa Murtinho: Eyes black above, passing into blue which in turn shades out to eray-green below; top of head metallic green. Face markings pale blue. Middorsal carina narrowly light brown, dorsum dark metallic green, bor- dered on either side with blue; mesepimeron brown with reddish reflec- tions; metepisternum light green; metepimeron very pale blue; below translucent yellow and reddish brown. Legs reddish brown, dark brown at joints. Abdomen above black with white or light blue basal rings; sides brown, with blue on 2 and with blue basal rings; beneath light and dark brown with some bluish on 7 and 8-10 light blue. Material examined: Brazil, Chapada, Matto Grosso, H. H. Smith, 2 males, 2 females (Calvert, Od. Neotropical Region), coll. Carn. Mus.; Villa Mur- tinho, Matto Grosso, creek at Km. 310 north of town, March 30 and April 3 and 7, 1922, creek at Km. 315 south of town and on a small tributary of it, March 31 and April 1, 4 and 6, 1922, and at the next creek south of the creek at Km. 315, March 31, 1922, a total of 50 males and 13 females, type male and allotype female taken April 4, 1922, on a small tributary, about one foot wide and a few inches deep, of the creek at Km. 315; Belém, Para, igarpé beyond Souza ear line, August 14, 1922, 1 male; all the speci- mens except those from Chapada in coll. E. B. W. At the creek at Km. 315 the three species of Perilestes taken at Villa Murtinho were found together, but while solutus was relatively abundant 34 E. B. WILLIAMSON AND J. H. WILLIAMSON only a single male of kahli and two males of cornwtus were taken. Moreover the last two species were not taken elsewhere at Villa Murtinho. Specimens of both sexes of solutus may be specifically recognized at once by the free, outwardly directed inferior end of the mesostigmal lamina, the character to which the specific name refers. 9. Perilestes minor, new species Abdomen male 40-41; hind wing male 20; stigma front wing male 1.10- 1.20; stigma hind wing male 1.20-1.30; abdominal appendage male 1.20; male anterior lamina from apex of sternum of segment 1 to apex of the posterior process 1.30. Male.—Labium as in solutus ; labrum black, narrowly blue across the base, this blue indented at the middle with a rounded backward projection of the black; anterior surface of the elypeus bright blue, the dorsal surface black, the line separating the two colors smooth. Prothorax similar to that of solutus, but the proepimeron pale like the adjacent parts. Thorax similar to that of solutus, the dorsal black wider, and with the lateral margins convex rather than straight as in solutus, above adjacent to the antealar sinus fading out to brown; the pale color on the mesepi- sternum, adjacent to the humeral suture, is reduced to about half the width or less of the corresponding pale area in solutus, and is seareely .3 wide; side of thorax similar to that of solutws, but the black superior spot on the humeral suture is less marked in minor, and the spot posterior to this spot in solutus is wanting in minor. Beneath similar to solutus but the dark line on the metasternum reduced to an anterior black spot and a pos- terior brown spot. Mesostigmal lamina nearly straight above, the lower end an acute nearly right-angled angle, this angle the homologue of the projecting point which characterizes solutus. Coxae pale. Femora pale, extensively dark marked as follows: largely black beneath, reduced toward the base, apically extending more or less onto the anterior surface, dorsal surface with base and apex dark shaded and with two transverse brown bands, the five bands of light and dark between base and apex thus formed are subequal in extent; this pattern present but not as well marked on the first femur. Tuibiae pale yellowish, black beneath on the first tibia along the comb. Venation similar to that of solutus; stigma brown to black, no trace of the reddish brown which is well marked in that of solutus. Abdomen seen from above: 1 pale brown, obscure and scarcely darker than the pale sides, this brown interrupting the apical black ring in the midline; 2 brown, slightly or scarcely darker at base and more distinctly so at apex, the two darker areas and the included paler area about equal in extent; 3-6 each with a narrow transverse basal pale blue ring, followed by brown which passes subapically into a narrow pale ring which is suc- ceeded by the two or three times more extensive apical black; the extensive brown area of each segment is darker, progressively from 3-6, until on THE GENUS PERILESTES 35 6 it is nearly or quite black and the anteapical pale area is almost obliter- ated ; on 3-6 the pale narrow median longitudinal stripe, present in solutus, is entirely wanting in minor, except for the merest trace of it on 3; 7 with a transverse basal ring, homologous and similar to that of the preceding segments, followed by brown or by black or almost black which passes into dark brown at the apex; 8-10 similar to those of solutus. Seen from the side, 1 and 2 patterned as in solutus, dorsum of 2 brown, not black as in solutus ; 3-6 with the pale basal ring produced posteriorly along the ventral border for only a short distance if at all, thus differing from solutus, but in minor the subapical pale spot, which in solutus is only the slightly more distinct apex of the pale ventral margin, is larger and is produced dorsally, while in solutus it is not so produced at all; on 6 the subapical pale area is obscure; 7 has a pale basal area and the ventral margin very narrowly pale, with this pale color widened very slightly subapically but not visible in dorsal view; 8 with a narrower pale basal ring but with the ventral margin broadly and distinctly pale for its entire length; 9 entirely dark to nearly black, the merest line of pale color on the extreme ventral margin; 10 largely blue, only the dorsum dark. Sterna of 1 and 3 similar to those of solutus, but brown on 3, rather than black; on 4-5 pale at base, shading into brown which is interrupted at the subapical pale spot of the terga; on 6 pale at base, shading into black which is continuous to the apex or may be narrowly interrupted subapically; on 7 broadly blue at base and ‘more broadly blue at apex, the intermediate region irregularly blotched with black; 8-10 entirely bright blue. Appendages colored as in solutus. Length of the wing or lobe on the expanded postero-ventral margin of the tergum of 2 about .7, and the distance from the apex of the lobe to the antero-ventral angle of the tergum about 1.6. Venational characters based on two males Postnodals front wing, 10 (25%), 12 (25%), or 13 (50%); hind wing 10, (50%), 11 (25%), or 12 (25%). M, in front wing and hind wing arising between the first and second postnodals (100%). Rs in front wing arising at the sixth postnodal (25%), at the seventh postnodal (50%), or at the eighth postnodal (25%) ; in the hind wing at the sixth postnodal (50%), or at the seventh postnodal (50%). M, in front wing arising at or near the seventh postnodal (25%), at or near the eighth postnodal (50%), or at or near the ninth postnodal (25%); in the hind wing at or near the seventh postnodal (50%), or at or near the eighth postnodal (50%). M,. in front wing arising at the distal end of the stigma (50%), or at the proximal end of the stigma (50%) ; in the hind wing at the distal end of the stigma (100%). Number of cells posterior to Cu, in front wing, 11 (75%), or 12 (25%); in hind wing, 9 (100%). Material examined: Brazil, Porto Velho, Amazonas, May 19, 1922, 2 males. one of these the type, coll. E. B. W. 36 E. B. WILLIAMSON AND J. H. WILLIAMSON The only two known males of this species were taken east of Porto Velho where the woods trail crossed a low stretch of ground. In January there was running water here but in May this had disappeared though the creek bed was damp and rankly overgrown, and only by stooping and crawling could one still follow it. In this jungle, associated with Subaeschna fran- cesca, minor was captured. The male abdominal appendages of minor are strikingly like those of solutus, from which species it is separated at once by the color of the face and the abdominal pattern, especially of segments 3-5. Minor is the smallest known species of the genus. ; ‘hole fader aa ; 7 ‘ . : F en } ” tar|i} A =| Tibet é oe Be ee af a f 7 -T | 7 ep: au F 38 E. B. WILLIAMSON AND J. H. WiiLIaMSON PLATE I Supero-internal views of tre left superior appendage of the males of the species of Perilestes. 1. P. cornutus, Porto Velho, May 2, 1922; 2. P. guianensis, type male; 3. P. remotus, type male; 4. P. magdalenae, type male; 5. P. attenuatus, Porto Velho, May 7, 1922; 6. P. kahlt, type male; 7. P. fragilis, Rio de Janeiro, September 14, 1922; 8. P. solutus, type male; 9. P. minor, type male. PERILESTES Puate [ De 1. P. cornutus 2. P. guianensis 3. P. remotus 133 2. © andeoalenae 5. P. attenuatus 6. P. kahli ne 7. P. fragilis 8.P.solutus ~ 9. P. minor “a yey ; ihe ‘ “fo - : c 5 “ ut \e : hs | a i: UNIVERSITY OF MICHIGAN, ie Ut MUSEUM OF ZOOLOGY. Miscellaneous Publications No. 1. Directions for Collecting and Preserving Specimens of Dragonflies for Museum Purposes BY EB. WILLIAMSON. + -- ANN ARBOR, MICHIGAN. PUBLISHED, BY THE UNIVERSITY. DECEMBER I, 1916, f Ve CANCE SAN a be Be UNIVERSITY OF MICHIGAN. - MUSEUM OF ZOOLOGY. Miscellaneous Publications No. 2. An Annotated List of the Odonata of Indiana BY kK, B. WILLIAMSON. ANN ARBOR, MICHIGAN. PUBLISHED BY THE UNIVERSITY. JUNE 5, 1917. Jb ah ‘ ety 4 Asti Seam J Vanes 3 aN ate as 7 My Pun RNP ay ith bi ee YR \ 7 uetyt ware yet) iy ) ¥ (@ \ ) ’ { ira oe i oe tes UNIVERSITY OF MICHIGAN MUSEUM OF ZOOLOGY Miscellaneous Publications No. 3 A Collecting Trip to Colombia, South America BY € E. B. WILLIAMSON ANN ARBOR, MICHIGAN PUBLISHED BY THE UNIVERSITY FEBRUARY 22, 1918 UNIVERSITY OF MICHIGAN MUSEUM OF ZOOLOGY Miscellaneous Publications No. 4 Contributions to the Botany of Michigan BY C. K. DODGE ANN ARBOR, MICHIGAN PUBLISHED BY THE UNIVERSITY FEBRUARY 23, 1918 yy, Ay oh i a i at Min awe 'y UNIVERSITY OF MICHIGAN MUSEUM OF ZOOLOGY Miscellaneous Publications No.5 Contributions to the Botany of Michigan, II BY C. K. DODGE ANN ARBOR, MICHIGAN PUBLISHED BY THE UNIVERSITY JULY 13, 1918 Pe ng et we ee : 7 <—e See =. — Se Sees aO=et SS = RS = Ren ; a a ae Sg oa . = SS ; = ~ = : — Ss = = = : _ * H) ki Res { i fk ie Hela at if - ht a 0 4 Nit as }) ayy! eI i i Han ‘ y Biv \ ay ih i i he fi Sn aE Te } ihe | ! Balt { iy Mp) Wy if Liye ui} ae ie cath aie f ty 7 i i: +) phy MAE TAT eed Wr bye 0.20 fi hi viet j q pal yet : ARIS Fy + ie UNIVERSITY OF MICHIGAN MUSEUM OF ZOOLOGY Miscellaneous Publications No. 6 _ASynopsis of the Classification of the Fresh- | Water Mollusca of North America, North of Mexico, A Catalogue of the More Recently Described Species, With Notes BY BRYANT WALKER ANN ARBOR, MICHIGAN PUBLISHED BY THE UNIVERSITY DECEMBER 30, 1918 ae cil + ieee aa A biel AN ig ¥ AS" No Hiv ap set ty ‘ DCPOP he ht | UNIVERSITY OF MICHIGAN MUSEUM OF ZOOLOGY Miscellaneous Publications No. 7 The Anculosae of the Alabama River Drainage BY CALVIN GOODRICH Published in Co-operation with the Geological Survey of Alabama ANN ARBOR, MICHIGAN PUBLISHED BY THE UNIVERSITY JULY I, 1922 UNIVERSITY OF MICHIGAN MUSEUM OF ZOOLOGY Miscellaneous Publications No. 8 The Amphibians and Reptiles of the Sierra Nevada de Santa Marta, Colombia BY ALEXANDER G. RUTHVEN WITH A DESCRIPTION OF THE REGION BY M. A. CARRIKER, JR. ANN ARBOR, MICHIGAN PUBLISHED BY THE UNIVERSITY SEPTEMBER 15, 1922 Cie be ih Ys AR gh ead UNIVERSITY OF MICHIGAN MUSEUM OF ZOOLOGY Miscellaneous Publications No. 9 Notes on American Species of Triacanthagyna and Gynacantha BY E. B. WILLIAMSON ANN ARBOR, MICHIGAN PUBLISHED BY THE UNIVERSITY JULY 2, 1923 or Wit ait w fet UNIVERSITY OF MICHIGAN MUSEUM OF ZOOLOGY Miscellaneous Publications No. 10 A Preliminary Survey of the Bird Life of North Dakota BY NORMAN A. WOOD ANN ARBOR, MICHIGAN PUBLISHED BY THE UNIVERSITY JULY 2, 1923 Spceargst . ry iy te eral hy iy Ww fh a 2 ‘ ik 2 # ‘ 9 “ N ; i a diy CAT ACRIED 4 wy Cie - DM sR yi i giv. ( ath as Une Aes ‘ ( UNIVERSITY OF MICHIGAN MUSEUM OF ZOOLOGY Miscellaneous Publications No. 11 Notes on the Genus Erythemis With a Description of a New Species (Odonata) BY E. B. WILLIAMSON ~The Phylogeny and the Distribution of the Genus Erythemis (Odonata) BY CLARENCE HAMILTON KENNEDY ANN ARBOR, MICHIGAN PUBLISHED BY THE UNIVERSITY JULY 14, 1923 404 He UNIVERSITY OF MICHIGAN MUSEUM OF ZOOLOGY Miscellaneous Publications No. 12 The Genus Gyrotoma BY CALVIN GOODRICH Published in Co-operation with the Geological Survey of Alabama ANN ARBOR, MICHIGAN PUBLISHED BY THE UNIVERSITY FEBRUARY 23, 1924 a eK Wt 4 $i y ae as 7 = a Ce scopes See oa = re UNIVERSITY OF MICHIGAN MUSEUM OF ZOOLOGY Miscellaneous Publications No. 13 Studies of the Fishes of the Order Cyprinodontes BY CARL L. HUBBS ANN ARBOR, MICHIGAN PUBLISHED BY THE UNIVERSITY January 18, 1924 Ay mie) i Fe an a Bee ie » NORA Wee Wa ativent nal Ae le Wi { a yaa aie Diy as 4 raat BANE tote UNIVERSITY OF MICHIGAN MUSEUM OF ZOOLOGY _ Miscellaneous Publications No. 14 The Genus Perilestes (Odonata) BY E. B. WILLIAMSON AND J. H. WILLIAMSON ANN ARBOR, MICHIGAN PUBLISHED BY THE UNIVERSITY JuLY 15, 1924 be bi Bi a Wy nyt Sa Y i a a , an all ele VK bate aed Gh tink ay Ouihie e Hodis tie Nis ne tu A Re PL AE 7 ; " , ; ‘ ip aa Ma wha ty ht j : tr an 7 ¥ 2 Bk eh) fl ae & { Pits u i ey. mt . i Pt apes if ; 40% Sy wy Y i j yee RS 4 ee te inn