» \sto-.n w?y h jo i®, z Vo's £ «F s | W s x^tP7 | s X I * Z o £ 05 ± 0) \ ± ,. THSON IAN “INSTITUTION NOlinilJLSNI NVINOSHIUAIS .S3 ! a V H 8 1 1_L I B R A R I ES ^SMITHSONIAN INSIST® . co Z — .... ^2 5 ■"-*'■ — z -'v ™ < •'V^S" ^ v S X o h \^4Mw z ^ s NSioc/ > '-#F' s > _ NOSHiiws^sa i a va a X u B RAR 1 EsX1THS0NIANlINST1TUT,0NtwN0linillSNLNvm°SHiltAJS^S3 ! dVS1 2 5 ITHSONIAN^INSTITUTION ~ NOlinillSNI^NVINOSHlIIAIS S3 I d Vd 9 11 LI B RAR I ES SMITHSONIAN INS 1 1 1 Ull t~ z r- ., z ■“ ^ m I m | *' n, 'XvHIl NOSHJLIWS^SH I a VH a n “LI B RAR I ESWSM!THS0NIAN“=INSTITUT10NWN0IinillSNl“NVIN0SHimS S3 I aVNi w z •, y> . z * <£> z ^ THSONIAN INSTITUTION NOIinillSNI_NVINOSHJLIWS S3 I a Va a 11 LI B RAR 1 ES~ SMITHSONIAN JNSTITl]! — to — ■'•*(/) — — •> vy z 0) r<-\ “ /JMirMPK H H /&5*»r^*A ^ /s/ , » Jx -h /b^ ,#/ ^ ' Z ~~ Q x,^vos^z — o ''V^' Z o ' £ o» £ c/>\“ ITHSONIAN~ INSTITUTION NOIlfUllSNI NVIN0SH1IWS S3 I avaa II LI B RAR I ES ^SMITHSONIAN INSTITf < XL 2 .< S -2: E = “i ./ jgy4 z -i , a-/ yX ' z -h 1 X 1 w 1 N5ssi57 I ^ I X X 1 1 noshiiws SHiavaan libraries smithsonian_institution NouniiiSNi nvinoshiiws saiavi 05 ^ 2 2 ,«X,' £C /o^ ’ 0 ITHSONIAN^INSTITUTION^NOlinillSNI^NVINOSHimS^Sa I ava 9 ll^LI B RAR I ES SMITHSONIAN INSTITU r- Z r~ j, z r* z A > m rn X#"' I rn S<^^/ \V'Y rn 05 ^ 0> — 05 _ ^ iNOSHims S3 1 ava an libraries Smithsonian institution NouniiiSNi nvinoshiiws^sh i av. 0> . z > w — z * <5 ^5 < 2 < X 05 o X 03 30 > llfUllSNl~NVINOSHll MS S3 I HVB 0 I1~L! B R AR I ES ' SMITHSONIAN "~ INSTITUTION ^ NOIlfUliSNI^NVINOSH < E , . V- ..^V\ - 5 E /- ^0? g = '«pr i ^ B RAR I ES^SMITHSONIAN^ INSTITUTION NOIlfUllSNI NVINOSHIIIAIS S3 I H V H 0 !1_ L 1 B R A R ! E s SMITHSO! — 03 ” 03 -s» _ q >\ ~ O lifllllSNl'^NVINOSHin/MS^S 3 IHVUan~LIBRARI ES^SMITHSONIANJNSTITUTION NOUfUliSNI^NVINOSH vS . ^ ^ Z (j) ^ CO BRAR I ESfy,SMITHSONIAN~INSTITUTION NOliniliSN! ~ NVINOSHIIIAIS S3 I H VH 8 H L! B RAR I ES^SMITHSC E w — z * <2 5 . £ < V s - - ^ l^ld^ o rn VS*» o IL, Jl x vl% 8 1 g ^ |- s >• HlfUllSNI^NVINOSHlIWS^Sa I H VH a ll\l B RAR S ES°°SM!THSONIAN INSTITUTION NOIlfUllSNI _ NVINOS! — •_ (O — to ^ w — . ~ v Z \, i jj ^STirTTJs. Z lj /SjM/a . fr XoaToX IjJ O BRARIES SMITHSONIAN INSTITUTION NOIlfUllSNI NVINOSHIIIAIS S3iavaail L I B R AR I ES SMITHS! —» ^ C/5 DlinillSNI- NVIN0SH1IWS SBIHVHOn LIBRARIES SMITHSONIAN INSTITUTION NOIlfUllSNI NVINOS -3S. f/T> 7 ... (/) v. . £/) ^ . < ^s°^X S ,| | /#S££\ 1 *jSfc s ^ =] * ;*\ir 8 «S*A I 8 •5 _ 5 X > X«Uiis«*X 2 -\ > 2E 2 co \ 2 ^ x z co BRARIES SMITHSONIAN INSTITUTION NOIlfUllSNI NVINOSHIIIAIS S3 I HVB 8 l"l_ Ll B RAR * Es SMITHS' c 5 5 ^ ™ o DiiniiisNi^NviNOSHiiiMs^sa i ava a n libraries Smithsonian institution NoiinuisNi nvinos r- , H r~ Z . . L_ __ ^ f: ^ BRARIES SMITHSONIAN INSTITUTION N0I1IUI1SNI NVINOSHIIIAIS S3 I H VB 3 11 1 1 B R AR I ES SMITHS g> ^ ^ Z r W ^ Z ^ % ,^v | V S s s - JW/- - ^ — /^sS*v°A z ,vX r. MX'/?,,, z H Division of Fishes, & t. National Museum BULLETIN OF THE UNITED STATES FISH COMMISSION. VOL. XIX, 1899. GEORGE M. BOWERS, Commissioner. WASHINGTON; GOVERNMENT PRINTING OFFICE. 1901. TABLE OF CONTENTS. Page. Bumpus, Hermon C. On the Movements of certain Lobsters liberated at Woods Hole 225-230 Clark, Hubert Lyman. The Synaptas of the New England Coast 21-31 Cobb, John N. The Lobster Fishery of Maine 211-265 Gorham, F. P. The Gas-bubble Disease of Fish and its Cause.: 33-37 Greeley, Arthur White. Notes on the Tide-pool Fishes of California, with a Description of Four New Species - 7-20 Green, Erik H. The Chemical Composition of the Subdermal Connective Tissue of the Ocean Sun-fish 321-324 Howe, Freeland. Report of a Dredging Expedition off the Southern Coast of New Eng- land, September, 1899 237-240 Jenkins, Oliver P. Descriptions of New Species of Fishes from the Hawaiian Islands, belonging to the Families of Labridse and Scaridte 45-65 Descriptions of Fifteen New Species of Fishes from the Hawaiian Islands 387-404 Jennings, H. S. Rotatoria of the United States with .especial reference to those of the Great Lakes '«* -. 67-104 A Report of Work on the Protozoa of Lake Erie, with especial reference to the Laws of their Movements *--- . 105-114 Jordan, David Starr, and John O. Snyder. Notes on a Collection of Fishes from the Rivers of Mexico, with Descriptions of. Twenty New Species 115-147 Kellogg, James L. The Clam Problem and Clam Otlltttrey. 1 39-44 Observations on the Life History of the Common Clam, Mya arenaria 193-202 Levene, P. A. Some Chemical Changes in the Developing Fish Egg 153-155 Linton, Edwin. Fish Parasites collected at Woods Hole in 1898 267-304 Parasites of Fishes of the Woods Plole Region 405-492 Mead, A. D. The Natural History of the Starfish 203-224 Hutting, C. C. The Hydroids of the Woods Hole Region 325-386 Shufeldt, R. W. Experiments in Photography of Live Fishes 1-5 The Skeleton of the Black Bass 311-320 Smith, Hugh M. Notes on the Florida Sponge Fishery in 1899 149-151 Tower, Ralph W. Improvements in preparing Fish for Shipment 231-235 Wheeler, William Morton. The Free-swimming Copepods of the Woods Hole Region 157-192 Woods Hole Biological Notes. No. 1 305-310 in LIST OF ILLUSTRATIONS. Facing page. Plate 1. The Marine Grotto in the United States Fish Commission building at Washington, U. C 1 2. Common Sun-fish (Eupomotis gibbosus) . ti 3. Long-eared Sun-fish ( Lepomis auritus). Group of Common Sun-fish ( Eupomotis gibbosus) C 4. The Large-mouthed Black Bass ( Micropterus salmoidcs). The White Perch (Morone americana) C 5. The Spotted Sea Trout or Squeteague (Cynoscion maculatum). The Tautog or Black-fish ( Tautoga oniiis) ... li 6. The Sea Bass ( Centropristes slriatus) - G 7. The Naked Star-gazer (Astroscopus guttatus). The Pike {Lucius lucius) 6 8. The Cat-fish (Ameiurus ncbulosus). The Brook Trout (Salvelinus fontinalis) - G 9. Young Rainbow Trout (Salnro irideus) G 10. (1) Synapta roseola. (2) Part of the same magnified four times. (3) Synapta inluerens 21 11. Figures illustrating parts of Synapta inhxrcns and S. roseola 24 12. Scup (Stenotomus chry sops). Three Views illustrating Gas-bubble Disease 33 13. Map showing Clam-fiats at Mouths of Essex and Castle Neck Rivers, Massachusetts 44 14-22. Figs. 1-46, illustrating Rotatoria of the United States, with especial reference to those of the Great Lakes. 104 23-26. Figs. 1-30, illustrating the Natural History of the Starfish 224 27. Map of the Region of Woods Hole, showing migrations of Lobsters 230 28. (1) The Sailing Smack Bar Bel , of Rockland. (2) The First Steam Smack to carry Lobsters in a Well 241 29. (1) The Steam Smack Mina and Lizzie landing her Cargo. (2) Fleet of Lobster Boats in Harbor 244 30. (1) Lobster Cars used in the Wholesale Trade at Portland. (2) Lobster Pots. (3) Fisherman’s Lobster Cars. 248 31. (1) Fishermen operating their Pots. (2) Inclosure for Live Lobsters at Vinal Haven, Maine 252 32. Boiling Live Lobsters preparatory to shipping in Ice, showing Boiler, Steam Tank, Cage, etc 254 33-43. Figs. 1-121, illustrating Fish Parasites collected at Woods Hole in 1898 304 44. Right Lateral View of the Entire Skeleton of Micropterus clolomieu 311 I-XXXIV. Figs. 1-379, illustrating Parasites of Fishes of the Woods Hole Region 392 TEXT FIGURES. Blennieottus recalvus 10 globiceps 11 Rusciculus rimensis 13 Dialarehus snyderi 15 Pseudoscarus jordani 64 Pseudocheilinus octotsenia 64 Views showing bands of teeth of Istlarius balsanus. . . 118 Istlarius balsanus 119 Notropis rasconis 121 Macropharyngodon aquilolo 46 Halichceres iridescens 47 lao 48 Coris lepomis 49 Hemicoris remedius 50 keleipionis 50 Thalas.soma pyrrhovinctum 51 Novacnlichthys woodi 52 calientis 122 Xystrosus popoche 123 Falcula chapalie 124 Characodon encaustus 126 Xenendum caliente 127 xaliscone 129 Pceeilia limantouri 130 Xiphophorus montezumae 132 Hemipteronotus umbrilatus 54 Iniistius leucozonus 55 verater 55 Cheilinus zonnrns 56 Anampses evermanni 57 Calotomus irradians 58 Cbirostoma Chapala? 135 Views of Head of Chirostoma promelas 136 Chirostoma diazi 138 erystallinum 138 Dorsal View of Head of Chirostoma erystallinum 139 Lateral and Dorsal Views of Head of Chirostoma gilberti 60 paluca 61 ahula 62 miniatus 63 Views of Head of Chirostoma lermse 142 Ciehlasoma steindachneri 142 Heros istlanus 145 Neetroplus earpintis 146 V VI LIST OF ILLUSTRATIONS. Calanus finmarchicus . minor Eucalanus attenuates monaehus Mecynocera clausii Paracalanus parvus Calocalanus pavo . . plumulosus Clausocalanus arcuicornis Centropages typicus liamatus 1 bradyi Temora longicornis Metridia hibernica Candace pectinata Labidocera cestiva Pontella meadii Anomalocera pattersonii Monops regalis Acartia tonsa Corynura bumpusii Oithona plumifera similis Setella gracilis Miracia eft'erata Clytemnestra rostrata Oncsea venusta - - Sapphirina gemma Coryceeus elongates. carinatus Mya arenaria. Ten camera outlines of shells Mya arenaria, with shell 0.4 mm. long Mya arenaria. Form 2.3 mm. long Lobster tag Lobster Pots, old style and patent head Left lateral view of cranium of Micrcpterus salmoides. Left lateral view of skull of M. salmoides Right lateral view of skull of M. dolomieu Left lateral view of mandible of M. salmoides Left outer aspect of upper jaw of M. salmoides Inner aspect of opercular bones of M. salmoides Inner aspect of left half of shoulder girdle Outer aspect of part of shoulder girdle Calva leptostyla , Cordylophora lacustris Syncoryne mirabilis Corynitis agassizii Bougainvillia carolinensis Perigonimus jonesii Eudendrium ramosum dispar carneum and tenue album Hydractinia polyclina Stylactis hooperi Pennaria tiarella Corymorpha pendula Tubularia couthouyi, larynx, and spectabilis crocea Hypolytus peregrinus Clytia bicophora and noliformis grayi Campanularia poterium, hincksii, and volubilis minuta and ed wardsi neglectaand verticillata amphora and angulata calceolifera and flexuosa Obelia flabellata and commissuralis [ Obelia diehotoma, geniculata, and gelatinosa 350 bicuspidata and longissima 351 Gonothyrrea loveni and tenuis 352 Ilebella pygmtea 353 Lovenella grandis 353 Opercularella lacerata 354 Calycella syringa 354 Lafoea dumosa and gracillima 356 Ilalecium halecinum, articulosum, and tenellum 357 beani and gracile 358 Sertularia pumila 359 cornicina and complexa 3G0 Diphasia fallax and rosacea 361 Sertularella abietina and tricuspidata 362 rugosa, polyzonias, and gayi 363 Thuiaria argentea : 363 cupressina and thuja , 364 Hydrallmania falcata 364 Monostsechas quadridens 365 Schizotricha tenella and gracillima 366 Antennularia antennina 366 americana and rugosa 367 Cladocarpus flexilis 368 Ilybocodon prolifer 371 Stomatocha apicata 371 Gemmaria cladophora 372 Corynitis agassizii 372 Perigonimus jonesii 372 Coryne mirabilis 373 Ectopleura ochracea 373 Pennaria tiarella 373 Dysmorphosa fulgurans 374 Stylactis hooperi 374 Turris vesiearia 375 Turritopsis nutricula 375 N emopsis bachei 376 Bougainvillia carolinensis and superciliaris 376 Lizzia grata 376 Orehistoma tentaeulata 377 Eutima mira 378 Hebella calcarata 378 Eucheilota duodecimals 378 Tima formosa 379 Obelia commissuralis and geniculata : . 380 Oceania singularis 380 Tiaropsis diademata 381 Epenthesis folleata . 381 Trachyriema digitale 382 Gonionemus vertens 382 Melicertum campanula , 382 Rhegmatodes tenuis 383 Sphyrtena helleri 387 snodgrassi 388 Anthias fuscipinnis 389 Aphareus flavivultus 391 Eupomacentrus marginatus 392 Chromis velox 393. Chsetodon mantelliger 394 sphenospilus 396 Ostracion camurum 397 Ovoides latifrons 398 Tropidichthys jactator 399 Eumycterias bitteniatus 400 Scorptenopsis cacopsis 400 eacopsis, dorsal view of head 401 Parapercis pterostigma 402 Brotula marginalis 403 165 165 166 166 167 168 169 170 171 172 173 174 175 176 177 179 180 181 182 183 185 1S7 187 1S8 1S9 189 190 191 192 192 194 195 195 230 248 312 313 314 316 316 317 318 319 327 328 328 329 330 330 331 332 333 334 335 336 336 337 339 340 340 343 344 345 346 347 347 348 349 J • X Bull. U. S. F. C. 1 899. (To face page 1 .) Plate 1 THE MARINE GROTTO IN THE UNITED STATES FISH COMMISSION BUILDING AT WASHINGTON, D. C. (Taken in July, 1897.) EXPERIMENTS IN PHOTOGRAPHY OF LIVE FISHES. By R. W. SHUFELDT. Captain , Medical Corps , United Stales Army. Up to the present time very few photographs of living fishes have been repro- duced and published, and, as compared with the photography of other living forms, attempts or successes in this line are extremely rare. There are a number of methods by means of which fish may be photographed in their natural element, with natural surroundings, as, for example, it is possible to accomplish it beneath the surface of the water by the use of some such contrivance as the subaquatic camera used by Dr. J. B. Romborsts, or that of M. Bouton, or the apparatus of Regnard. By the employment of instantaneous photography some fishes have been taken in the air, in the act of “leaping,” as in the case of salmon, or in the act of flight, as in the case of the flying-fish. Such pictures as these latter, however, illustrate peculiar habits rather than topographical characters of the forms thus secured. To obtain these we must resort to the photography of living fishes in suitable aquaria and under proper conditions. In such receptacles the types to be photographed may be taken either through the glass sides of the aquarium (with or without background) or the exposure may be made from above. This, of course, would depend upon the form of the fish and its habits in nature, or, in other words, whether the subject was a bass or a flounder. Again, certain fish in nature have the habit of occasionally resorting to the dry land, and when the opportunity offers species of this kind may be taken upon terra firma in various situations, as in the case of the peculiar gobioid Periophthalmus. The experiments to be described in the present contribution, however, will be restricted to a few the author has made at the aquaria of the U. S. Fish Commission building in Washington in July, 1897, and upon various occasions since. The fish in these cases were all medium-sized teleostean types, and the photographs were first taken through the glass sides of the aquaria in which they are kept in the “ Marine Grotto”; and afterwards in a special aquarium placed in the court yard of the build- ing during the forenoon of a perfectly clear day in July (1898) — two very different conditions. In the first instance the aquaria consist of a series of tanks arranged around a rootless corridor, thus admitting sunlight, when protective awnings are not in use, only from above. Within the grotto, this series of aquaria comes flush by glass-fronts with the wall of the long room, so named. Here they are of glass, 4 or 5 feet above the floor, and as one enters the grotto the impression is given of mural pictures wherein the fish-subjects are alive and moving about. The walls of the grotto 1 ' 9 BULLETIN OF THE UNITED STATES FISH COMMISSION. and its entrance are of tin, so modeled, painted, and sanded as to give the appearance of having’ been built in solid freestone. Practically all the light that gets into tin- place is through the glass fronts of the series of aquaria and the doorway passage It is an admirable arrangement and admits of the study of the forms of many kinds of fish and plants, and certain iuvertebrata as well. To a limited extent it likewise permits the study of some of the habits of the forms exhibited. To one having but little knowledge of the use of the camera, it would appear to be but a simple matter to photograph under such apparently favorable conditions, but such is by no means the case. In the iirst place, in most instances the incessant, rapid, and often erratic movements of the tish themselves have to be taken into account; the aquaria being large, we have in the second place the difficulty of prompt focussing to contend with, due to the latitude enjoyed by the smaller and more active forms. Thirdly, there is the question of reflection, and this, taken in connection with the light, is a serious problem. Reflections are especially troublesome, as the glass fronts of the aquaria receive them from all directions, so that, after focussing, a careful study of the image upon the ground-glass will show these reflections not only from some of the other aquaria, but possibly the photographer and his camera besides. All this must be carefully guarded against. In the early part of July, 1897, I made a number of attempts to photograph the fish contained in these aquaria through the glass-fronts, and in several instances I was successful. Where failure resulted it was due to some of the causes enumerated above, or, as in the case of a catfish, due to the high light upon the fish itself. High lights on the bodies of fish, if present at the time the exposure is made, will in the prints made from such a negative produce areas of white wherein all detail is absent. This is to be especially avoided, and often can only be overcome by shielding the aquarium from the suu above. An umbrella will in nearly all cases serve this purpose. The camera employed upon this occasion was an old-model Blair tomograph, with a Voigtlauder lens (No. 1) (27,967), an instantaneous shutter of the Low pattern, Seed’s gilt-edge plates (5 by 8). I used stops as any special case demanded. A tripod is absolutely essential to success in this kind of work. The instrument was set up in front of one of the more favorable aquaria and focussed upon the part desired and an inch or two beyond the surface of the glass. An armed plate-holder was inserted in place and the “ snap ” set. Patient waiting for an exposure when the fish swims to the place where you want it is necessary. Care must be taken in drawing or pushing back the slide to the plate-holder, and some of niy failures were due to complications of this nature. The first exposure was made upon a large pike [Lucius lucius), some 18 or 20 inches long and in good color and condition. It had a duration of about 2 seconds, at which time the plane of the left side of the fish’s body was nearly parallel to the plane of the glass, and about 3 inches from its inner surface. A quarter of an inch diaphragm was used, and the subject remained practically motionless during the time of exposure. Overhead the light was somewhat diffused, and an additional disadvantage presented itself in the fact that the color of the pike closely simulated the shade of the metal-back of the aquarium, thus rendering strong outlines of the resulting negative a matter of doubt. However, the picture (plate 7, lower figure) was fairly good, and on com- paring it with the figure of this species in “The Fisheries and Fishery Industries of the United States” (plate 183, upper figure) it is to be observed that in the living fish the pectoral fins are extended almost directly downward; and further, that the EXPERIMENTS IN PHOTOGRAPHY OF LIVE FISHES. 3 extremities of the forks of the tail are distinctly rounded and not acute, as in the aforesaid drawing. In fact, the caudal flu, or tail, in the latter is incorrect in outline, and there are still other differences to be observed upon comparing the figure of the present paper with the figure given us by Goode, pointing to inaccuracies in the latter. Here is where the great value of the camera comes in. In time, with suitable subjects taken under the most favorable conditions, pictures of fish (as in the case of other animal forms), produced by half-toning processes from faultless photographs, will surely supersede in biological literature the often inaccurate figures that now illustrate it. This is what we strive to accomplish in our efforts to obtain the best possible photographic negatives of fish — live fish in their natural element, with normal surroundings. On the same day I attempted to photograph the two species of sun-fish then in the aquaria. One of these was the common pumpkin-seed ( Eupomotis gibbosus) and the other the long-eared sun-fish [Lepomis auritus , plate 3, upper figure). In the aquarium at the south end of the grotto there were upward of two dozen specimens of the former, while a handsome male of the latter species, with three or four females, were living in another tank at the side of the room, where the light was much more favorable. By instantaneous exposure I secured two fine negatives of the common sun fish. One of these had twenty fish in it, all of which were swimming at the time, but the resulting picture shows not the slightest degree of motion in any one of them. There were nearly as many specimens on the second negative, here shown in plate 3, lower figure, and published for the first time in the Photographic Times , of New York. These results exemplify what may be expected from a highly colored fish, though rather a dark one, attempted under by no means favorable conditions, and where reliance was mainly placed upon tact, patience, and the best material that the market afforded. It will be observed that those specimens which were deep down in the water took the darkest, while those nearer the surface showed better definition. Nearly all of them, however, give the external characters of the species pretty well, and surely are far more interesting than many illustrations frequently seen in zoological works. In the case of Lepomis auritus the subject selected was the single male fish, and for fully two hours, upon an intensely sultry afternoon, I was obliged to wait before this beautiful specimen came into the proper place to be photographed. The result, however, fully compensated for the delay, and the photograph is an absolutely accurate representation of the male long-eared sun fish of our American ichthyfauna. About a week after making these experiments very good results were also obtained with the striped sea-robin [Prionotus evolans) and with the naked star-gazer (Astro- scopes guttatus). The former was taken while resting upon the bottom of the aquarium, while the latter was secured in two positions, the one where it had settled down upon a piece of fiat stone, and the other an instantaneous exposure, showing the fish in the act of hiding itself in the sand, a very interesting habit that it constantly exhibits. The reproduction of my photograph of the star-gazer is shown in plate 7, upper figure, and it is a very accurate representation of this species as it appears in life. This work was not resumed until .July of the following year, when the present Commissioner, Hon. George M. Bowers, extended me additional facilities. Mr. Edw. S. Schmid had also had a special aquarium manufactured for my use, and I had the kind assistance in the experiments of Mr. Leighton G. Harron, the superintendent of the Aquaria at Central Statiou in Washington. I made a number of exposures upon peedle-gars, black bass, and crappie. With both the former species I failed for not 4 BULLETIN OF THE UNITED STATES FISH COMMISSION. having used a shutter of a sufficient degree of rapidity in closing. With the crappie, however, I obtained three serviceable negatives, two of them being very good. On July 17, 1898, another day was given to this work, at which time the same methods and materials were used; but by the aid of former experience the results were more successful, and excellent negatives of series of three species of fish were obtained. Special good fortune attaiued the taking of the large-mouthed black bass ( Micropterus salmoides), of which several negatives were made (see plate 4, upper figure). Yiews of the common sun-fish ( Eupomotis gibbosus) were also obtained, far better than those secured during the year previous. These show in great detail the external appearances of this well-known fish (plate 2). The cat-fish (plate 8, upper figure) proved to be another fair result, but this form is a difficult one to photograph on a number of accounts. Success was attained in the case of the white perch ( Morone americana ) on October 1(3, 189S. On this date there were two of these fish, with several sea trout ( Cynoscion maculatum ), in the northwest aquarium of the marine grotto. At the time the instantaneous exposure was made the finer of the two perch was swimming slowly over the bottom in search of food, while a broad ray of light lit up the sand just beyond him. Plate 4, lower figure, therefore, not only gives a truthful representation of this well-known species as it appears in nature, but the illustration possesses peculiar artistic merit besides, a piece of good fortune that sometimes befalls the operator. While thus occupied, this fish lowers its anterior dorsal fin and draws up its ventral ones, while the pectoral fins stand out perpendicular to the surface of the body. A week later (October 23, 1898) a number of very successful exposures were made, and negatives were secured of the sea trout ( Cynoscion maculatum ), the tautog or black-fish ( Tautoga onitis), and the sea bass ( Gentropristes striatus). The sea trout, a young specimen, was in slow movement forward at the time of the exposure, and shows the anterior dorsal and ventral fins slightly drooping. The mouth is open, and the fish was doing nothing at the time beyond watching its com- panions in other parts of the same aquarium. * The light marks on the anterior part of the body of this specimen, as well as the emargination of the tail, are due to injury and inflammation, resulting from injuries received during transportation from the seaboard to the Pish Commission building. (See plate 5, upper figure.) The tautog (plate 5, lower figure) is a medium-sized specimen, photographed while resting in a vertical position upon the side of a little mound of sand and very close to the surface of the glass. It was in one of the west side aquaria in the marine grotto, which also contained some ten or twelve more specimens of the same species, of various sizes. These fish in confinement exhibit all their natural traits, and in their aquarium some will be seen swimming about not far above the bottom; others will be lying upon their sides, and still others attempting to secrete themselves beneath the rocks, while occasionally exhibiting a peculiar method of combating each other. This consists in two fish (males?) coming at each other face to face, opening their mouths, and, the one bringing his teeth in contact with those of his antagonist, each attempts to force his adversary backward, or if he or the opponent be taken off guard for the instant, the more watchful fish of the two will make the attempt to bite the other. Sometimes there seems to be a certain playfulness about the grotesque maneuvers, while at others an earnest combative nature is quite apparent. Whether the opposite sexes ever engage in this procedure I am unable to say at this writing. At the best the tautog is a peculiar fish in its habits, and their behavior together often EXPERIMENTS IN PHOTOGRAPHY OF LIVE FISHES. 5 reminds me of that of a litter of little pigs, with some of the movements characterized by a certain kind of cat-like fawning. They feed voraciously and take with avidity their natural food, but in the aquarium they usually receive crushed crabs. The specimen represented in plate 5, lower figure, shows the handsome vertical markings and mottlings that the tautog frequently assumes. At other times it is ot a dull leaden-black all over, while some specimens show every variety of intergradation between these two extremes. These changes, it would appear, are almost wrought at the fish’s will, or they may be indicative of the humor it is in, or a color may be assumed that renders the fish less likely to be observed, and this is doubtless of value to it in its native waters, where all fish have enemies of one kind or another. On the same day this tautog was photographed I secured two successful nega fives of the young of the sea bass [Centropristes striatus). There were a number of these fish of various sizes in one of the aquaria on the west side of the grotto, and the light at the time of taking was excellent. Instantaneous exposures were given, and iu one instance the specimen was takeu just as it came to rest upon the bottom (plate 6, lower figure), while iu the other it had assumed that remarkable attitude of resting upon its pectoral aud anal fins that it has in nature (plate 0, upper figure). Both of these results present us with all of the external characters of these fish, aud are valuable on that account. This species undoubtedly has the power of changing its color at will, both for the purposes of protection as well as to indicate the play of the humor it may happen to be in. The various shades are assumed very suddenly, quite as much so as I have seen them to be in the American chameleon ( Anolis principalis ) of the Southern States. Sea bass have beautiful eyes, that change color a little at times, though usually they are of a brilliant emerald green, which unfortunately photographs very dull aud pale. Their large and handsome fins are almost constantly in motion, rendering it extremely difficult to catch this species with the camera. The distal extremity of the upper lobe of the tail is seen to project slightly as a blunt point, but is never in the young extended as an elongated ornamental filament. In the adult, however, of a southern species ( Centropristes ocyurus) both the upper and lower lobes of the caudal fin are thus filamentously produced. I have also examined specimens of sea bass wherein the middle three rays of the tail were likewise somewhat drawn out in this manner. These examples were in the Washington Center Market ^October 26, 18!)S), and were said to have come from New York. The upper and lower caudal lobes were light-colored, and it was only in the former that any indication of a filament was noticed, and that only in some few of the specimens. These fish were probably Centropristes ocyurus , wherein the tails had suffered mutilations due to transportation after capture. Some of the difficulties which attend the photography of living fish are seen in plate 9, from an instantaneous exposure (made October 23, 1898) upon the north end aquarium of the grotto, when there were swimming in it 450 rainbow trout ( Salmo irideus ). Necessarily some of the number were out of focus. The lower ones show but little detail, owing to being in the shadow caused by the great mass of fish above them, others are indistinct from lateral shadows, and at the best the light at the time of exposure was not of the kind to insure the most perfect success; nevertheless, this result is a very interesting one, and probably not many photographs extant, if any, show so mauy examples of swimming fish upon the one plate, where not a single individual of them exhibited the least movement in its photograph. Plate 2. Bull. U. S. F. C. 1 899. (To face page 6.) COMMON SUN-FISH (Eupoiliotis gibbosus). Upper figure, natural size ; lower figure, slightly reduced. GROUP OF COMMON SUN-FISH (Eupomotis gibbosus). Bull. U. S, F. C. 1899. (To face page 6.) Plate 4. THE LARGE-MOUTHED BLACK BASS ( Micropterus salmoides) . Slightly reduced. THE WHITE PERCH (Morone cunericanCL). Two-thirds natural size. Represents the fish searching for food along the bottom of the aquarium, similar to the habit it has in nature. Plate 5. |||. U. S. F. C. 1899. (To face page 6.) THE SPOTTED SEA TROUT OR SQUETEAGUE ( Cynosdon maculcitum). One-half natural size. THE TAUTOG OR BLACK FISH ( Tautoga 0)iitis). Nearly natural size. Exhibiting the fish resting on the bottom, a habit it has commonly in nature. THE SEA BASS (Centropristes striatus). Different specimens. Plate 6. Bull. U. S. F. C. 1899. (To face page 6.) THE PIKE ( Lucius lucius). Left lateral view. THE BROOK TROUT {Salvelinus fontinalis). Natural size. Taken July, 1897. YOUNG RAINBOW TROUT ( Salmo irideus). Four hundred and fifty specimens in the aquarium when instantaneous exposure was made. Bull. U. S. F. C. 1899 (To face page 6.) Plate 9. i NOTES ON THE TIDE-POOL FISHES OF CALIFORNIA, WITH A DESCRIPTION OF FOUR NEW SPECIES. By ARTHUR WHITE GREEEEY, Teacher of Biology , San Diego State Normal School. This paper is based on collections made on several trips along the coast of Cali fornia from San Francisco Bay to Point Sur, in Monterey County, in 1897 and 1898. The fishes were taken exclusively in the tide-pools exposed at low water and were captured with small hand nets. .Calcium hypochlorite or ordinary chloride of lime (bleaching powder) was used with excellent effect for stupefying the fisbes in small, isolated tide-pools. The fishes were taken out as soon as they came to the surface and were killed in dilute alcohol or formalin. The following new species were obtained : Eximia rubellio , Rusciculus rimensis, Dialarchus snyderi , and Blennicottus recalvus , three of them representing new genera, Rusciculus, Eximia, and Dialarchus. One of these species, Dialarchus snyderi, is mentioned in the addenda of Jordan & Evermann’s Fishes of North and Middle America as Oligocottus snyderi, but it is now made the type of a new genus and is fully described and figured for the first time. The group of tide-pool cottoids, the allies of Oligocottus, are here subjected to a critical revision, in view of the confused state of the literature concerning them. The group of Gottidce of the type Oligocottus , comprising the genera Oligocottus , Blennicottus , Glinocottus, and Oxycottus , and the new genera Eximia, Rusciculus, and Dialarchus , is distinguished from the rest of the family by the separation of the gill membranes from the isthmus, the presence of palatine teeth, and either the entire nakedness of the body or the presence of only rudimentary, prickly scales. They are all strictly tide pool fishes of the Pacific coast, ranging from Bering Sea to Lower California aud never wandering far from shore. Each species inhabits, with surprising regularity, only certain kinds of tide-pools, its distribution depending almost entirely upon the character of the rocks aud the kind of algte present. They all imitate very closely the color of their surroundings, and two or three species show parallel color phases, each copying after a certain kind of alga. Thus, depending on these condi- tions of rock and plant life, there are along almost any part of the coast two or three zones of vertical distribution, one species inhabiting the deeper tide-pools, another the shallower, and so on, as will be seen by reference to the descriptions. Glinocottus analis offers a marked exception to these generalizations, however, as it is found in every kind of tide pool within its range. The first known species of this group were described by Girard as follows: Oligo- cottus maculosus in 18.56, 0. analis in 1857, and 0. globiceps in 1858. These last two species were made types of new genera by Gill in 1861, giving them the names Glino coitus and Blennicottus. These two genera of Gill were not recognized by Jordan A 7 8 BULLETIN OF THE UNITED STATES FISH COMMISSION. Gilbert in their Synopsis of 1883, the genus Oligocottus being made to include all three species; but they were finally restored in Jordan & Evermann’s Check-List of 1896. To the genus Oligocottus there have been since added the species acuticeps (Gilbert, 1893), embryum (Jordan & Starks, 1895), and borealis (Jordan & Snyder, 1896); to Blennicottus the variety B. globiceps bryosus (Jordan & Starks, 1896). The status of these species has remained unchanged, except that Jordan & Evermann, in The Fishes of North and Middle America, have considered Oligocottus acuticeps the type of a new genus, Oxy coitus, to which they have transferred also Oligocottus embryum. This nomenclature is here adopted, except in the genera Oligocottus and Blennicottus , where confusion in specific identification has taken place and a reassign merit of specific names is necessary. Girard’s original species, Oligocottus maculosus and Blennicottus globiceps, were described respectively from Tomales Bay, north of San Francisco, and from the Farallon Islands, off San Francisco. Both are now shown to be forms of northerly distribution, the type locality being, in each case, near the southernmost limit of the range. Southward along the coast each is replaced by a distinct species, both of which are abundant at Monterey Bay. Recent authors have, unfortunately, identified Girard’s names with specimens from Monterey Bay, while the northern species to which his names should apply have been rechristened Oligo- cottus borealis and Blennicottus globiceps bryosus. Therefore these two names are now abolished, the species becoming Oligocottus maculosus and Blennicottus globiceps, and the southern forms are described as new species, Dialarchus snyderi and Blennicottus recalvus, the former being also made the type of a new genus. The different species of this group resemble each other to a remarkable degree in external appearance, yet most of them are separated by characters which we now consider of generic importance. The characters are all remarkably constant except that of color, which varies greatly with the surroundings and can not be described with great exactness. The color descriptions given in this paper are all from life, and have been made broad enough to cover all the specimens examined. The character and arrangement of the cirri afford perhaps the best specific distinguishing features, and not the slightest variation from the adult plan has ever been discovered in these species. The features of generic importance in this group are the character of the preopercular spines, the presence or absence of scales, the shape and size of the head and mouth, and the nature of the first three or four anal rays of the male. There may be one, two, or several rays enlarged, or they may be all of normal size, the number of modified rays and the amount of enlargement always remaining the same in any one species; furthermore, these enlarged rays may or may not be separated from the rest of the fin. I have used the form and size of the preopercular spines for the primary divisions of the key, and this arrangement brings together the species nearest alike in geographical range and external appearance. The presence or absence of scales can not be considered a mark of less importance, however, and the two together stand out distinctly as dividing the species into natural groups. All of these species have a slit behind the last gill except Blennicottus embryum and Rusciculus rimensis. This seems to be an important character in determining the relationships of the species. The tables accompanying the descriptions give the various dimensions of the body in hundredths of the total length to base of caudal. The author is under deep obligations to Dr. Charles H. Gilbert, in whose laboratory and under whose direction the work was carried on, and to President David Starr Jordan, whose suggestions and advice have been of great help. THE TIDE-POOL FISHES OF CALIFORNIA. 9 Key to Genera and Species allied to Oligocottus. I. Preopereular spine simple, not forked or branched. a. Scales none; no enlarged anal rays in the male; anal papilla present in the male. b. Blennicoitus. Head very wide and blunt; mouth terminal, mainly transverse, c. Cirri of head few, none on interorbital space, two regular rows above the eyes, and a few on sides of head; preopereular spine scarcely appreciable . B. recalvus. cc. Cirri of head numerous, entirely covering the occiput and interorbital space, a thick bunch above preopereular spine and opercular flap ; preopereular spine nearly half diameter of eye, curved upward B. globiceps. bb. Oxycottus. Head pointed ; mouth extending laterally below eye. d. Cirri of head mossy or joined at base; four occipital bunches; dorsal fins separate. O. EMBRYUM. dd. Cirri of head single or double; three occipital bunches; dorsal fins slightly joined at base; anal papilla of male very large, situated between ventrals O. acuticeps. aa. Rmciculus. Dorsal half of body closely scaled; first two anal rays of male enlarged; no anal papilla; cirri of head small, single, or rarely doubled; one supraorbital, three occip- ital, one nasal, one maxillary, and two or three preopereular cirri; lower margin of preopercle bluntly toothed R. rimensis. II. Preopereular spine forked at tip. e. Scales none; one or more anal rays enlarged in the male; anal papilla inconspicuous. /. Dialarchus. First anal ray of male enlarged, joined to second, the two widely separated from rest of fin; cirri of head joined at base, two supraorbital hunches, three on occipital, a thick preopereular bunch, an opercular one, and a few on sides of head; a row of cirri along the dorsal fin close to its base D. snyderi. ff. Oligocottus. Three or four anterior anal rays of male enlarged, not separated from rest of fin and becoming progressively smaller posteriorly ; cirri of head irregular, usually one supraorbital and three or four occipital bunches in which all the cirri are joined at base ; no cirri along dorsal fin, except in the young or one or two scattering ones in the old; maxillary reaching a vertical below center of pupil O. maculosus. ee. Clinocottus. Scales present, sometimes obscure in adults; no enlarged anal rays; anal papilla present, large; cirri scattered irregularly over top of head in four irregular occipital rows, and in old individuals extending back on body along nearly entire length of dorsal fin, and down over anterior and dorsal half of body, none below lateral line, no scales on anterior half of body in the old; a fringe of cirri on preopercle, a thick bunch on opercle, and five or six cirri on maxillary C. analis. III. Eximia. Preopereular spine three-pointed. Scales none; eye and nasal spines large; first anal rays of male enlarged, second slightly elongated, not separated from fin ; anal papilla inconspicuous; cirri not joined at base, three supraorbital and three occipital pairs of bunches of two or three each; a maxillary bunch, a preopereular row, and an oper- cular bunch of cirri, and a few scattered ones on side of head ; a row along dorsal fin bending downward at end of spinous dorsal; scattered cirri between the dorsal row and lateral line, and below lateral line behind pectoral fins E. rubellio. LIST OF FISHES. Blennicottus recalvus Greeley, new species. Fig. 1. Centridermichthys globiceps Giinther, Cat., u, 171, 1860. Oligocottus ylobiceps Jordan & Gilbert, Synopsis, 718, 1883. Blennicoitus globiceps Jordan & Starks, Proc. Cal. Ac. Sci. 1895, 808; Jordan & Evermann, Fishes of North and Middle America, ii, 2017, 1898; not Oligocottus globiceps Girard. Head 3.66; eye 4.75 in head ; snout 3; D. ix, 15 or 16; A. n, 12; P.14. Body short, stout, broad anteriorly ; head very broad, short and blunt; snout obtuse ; interorbital space five-sixths of eye, grooved, the groove leading into a depressed space behind eyes; mouth distinctly terminal, maxillary reaching a vertical below anterior edge of orbit, lower jaw included; minute conical teeth on jaws, vomer, and palatines; nasal spines very small; no preopereular spine apparent in adults ; edge of preopercle rounded ; opercle ending in a rounded flap; branchiostegals 6, membranes broadly united, free from isthmus; gills 31, a slit behind last gill. 10 BULLETIN OF THE UNITED STATES FISH COMMISSION. Dorsal tins very long, slightly joined at base, origin of first dorsal directly over tip of opercle, that of soft dorsal in advance of origin of anal; first dorsal slightly rounded, middle ray longest; pectorals reaching origin of anal, membranes of first seven rays deeply emarginated ; ventrals reaching vent; anal papilla of male very large; anal low, membranes of all the rays except last three deeply emarginated, none of rays enlarged in male; caudal short, slightly rounded. Cirri few and small, those of top of head joined at base in conspicuous bunches, two irregular occipital rows, a few below these on sides of head and on margins of preopercle and opercle; a few above origin of pectoral, and a weak row along anterior third of lateral line. Color of body light brown, vermicnlated with white, and marked dorsally with four or five wedge- shaped spots of dark brown, edged with white, and more distinct posteriorly; two pinkish spots on dorsal side of caudal peduncle, and a faint shading of same color on sides of head and along anterior fourth of lateral line; entire undersurface dull brown, tinged with olive; fins indistinctly barred with grayish-white; tail faintly tinged with pink. In some specimens the color is an almost uniform dull brown, while in others light markings are prominent. Some young individuals from among green algse are uniform light green. Fig. 1. — Blennicottus recalvun Greeley. Type. Comparative measurements of Jive specimens of Blennicottus recalvus. Measurements. Collectors and localities. Pacific Grove, L. S. Jr. U. M. (4245). Greeley & Cowles, Pacific Grove. Greeley & Spaulding, Santa Cruz. Greeley '. globiceps bryosus was based upon this northern form, the typical globiceps, and to it the original name is now restored. Head 3.66; eye 4.75; snout 3; D. ix, 16 or 17 ; A. 11 or 12; P. 13 or 14. Body short, stout; head deep, blunt, with a short decurved snout; interorbital space five-sixths of eye, deeply grooved; mouth nearly terminal and transverse, with slight lateral cleft, the maxillary reaching a vertical through the center of orbit; lower jaw included. Preopercular spine single, curved upward, nearly half diameter of orbit; nasal spines prominent, nearly half diameter of eye; opercle ending in a rounded flap. Dorsal fins long, slightly joined at base; no anal rays enlarged in male; anal papilla large. Cirri very numerous on top and sides of head, extending through the interorbital groove to nasal spines, two rows of prominent joined cirri on top of head, others between these, still others on sides of head; a large hunch above preopercular spine and on dorsal margin of opercle, a V-shaped row above pectorals, and a thick row along anterior half of lateral line. Color, light plumbeous brown, with traces of four or five more or less distinct black transverse bands, the whole body more or less vermiculated with white; some specimens largely streaked with white vermiculations, others nearly uniform brown; under parts dull white tinged with brown on under lip, and with yellow posteriorly; fins barred with yellow. The northern form of this species 12 BULLETIN OF THE UNITED STATES FISH COMMISSION. lighter in color; young individuals frequently uniform plumbeous black with a conspicuous transverse band of silvery white on anterior part of body. Four to 5 inches long, specimens from Puget Sound 7 inches. Very close to B. recalvus, which is its southern representative. Known from Puget Sound to Pigeon Point, San Mateo Co., Cal., where it stops abruptly, B. recalvus succeeding it immediately to the south. A few specimens have been taken, however, immediately south of Monterey Bay, within the range of B. recalvus. Its occurrence to the south needs further investigation. Comparative measurements of six specimens of Blennicottus globiceps. Measurements. Collectors and localities. E. 0. Starks, Neah Bay, Washington, L.S. Jr. U.M. (3404). E. C. Starks, Neah Bay, Washington, L. S. Jr. U.M. (3404). Greeley & Cowles, Pillar Point. Greeley osterior to a point midway between anal 14 BULLETIN OF THE UNITED STATES FISH COMMISSION. and base of pectorals in male, anterior to it in female, the difference caused by enlargement of first two anal rays in male; anal fin small, rays slender, membranes of all deeply emarginated ; first two anal rays of male greatly enlarged, joined by membrane to each other and to rest of fin; posterior edge of tail nearly straight ; anal papilla inconspicuous. Cirri small and scarce, always occurring singly, never in bunches or joined at the base, except a few pairs along anterior third of lateral line, one above each orbit, two rows of three each behind these on top of head, one cirrus on inside of each uasal spine; a cirrus on end of maxillary, two or three on margin of preopercle below preopercular spine, and a row along anterior half of lateral line. Color, light olive or reddish brown tinged with lavender, marked dorsally with four or five ° wedge-shaped indented spots of black, a broken band of same color along lateral line, sometimes sending branches below it, which show a tendency to inclose round spots; a more or less distinct spot of black on top of head; a faint postocular line, a spot below eye, and a preocular line running from eye to snout, all of same color; pectorals and caudal indistinctly barred with brown, anal tinged with it, and dorsal covered with fine brown or black spots, sometimes very faint; throat and belly pale yellowish- white, unspotted. This species is most closely related to Oxycottus embryum, with which it agrees in general colora- tion, but differs decidedly in the presence of scales, the slenderer body, the larger number of soft dorsal and anal rays, the serrated margin of the preopercle, and the arrangement of the cirri. Described from two specimens taken at Point Lobos, Monterey County, Cal., by A. W. Greeley. (Tyjte, No. 6067, L. S. Jr. Univ. Museum.) Eare ; only two other specimens from Point Lobos are known. It inhabits tide-pools lined with corallines, and in coloration imitates very closely these alg®. Length, 40 mm. The smallest of our tide-pool fishes. Comparative measurements of two specimens of Kusdculus rimensis. Collectors and localities. Measurem en ts . Gredev, Greeley, Point, Lobos Point Lobos, (type), June, 1898. June, 1898. Extreme length, in millimeters — to 39 Greatest height of body 20 22 Least height of caudal peduncle 6 7 Length of caudal peduncle 17 16 Length of head 29 28 Width of head 23 23 Width of interorbital space. 7 7 Height atpupil 15 16 Length of snout 9 9 Diameter of orbit 7 7 Distance from snout to spinous dorsal 28 28 Length of spinous dorsal at base 24 27 Greatest height of spinous dorsal 10 9 Length of soft dorsal at base 48 47 Height of longest ray of soft dorsal 15 15 Distance from snout to anal 43 48 Height of longest anal ray 15 14 Length of caudal 19 19 Distance from snout to pectoral 30 27 Length of pectoral 39 38 Distance from snout to ventral 32 32 Length of ventral 18 18 Number of dorsal spines - - . 9 9 Number of dorsal rays 17 18 Number of anal rays 14 14 Number of pectoral ravs 14 14 21 DIALARCHUS Greeley, new genus. Preopercular spine forked at tip; scales none; first anal ray of male enlarged, joined to second, the two widely separated from rest of fin. Closely allied to Oliyocottus, differing only in character of anal rays of male. Dialarchus snyderi Greeley, new species. Fig. 4. Centridermichthys maculosus Gunther, Cat. Fishes, II, 171, 1860; not Oliyocottus maculosus Girard. Oliyocottus maculosus Jordan & Gilbert, Synopsis, 718, 1883; Jordan & Evermann, Fishes of North and Middle America, n, 2013, 1898. Oliyocottus snyderi Greeley, in Jordan & Evermann, Fishes North and Mid. Amer., in, 2871, 1898. THE TIDE-POOL FISHES OF CALIFORNIA. 15 Head 3^; eye 4^ in head; snout 34; D. vm, 18 or 19; A. 13 to 15; P. 13 to 15; V. i, 3. Body elongate, slender ; snout pointed, compressed ; minute conical teeth on jaws, vomer, and front of palatines; jaws equal; mouth horizontal, maxillary 3 in head, reaching a vertical below anterior edge of pupil. Interorbital space live-sixths of eye, shallowly grooved, the groove leading into a depressed space between occipital ridges; nasal spines large. Margin of preopercle armed with a strong spine, half as long as eye, from upper border of which at base extends a second spine pointing abruptly upward and inward; both spines covered with skin in life; margin of opercle ending in a pointed flap, entirely unarmed. Branch iostegals 6, the membranes broadly united, free from isthmus. Gills 34, a slit behind last gill. Dorsal fins large, separated by half diameter of eye, whole length equaling that from caudal to base of pectoral; first dorsal beginning slightly in advance of margin of opercle, upper edge nearh straight, curving abruptly downward from sixth spine; origin of soft dorsal in advance of anal; pec- torals large, reaching well beyond origin of anal; ventrals almost midway between base of pectorals and anal; anal fin small, rays all feeble in female, in male the first ray only greatly enlarged, joined to second, the two distinctly separated from rest of fin, membranes of all except last three or four rays deeply einarginated. Anal papilla small, present in male only. Cirri very numerous, usually occurring in hunches of three or four, those of head joined at base, forming a comb; two pairs of bunches above orbits, with rudiments of a third bunch in front of these, three on top of head, behind orbits, two or three bunches just below these on sides of head, two or three single cirri on margin of preopercle, a thick bunch above preopercular spines, four or five on lower margin of opercle, with a thick bunch on its upper margin; a short row above base of pectorals; a row of bunched cirri along anterior two-thirds of lateral line, another well-defined row along dorsal fin from third spine to six- teenth or seventeenth ray of soft dorsal, this row containing a hunch at base of each spine ami ray, with the occasional exception of the first ray ; five or six scattered bunches between dorsal and lateral rows on each sile of the body ; a cirrus at tip of each dorsal spine. Color, light reddish-brown, sometimes almost pink, thickly spotted with fine indistinct white spots; four or five irregular dark-brown spots along base of dorsal, a band of same color along lateral line, sometimes very much broken and extending ventrally, shading into uniform reddish-brown below, and including three or four round pinkish spots; a dark-brown postocular line, another running forward from eye, a patch of same color on top of head, another on sides of head, and two or three on edge of opercle; throat reddish-brown, with several distinct white spots ; belly bluish- green; a silvery white patch between bases of pectorals; dorsal fins pale reddish-brown, with black and clear spots; pectorals crossed irregularly with white; anal fin pale pink, crossed with dark brown. There are two or three perfectly distinct types of coloration, as follows: Some specimens from pools containing green alga? are pure light green ; others from coralline pools are tinged with lavender, as B. embryum. This species resembles most closely O. maculosus, which name has been erroneously applied to it, but it differs markedly in its slenderer body, more pointed snout, the arrangement of its cirri, and the perfectly distinct coloration, also in greater length of dorsal fins, the enlargement of only one anal ray in male, and shortness of maxillary. BULLETIN OF THE UNITED STATES FISH COMMISSION. 16 Hern described from a specimen taken at Pacific Grove, by Greeley & Maddren. (Type, No. 5846, L. S. Jr. Univ. Museum.) Five cotypes were taken at the same time. Specimens are at hand from Orescent City, Bolinas Bay, Half Moon Bay, Monterey Bay, and San Luis Obispo, Cal. Found in all kinds of pools from San Francisco to Monterey Bay, but nowhere common. Length, 60 mm. The most beautiful and active of the tide-pool fishes, extremely variable in color. Comparative measurements of eleven specimens of Dialarchus snyderi. Measurements. © ~ o . S3 o . © Hz u . & Albatross, Monterey, L. S. Jr. IJ. M. (3642). Snyder, Pillar Point. Colic P ■ ic- £•5 73 6 © a t* Oh Greeley & Cowles, Pa- g. cifie Grove. £ Greeley & Cowles, Pa- ^ cific Grove. a Greeley & Maddren, Pac. Grove (type). § ilitii s CO © . 1h © 73 © O 3S. bJU a § 5 £ tc5 s © © s-t 0 oh 2 3 N jL E XO O C3 o JT'tS © fH 0 Greeley & Spaulding, Santa Cruz. Extreme length in millimeters 58 64 59 56 61 54 60 36 65 65 62 Greatest height ot' body 26 24 24 23 25 23 24 26 28 24 25 Least height of caudal peduncle 8 8 8 7 8 8 7 8 8 8 8 Length of caudal peduncle 17 17 17 17 17 15 15 18 16 18 17 Length of head 33 31 31 32 31 31 32 34 31 31 30 Width of head 26 24 23 24 23 24 25 26 9 '2 9.2 25 Width of interorbital space 5 5 5 6 5 6 5 6 6 5 5 Height of head at pupil 14 14 14 15 15 15 15 16 15 15 15 Length of snout 9 9 9 9 8 9 9 9 9 8 9 Diameter of orbit 6 6 5 6 6 6 6 6 6 6 6 Distance from snout to spinous dorsal 30 29 27 27 26 29 29 30 28 27 28 Length of spinous dorsal at base 23 25 23 21 25 22 23 21 26 24 25 Greatest height of spinous dorsal 10 10 10 10 11 10 9 9 9 9 10 Length of soft, dorsal at base 44 45 44 43 44 43 43 43 44 43 42 Height of longest ray of spinous dorsal 12 13 13 14 13 16 13 17 12 12 14 Distance from snout to anal 43 50 44 50 48 45 50 50 45 43 47 Height of longest anal ray 14 13 13 13 13 13 12 13 13 13 13 Length of caudal 22 21 21 21 22 21 20 22 18 20 20 I >istance from snout to pectoral 32 32 30 31 29 32 32 33 29 29 29 Length of pectoral 34 33 31 31 31 34 32 34 32 36 35 1 >i stance from snout to ventral 32 31 32 32 29 33 33 32 27 29 29 Length of ventral 17 17 17 17 18 37 17 17 17 17 17 8 8 8 8 8 8 8 8 8 8 8 Number of dorsal rays 19 in 19 18 18 19 19 18 19 18 19 Number of anal rays 14 14 15 14 13 14 14 14 15 13 14 Number of pectoral rays 14 14 14 14 13 14 14 14 14 14 15 24 22 23 23 1 Oligocottus maculosus (Girard). Oligocottus maculosus Girard, Proc. Ac. Nat. Sci. Phila. 1856, 153; Girard, U. S. Pac. R. R. Sur., x, Fishes, 56, 1858. Oligocottus borealis Jordan & Snyder, Proc. Cal. Ac. Sci., series 2, vol. vi, 1896, 225, Neali Bay (coll. E. C. Starks, type, No. 3396, L. S. Jr. Univ. Museum); Jordan & Evermann, Fishes of North and Middle America, n, 2014, 1898. This species was described by Girard in 1856, from specimens taken at Tomales Bay, but this account was erroneously associated with another fish, now recorded as Dialarchus snyderi. It was rediscovered and described as a new species, O. borealis, by Jordan & Snyder, from a large series of specimens taken at Puget Sound; hut Girard’s original name is now restored, as there is no doubt that this is Girard’s species. Inthisopinion Professors Jordan and Gilbert and Mr. Snyder fully concur. A series taken by Dr. Jordan at Sitka and a specimen collected by me at Prince William Sound extend the range northward to the latter point. It is the most common tide-pool fish at Crescent City, Del Norte County, Cal., and also at Half Moon Bay, where the shallow open tide-pools with little algae suit the species very well. South of Half Moon Bay it gradually becomes scarcer, but extends as far as Pigeon Point. None was taken at Santa Cruz or Pacific Grove. The absence of this species south of Pigeon Point is probably due to a change in the character of the tide-pools. Below Pigeon Point the coast is made up of great shelving ledges of very hard sandstone, with few pools. North of Pigeon Point the rocks are much softer and contain shallow jiools. Wherever found, this species inhabits all kinds of tide- pools, but especially those with dull surroundings, either bare rocks or rocks covered with Fueus, the brown seaweed. THE TIDE-POOL FISHES OF CALIFORNIA. 17 Young individuals occasionally have a few cirri along the dorsal fin, on the opercle, and above the pectoral fins, probably showing a reversion to a type resembling Dialurclius snyderi. Color, light brown, a very soft gray in some specimens, varying from almost olive to a dull brown in others; marked dorsally with fine white spots, which become larger below and shade into the pale olive of belly ; some specimens tinged with milk-white or lavender, with a series of spots of same color along lateral line; fine dark-brown transverse bauds, varying greatly in distinctness, extend downward from dorsal fin, first two reaching the belly, last three interrupted; a white spot at base of tail, usually one on dorsal side of caudal peduncle; head uniform with body, but more or less vermiculated with white or olive; throat olive, spotted with white, the ground color becoming pale on belly and along the sides of the anal; fins light olive, barred with brown. Comparative measurements of nine specimens of Oligocottus maculosus. Measurements. Collectors and localities. Jordan, Sitka, L. S. Jr. U. M. (5658). (H •"D (n . - CO ci .o, |P o Starks, Port Ludlow, L. S. Jr. U. M. (5027). Snyder, Crescent City. Snyder, Crescent City. Greeley & Cowles, Pil- lar Point. Greeley & Cowles, Pil- lar Point. Greeley & Spaulding, Pigeon Point. Greeley & Spaulding, Pigeon Point. Extreme length in millimeters 51 44 56 59 61 58 61 58 59 Greatest height of body 11 22 26 26 24 26 23 27 24 Least height of caudal peduncle 8 8 10 10 9 9 9 9 9 Length of caudal peduncle 16 19 18 17 17 16 17 17 16 Length of head — 31 32 34 35 33 34 32 34 33 Width of head 24 22 27 24 26 24 24 25 25 Width of interorbital space 5 6 5 5 6 5 5 5 5 Height of head at pupil 16 16 15 15 15 15 14 16 15 Length of snout 9 9 9 9 9 9 9 9 9 Diameter of orbit 6 6 6 7 7 7 6 7 6 Distance from snout to spinous dorsal 30 32 31 32 32 31 30 31 31 Length of spinous dorsal at base 25 24 28 28 24 25 22 26 24 Height of spinous dorsal 10 10 10 13 10 9 10 11 11 Length of soft dorsal at base 42 42 40 44 43 41 42 40 43 Height of soft dorsal 14 14 14 14 14 13 16 14 15 Distance from snout to anal 46 45 50 47 46 57 53 51 48 Height of anal 14 15 15 16 15 14 14 13 13 Length of caudal . . 23 23 22 24 26 22 23 21 21 Distance from snout to pectoral 33 31 32 34 33 32 32 33 32 Length of pectoral 37 33 34 36 36 34 36 34 35 Distance from snout to ventral 31 30 30 32 32 32 35 32 31 Length of ventral 18 20 19 19 20 19 19 19 18 Number of dorsal spines 8 8 8 8 8 8 8 8 8 Number of dorsal rays 16 17 16 17 17 17 17 16 16 N umber of anal ravs 13 13 13 13 13 13 13 13 13 Number of pectoral rays 14 14 14 14 14 14 14 14 14 17 17 20 19 17 18 Clinocottus analis (Girard). Oligocottus analis Girard, Proc. Ac. Nat. Sci. I'hila. 1857, 201, Monterey ; Girard, U. S. Pac. R. R. Sur., x, Fishes, 57, 1858; Jordan & Gilbert, Synopsis, 718, 1883. Coitus criniger Giinther, Cat., ii , 522, 1860, Monterey. Centridermichthys analis Gunther, Cat., II, 171, 1860. Clinocottus analis Jordan & Evermann, Fishes of North and Middle America, ii, 2012, 1898. Body light brown, crossed by seven or eight irregular dark bands, broken anteriorly and covered everywhere with fine white, yellow, and pink spots, occasionally olive in some specimens, and in others uniform white, which may be united in irregular blotches; a faint reddish spot on dorsal fin, another on dorsal side of caudal peduncle; head, dull black, shaded with light brown, and thickly spotted as body ; top of head lighter, throat silvery white, with irregular dark- brown and black spots ; belly light yellow or white, pectorals and I ail barred with broad bands of olive green, and sometimes edged with yellow or red; soft dorsal olive, rays spotted with white and black, and whole fin of breeding males edged with a row of red spots varying greatly in extent, but always conspicuous. Some young specimens have a distinctly reddish tinge over head and dorsal part of body, others are or F. C. B. 1899—2 18 BULLETIN OF THE UNITED STATES FISH COMMISSION. a uniform green color, broken only by dark markings, and agree closely in color with the green algae ( Viva), among which specimens exhibiting this type of coloration live. This is the largest and most common cottoid in the pools of Monterey Bay, and is found every- where about the bay from low to high water mark, but especially in the shallow exposed pools high up on the coast, which contain almost no algae. Its dull, almost uniform, color agrees very well with the bare rocks and sand. It is much less common at Pillar Point, San Mateo County, where its place is partly taken by Oligocottus maculosus. The young have much larger scales proportionally than the adults, and the scales extend forward as far as the origin of the spinous dorsal, while in old indi- viduals these anterior scales have entirely disappeared, and their place is taken by cirri as far hack as the soft dorsal, while the posterior scales are greatly obscured. EXIMIA Greeley, new genus. Allied to Oligocottus, but differing in the presence of a large three-pointed preopercular spine instead of the simple forked spine of Oligocottus. Skin smooth. A slit behind last gill. Eximia rubellio Greeley, new species. Fig. 5. Head 2.83; eye 3.75 in head; snout 3.75; D. vii or vm, 15 or 16; A. 12 or 13; P. 13 or 14; V. i, 3. Body compressed, snout pointed and compressed, head deep, occiput narrow, slightly concave; interorbital space narrow, half the large eye, shallowly grooved. Nasal spines prominent, very large, and pointed. Teeth small, pointed, on jaws, vomer, and palatines; jaws equal, mouth horizontal, maxillary 3 in head, reaching a vertical below anterior edge of pupil. Margin of preopercle armed w7ith a very strong spine as long as eye, extending backward and downward, hearing on its upper surface a second and third spine, both pointing back and up; all the spines covered with skin in life; opercle ending in a rounded flap. Branchiostegals 6, not united to isthmus; gills 31, a slit behind last gill. Anal papilla inconspicuous. Dorsal fins not joined; first dorsal beginning in advance of margin of opercle, first two spines short, upper margin slightly rounded, soft dorsal beginning in advance of origin "of the anal; all rays and spines very slender ; pectorals reaching well beyond origin of anal; anal fin small, rays slender, membranes emarginated between each ray; in males the first ray enlarged, the second slightly elongated, the two united, and not separated from rest of fin, as in Dialarchus snyderi; ventrals situated below upper edge of base of pectorals, just reaching anus. Cirri all distinct, never joined at base in a comb as in Dialarchus snyderi; three pairs of bunches of two or three cirri each above orbits, first bunch directly above nasal spines, three pairs on top of head behind orbits, a few scattered cirri below these on sides of head, a bunch of two or three ou end of maxillary, a row on lower margin of preopercle, a large bunch above the preopercular spines, and several scattered cirri on margin of opercle; a row of bunched cirri along anterior half of lateral line, a row along base of dorsal, including a bunch of three or four for each spine and ray, the row bending downward at last spine of first dorsal, leaving a space between cirri and base of dorsal spines; a few scattered cirri between dorsal and lateral rows and below lateral line behind pectoral fins. Color light brown to all shades of light red, pink, or lavender, spotted everywhere with white spots extremely minute on dorsal half of body, but more conspicuous ventrally ; five wedge-shaped spots of dark brown along dorsal side of body; head dark brown, sometimes blotched with red or THE TIDE-POOL FISHES OF CALIFORNIA. 19 green, becoming lighter on sides, leaving a dark postocular line extending from eye to preopercular spine and a dark spot on lower margin of preopercle, everywhere very finely marked with white and bine; a white spot with a brown center just in front of first dorsal; throat and belly a very light bluish-green, shading into a faint yellow behind pectorals, and a brownish-green on each side of anal; all the fins, except ventrals, light brownish-green barred with dark brown; caudal light reddish-brown, anal and tip of pectorals tinged with pinkish. A young example is lighter and more brilliantly colored. This species is most closely allied to Dialarchus snyderi, from which it differs in presence of a third preopercular spine, greater depth and comparative length of head, larger eye and nasal spines, and arrangement of cirri. Many specimens taken at Monterey Bay, but not recorded from any other locality. The most brilliantly colored of the tide-pool fishes, inhabiting only deep pools rich in plant life. Comparative measurements of six specimens of Eximia rubellio. Measurements. Extreme length in millimeters. . - Greatest height of body Least heigh; of caudal peduncle. Length of caudal peduncle Collectors and localities. Diameter of orbit. Length of spinous dorsal at base . •Height of spinous dorsal Length of soft dorsal at base Height of soft dorsal Distance from snout to anal Height of longest anal ray Length of caudal Distance from snout to pectoral. . Length of pectoral Distance from snout to ventral. . . Length of ventral Number of dorsal spines Number of dorsal rays Number of anal rays Length of first anal ray, male Albatross , Monterey, L. S. Jr. U. M. (3642). McGregor, Monterey, L. S. Jr. U. M. (4048). McGregor, Monterey, L. S. Jr. U. M. (4048). a » § ^ p^ © O 6 a Pd © O) £ o „• ■8 © © © <3 Greeley & Cowles, Pac. Gr. 53 53 52 53 55 45 30 28 25 27 27 27 7 7 8 7 8 8 16 17 15 14 15 15 37 35 36 37 37 37 25 28 28 26 28 27 5 6 6 7 6 6 J9 18 18 19 20 20 10 10 10 10 10 10 9 8 9 8 8 8 32 32 32 31 33 33 21 22 24 25 20 22 10 11 10 12 10 10 37 41 41 37 39 39 lfi 15 15 16 17 16 57 57 52 54 55 54 15 15 15 13 13 14 21 21 21 21 21 21 35 35 34 34 36 36 32 32 32 34 34 34 36 37 32 32 37 37 17 18 17 18 18 18 7 8 8 8 8 8 15 15 16 16 16 15 12 13 13 13 13 12 13 13 13 14 14 14 18 Artedius lateralis (Girard). Scorpanich t hys lateralis Girard, Proc. Ac. Nat. Sci. Phila. 1854, 145, San Luis Obispo and San Francisco. Artedius lateralis Girard, Proc. Ac. Nat. Sci. Phila. 1856, 134; Girard, U. S. Pac. R. R. Surv., x, Fishes, 70, pi. 229, figs. 5 and 6,1858; Gunther, Cat., II, 174, 1860; Jordan & Evermann, Fishes of North and Middle America, II, 1902, 1898. Ground color brown, tinged with olive and broken by many white spots below, lower row shading into yellow of ventral side, leaving a scalloped margin ; body crossed by several lavender bands, the extent of which is extremely variable, depending on color of algae; usually a broad baud extending front just behind eye’s to the middle of first dorsal, and reaching edge of opercle on sides, with two or three less distinct bands behind this; a spot of same color at end of soft dorsal, continued on the two sides; another more distinct spot on base of caudal at end of lateral line; three or four smaller spots along lateral line which embrace a corresponding number of broad lamellaj-skaped cirri, much larger than the rest; a blotch of pink at base of pectorals, with very fine vermiculations of same color on soft dorsal; spinous dorsal reddish ; soft dorsal brown, shaded with red and irregularly spotted with 20 BULLETIN OF THE UNITED STATES FISH COMMISSION. white; under side of throat brown, with tine light spots shading into white posteriorly, and shaded with green, which extends about the inside of the mouth, hack of pectorals, and down middle line of belly; sides of belly and ventral side of body on each side of anal yellow. The most brilliant and sluggish of the tide-pool cottoids, usually lying close to the rocks and rarely moving rapidly. Distinguished from A. asperulus by the shape and depth of head and length of the band of scales, and not by number of rows in the band. A number of my specimens have nine rows, and are identical in this respect with A. asperulus. A. lateralis occurs in all kinds of tide-pools, high or low, but is usually found singly, and never are there many in the same pool. These solitary habits differ very noticeably from those of the other tide-pool cottoids in their strikingly social manner of life. Caularchus maeandricus Girard. Lepidogaster reticulatus Girard, Proc. Ac. Nat. Sci. Phila. 1854, 155, San Luis Obispo, Cal.; name preoccupied. Lepidogaster nurandricus Girard, Pacific R. R. Surv., x, Fishes, 130, 1858, San Luis Obispo, Cal. : substitute for reticulatus, preoccupied in Ledadogaster ; Gunther, Cat., in, 505, 1861. Gobiesox reticulatus Jordan & Gilbert Synopsis, 749, 1883. Caularchus nurandricus Jordan & Evermann, Fishes of North and Middle America, in, 2328, 1898. Color yellowish green, with faint vermiculations of brown, and spots of clear yellow more apparent on top of head; a distinct interorbital line of same color; tip of snout tinged with black; dorsal, anal, and caudal light-brown, edged with bluish- white. The only fish of this family ( Gobiesocidce ) found in tide-pools of Monterey Bay and northward. Gibbonsia elegans (Cooper). Myxodes elegans Cooper, Proc. Cal. Ac. Sci., in, 109, 1864, San Diego and Santa Barbara. Climes ocellifer Mocquard, Bull. Soc. Philom. Paris, 1886, 44, California. Clinus evides Rosa Smith, Proc. U. S. Nat. Museum 1883, 235, specimens from Todos Santos ; not of Jordan & Gilbert. Gibbonsia elegans Jordan & Evermann, Fishes of North and Middle America, hi, 2353, 1898. Originally described from Point Loma and its geographical range given as from Point Concepcion south to Todos Santos. In 1883 Jordan & Gilbert described G. evides, ranging from Point Concepcion northward. The two nominal species were distinguished as follows by Jordan & Evermann: G. evides: Dorsal v-xxxi, 10; anal n, 26; soft dorsal low; coloration comparatively plain, the soft dorsal without pellucid area. G. elegans : Dorsal v-xxvm, 7 ; anal ii, 24 ; soft dorsal high ; coloration more or less variegated, soft dorsal with a high pellucid blotch posteriorly. My collection from Monterey Bay contains some specimens which are undoubtedly G. elegans, thus extending the range of that form to Monterey Bay. Others are exactly intermediate between the two species, with some of the characteristics of one and some of the other, thus indicating that these two species are not really distinct. The characters of these intermediate forms are: Dorsal v-xxx, 9; anal n, 26; a small pellucid spot on soft dorsal; coloration brilliant black with longitudinal yellow bands. Dorsal v-xxx, 8; anal ii, 27; a large pellucid spot on soft dorsal; color light reddish-brown, crossed by narrow bands of dark brown; fins yellowish, barred with brown. Dorsal v-xxix, 8 ; anal ii, 25 ; a large pellucid spot on soft dorsal ; coloration perfectly plain. Dorsal v-xxx, 9; anal n, 28; small pellucid spot on soft dorsal; coloration perfectly plain. I find similar intergradations in specimens from San Diego. It appears that the two forms can not be distinguished at any point throughout their range. The number of fin rays and the coloration will not serve to separate the species; therefore G. evides must be identical with G. elegans, which has the priority of description. BULL. U. S. F. C. 1899. PLATE 10. 1. SYNAPTA ROSEOLA . NATURAL SIZE. 2. A PART OF THE SAME MAGNIFIED FOUR TIMES. 3. SYNAPTA INHAERENS (O. F. MULLER). NATURAL SIZE. Contributions from the Biological Laboratory of the U. S. Fish Commission Woods Hole, Massachusetts. THE SYNAPTAS OF THE NEW ENGLAND COAST. By HUBERT LYMAN CLARK, Professor of Biology , Olivet College , Michigan. Through the kindness of the United States Commissioner of Fish and Fisheries special opportunities were enjoyed during the spring and summer of 189S for carrying- on biological investigations at the laboratory of the Fish Commission at Woods Hole, Mass. The few weeks at my disposal were devoted to the study of the two liolothur- ians of the genus Synapta common at that place. Primarily the object in view was to determine the systematic position of our New England synaptas, their relation to each other and to European forms. At the same time experiments were carried on designed to throw light on the function of certain organs and on the possibilities of regeneration in Synapta. The latter were, however, limited by the shortness of my stay. For the privileges of the laboratory I desire to express my thanks to the Commis- sioner, and particularly to the director of the laboratory, Prof. IP. C. Bumpus, whose constant kindness, and sympathy made the work doubly pleasant. I desire also to acknowledge my indebtedness to Mr. Charles M. Pratt, of New York City, for a very fine lot of synaptas from Naples, without Avbich the relation of the American to the European species could not have been positively determined. THE SYSTEMATIC POSITION OF THE NEW ENGLAND SYNAPTAS. On February 5, 1851, at a meeting of the Boston Society of Natural History, Mr. W. O. Ayres (’51) described under the name tenuis a synapta which he had found abundantly in Boston Harbor and also at Provincetown, Mass., and Sag Harbor, Long Island. He separated it from the European species on account of slight differences in the “ hooks” and “ plates.” How little the anatomy of the animal was understood is shown by the description of the calcareous ring which he says consists “ of 12 pieces of granulated structure, some of which are pierced with holes for the admission of water in respiration.” The same year Pourtales (’51) read a paper at the meeting of the American Association for the Advancement of Science, in Cincinnati, “ On the Holo- thurise of the Atlantic coast of the United States,” in which he described the common New England synapta under the name girardii , but he says that the only differences he can find between it and the European S. inhaerens (O. F. Mull.) are that the anchors are less curved and the plates more rounded. He seems to have been unacquainted with Ayres’s paper, which was presented three months before his own. In 1867, Verrill (’67) refers to our common synapta under the name of tenuis, and proposes to make it the type of a new genus Leptosynapta, “distinguished by their more slender form, the absence of prominent verruca;, fewer (12), shorter, and more 21 22 BULLETIN OF THE UNITED STATES FISH COMMISSION. digitate tentacles, etc.,” from “the typical species of synapta,” which “have 15 tenta- cles and prominent verrucse.” This proposed genus has not been accepted by any writer on holothurians, because the form of the body and the prominence of the verrucfe are so closely connected with the degree of contraction of the muscles that they are virtually worthless as characters, while the number and shape of the tenta- cles differ in different species to such an extent that generic distinctions can not be based exclusively on them. Since Yerrill proposed his genus Leptosynapta three species of synapta with normally 13 tentacles, one with 11, and two with 10 have been described, and species with 20 and 25 are also known. Doubtless the genus Synapta as at present constituted includes two or more natural genera, but Yer rill’s genus Leptosynapta is not sufficiently well characterized to stand. The same year in which this change was proposed saw the publication of Selenka’s (’67) well-known mono- graph'on the Holothurians. He pointed out that the name tenuis had been used for a synapta by Quoy & Gaimard in 1833, and accordingly suggests ayresii as a name for the form Ayres had called tenuis. He gives girardii Pourtales as a form from Cape Florida, evidently having read the original description hastily or carelessly, and he entirely overlooks its similarity to tenuis Ayres. In addition, he describes a species gracilis from Boston Harbor, which, as Theel (’86) has pointed out, is obviously based on specimens of tenuis Ayres, in which the calcareous bodies have evidently undergone some change due to the alcohol or other preservative used. In 1874, Yerrill (’74) gives the commonJSTew England synapta the name Lepto- synapta girardii , accepting Selenka’s objection to the name tenuis and recognizing the fact that Selenka’s names ayresii and gracilis are but synonyms of Pourtales’s name girardii. At the same time lie gives a very brief and hasty description of a species of synapta from the New England coast which he calls Leptosynapta roseola. Thii'teen years later Lampert (’85) gives gracilis Sel. the rank of a good species, but places tennis Ayres, girardii Pourt., and ayresii Sel. as synonyms under 8. inhcerens (O. F. MiilL). He ignores Yerrill’s roseola altogether, not even giving it the place of a synonym, though in his literature list he gives Yerrill’s paper (1874) as containing the description of one new synapta. The following year Theel (’86) placed the names tenuis and girardii in the list of synonyms of inhcerens. He says of Selenka’s gracilis that it “seems very doubtful; doubtless identical with inhcerens ,” and of Yerrill’s roseola , “doubtless not a distinct species.” The very best authority on the group, Ludwig (’92), gives gracilis Sel. and roseola Ver. as good species, but ignores tenuis Ayres, girardii Pourt., and ayresii Sel., apparently regarding them as synonyms of inhcerens. American zoologists have generally followed Yerrill in recognizing two American species of synapta, both distinct from the European species, and the names Leptosynapta girardii and Leptosynapta roseola have been in common use in this country. European zoologists, on the other hand, have generally credited us with but a single species, and that one identical with the common European form, inhcerens. A careful comparison of numerous specimens convinces me that truth lies between these two positions. American zoologists are right in supposing we have two species, but European writers are correct in considering girardii identical with inhcerens. The identity of the two forms seems to me to be beyond question, for there are absolutely no constant differences in either the gross or microscopic anatomy. A comparison of anchors and plates, both in size and shape, shows that while the individual differences may be great, they are entirely inconstant, and I could not find a single point by which the specimens from Naples could be distinguished from those collected at Woods Hole. THE SYNAPTAS OP THE NEW ENGLAND COAST. 23 The name and synonymy of the common white synapta of the New England coast are therefore: Synapta inhaerens, (O. F. Miill.), Zool. Dan. 1779-1784. Synapta tenuis, Ayres, Proc. Bos. Soc. Nat. Hist., vol. IV, p. 2. girardii, Pourtales, Proc. Am. Ass. Ad. Sci., 1851, p. 14. ayresii, Selenka, Zeit. f. wiss. Zool., 1867, vol. xiv, p. 362. gracilis, Selenka, Zeit. f. wiss. Zool., 1867, vol. xiv, p. 363. Leptosgnapta tenuis, Verrill, Trans. Conn. Acad., vol. i, p. 325. girardii, Terrill, Rep. on Inv. Ani. of Vineyard Sound, p. 422. The status of Verrill’s Leptosynapta roseola is quite different, and I believe that it must be accepted as a valid species. In living specimens there is never the slightest difficulty in distinguishing between roseola and inhcerens. I have never seen but one specimen in which the color alone was not sufficient to distinguish them, and in that specimen even a very superficial examination showed that it was roseola. Alcoliol makes the color differences even more marked, so that in preserved material the two species are very' easily separated. But there are three points in the microscopic anatomy of roseola which serve to distinguish it readily from inhcerens , and on these three points its claim to good standing as a species must rest: (1) The most important point is found in the radial plates of the calcareous ring. In inhcerens these plates are about three-fourths as high as broad, and each is perfo- rated through the center for the passage of the radial nerve. In roseola , on the other hand, the radial plates are oidy about half as high as they are wide, and they are not perforated, the radial nerve passing over the top of each in a shallow notch. At first I thought this was simply a stage of development in the formation of the plate, but I was finally convinced that such could not be the case. No specimen of inhcerens, however small and immature, had a radial plate which was not distinctly perforated through the center, and no specimen of roseola , however large, showed any condition beyond the notch. The differences in the calcareous ring are shown in figs. 1 and 2, plate 11. (2) The calcareous spicules of the body wall present the second distinguishing characteristic: of roseola. In form, proportions, and distribution of anchors and plates I could find no constant differences between the two species; and Verrill errs in saying that in roseola “the perforated plates are smaller and the anchors relatively much longer, with a very slender elongated shaft.” Careful measurements, with the aid of a camera lucida, of a large number of anchors and plates taken at random in specimens of roseola and inhcerens from Naples and from Woods Hole, gave the following results: Species and locality. Length of anchor. Breadth of shaft. Ratio of breadth to length. Length of arms. Ratio to length of anchor. Breadth of arras. Ratio to length of anchor. Length of plate. Ratio to length of anchor. Breadth of plate. Ratio to length of anchor. Inhaerens, Naples 177 m 17+ m . 10 44 ix .25 83 m .47 131 fx .77 ioom .56 Inhaerens, Mass 161 17+ 108 49 .30 92 .56 135 .82 93 .57 Roseola, Mass 142 16 11 38 .27 76 .53 115 .81 72 .57 ix Abbreviation for micron. In a still larger series: Species and locality. Average length of anchor. Average length of plate. Ratio of plate to anchor. Inhserens, Naples 179m 133 m .74 Inhasren^, Massachusetts 177 139 .78 Roseola, Massachusetts 164 131 .79 24 BULLETIN OF THE UNITED STATES FISH COMMISSION. It would be absurd to attempt to make any specific character dependent on such slight differences, but in the other calcareous particles in the body wall, roseola dif- fers markedly from inhcerens. In the skin of the tentacles, especially on their inner side, we find in inhcerens some small, simple particles, not at all branched (fig. 3). In roseola these particles are more numerous and are always very much branched and perforated (fig. 4). In the longitudinal muscles of inhcerens are numerous small, round, or sometimes dumb-bell shaped, particles; they are never branched or perfor- ated (fig. 5). In roseola the concretions of the longitudinal muscles are either C-shaped or perfect circles; more rarely they are somewhat branched (fig. 6). These differences, though seemingly slight, are remarkably obvious and constant and no signs of intergradations were found. (3) The third characteristic of roseola is found in the size and shape of special ciliated funnels. In both species these large ciliated funnels seemed to be confined chiefly to the mesentery of the left dorsal interradius and occur singly every milli- meter or two. Semon (’87) speaks of two sorts of ciliated funnels in digitata , but he does not seem to have observed them in inhcerens. In the latter species the large fun- nels are from 400// to 1,200// high, aud from 150// to 400// in diameter. Their shape is shown in fig. 7. In roseola they are smaller and much more slender, measuring less than 300// high and 80// in diameter (fig. 8). In addition to these points of anatomy, there are noticeable differences in habitat between the two species; inhcerens occurs in sand or sandy mud, or even in pure mud, less commonly along gravelly shores; roseola occurs on rocky or gravelly shores under stones or among the pebbles, and never in pure sand or mud. I have never found roseola except where there was sufficient iron present in the soil to give it a decidedly rusty color, and it has occurred to me that there might be some connection between the very unusual amount of pigment developed in roseola and this excess of iron. Some specimens of inhcerens show more or less pigment wheu carefully examined, and several specimens among those received from Naples were as rosy in color as the average roseola , but none of them show any approach to that species either in the calcareous ring, the concretions in the body wall, or the large ciliated funnels. Until intergradations are shown and some better explanation is offered of the constancy with which these characters separate roseola and inhcerens , it seems to me they must be regarded as distinct species. As Verrill’s description of roseola is so incomplete and is also erroneous, I venture to give the following summary of its characters: Synapta roseola (Ver. ). Leptosynapta roseola, Verrill. Rep. on the Inv. Ani. of Vineyard Sound, 1874, p. 422. Synapta roseola, (Ver.), Thdel. Report of the Challenyer, The Holothurians, vol. xiv, 1886, pt. xxxix, p. 25. More slender than inhcerens and general appearance much more soft and delicate; usually much smaller, rarely exceeding 100 mm. in length, even when extended. Body wall white or colorless, thin, hut thickly covered with verrucae, which contain numerous pigment granules of a reddish color, giving a generally bright rosy color to the animal. The pigment resembles that found in S. digitata (Mont.) (see Semon, ’87) in that it is scarcely at all bleached by alcohol even after mouths, but it is entirely destroyed by acids or corrosive sublimate. Rarely the body wall is yellowish or pale buff, making the general effect reddish-yellow. Tentacles 12, each with 2 or 3 (rarely 4) pairs of digits and with 7 to 15 sensory cups on the inner side. Genital glands much branched, and when filled with the sexual products very conspicuous through the pink skin. Polian vessel generally single. No cartilaginous ring. Calcareous ring narrow, the radial pieces not perforated for the passage of the nerves, but simply notched on the upper edge. Ciliated funnels numerous, of two kinds; the largest ones, measuring about 300// high by 60// in diameter, infrequent, confined almost exclusively to the left dorsal mesentery; the smaller ones measure hardly one-fourth as much in height but Bull. U. S. F. C. 1899. (To face page 24.) Plate 1 1 . EXPLANATION OF PLATE 1. Part of calcareous ring of Synapta inhcerens. 45 X- 2 Part of calcareous ring of Synapta roseola 45 X. 3. Calcareous particles from the tentacles of S infuerens. 450 X 4. Calcareous particles from the tentacles of S- roseola. 450 X. 5. Calcareous particles from the longitudinal muscles of Synapta inhcerens. 450 X . 6. Calcareous particles from the longitudinal muscles of Synapta roseola. 450 X. 7. Large ciliated funnel of Synapta inhcerens. Seen from behind 1 25 X . 8. Large ciliated funnel of Synapta roseola. Seen from in front. 337 X. THE SYNAPTAS OF THE NEW ENGLAND COAST. 25 nearly as much in diameter. Anchors and plates not essentially different from those of inlicerens. Calcareous rods in the sides of the tentacles and in the digits slightly curved and knobbed as in inhcerens; besides these, numerous branched, curved, and perforated rods and plates occur abundantly on the inner surface near the base of the tentacles. Calcareous particles in the longitudinal muscles C or doughnut-shaped, rarely branched. This species occurs between high and low water mark in gravelly banks or on rocky beaches, where there is considerable iron in the soil. Usually found near the surface and often underneath rocks. PHYSIOLOGICAL NOTES. Contrary to previous experience and to several writers on holothurians, I found during the summer of 1898 that Synapta inlicerens is very easy to keep in aquaria and roseola is about as hardy. On account of the greater abundance and larger size of the former, most of my observations were made on that species. When left in a vessel containing sea water only, the synaptas crawl about restlessly on the bottom, and unless fresh sea water is supplied they soon begin to constrict off parts of the body, beginning near the posterior end, and after a time nothing but small pieces will remain, and these soon die. If the supply of sea water is abundant and well aerated, this process may be delayed some hours, but it usually occurs in less than half a day. When, however, there is a sufficient amount of clean sand in the dish to allow the animals to burrow at will, they will live indefinitely if the supply of water is con- stantly renewed. By filling glass jars half full of sand, one is able not only to keep synaptas alive, but to study more or less of their underground habits. I agree entirely with Cu^not (’91) in believing that autotomy is not normal or defensive, but is due entirely to pathological conditions. I never saw a case of it in synaptas supplied with plenty of sand and an abundance of sea water. The fact that portions constricted off can not live is good reason for supposing the process is abnormal. An excess of magnesium sulphate in the water causes stupefaction and ultimate death, so that excellent specimens, either for laboratory purposes or for the cabinet, may be obtained by narcotizing with this salt and killing in strong alcohol or corrosive sublimate. Both species of synapta breed during the spring and early summer. The sexual glands are well developed by the last of April, and individuals with ripe ova may be found well into August. About the last of June or early in July seems the height of the breeding season at Woods Hole. Personally I have had no success with artificial fertilization of the eggs, but Dr. W. R. Coe, of Yale University, states that he has found no difficulty in fertilizing the eggs of 8. roseola, artificially, though he has made no attempt to carry their development beyond the segmentation stages. In that species the ripe genital glands show plainly through the body wall, and individuals in which the male elements fill the glands are easily distinguished from those in which the ova are mature. Further investigations into the breeding habits and embryology of our two synaptas are very much to be desired. Synaptas burrow into the sand head first and almost always go straight downward for some distance, but when once completely buried, they turn in any direction up or down or on the horizontal plane. They can and do turn in their burrows, but as a rule they make new tubes when they come to the surface. They are seldom still, and the old idea that they remained in the tube they have formed with their tentacles just above the surface is scarcely true. Sometimes they assume that position, but seldom remain so very long. They rarely leave their burrows and come out on the surface of the sand, and I doubt if they ever do so under normal conditions. Passage through the sand is chiefly accomplished by means of contractions and extensions of the body, but is materially assisted by the tentacles. With the latter, BULLETIN OF THE UNITED STATES FISH COMMISSION. 26 which are almost continually in motion, the sand is loosened and the grains more or less separated. By the contraction of the longitudinal muscles the rear of the body is brought up nearer to the head, and then the circular muscles contract and extend the body again. It is prevented from slipping back by the anchors, which are elevated by the contraction of the circular muscles and hold against the sand. Since the contraction begins next to the rear end and moves forward, the head end is pushed onward, the anchors there lying flat in the skin. This process of alternate contractions of the two sets of muscles is very obvious to an observer, and takes place very continuously, though not rapidly. In this way a synapta can move through the sand from 2 to 3 centimeters a minute, and an inhcerens of average size can get entirely out of sight in 5 or 6 minutes. One of the most remarkable provisions for the use of the anchors in locomotion is their much greater abundance and their considerably greater length in the posterior part of the body. The use of this is clear when one realizes how the rear of the body acts as the resisting base against which the muscles work in pushing the anterior end forward. This dif- ference between the anchors of the anterior and posterior ends of the body seems to have been overlooked hitherto, so great an authority as Ludwig (’98) saying of inhce- rens “Anker und anker platten differieren in vorderen und hinteren korperabschnitt nicht merklich von einander.” In the same paper, however, he calls attention to the fact that in 8. digitata the anchors of the posterior end of the body are about 50 per cent longer than those in the anterior end. While the difference in inhcerens is not quite so great as that, yet it is very noticeable, being above 33 per cent. In both species there is an increase also in the length of the plates, but, as would naturally be expected from the passive part they play, it is not nearly as great. The following figures will bring out the condition in inhcerens very plainly. A specimen 14 cm. long was chosen at random and cut into seven approximately equal pieces, and in each piece 10 anchors and plates, selected entirely at random, were measured with these results, No. 1 being the most anterior, No. 7 the farthest back; measurements are all in microns: No. 1. No. 2. No. 3. No. 4. No. 5. No. 6. No. 7 Average length of anchor. Average length of plate. Average length of anchor. Average length of plate. Average length of anchor. Average length of plate. Average length of anchor. Average length of plate. Average length of anchor. Average length of plate. Average length of anchor. Average length of plate. 1 / Average length of anchor. Average length of plate Number 141 124 153 134.5 162.5 135.6 164 138 202 157 204 145 197 135 88 . 87 .83 84 . 75 . 71 .68 Per cent i ncrease in length . ' .09 .08 .06 .008 .009 .017 .23 . 14 .009 — . 077 —.03 —.07 Maximum length of an- - chor in each section 155 171 178 197 220 220 224 The most striking feature of this table is the abrupt and marked increase in the size of both anchors aud plates in the fifth section. The average length of anchors for the first four sections is 155/<; of plates, 133/1; ratio = 86 per cent. For the last three, 201 it is the average length of the anchors, 144 of the plates, and the ratio - 71 per cent. That is, the anchors increase 29 per cent and the plates 8 per cent. The slight decrease in the last section is probably due to the fact that comparatively little strain could be brought upon the anchors situated so very near the tip of the body, but it is noticeable. that the longest anchor of all was found in that section. The THE SYNAPTAS OF THE NEW ENGLAND COAST. 27 increase in the leugth of the anchors of the last two sections over those of the first one is over 40 per cent. In two other specimens of inlicerens and one of roseola, also selected at random, the following measurements in microns were made: Measurements taken. Inliaerens. Roseola. Length of longest anchor anteriorly 160 160 151 Length of longest anchor posteriorly 226 249 195 Percentage of increase in length of anchors 41 56 29 Length of longest plate anteriorly - 144 131 125 Length of longest plate posteriorly 169 178 153 Percentage of increase in length of plates 17 36 22 Average length of anchors anteriorly 149 140 142 Average length of plates anteriorly 134 121 118 Ratio of plates to anchors anteriorly .90 .86 .83 Average length ot anchors posteriorly 200 220 188 Average length of plates posteriorly 148 156 145 Ratio of plates to anchors posteriorly .74 .70 .77 Percentage of increase in average anchor’s length 33 57 38 Percentage of increase of average plate’s length 10 29 23 These figures prove conclusively that the anchors at the posterior end of the body are from 30 per cent to 40 per cent longer than those near the head, while the plates only increase 10 per cent to 25 per cent, and, consequently, there is a decided drop in the ratio between anchors and plates in the two regions. Now, in regard to the relative abundance of anchors and plates anteriorly and posteriorly, the following figures show that not only are the anchors shorter near the head but they are decidedly less frequent: Species. Average number of anchors per s(j. mm. anteriorly. Average number of anchors per sq. mm. posteriorly. Pecent- age of increase. Xnhserens 10.3 14 36 Do 6.8 11.2 65 Roseola 12 23.6 96 Average of a large series 9.8 15.6 59 There does not seem to be any distinction in the size or abundance of the anchors between the dorsal and ventral surfaces of the animal. Semon (’88) has expressed the opinion that synaptas are not subterranean in their mode of life, and he bases his opinion largely on the color of S. digitata. The question, however, does not admit of debate 'so far as inhcerens and roseola are concerned, for not only do they never appear normally on the surface of the sand either along shore or in aquaria, but the arrangement of the anchors just described could only be of much practical service in a close fitting burrow. The whole structure of the animal shows modification adapting it to underground life and the increase in size and number of the anchors posteriorly is one of the most interesting. When placed in a glass dish without sand, synaptas soon cease to contract the muscles of the body wall but drag themselves slowly along by means of the tentacles, thus showing that the anchors are of little use on a smooth surface. In crawling by means of the tentacles, the glandular outside of the tentacle tips and digits is used, so that many of the tentacles are 28 BULLETIN OF THE UNITED STATES FISH COMMISSION. continually twisted around, making the process appear somewhat awkward. When in their tubes, however, the outer side of the tentacles would be most naturally used, and it is by their adherence to that side of the tentacles that particles of food and sand are brought to the mouth. This use of the outer side of the tentacles in locomotion was observed by Pourtales (’51) in the synapta from Florida, which he called 8. viridis. The New England synaptas seem to have organs of special sense of two kinds, olfactory and equilibratory or positional; no evidence of ability to distinguish between light and darkness could be detected and there are no anatomical structures to which this sense could be ascribed. The sense of smell is centralized in the cups on the inner side of the tentacles; that of position in the so-called “otocysts” or “auditory” organs. Semon (’87) has shown that 8. inhcerens possesses the sense of smell, and similar experiments made at Woods Hole confirm his results. If a piece of any rank- smelling substance is placed near the tentacles of a synapta they are immediately retracted and the head is turned away. Small bits of decayed starfish were used and it was found that the synaptas would avoid them even when they did not actually touch them, and they seemed to avoid sand with which decayed starfish had been mixed. While the experiments were not conclusive by themselves, they were satis- factory as confirmation of Semou’s work. In regard to the so called “otocysts” more numerous and more careful experiments were made, with results which seem to show conclusively that these organs are used to show the position of the animal in the water or sand. Semon (’87) has demonstrated the fact that synaptas appear to have no sense of sound and no ability to detect even strong vibrations in the water. He, however, left the function of these “otocysts” undetermined, but in a previous paper (Clark, ’98) I suggested that in S. vivipara they determined the animal’s position in the water. Experiments on inlmrens and roseola have fully confirmed this opinion and there can no longer be any doubt that this is their true function. In regard to their structure there is little to be added to Semon’s (’87) description, as it has not yet been possible to demonstrate the cilia which line them. That they are lined with cilia, which are in constant motion, can be easily seen in the living animal under the microscope. The siugle vesiculated cell or “otolith” which each “otocyst” contains may be seen to be constantly revolving and never actually rests against the wall of the sac, but is kept out from it, apparently, by these cilia. Cuenot (’91) in his description of these organs says that there are “un grande nombre d’otolithes spheriques” in eacli “otocyst,” and Semon (’87) says there are one or more (often six or more) “blaschen” in each sac. I have examined a large number of living specimens of both inhcerens and roseola aud I have never yet found a case where there was more than one. Whether these so-called otoliths are calcareous, as Cuenot thinks, or vesiculated cells filled with fluid, as Semon says, I have not been able to determine positively, though I incline to the latter view, as they do not appear calcareous when compared either by transmitted or reflected light with any undoubtedly calcareous body. But there can be no doubt that they are heavier than the fluid which surrounds them in the sac, as may be readily shown by the following simple experi- ment: If a pair of the sense organs are cut from a living synapta and placed under a cover-slip and examined under the microscope, the otolith will be seen in the center of the sac so long as the slide is perfectly horizontal. But if the microscope be tipped, so that the slide approaches a perpendicular position, the otolith will be seen to sink slowly to the lower side. (Owing to the reversal of upper and lower sides under the THE SYNAPTAS OF THE NEW ENGLAND COAST. 29 microscope, of course the otolith appears to rise.) If the slide be turned around slowly the otolith keeps constantly at the lowest point, though kept out from the wall and in motion by the cilia. In the case of 8. vivipara I expressed the opinion that this otolith or vesiculated cell floated in the liquid and by touching the cilia or sensory hairs at the highest point gave rise to sensations of changed position. Now I am convinced that the inclosed cell does not float at all but always rests upon the cilia at the lowest point, thus arousiug new sensations with every change of position. That the animal is affected by change of position was proven by the following experiment: Synaptas were placed on a piece of thin board which sloped sharply to the bottom of a dish of sea water. They always sought the bottom of the dish, no matter in what position they were placed on the board. Not a single instance occurred of the animal crawling upward. A single synapta was placed on the board and after it was well started on its way down the slope the board was very gently reversed so that the lower end became the higher. As soon as the board passed beyond the horizontal plane the synapta would stop and as the slope became greater it would turn and start back in the opposite direction, and every change in the slope of the board caused a reversal of direction in the movement of the synapta. This experiment was tried a number of times and on different individuals, but with unvarying results. The change in the position of the board was accomplished with the least possible disturbance of the water, and no one who saw the actions of the synaptas could doubt that it was the change of position which caused the change in the movement of the animal. In the light of these facts and the entire absence of any evidence to show that sound vibrations of any sort can be detected by these or any other organs, it would seem out of place to speak of them as “auditory” organs or even as otocysts , and I suggest the name positional organs. A number of experiments were made to determine if possible the function of the ciliated funnels, and while the results were not entirely satisfactory they throw some light, I think, on the use of these curious organs. A large amount of carmine was thoroughly mixed with clean sand and, after it had settled and fresh sea-water had run over it for a little while, half a dozen specimens of inlmrens were placed in the dish. They lived in the brightly colored sand for one week and were then washed and killed. They showed a very decided pink tinge, which was due only in a very slight degree to the carmine attached to the skin externally. Microscopic examination showed that in the connective tissues there were numerous reddish-brown granules and these were most abundant near the lines of ciliated funnels. No such granules were ever found in synaptas living in sand free from carmine. While this experiment seemed to indicate that carmine taken in with the sand was in some way absorbed and formed the granules, it does not show that the ciliated funnels were in any direct way connected with the process. Accordingly another line of experiments was begun, carmine mixed with sea-water or the body-cavity fluid of other synaptas being injected directly into the body cavity by means of a fine canula through the body wall. Synaptas so treated were then killed at intervals of about twenty-five minutes for two or three hours, and then at intervals of several hours up to twenty-four. It was found that in a very few minutes the carmine began to gather along the lines of funnels and in a shorter or longer time, according to the individual synapta, the body would resume its normal more or less transparent whiteness, except along those lines which appeared as dark-red longitudinal stripes. An examination 30 BULLETIN OF THE UNITED STATES FISH COMMISSION. under the microscope showed that the funnels were actually choked up by the excess of carmine, sometimes being almost buried in it. After a few hours these dark-red lines became less prominent and in a few days they generally were completely obliterated. Microscopic examination of the various stages showed plainly that the ciliated funuelswere very closely concerned with this change, and that with the disappearance of the carmine from the body cavity the reddish-brown granules appeared in the connective tissue. In spite of careful study and long search I was never able to dis- cover a grain of carmine or one of the granules actually in the stalk of the funnel, yet in some way the carmine must pass from the inside of the funnel into the body wall. Since it is an undoubted fact that the stalk of the funnels is solid and not a tube, the only way by which particles could be conveyed from the cavity of the funnel into the connective tissue would be by means of “wandering cells,” and I believe that is the process which actually goes on. Semon’s (’87) view of the funnels, as “grosse und complicirt gebaute Lymphstomata der Leibesholhe,” seems to me a correct one, and I believe they perform their excretory function not only by sweeping up and collecting the waste matter in the body-cavitv fluid, but also, as Semon suggests, by acting as starting points for the movement of ‘-phagocytic” wandering cells which destroy or carry into the connective tissue of the body the waste matter collected by the funnels. It seems to me that in view of these facts the ciliated funnels are evidently associated with the function of excretion and it is proper to regard them as excretory organs. Cuenot (’91) considers that their function is to keep up currents in the body-cavity fluid, but every movement of the animal starts new currents so that special organs for that purpose would be superfluous. A number of experiments were made to test the tenacity of life and the possi- bility of regeneration in synaptas, and the results show that inhcerens is not a very sensitive animal. If an individual is cut in two the anterior end will live and grow as well apparently as any normal specimen, but the posterior end will only live for a few hours, or perhaps a day. So far as I could see, the only reason for its death was its inability to take in food ; and I am inclined to think that if food could be provided the posterior half would live as well as the anterior. It seemed to make no difference whether the bisection occurred near to the head or far from it; the head always lived, and in the course of two weeks would show perceptible signs of growth. All that seemed to be necessary was the mouth and a small part of the digestive tract. That it was not the tentacles which were essential was shown by the fact that synaptas lived all right without them. Two or more tentacles were cut from a number of syn- aptas— in two cases every one being removed — yet they all lived and burrowed in the sand with more or less ease; and not only did they live, but regeneration began at once, so that in two weeks the new tentacles were large enough to bear a digit on each side. At first I thought the nerve ring was the essential part, but that seems to be doubtful, for the nerve ring was carefully and completely severed in a synapta without apparently causing any inconvenience. In other specimens it was cut in two or even three places, but with the same result. In no case were any serious effects shown, and the animals lived and burrowed in the sand with apparently as much ease as ever. I made no microscopical examination to show whether the nerve ends reunited or not, but the cuts apparently healed in a short time. The most striking- fact was that with the cutting of the nerve ring there was not the least evidence of any lack of coordination in the movements of the tentacles, nor of the muscles of the body. THE SYNAPTAS OF THE NEW ENGLAND COAST. 31 CONCLUSIONS. For the sake of clearness and conciseness the results of these investigations may be summarized as follows: 1. The common white synapta of the New England coast is S. inhcerens (O. F. Mull.). 8. girardii Pourt. and 8. tenuis Ayres are synonyms of that species. 2. Verrill’s genus Leptosijnapta can not stand. 3. There is a second species of synapta found in suitable places along the New England coast easily recognized by its reddish color, 8. roseola Ver. 4. 8. roseola Ver. differs constantly from 8. inhcerens in the calcareous ring, the deposits in the tentacles and longitudinal muscles, and in the large ciliated funnels. 5. There are no important differences between the anchors or the plates of the two species. 6. The anchors are of real use to the animal in moving in its burrow, and those at the posterior end are most so. Accordingly we find they are there longer, larger, and more abundant. 7. The so-called otocysts are not auditory but are undoubtedly positional organs. 8. They never contain but a single otolith. 9. The sensory cups on the tentacles are seemingly olfactory organs.- 10. The ciliated funnels are large and complex lymphstomata and may properly be called excretory organs. 11. Regeneration takes place readily and rapidly in synaptas where the mouth and part of the adjacent digestive tract are left intact. Amherst College, Amherst , Mass., October , 1898. LITERATURE consulted. Ayres (’51). Notices of llolothuridae. Proc. Bost. Soc. Nat. Hist., vol. iv. Ci.ark (’98). Synapta vivipara: A Contribution to the Morphology of Echinoderms. Memoirs of Bost. Soc. of Nat. Hist., vol. v, No. 3. Cuenot (’91). Ltudes morphologiques sur les Echinodermes. Arch, de Biol., vol. xi. Lampert (’85). Die Seewalzen, etc. Wiesbaden, 1885. Ludwig (’92). Die Seewalzen: In Bronn’s Klassen nnd Ordnungen des Thier-Reichs. Leipzig, 1889-92. (’98). Einige Bemerkungen fiber die Mittelmeerischen Synapta-Arten. Zool. Anz., No. 549. Pourtales (’51). On the Holothurise of the Atlantic Coast of the United States. Proc. Am. Assoc. Adv. Sci., 5th meeting. Selenka (’67). Beitriige zur Anatoime nnd Systematik der Holothurien. Zeit. f. wiss. Zool., vol. xvii. Semon (’87). Beitriige zur Naturgeschichte der Synaptiden des Mittelmeeres. Mitth. Zool. Sta. Neap., Bd. vn. (’88). Die Entwickelnng der Synapta digitata, etc. Jena. Zeit. f. Naturw., Bd. xxn. The&l (’86). Report of the “Challenger,” vol. xiv, pt. xxxix. The Holothuroidea. Verrili. (’67). Notes on Radiata. Trans. Conn. Acad., vol. i, pt. 2. (’74). Report on the Invertebrate Animals of Vineyard Sound. Washington, 1874. Bull. U. S. F. C. 1899. (To face page 33.) Plate 12. SCUP ( Stenotomus chrysops). Three views of the same fish taken while alive in the aquarium of the United States Fish Commission at Woods Hole, Mass Contributions from the Biological Laboratory of the U, S. Fish Commission, Woods Hole, Massachusetts. THE GAS-BUBBLE DISEASE OF FISH AND ITS CAUSE. By F. P. GORHAM, A. M., Assistant Professor of Biology , Brown University. For several years it has been noticed that many salt-water fish when kept in aquaria develop, after a longer or shorter period, a disease which is first manifested in the formation of vesicles of gas on the fins or other parts of the body. These vesicles gradually increase in size and number, and finally invade all superficial parts of the animal. The fins, one after another, become affected, and the vesicles frequently form in the eyeball, beneath the cornea, or in the loose connective tissue of the orbit, so that the eyes are forced from their sockets; less frequently the bubbles gather beneath the mucous membrane lining the mouth and gill-arches, or beneath the integument, partic- ularly along the lateral line, so that the scales are raised from flie surface. The pres- ence of these vesicles often disturbs the equilibrium of the fish so that it swims about with its head elevated, or, more frequently, directed downward or tilted to one side. The disease has been noticed only in fish kept in the aquaria. It has not been found in specimens taken from the sea, nor lias it been observed in fish kept in “cars” in deep water, nor in those retained in “pounds,” the water of which is from <> to 12 feet in depth. The affected fish live a shorter or longer time after the vesicles begin to appear, some succumbing in a few hours, others resisting the disease for several weeks. Young “ puffers” usually die in less than 24 hours after being placed in the aquaria, but several scup lived for weeks after their eyes had actually dropped from their sockets. The following fish were affected by the disease during the spring and summer of 1898, and the list includes all the fish, with two exceptions, which were kept in the aquaria of the Fish Commission at Woods Hole, Mass., from March to September: Common name. Scientilic name. Bathymetric range. Fathoms. 1 to 1 1 1 1 1 1 30 30 30 30 15 300 150 80 80 10 35 100 40 10 F. C. B. 1809—3 33 34 BULLETIN OF THE UNITED STATES FISH COMMISSION. The following invertebrates showed signs of the disease: Squid, naked mollusks, scallops ( Pecten tenuicostata), hydroids ( Paryplia crocea). Squid egg-sacs and some of the green alga: also developed internal bubbles apparently of the same nature. Minnows (Fundulus heteroclitus and Cyprinodon variegatus) and skates ( Raja erinacea ) in no case were affected. In most fish the disease manifests itself first on the fins, but in the scup it attacks the socket of the eye and gradually forces the eyeball from the orbit (plate 12). In the adult puffer the dorsal fin is first attacked; in the young puffer the base of the tail. In the pipe-fish the bubbles appear about the snout and later spread to other parts of the body. In young winter flounders (about 1 centimeter long) the body cavity frequently contains a large bubble. The disease is not confined to the aquaria at Woods Hole. Mr. L. B. Spencer, who is in charge of the Battery Park Aquarium in (New York City, writes me that he has noticed it there for several years. Prom the appearance of the fish, from the rapid development of the disease in fish introduced into the aquaria, and from the nonappearance of the disease in fish outside of the aquaria, it at first seemed to me that the aquaria had become infected with some disease producing organism, that the organism was a “gas producer,” and that it spread rapidly through the tissues of the fish and produced the gas which collected as bubbles in various parts of the body. With this idea in mind the task of finding, isolating, and studying the organism was begun. All of the tissues of the fish were searched thoroughly for foreign organisms; hundreds of cultures were made on the ordinary and also on special media, such as fish-bouillon, fish-gelatin, and fish -agar. The cultures were kept warm, cold, and at the temperature of the water in the aquaria; cover-glass “ smears” were repeatedly examined; but all to no purpose. Not a sign of any pathogenic organism was found. The only pathological change noticed in the tissues was a remarkably emphysematous condition of epidermis, muscles, connecti ve tissue, fat, etc., in the neighborhood of the vesicles. I became convinced that the disease could not be attributed to the invasion of micro-organisms. Two phenomena suggested a new line of inquiry: (1) None of the small shallow- water fish developed bubbles; (2) the deep-sea fish when brought to the surface by line or dredge often show a protrusion of the eyes and an expansion of the tissues, comparable to the conditions under consideration. Could not the disease be the direct result of the reduction of pressure upon the tissues of the fish, a reduction which must occur when animals habituated to a life in deeper water are compelled to live in the shallow water of the aquaria? To answer this question the following experiment was made: Several young scup were placed in water in a flask; the air was then exhausted from above the water, thus reducing the pressure. As soon as the pressure was at all diminished, the fish immediately sought the bottom of the flask, heads down, and made every effort to seek deeper water and thus regain their normal pressure. By their continued struggles they soon became exhausted and came to the surface. The atmospheric pressure was then restored and they became quiet and swam about naturally. The reduction of pressure was repeated three or four times at intervals of 30 minutes. After the experiment the eyes of the fish showed well-developed bubbles. The experiment was repeated on the following day, with the same fish, with the result that the bubbles enlarged until the eyes began to protrude. THE GAS-BUBBLE DISEASE OF FISH AND ITS CAUSE. 35 Other experiments, indicated in the following table, were made: Fish. Total length of time subjected to reduction of atmospheric pressure. Results. Scup (5 cm. long) Mrs. 1 Mins. 0 Bubbles in eyes. Same fish i 0 Bubbles enlarged ; eyes protrude. Soup (5 cm. long) 1 12 One eye shows bubbles and hemorrhage. Bubbles on pectoral fin and on “feelers. ” Sea-robin (6 cm. long) 3 49 Sea-robin (6 cm. long) 0 14 Bubbles on lower jaw. Puffer 4 4 Bubbles on tail. Minnow ( Cyprinodon variegatus) 6 0 No change. Normal fish were kept as controls in all cases and showed no change. Whenever scup were subjected to this reduction their eyes were affected first; bubbles formed and the eyes protruded in much less time, but in exactly the same way as in scup placed in the aquaria. In the young sea-robins bubbles appeared on the pectoral fins and on the finger-like appendages, while in the young puffers they appeared at the base of the tail. In all these cases the parts first affected were the same as those subject to the disease in the aquaria. The minnows unaffected in the aquaria were unaffected by the removal of the atmospheric pressure. Now, the question naturally arises, will the increase in pressure improve the condition of the fish already suffering from the disease? Small scup taken from the aquarium, already showing protrusion of the eyes and bubbles on the head and fins, were subjected to a pressure equal to 10 feet of water. In 24 hours many of the bubbles had disappeared, and the eyes had returned to their normal condition. Under this treatment puffers recovered from a decidedly diseased condition in 24 hours. The data of these experiments may be tabulated as follows: Fisb. Length of time subjected to pressure of 16 feet of water. 1 Results. Scup with bubbles on head Scup with protruded eye Scup with bubbles on head and protruded eye Puffers (3.5 cm. long). Bubbles at base of tail Puffer (3.5 cm. long). Bubbles on dorsal fin Puffer (3.5 cm. long) . Bubbles on eye and on fin Puffer (3.5 cm. long). Large bubble under right fin. Sours. 24 24 24 24 24 24 24 Bubbles disappeared. Eye nearly normal. Entirely normal. Do. Normal. Eye nearly normal. Bubbles on fin disappeared. Normal. Control animals checked the results of these experiments. In both the reduced and the increased pressure experiments above tabulated young fish were used, but whenever adult animals were employed similar results were obtained. The change in pressure which fish must undergo when transferred from the ocean to aquaria is not small in amount. We have only to consider that at the surface the pressure is about 14.7 pounds per square inch, and that for every foot in depth it increases at the rate of 0.445 pound per square inch. At a depth of 5?} fathoms the pressure is just double that at the surface, and at a depth of 300 fathoms it would amount to over 800 pounds. 36 BULLETIN OF THE UNITED STATES FISH COMMISSION. Now, as may be seen by referring to tlie table of the depths from which the fish under consideration were taken, a reduction of pressure must take place when fish are confined in shallow aquaria. Their tissues relax under the reduction, and any gas present must expand. Since, according to the law of the expansion of gases, a volume of gas occupies just twice the space at the surface that it does at 54 fathoms, a fish drawn from 300 fathoms to the surface must suffer an expansion of its contained gases to over 54 times their original volume. Gas is present in all fish, in the alimentary tract, in the air bladder, and in the blood and other fluids of the body. As this gas expands and seeks an outlet the tissues are loosened and torn apart; the intestine and air bladder are greatly distended and perhaps ruptured, the circulation of the blood is impeded or stopped, and vesicles of gas form in various parts of the body. These phenomena have been noticed for a long time in fish taken from great depths. Such fish are usually dead when they reach the surface, their eyes are pro- truded, their air bladder is ruptured, their intestines are everted from mouth or anus, and their scales are often lifted from the skin. The greater the depth from which they come the more pronounced are the changes. So great is the pressure to which some of the deep-sea fish are subjected ( Melampliaes beanii from 2,940 fathoms sustains a pressure of 3.9 tons per square inch), that the structure of the bones, connective tissues, and muscles is peculiarly modified. Even in fish taken from 15 or 20 fathoms one may note the protrusion of the eyes, the eversion of the intestine, and the noise of the expanding gas working through the tissues. We are thus led to the conclusion that the “gas bubble disease” is caused by a reduction of pressure. Naturally, the disease has never been noticed except in the aquaria; normally shallow- water fish are never affected, and the greater the normal depth of a fish the more severely is it affected. Occasionally certain deep-water fish voluntarily seek the surface, but it is only when they can not return to deeper water that the changes brought about by the expansion of gases become evident. It is interesting to compare these changes in fish with those observed in higher animals when subjected to alterations of pressure. Animals at the surface of the earth are subjected to an atmospheric pressure of about 14.7 pounds per square inch. When this pressure is diminished or increased, physiological changes result. Moun- tain climbers, at an elevation of 2 miles, suffer from increased rapidity of respiration, quickened pulse, painful, throbbing headache, flow of blood from the nose, eyes, and mouth, nausea, and vomiting. On the other hand, when the pressure is increased to from 50 to GO pounds per square inch, as in caissons, workmen experience an annoy- ing sensation about the tympanum, an irritation of the skin called the “puces,” an alteration of the voice, deep, slow, and easy respiration, active digestion, an absence of thirst, and an increased secretion of saliva and urine. Return to normal pressure has the same effect as going into rarefied air. The “puces” is replaced by a chill, respiration becomes difficult, the pulse is rapid and hard, there are pains in thejoints and muscles, and persistent cramps, paralysis, and coma are frequent; death may result in a few minutes. In such cases autopsies have shown congestion of the viscera, emphysematous spots on the lungs, and other indications that the blood contained bubbles of gas.* *1 am indebted to Mr. Frederick T. Lewis for calling my attention to tliese facts. See “The physiological effects of compressed air,” in the Boston Medical and Surgical Journal, October 6, 1898. THE GAS-BUBBLE DISEASE OF FISH AND ITS CAUSE. 37 Experiments have shown that under compressed air the amount of oxygen absorbed by the blood increases with the pressure, according to the law of the absorption of gases. It may be the liberation of this oxygen and other gases under reduced pressure that causes death in the above cases. In recording experiments upon the influence of high pressures upon various animals, Monsieur P. Eegnard says: * A (ish without an air bladder, or one in which the air bladder has been emptied of gas, can be submitted to a pressure of 100 atmospheres, 1,470 pounds per square inch, without injury. When the air bladder is not emptied a very curious phenomenon is observed. Under pressure of several atmospheres the gas of the air bladder dissolves in the blood, and at, the moment of decompression it suddenly becomes disengaged in the blood vessels, forming a foam which stops all circulation, and the animal dies; for the same reason fish drawn from the depths of the sea are dead when they reach the surface. The bearing of this matter of pressure on geographical distribution should be considered. It might seem that there are no barriers to universal distribution in the sea save those indicated by the isotherms, but the isobars must indicate the location of effectual barriers for certain species. Since even slight changes of pressure are detrimental, and each fish swimming in the sea tends to remain at about one level, it is reasonable to suppose that, when a fish leaves this level the difference of pressure becomes uncomfortable and there is a tendency for it to return to its normal habitat. The air bladder is at present considered not an organ under the influence of the muscles of the fish, by means of which it alters its specific gravity, but rather a delicate gauge, which notifies the animal of changes of pressure above or below the normal. Were a fish to pass too far beyond its normal depth, either up or down, the expansion or contraction of the gas of the air bladder would so change the specific gravity of the animal that its muscular activity would no longer be able to cope with the upward or downward tendency, and the fish would be carried to the surface or the bottom. Tims the very presence of an air bladder works the destruction of the fish. Cases are reported where deep-sea fish have come to the surface in this way, evidently having- been carried too far from their normal level in struggles with their prey. It is true that slight changes of level may be brought about by the secretion or absorption of gas by the walls of the air bladder. That the walls of the air bladder secrete and absorb gas has been shown by experiment. 1 have repeatedly emptied the air bladders of fish and found a small amount of gas secreted in 24 hours. But this process of secretion and absorption is relatively slow, and permanent changes of habitat could be brought about only very gradually; moreover, only very slight changes are possible for each species. Thus we see that fish are surrounded by barriers of pressure, and that each species and probably each individual is constrained to remain at a certain level, and the restricted distribution of certain fish is thus explained. * Rechercliea experi men tales aur l’in 11 uence des tres li antes pressions sur les organismea vivants. Conipt. lien., xcviii, March 21, 1884, p. 745. ‘ THE CLAM PROBLEM AND CLAM CULTURE. By JAMES L. KELLOGG, Professor of Biology in Williams College. At the present day the public seems to recognize quite fully the value of the work of our fish commissions — State and national. They know that through the artificial rearing of young shad from the egg the United States Fish Commission has saved our shad industry on the Atlantic coast from inevitable ruin. They have also seen the direct benefit, in thousands of instances, of the stocking of inland streams and lakes with fish which have been hatched and cared for until they were old and strong enough to care for themselves. There does not, however, seem to be so general an appreciation of the fact that the shellfish — the oysters, clams, and scallops — of our eastern coast also need to be very carefully conserved in the immediate future. For some years past there has been considerable study and discussion of the artificial methods of oyster propagation, and in some of the North Atlantic States, especially Rhode Island and Connecticut, “oyster farming” has been the means of saving the industry. But the oyster question is not yet solved, and much remains to be done in the development of new methods of artificial propagation, if the supply is to meet the increased demands of the future. The appropriations for State and national commis- sions are much too small to allow extended investigations along these lines, though public money could probably not be expended with better results to the whole people than in this way. Quite unexpectedly we are confronted with a new problem. Over the greater part of the iNew England coast the supply of clams has suddenly diminished to an extent which has become alarming. Extensive areas which four or five years ago produced great numbers of clams are now practically barren. The explanation is simply that the demand has increased at such a rate that too large a number of the natural “seed” clams have been removed, and extinction suddenly follows. The beds should recover themselves quickly, but one man, in wandering day after day over au area of many acres thus impoverished, is easily able, by digging up the few large clams which he may find here and there, to absolutely prevent the possibility of establishing a new supply. Not only are the larger clams now sought for, but in some localities those which will measure little more than one inch in length are dug up and sent to market. The demand is increasing and prices are rising.* When a certain locality is exhausted, the amount taken from others still productive is conse Large clams sell in the Fulton market for 6 cents apiece. 39 40 ■ BULLETIN OF THE UNITED STATES FISH COMMISSION. quently much greater. We may thus understand why the calamity — for such it really is — has suddenly fallen. In certain localities, of course, this exhaustion of the clam beds took place many years ago, and it is interesting to notice that, for the reasons given above, they have ever since remained practically barren. The history of one of these regions demands especial attention, being particularly instructive at present, because a considerable effort has been made there to reclaim the barren hats. At the mouth of the Essex River, just north of Gloucester, Mass., are some very extensive hats, upon which immense numbers of clams were formerly found. Several years ago these hats were rendered practically unproductive through excessive digging, and have remained so until the present time. About ten years ago the only serious experiment at clam culture of which we have any record in this country was attempted here. It proved to be a failure, and yet the reasons for the failure are easily found, and the fact is demonstrated that the methods employed would have been entirely successful if the experimenters had been protected from outsiders and from each other. In the report of the United States Commissioner of Fish and Fisheries for 1891 we find the following reference to the Essex experiment, quoted from Mr. Ansley Hall : I found quite an interesting feature in connection with the clam fisheries at Essex, Mass., in the shape of clam culture. In 1888 an act was passed hy the legislature authorizing the selectmen of the town to stake off, in lots of 1 acre or less, each of the flats along the Essex River, and let them to persons desiring to plant clams, for a rental of $2 per acre or lot for five years and an additional fee of 50 cents. Thus far 371 acres have been taken up and seeded with clams. Small clams are dug on the natural heds and planted on these hitherto unproductive flats. Mr. J. Bennett Fuller states that about 500 bushels are required to plant an acre properly. During the first two years (1889 and 1890) the people were slow to avail themselves of the privilege of planting for fear that after they had spent their time and labor they would not he able to secure protection from trespassers. But in 1891 and 1892 lots were obtained and planted. The principal difficulty encountered has been the loss of clams by the sand washing over them, the bottom in some localities being soft and shifting. In 1892 there were 25 acres that were quite productive, about one-third of the entire catch of the section being obtained from them. The catch from these lots is not definitely known, but is estimated at about 2,500 barrels. The cultivated clams possess some advantage over the natural growth from the fact that they are more uniform in size and are as large as the best of the natural clams. They bring $1.75 per barrel, while the natural clams sell for $1.50 per barrel. This is the price received by the catchers. One acre of these clams is considered to be worth $1,000, if well seeded and favorably located so as not to bo in danger of being submerged with sand. This valuation would be too high for an average, since all the acres are not equally well seeded and located. The clammers are generally impressed that the indus- try can be extensively and profitably developed, and their only fear is that they will not be able to securelots permanently. The greater part of the land available for this purpose is covered hy the deeds of people owning farms along the river, and the consent of the land-owners has to be obtained before lots can be taken up. It seems probable, however, that the business will continue to progress unless checked by complications that may arise relative to the occupancy of the grounds. This report is in the main correct, but there are some points in which it seems to be in error. Perhaps its author in stating that the flats were previously unpro- ductive did not mean that they had always been so. The testimony of the older inhabitants is that at one time most of the flats, and the river banks back to the town of Essex, were covered with clams. Another statement is that many planted clams were lost by shifting sand. In one or two instances trouble of this kind occurred on some of the river clam banks, but it was very insignificant, considering the total area seeded. For the last two THE CLAM PROBLEM AND CLAM CULTURE. 41 years at least, so far as I am able to determine, there lias been very little change in the contour of the Hats. There has been some shifting of sand and extension of the thatch plants, but these changes have been relatively unimportant. Very much less than the estimated 500 bushels were put upon an acre of ground, and the clammers generally believe that half of that amount would be sufficient. As a matter of fact, few of the areas leased were properly plauted, and for this and other reasons all estimates of the amounts which should be obtained as a result of planting under the most favorable circumstances are of little value. As stated in the report, the cultivated clam rapidly became large and uniform in size, and hence had a high market value. We have much evidence that the clam industry in Essex has in the past been extensive. From a curious little volume, which was published in 1SGS by a local clergyman, on the “History of the Town of Essex from 1634 to 1868,” I find the following paragraph: For the last twenty years * about 50 men and boys have been employed, chiefly in the spring and fall, in digging clams for fishing bait. For this purpose the clam flats in each town (Essex and Ipswich) are, by law, free to all its residents, and to no others. Five bushels of clams in the shell, it is usually reckoned, make one bushel of “ meats’ ; about 2} bushels of the latter are put into each barrel, and this quantity an able-bodied man can dig in three tides. One bushel of dry salt is used for each barrel. During this period of twenty years about 2,000 barrels of clams have been dug yearly, on the average, and sold at an average price of $6 per barrel. Deducting the cost of the barrel, $1, and of the salt, 75 cents, the sum of $4.25 per barrel, or $8,500 per year, has been earned in this business. The bait has been marketed chiefly in Gloucester. Mr. J. B. Fuller, an old resident of Essex, has this to say of the former industry: When I was a boy there were about 100 men who were making a business of digging clams in Essex, while to day there are not 10 who get their whole living by it. In those days a man could make from $2 to $5 and some times $7 or $8 a day. Now they obtain from 50 cents to $1.50 per day, with .just as many acres capable of bearing clams as formerly and with a much better market. The demand, also, is rapidly growing. Then, again, the amount of labor now necessary is only about two- thirds of that formerly required, for then clams usually had to be taken out of the shell, and now they are shipped as they are taken from the bed. Much more testimony of a similar character may be had to show that the Hats, once very productive, have almost entirely failed, and in spite of the effort made to reclaim them. It is not difficult to determine the reasons for the failure of the culture experiment at Essex. The areas upon which clams were planted were those which were at the time unproductive. The beds still containing clams — the “town fiats” — were free to any native of Essex. The one thing which was absolutely necessary to the success of any planter was that the clams on his leased ground should not be disturbed by other diggers. This protection was apparently not given in any case by the town authori- ties, and, as no person lived within sight of the majority of the beds, it was quite impossible for any man to guard his property much of the time. As to what followed it is not easy to obtain definite testimony from the clammers themselves. Other citizens of the town, however, and some few clammers, intimate that most of the men began to take clams from any property but their own, and that Presumably datiug from 1868. 42 BULLETIN OF THE UNITED STATES FISH COMMISSION. in this way the full result of no man’s labor in planting was ever realized. Others who did not make clam-digging a regular business, but only dug occasionally, are said to have had no respect for the rights of those who had leased property. It was said that at times when vessel builders and the shoe factory released employees, many of them, for lack of other occupation, turned their attention to clam digging, with the result that too many clams were at the time taken from the Hats. Another reason for the failure of the Essex experiment is that a number of short- sighted clammers began to fear, after the clams had been planted, that the production might suddenly become so great as to glut their market and, as a consequence, force prices down. Some few individuals, inspired by this fear, are reported to have said and to have done everything in their power to prevent the success of the experiment. In all cases, it is said, the selectmen of the town, who issued the leases, refused their aid in the prosecution of trespassers. In spite of the fact, which had been demonstrated in the experiment, that when properly planted the clams grew much more rapidly and became much larger than on the natural beds, no applications for a renewal of the leases were made when the first ones expired. No change in the condition at Essex may be hoped for until there is some evidence that a law protecting the planter will be strictly enforced. With proper protection a great industry might, and probably would, be quickly established, not only in Essex, but in any region where clam flats are now unproductive because of excessive digging. It would be comparatively easy to formulate plans which, if carried out in this region, should reclaim the Essex Hats. No single method would, however, be accept- able to all who are interested in the clam industry, and we have reason for believing that the few dissatisfied might easily defeat the efforts of the majority. The whole problem, as shown in the history of the oyster industry in this country also, narrows itself down to the simple question of protection. The leasing of lands to individuals is the necessary first step, and when the town or State authorities are willing to protect the lessee, the problem will be solved. This plan of leasing to individuals would seem to be the best one. Much may be said positively in its favor. A strong negative argument we have also, when we consider that any other scheme must depend for its success upon the cooperation of all concerned for the common good. We might know that this would hardly be possible, even if we had not the history of past events on these flats to guide us. Mentioning a few of those plans which do not consider the lease, the first is a closed season. If all digging could be prohibited for one or two years, many clams would come to maturity, the young would establish themselves, and the beds would once more become productive. Undoubtedly this recovery would be rapid. But it must be remembered that in this process the clams would probably crowd each other closely, as they have done on the few natural beds existing to-day, and a season or two of thorough digging and thinning out would be necessary before many clams reached a large size. This thinning process would not be immediately advantageous, on account of the small size of the clams, and hence would probably not be done at all. Another plan, proposed by some of the clammers, is that every man who makes a business of digging clams be required by law to deliver and plant on barren flats, set aside for the purpose, a certain number of small clams in the months of April or May — this planted area to be protected for a season, and eventually extended until THE CLAM PROBLEM AND CLAM CULTURE. 43 all the barren flats are covered, all flats upon which clams had matured to become common property. The fatal objection to this, as well as to the closed season, is that every man’s interest is bound up in that of his neighbor, and he would be constantly haunted with the fear of not realizing his full share of the profits. This would prompt him to make sure of his own, and the result would be the quick and certain defeat of the whole i>lan. These communistic schemes have been tried often in our country in recent years, and we are able to know from their almost universal failure that this would certainly fail. Besides, the difficulty of determining who were really engaged in the clam business— making it a means of livelihood — would be very great. How much work, in planting for the public, should be asked of the man who digs clams for market only occasionally? It would also be difficult to determine when the public work had been done. The lease, with swift and certain enforcement of a law against trespassers, would at once establish a great industry where there is now only the prospect of continued desolation. The flats at Essex afford an opportunity for producing an immense number of clams. Great tracts, Avith here and there a growth of thatch plants, are exposed at low tide. Down near the opening to the sea the sands shift to some extent, but in almost every other locality the changes are so slow aud so slight that clams are not affected by them. A glance at the appended map will give some idea of the immense size of this tract. But the map itself shows only what are known as the south flats. Stretching in a northwesterly direction from the region about Choate Island is an area of flats which many of the clammers claim to be larger than the one shown in the map. Unfor- tunately 1 was not able, when I visited Essex in August of 1898, to go over this ground as I did the south flats, but I could see from Choate Island that it was extensive. In constructing my map I have represented the natural beds — areas where clams can now be found — by stippling. The beds which were planted in the culture experi- ment are again almost entirely barren. Of course clams may be found here and there on these tracts, but they are few in number. The clams of the natural beds are generally too small to market, chiefly because they are so closely packed together. If the majority of such clams were removed for planting elsewhere the natural beds themselves would soon produce an abundance of large clams. The most important feature shown by the map is the extent of nonproductive ground, where every natural condition is favorable for the growth of clams. These areas were formerly, as I have indicated, natural beds. They are represented by the oblique shading lines. Almost every foot of ground for hundreds of yards about the point where the Essex Biver widens out into the great flats, might and should be yielding great quantities of the finest of clams, and this without glutting the markets near at hand, which are now forced to obtain their small supply from the Maine coast. These barren beds extend far up the river toward the town and into the mouths of the numerous creeks emptying into it, some of which are indicated in the diagram. I have been told by old clammers that at one time the mouths of these creeks were stocked with fine clams, and that in the late fall and early spring they afforded good shelter from the wind for men engaged in digging. The thatch banks, represented by short lines and stipples, contain at present great numbers of clams. The tough roots of the thatch vegetation, however, prevent digging, except along the edges and where the plants are scattered. 44 BULLETIN OF THE UNITED STATES FISH COMMISSION. The map represents the extensive clam hats at the mouths of the Essex and Castle ISTeck rivers within the townships of Essex, Ipswich, and Gloucester. To the north and west of Choate Island, and connected with this, is an area of beds nearly as extensive as those represented in the map.* The main features illustrated are — 1. The present extent of the “natural beds” (this was determined in August of 1898). These are represented by stippling. Though the beds appear to be extensive, the clams are in some places much scattered, and in others so densely packed as not to be able to grow. Very few marketable clams are taken from these beds. 2. The great areas — represented by oblique shading lines — wliere every natural condition is favorable for the growth of clams. These beds are and have been for years practically barren. At one time producing an immense harvest, they were reduced and have since been kept down by excessive digging. The flats planted in the culture experiment are those directly west of Conomo Point and along the river banks. 3. The thatch banks, submerged at high tide, which, on account of their tough roots, hold great numbers of clams safe from molestation. They probably produce great numbers of young, which under favorable conditions should replenish the losses, from judicious digging, of the entire area. 4. The extent of the flats as a whole, indicating how great an industry might be supported in this region. * While ninny kindly gave their assistance, I am especially indebted to Mr. J. B. Fuller and to Mr. E. Hobbs, of Essex, for their aid in collecting Ihc data used in text and map. MAP SHOWING CLAM FLATS AT MOUTHS OF ESSEX AND CASTLE NECK RIVERS, MASSACHUSETTS. DESCRIPTIONS OF NEW SPECIES OF FISHES FROM THE HAWAIIAN ISLANDS BELONGING TO THE FAMILIES OF LAB RIDE AND SCARIDSE. By OLIVER P. JENKINS, Professor of Physiology, Leland Stanford Junior University. Tlie fishes here described are part of three collections from the Hawaiian Islands. The largest collection, containing over 200 species, was made by me in 1889 under the generous auspices of De Pauw University. The second, of about 45 species, w;is obtained, under the direction of Dr. David Starr Jordan, by the United States Fish Commission steamer Albatross in 189G, on the return of the vessel from the work of the Fur Seal Commission of Investigation. The third contained some 80 species, and was made by Dr. Thomas Denison Wood, of Stanford University, in 1898, for the museum of the university. To this a small but valuable collection was added by his brother, Mr. A. B. Wood, of Honolulu, in 1899. A fuller account of these collections will appear in a subsequent paper. Of the 22 new species here described, 17 were first obtained by me in 1889 and 5 were first seen in Dr. Wood’s collection. With but one exception the fishes collected by me were examined in a fresh state and careful notes of the color were taken. Of the 5 from the collection made by Dr. Wood, all were seen after they had been in formalin for some weeks. While the markings were fairly well preserved, and in some cases brilliant coloration still remained, still the quickness with which the colors of these groups fade or change makes definite statements of color of these 5 species impossible. In all 42 species of labroids and scaroids were obtained. The large percentage of these that appear to be new would indicate that these islands and their neighbors lying west and south form a most interesting field for the study of these forms. While this fauna has already received the attention of Bleeker, Gunther (on Garrett's col lections and drawings), Steindachner, and other ichthyologists, our knowledge of it is only in its beginning. The fishes of these groups were mainly obtained from the market in Honolulu, which is supplied by the native fishermen. The species are all valued highly by the native Hawaiians as food, the scaroids especially so. Some small specimens of the latter bring large prices. They are eaten raw by the natives, as are most of the other fishes, cooking being regarded as especially inimical to the food qualities of a scaroid. The Honolulu fish-market is the richest source of the collector’s material. Fishermen, who know the habits aud haunts of these fishes, and who have great skill in catching them, are urged to bring them here by the high price and ready sale of these forms. The types of these species are deposited in the museum of Leland Stanford Junior University. Series more or less complete are also placed in the De Pauw University museum, the United States Fish Commission, and the United States National Museum. ■1.3 46 BULLETIN OF THE UNITED STATES FISH COMMISSION. LIST OF SPECIES. Macropharyngodon aquilolo Jenkins, new species. Native name, Hinalea akilolo, applied also to other fishes. Fig. 1. Head with flap 3.33 in length of body to base of caudal ; 4 in total ; depth 2.5 ; in total 3. D. ix, 11. A. hi, 11. Scales 2-28-10. Tubes two to three branched, or single. Head naked. Four front teeth in each jaw, strong, the second on each side of upper jaw turned backward, prominent posterior canines at each angle, two at one angle in one specimen. General color in life a light-brown, sometimes darker; a very bright-blue spot on each scale, this spot with a black posterior border; five bright-blue lines on side of head running obliquely downward and forward; two parallel ones from eye to mouth, the third from just below lower border of eye to chin below angle of mouth, the remaining two lower; bright-blue spots and short lines on top of head and on opercle. The positions of these lines are constant in the five specimens examined. Dorsal and anal with three longitudinal series of bright-blue spots with black borders, the inner and middle series being large and distinct spots; anterior dorsal spot between first and third spines bright-scarlet, bordered below with bright-yellow, and below this with dark-green ; pectoral plain-brown ; ventrals bright-brown with about six transverse series of bright-blue ocelli ; caudal with about five transverse series of bright-blue ocelli. The coloration is quite constant in the five specimens taken, varying only in darker or lighter shade of brown of the general color. Tig. 1. — Macropharyngodon aquilolo Jenkins, new species. Type. Allied to M. geoffroyii Bleeker, Proc. Zool. Soc. 1861, and Atl. Ichth , p. 129, tab. 37, fig. 5, from which it differs in form of body and pectoral and ventral fins, and in coloration of body, head, and fins. The peculiar form of the pharyngeal teeth seems to be a sufficient character upon which to found Bleeker’s genus Macropharyngodon , which is here retained. Type No. 6130, L. S. Jr. University Museum. This very beautiful fish is not common at Honolulu; some fishermen had never seen specimens before. The five obtained by me were taken with the hook. Their measurements were as follows : Macropharyngodon aquilolo. No. 1. No. 2. No. 3. No. 4. No. 5. Macropharyngodon aquilolo. No. 1. No. 2. No. 3. No. 4. No. 5. mm. mm. mm. mm. mm. mm . mm . mm. mm . mm. Length to base of caudal. 99 103 112 90 105 Height of first dorsal 17 23 23 22 25 10 8 10 9 1 o Total length 116 126 135 112 130 Height of soft dorsal 15 19 18 15 1 6 30 34 34 30 34 15 19 18 15 1 6 40 44 46 40 41 25 27 29 24 26 7 7 y 6 7 25 27 29 24 26 Snout 9 ii ii 10 ii Width of body ii ii 15 ii ii FISHES FROM THE HAWAIIAN ISLANDS. 47 Halichceres iridescens .Jenkins, new species. Native name, Olma paawela. Fig. 2. Head witli flap in length of body to base of caudal 3.5, in total length 1 ; depth 3.1(1, 3.66. D. ix, 12 A. hi, 12 (in one specimen in, 11). Scales 3-27-8. Outline elliptical, body compressed, somewhat elevated; teeth 2-2, posterior canine in each jaw. Type No. 6131, L. S. Jr. University Museum. Coloration in life : Each scale with a dark-red crescent spot, convex anteriorly, this spot bordered posteriorly with blue; head bright-red, with bright-green stripe on median line from snout to base of caudal, a bright-green stripe along upper margin of eye on to body, where it continues more or less distinctly just above lateral line about half the length of body; a narrow green line from snout to middle of anterior margin of eye, a bright-green stripe from near angle of mouth along lower margin of eye to opercular flap, where it is confluent with a wide bright-green band from the chin backward over lower limb of preopercle and the subopercle and opercle; this green gradually shading into blue on throat and belly; a black spot behind eye surrounded by bright-green, with red spots in some specimens; dorsal tin dark-red, with a row of dark-green oblong spots on proximal border, one spot on each iuterspinous membrane, a green longitudinal band on outer portion, with a very narrow outer margin of light-blue; a black spot in membrane between first and second spines; in one specimen a double blotch between first and third soft rays, a median row of green spots. The detail of markings in this flu varies in the three specimens. Anal dark-red with a green band along central portion, outer margin with narrow blue line; caudal dark-red with a crqssband of bluish-green at the base, with two or three other crossbands often broken up into spots of green with orange centers; ventrals blue, with dark line on anterior portion ; pectorals pale-red, with base and axils bright-green. Coloration similar to that of Jitlis ornatissimus Garrett. Garrett’s description is incomplete, but shows that this fish differs in marked points of coloration, notably in conspicuous black spot behind eye, spot on anterior portion of dorsal, form of markings on the scales, and in number of anal rays. A very brilliant tish, taken from the coral reefs, apparently not very common. Three specimens were taken by me, measuring 140 mm., 130 mm., and 120 mm. Detailed measurements of the last are given below : Halichoeres irideseens. Milli- meters. 120 20 140 34 30 Depth of body 38 Halichoeres irideseens. Milli- meters Longest soft ray of dorsal 18 Longest solt rayol anal 17 22 Ventral 30 Eve 6 Snout 12 48 BULLETIN OF THE UNITED STATES FISH COMMISSION. Halichceres lao Jenkins, new species. Native name, Lao. Fig. 3. Head with Hap 3.4 in body to base of caudal; in total 4. Depth 3.4. D. ix, 12. A. in, 12 (the first anal spine showing only on dissection). Scales 2-29-9. Lateral lino continuous, tubes branched. Posterior canines present, two prominent canines in front portion of each jaw. Body compressed, not elevated; fins moderate; caudal rounded; anal reaching vent; soft dorsal and soft anal equal in height and are one-half length of head with flap. Coloration, body pink, lower parts blue; dark-brown spot on each scale; a green line on middle lino of head from snout to origin of dorsal; a bright-green line from snout backward just above eye to body, where it is continued as a row of indistinct spots just above lateral line nearly to caudal; another green lino from month just under eye to opercular flap; another broad green band covering lower limb of preopercle and subopercle, and lower portion of opercle joining the last on posterior portion of opercle; space between last two green bands a bright pink; coloration of head similar to that of II. iridescens; a black spot behind eye; dorsal fin with three black ocellatcd spots; ilie first, small and least distinct, is between first and second dorsal spines; the second, larger than eye, from first to third soft rays; the third, smaller, from tenth to eleventh soft rays; base of pectoral green; base of caudal orange and green. Allied to E. iridescens, from which it differs in the more slender and less elevated body, somewhat longer head, and in coloration. The most conspicuous point of difference is the presence of black ocellated spots on dorsal. Type No. 6132, L. S. Jr. University Museum. One specimen was taken by me, the measurements of which were as follows: Halickoeres lao. Milli- meters. Halichceres lao. Milli- meters. 82 12 13 12 95 10 24 Ventral 18 22 0 24 8 Coris lepomis Jenkins, new species. Native name, Hilti lauli. Fig. 4. Head, with flap, in length to base of caudal 3.16; in total 3.66; depth 3.66. D. ix, 12. A. in, 12. Scales 7-85-33. Lateral line continuous. Teeth, two strong canines in front of each jaw, followed by smaller teeth in single series; a posterior canine. Head naked; body elliptical, scales on breast and before dorsal smaller; two anterior dorsal spines slightly produced and flexible, remaining spines pungent; caudal fin slightly rounded; ventrals not produced, not reaching vent. FISHES FROM THE HAWAIIAN ISLANDS., 49 Coloration in life: General color bright-blue, sides rosy, a series of about live or six short (about one-fourth depth) oblique indistinct dusky bars on sides of body just behind pectoral fin, behind this an indistinct rosy patch; head, throat, breast, and belly covered with bright-golden spots and short irregular bars of same color. All the fins bright-blue with golden reticulations, which are especially bright on vertical lins; opercular flap large, terminating with a black spot as large as eye, posterior border of this spot with a narrow yellow border, anterior golden border. This spot is very similar to that seen on a suufish {Lepomis) . One fine specimen of this beautiful fish was obtained by me. Type No. 12141, ~L. S. Jr. Uni- versity Museum. It gave the following measurements : Coris lepomis. Milli- meters. Coris lepomis. Milli- meters. 123 45 Flap 23 65 388 72 450 15 127 40 Height ol' soft dorsal 45 Fig. 4. — Coris lepomis Jenkins, new species. Type. Figure not quite accurate in proportional measurements. Hemicoris remedius Jenkins, new species. Fig. 5. Head with flap 3.66 in body to base of caudal; in total length 4.33; head without flap 4.40 and 5.25. Depth 2.66 and 4.5. D. ix, 12. A. in, 12. Caudal rounded; first rays of ventral produced, reaching second anal spine. Scales 3-54-17; tubes simple; scales with tubes 50; scales on throat small, those on belly not much smaller than on body; head naked; rows of small scales extending on membranes between the caudal rays. Teeth, four prominent canines in upper jaw, two in lower, the second pair not much larger than those of series following; posterior canine tooth present. Coloration in life: Body brown, dark above, light below; a bright-red baud from eye aloug lateral line to its downward bend; a bright-red band from opercular flap extending backward to tip of pectoral, from which it is broken up into a series of short oblique crossbars on every alternate row of scales, becoming less distinct toward caudal peduncle; a bright-red band curving downward just below anterior portion of middle line of body, becoming indistinct about below a vertical from twelfth dorsal spine; another bright-red band from axil curving downward and extending backward to about second soft ray of anal; a bright-red band from angle of mouth curving upward to lower margin of eye, and down to edge of opercle just below flap; a bright-red band from a short distance behind symphysis curving upward over cheek and downward across middle of posterior margin of opercle on to breast to a point just behind base of ventrals; a bright-red band on each side of middle F. C. B. 1899—4 50 BULLETIN OF THE UNITED STATES FISH COMMISSION. line of throat; a short red har at symphysis; a red spot just above and hack of eye; anterior portion of opercular flap scarlet, bordered posteriorly first with black, then with bright yellow; spinous dorsal with a longitudinal band of red on middle portion, which on soft portion is broken up into small wavy lines and reticulations. In one young individual there is a black spot at root of sixth dorsal ray, one at root of the third from last, and one at roots of last two. These spots do not appear in the four large specimens. Caudal with crossbars of red alternating with yellow, somewhat wavy; pectorals plain yellow, base red; ventrals golden-yellow. A very bright fish. Fig. 5. — Hemicoris remedius Jenkins, new species. Type. Fig. 6. — Hemicoris keleipionis Jenkins, new species. Type. Figure not quite accurate in proportional measurements. Formerly valued by the natives as a medicine, the purpose not clearly stated. This species seems to be well known by the natives. Type No. 6133. L. S. Jr. Univ. Mus. Five specimens, taken by me, had total length, including caudal fin, 15. cm., 15.5 cm., 14.3 cm., 15 cm., 12 cm., respectively. Measurements of one are given below. Hemicoris remedius. Milli- meters . Bofiv to base of caudal 133 24 157 Head with flap 36 Head without flap 30 Depth of body 37 Hemicoris remedius. Milli- meters. 19 19 25 32 7 Snout 9 FISHES FROM THE HAWAIIAN ISLANDS. 51 Hemicoris keleipionis Jenkins, new species. Fig. 6. Head 3.2 in body to base of caudal, 3.8. in total length; depth 3.6 to base of caudal, 4.5 in total length. D. ix, 12. A. ill, 12. Scales 56, with tubes 50. Head entirely naked, bases of vertical fins not sheathed. Teeth pointed, two rows in each sideof upper jaw, a single series m lower; no posterior canine, the two anterior canines strong and curved outward. First dorsal spine somewhat extended, spinous portion lower than soft rays; the caudal rounded; anal less high than dorsal; ventral not quite reaching vent. Coloration in alcohol (not seen fresh): General color of body pale with tinge of pink, with about 21 distinct narrow longitudinal stripes, these Tunning through upper border of each row of scales and lower border of the adjacent row; head pale without markings; ventral fins white; dorsal show- ing trace of a longitudinal line about one-tliird its height from outer margin; anal with no markings except faint trace on portions of outer margin ; caudal showing remains of about five irregular cross- bands; pectoral and veutrals colorless and without markings. This species is dedicated to Mr. Keleipio, of Honolulu, who rendered very valuable aid in securing the fishes of Dr. Wood’s collection. One specimen was obtained by Dr. Wood at Honolulu. Type No. 6049, L. S. Jr. University Museum. The following are its measurements: Hemicoris keleipionis. Milli- meters. Hemicoris keleipionis. Milli- meters 224 25 190 20 34 38 59 30 50 22 First dorsal spine 22 Fig. V. — Thala8soma pyrrhovinctum Jenkins, new species. Type. Thalassoma pyrrhovinctum Jenkins, new species. Native name, Hinalea lauli. Fig. 7. Head with flap in body to base of caudal 3.4, in total, 4; depth in body to base of caudal 3.25, 3.6. D. viii, 13. A. ii, 11. Scales 2-28-9. Lateral line continuous, tubes three-branched. Teeth, two prominent canines from which the remaining ones of the series are graduated in size, no posterior canine. Caudal lobes produced, veutrals somewhat prolonged. Coloration in life: Colors very bright; the whole head dark purple without markings, this color extending a very short distance on body ; behind this coloration, a broad orange area extending around body from head as far back as third dorsal spine; remainder of body, with dorsal, anal, and caudal fins dark purple, with (in some specimens) reddish coloration; upper lobes of caudal sometimes distinctly reddish-brown ; upper lobes of candal much produced, amount varying, in some equaling three-fourths length of head ; pectoral with a broad dusky bar extending from middle of fin to tip ; axil and base of fin a dark purple; ventrals produced. Mrs. Whitney, of Honolulu, has shown me a series of colored drawings of Hawaiian fishes made by her, in which the colors represented are as contained in my notes with the exception of representing color of body behind the orange bar as dark green, each scale with a vertical brown line; dorsal and anal fins very dark purple. This fish is similar to T. melanoptera (Gunther), differing in conspicuous orange band from head to third dorsal 8i>ine extending around body, in dark-purple color of caudal, and in veutrals being pro- 52 BULLETIN OF THE UNITED STATES FISH COMMISSION. duced. The fish identified by Streets, Bull. U. S. N. M., No. 7, 1877, as ? Julis melanoptera, is probably this species In his specimen the colors were faded, which may account for absence of orange area. This fish is common. Mauy specimens were obtained by me aud three by Dr. Wood, the largest being 165 mm. in length. Type No. 6138, L. S. Jr. Univ. Museum. The measurements of one are given : Thalassoma pyrrliovinctum. Milli- meters. Thalassoma x>yrrhovinctum. Bod v to base of caudal 137 22 159 40 35 44 Height of longest soft dorsal Milli- meters, j 15 14 31 ! 23 8 7 Novaculichthys woodi Jenkins, new species. Fig. 8. Head 3 in body to base of caudal (3.5 in total) ; depth 2.6 in body to base of caudal (3.1 in total) ; D. ix, 12. A. hi, 12. Scales 27 (21 with tubes single). Body deep, compressed; profile of head ohtuse, evenly curved; eye but small distance above axis of body; distance from margin of eye to angle at mouth 3J in head. Scales 2-27-9; head naked; the area in front of ventrals naked, or with few deeply imbedded scales; scales on sides of breast somewhat smaller than on sides of body. Teeth conical, in a single series, two anterior canines in each jaw, the lower two fitting in between upper two; no posterior canine. Anterior spine flexible, next less so, the remaining pungent ; origin of dorsal in advance of base of pectoral; height of the soft dorsal nearly 3 iu head; ventral spine short, strong, pungent; first soft ray filamentous, reaching somewhat beyond origin of anal; caudal rounded ; pectoral 14 in head. Coloration: In alcohol, general color body pale without markings; tins white, except membrane of spinous dorsal, which is black; black dots on membrane between first 3 or 4 soft rays. Similar to Novaculichthys entargyreus Jenkins. This species is named for Dr. Thomas Denison Wood, professor of hygiene in Stanford University, who obtained two specimens at Honolulu. Type No. 6029, L. S. Jr. Univ. Mus. The measurements of two specimens are given: Novaculichthys woodi. No. 1. No. 2. Novaculichthys woodi. No. 1. No. 2. Bodv to base of caudal mm. 130 25 155 44 49 12 mm. 131 27 158 44 50 13 Longest soft ray ol' dorsal mm. 10 29 37 9 10 mm. 15 30 37 9 11 Depth of bodv Lougest spine of dorsal Snout (from margin of eye) FISHES FROM THE HAWAIIAN ISLANDS. 53 Novaculichthys entargyreus Jenkins, new species. Fig. 9. Head 3.2 in body to base of caudal (4 in total); depth 2.7 to base of caudal (3.2 in total). D. ix, 12. A. hi, 12. Scales 27 (24 with tubes, tubes single). Body deep, compressed ; profile of bead obtuse, evenly curved; eye only moderately high; distance from margin of eye to angle of mouth about 4 in head. Scales 2-27-9; head naked; area in front of centrals naked or with few deeply imbedded scales; scales on sides of breast somewhat smaller than on sides of body. Teeth conical, in a single series, two anterior canines in each .jaw, the lower two fitting in between upper two; no posterior canine. Dorsal fin continuous, the anterior spines not protruding, the spinous portion lower than soft portion; anterior spine flexible, the next less so, remaining ones pungent; origin of dorsal in advance of base of pectoral ; soft dorsal 2.5 in head; ventral with a short, pungent spine; first soft ray filamentous, just reaching vent; caudal rounded; pectoral 1.5 in head. Coloration (in alcohol) : Body and tins pale, without distinct markings; two broad longitudinal silvery bands along sides of body, the pigment being in the skin below the scales but showing through ; transverse narrow silvery markings on peritoneum show through between ventrals and vent. One specimen of this fish, 140 mm. in length, is in the collection made hy Dr. Wood, at Honolulu. Type No. 5984, Leland Stanford Jr. University Museum. Its measurements follow: Novaculichthys entargyreus. Milli- meters. Novaculichthys entargyreus. Milli- meters. 117 14 23 j 13 140 25 37 28 44 8 Longest spine of dorsal 9 Snout (from margin of eye) 8 Hemipteronotus umbrilatus Jenkins, new species. Native name, Lae nihi. Fig. 10. Head, with flap, in length to base of caudal 3.5, in total 4; depth 2.8, 3.2. D. ii-vii, 12. A. Hr, 12. Scales 2-25-9, tubes simple, lateral line interrupted; scales behind eye, cheeks covered with small scales, opercles and preopercles scaleless; the two anterior spines of the dorsal not produced, separated from remainder of dorsal by narrow space not quite twice eye. Teeth, two prominent canines in anterior part of each jaw, the lowrer fitting between the upper. Coloration in life; general color a light-drab, with posterior portion of each scale white or whitish; a large dark-brown blotch, large as head, on middle portion of hody, in which blotch the posterior half of each scale is white; the anterior separated portion of dorsal dusky; no distinct markings on the fins. Type No. 6135, L. S. Jr. University Museum. 54 BULLETIN OF THE UNITED STATES FISH COMMISSION. One specimen obtained by me, of which the following table of measurements was taken: HemipteroDotus umbrilatus. Milli- meters. 107 16 Total length - 123 31 27 38 13 Hemipteronotus umbrilatus. Milli- meters. 13 22 20 7 15 12 Fig. 10. — Hemipteronotus umbrilatus Jenkins, new species. Type. Iniistius leucozonus Jenkins, n. sp. Native name, Lac nihi, applied to other fishes, also. Fig. 11. Head in body to base of caudal 3.2, in total 3.7 ; depth 3.5. D. ii-vii, 12. A. in, 12. Scales 2-26-9 ; lateral line interrupted, tubes simple. Two prominent canines in anterior part of each jaw, the pair in lower jaw fitting between those in upper and projecting above the jaw; no posterior canine; gill- rakers minute; upper edge of head sharp; the two anterior rays of dorsal produced, and separated from remainder of dorsal by a space equal to snout; a very few scales just behind eye; rest of head, cheeks, opercles, and preopercles scaleless. One specimen was obtained by me, from which were taken the following measurements: Iniistius leucozonus. i Milli- meters. Iniistius leucozonus. Milli- meters. 112 18 Cauclal 18 36 130 22 35 24 29 7 37 14 15 31 Coloration in life: Body with four white bands alternating with four bands of brown to olivaceous; first brown band covering posterior portion of opercle and body just posterior below, being in width from base of ventral nearly to anal; second brown band in width from seventh spine to fifth soft ray of dorsal, extending on anal; third, from ninth to last soft dorsal ray, extending on anal; fourth brown band on base of caudal; a brown dart from eye to angle of opercle; middle line of snout, chin, and throat yellow; two dark brown lines from eye to eye over head; small black spots above and behind eye; several scales in upper part of second brown band on body with distinct brown dots; a scale in second row just below fourth spine jet black ; anterior dorsal with alternating dark and white spots; second portion of dorsal with three ocellated black spots; first between second and third spines, the longest as long as snout; second between first and fourth soft rays; third between eighth and tenth soft rays, lying on ninth; peotorals colorless; ventrals olivaceous. FISHES FROM THE HAWAIIAN ISLANDS, 55 Allied to Inistius tetrazonus Bleeker, which it resembles, bnt from which it differs in having three ocelli on dorsal, and a distinct black spot on scale below third or fourth dorsal spine also. Type No. 6137, L. S. Jr. University Museum. Fig. 11 Iniistius leucozonus Jenkins, new species. Type. Tlie first anal ray in figure should be a spine. Iniistius verater Jenkins, new species. Fig. 12. Head 3 to base of caudal, 3.5 in total; depth 2.75, 3.1 in total. D. ii-vii, 12. A. in, 12. Eye 6 in head; snout 2.2 in head; body deep, compressed; profile of head vertical; eye high near upper margin of head; posterior angle of mouth in a vertical line below front margin of eye, distance of eye from this nearly half the length of head. Scales 2-28-11 ; head naked exceyff deeply imbedded 56 BULLETIN OF THE UNITED STATES FISH COMMISSION. scales on nape and faint trace of margins of two or three scales deeply imbedded just behind eye; scales on breast smaller than on sides of body; each scale with a delicate vertical ridge running across its center; tubes in lateral line single. Teeth conical, in a single series, the two anterior in each jaw enlarged canines, the two lower fitting between those of upper; the core of each tooth black, showing through. Two anterior dorsal spines stand over eye free from rest of dorsal, the space between them and remaining dorsal equaling half length of second spine, these two spines prolonged, the first filamentous, its height nearly equaling head; ventrals just reaching vent, caudal rounded. Coloration in alcohol : The whole body and tins black except tips of posterior two or three rays of dorsal and anal fins, posterior half of caudal, and most of the pectoral tins, which are white, being black-blotched at the base. This descrijition is based on two specimens collected in the Honolulu market hy Dr. T. D. Wood in 1898. Type No. 5990, L. S. Jr. University Museum. The measurements of the two are given below: Iniistius verater. No. 1. No. 2. Iniistius verater. No. 1. No. 2. Body to base of the caudal Caudal mm. 135 27 162 46 39 52 41 mm. 120 24 144 42 31 44 39 Longest soft ray of dorsal Longest soft ray of anal mm. 17 19 28 26 8 22 mm. 17 19 25 23 7 19 Depth of body Longest ray of anterior dorsal. . Snout (from margin of eye) Clieilinus zonurus Jenkins, new species. Native name Poou. Fig. 13. Head, with flap, to base of caudal 2.9, in total 3.4; depth 3 (3.8 in total length). D. ix, 10. A. hi, 8. Scales 2-21-6; lateral line interrupted, 23 or 24 in upper portion; 14 or 15 scales with tubes, somo simple, some branched. Two strong canines in anterior portion of each jaw, no posterior canines ; lower jaw produced, its canines fitting in between those of the upper; gillrakers, on lower limb 10, short, conical; on upper about 6, mostly rudimentary. Body somewhat compressed, elliptical, con- tracted suddenly at caudal peduncle; lower jaw produced; head concave above eye; caudal rounded; soft dorsal like soft anal, posterior portion of each the higher; two rows of scales on cheek, preopercle naked, opercle with two, subopercle one row of scales. General color a dull red to drab, each scale with a vertical line of brighter red ; a series of orange lines running obliquely downward and backward on each side of face to edge of opercle, and from eye forward to mouth; a light-rod band around caudal peduncle just behind dorsal and anal fins; spinous dorsal mottled with dark, with two longitudinal orange stripes more distinct anteriorly; posterior portion of soft dorsal light pink; anal fin rosy, brighter anteriorly; caudal with many dull shadings, with olive prevailing toward outward margin ; pectoral pale-red, with yellow base; ventral rays rosy, membranes white, a large dusky-red spot on anterior portion, showing only on upper surface. FISHES FROM THE HAWAIIAN ISLAND'S. 57 This fish appears to he the same as that identified by Smith &. Swain as Cheilinus diagramma Lacepode, Proc. U. S. Nat. Museum 1882, 133, in a collection of fishes from Johnson Island, Pacific Ocean. Type No. 6134, L. S. Jr. University Museum. The measurements follow: Clieilinus zonurus. Milli- meters. Clieilinus zonurus. Milli- meters. 188 26 38 33 226 20 66 Eye 19 60 27 Longest soft ray of dorsal 26 Anampses evermanni Jenkins, new species. Fig. 14. Head, with flap, in body to base of caudal 3 (in total 3.6) ; depth 2.2 (2.5). D. ix, 12. A. m, 12. Scales 4-26-11; scales with tubes 28, tubes mainly simple. Lateral line continuous. Two prominent canine teeth in each jaw, flattened with cutting edge. Body oval, deeper in front; scales large, those in front of dorsal and on breast small; a row of small scales at base of anal and at base of anterior part of dorsal; caudal truncate; venlrals not much produced, reaching origin of anal. Opercles with prominent striations radiating from upper anterior angle ; striations also on suhopercle, preopercle, and interopercle; dorsal spines flexible at tips. Coloration, brownish-red, with a vertical blue line on each scale; snout blue; chin, throat, and sides of face with blue reticulations; dorsal fin color of body, outer margin blue, fin covered with 7 or 8 longitudinal wavy blue lines and rows of dots, some running into each other and making reticulations; anal, ground color red, outer margin blue, fin with 4 or 5 distinct blue longitudinal lines; ventral, anterior margin blue, fin red with blue lines and dots; pectoral olive, anterior margin blue; caudal red, with longitudinal blue line. The coloration of the first is almost exactly the same as in Anampses cuvieri. Allied to Anampses godeffroyi Gunther, differing in coloration and number of scales. Four specimens taken hy me ; total length of each, respectively, 297, 260, 295, 295 mm. Type No. 6136, L. S. Jr. University Museum. The measurements are given below: Anampses evermanni. Milli- meters. Anampses evermanni. Milli- meters. 250 47 32 297 11 82 30 74 42 112 62 58 BULLETIN OF THE UNITED STATES FISH COMMISSION. Calotomus irradians Jenkins, new species. Fig. 15. Head 3.33 (in total 3.9) ; depth 2.33 (2.9). D. ix, 10. A. hi, 9. Scales 25. Teetli pointed, both tips and margins free, posterior surface adnate to dental plate, of about equal size in both jaws; at least one strong tooth at posterior angle of each upper jaw, curved backward and downward; teeth in oblique series of about 5 series in each half of upper, and of about 8 series in each half of lower jaw ; 3 to 4 teeth in each series. Scales on cheek in a single series, 6 in number, 4 scales in median line before dorsal, 3 scales before ventral, this line continued by 2 scales between bases of ventrals, form- ing a sort of appendage in this position; scales not reduced on breast; one and one-half series of scales between lateral line and base of dorsal, the half series only slightly sheathing base of dorsal fin. Dorsal spines flexible; origin of dorsal, base of pectoral, and base of ventral fins in a vertical line; posterior margin of caudal a straight lino, except moderate prolongation of upper rays, longest ray 2.4 in head; ventrals reaching halfway to vent; pectoral fin broad as long, longest rays (second and third) 1.5 in head. Posterior margin of fin slightly convex. Color in life: General color of body and fins blue; head, chin, and throat bright-blue, marked with pink, arranged as follows: About 8 bands radiating from eye and a number of hands and irregu- lar figures on snout, face, and occiput, and a few small spots on chin ; of the pink bands radiating from eye 2 reach region of angle of mouth, 2 join the irregular markings on face and occiput, and 4 radiate over the region of cheek and behind eye ; a vertical bar of pink on each scale ; in some regions this bar is partly concealed by overlapping scale; vertical fins bright-blue, with reticulations and spots of pink; ventrals blue; pectorals blue, olive, and pink. This very beautiful fish is somewhat rare and is very highly prized by native fishermen for virtues they suppose it to possess. They usually ask an exorbitant price for it. The description is based on two specimens, one obtained by me in 1889 and one by the Albatross in 1896. Type No. 12142, L. S. Jr. University Museum. The measurements of the two are given below: Calotoimis irradians. No. 306. No. 1298. Calotomus irradians. No. 306. No. 1298. mm. mm. mm. mm. Length total (to end of middle Depth 150 150 of caudal) 415 410 Height of longest soft dorsal . . . 45 45 Length to base of caudal 350 345 Height of longest soft anal 47 45 Caudal (middle) 65 55 Pectoral 73 73 79 75 00 00 108 100* 47 40 90 94 10 10 FISHES FROM THE HAWAIIAN ISLANDS. 59 Scarus brunneus Jenkins, new species. Fig. 16. D. ix, 10. A. ii, 9. P. 13. V. 6. Head 3 in length to base of caudal; in total (to middle of caudal) 3.5. Depth 2.6 in length to base of caudal; in total (to middle of caudal) 3.1. Scales 24; lateral line interrupted, scales slightly roughened by striations of minute tubercles; tubes, some single, some once or twice branched. Body deep, compressed. Teeth whitish, a tooth at posterior angle of upper jaw ou one side in one specimen, the largest, none in 7 others. Upper lip double the whole length, broad, nearly covering dental plate; lower lip covering more than half of lower dental plate. Cheek with two rows of scales, 6 in upper and 5 in lower row; no scales on anterior limb of preopercle; a row of scales along margin of opercle; a series of 4 scales on median line before first dorsal spine. Dorsal spines flexible, spinous portion nearly equaling the soft portion in height; margin of caudal fin convex in the smaller specimens, with lobes not produced; lobes produced in larger ones; height of caudal peduncle 2 in head; pectoral 1.2 in head, its breadth about If its own length; ventral 1.9 in head, not reaching the vent by a distance equal to half the whole length of the fin, inserted on a vertical through the anterior point of insertion of base of pectoral. Coloration in life: Body, head, and vertical fins dusky-brown; pectorals and ventrals lighter; no distinct markings anywhere. This fish appears to be similar to Scams dnbius Bennett, whose description is very meager. In that given, it differs in not having scales on anterior limb of the pre- opercle, in having a pointed tooth at angle of jaw, in having 13 rays in pectoral, and in the caudal not being white on posterior edge. Type 6139, L. S. Jr. University Museum. Seven specimens were obtained by me. Measurements of three of these are given below : Scarus brunneus. Total length to middle of caudal Length to base of caudal Head Depth Eye Snout No. 1. No. 2. No. 3. Scarus brunneus. No. 1. No. 2. No. 3. mm. 220 mm. 185 mm. 143 Height of first dorsal spine mm. 18 mm 16 mm. 13 182 152 116 Height of soft dorsal rays 22 19 15 62 53 89 Height, of soft anal rays 18 17 14 70 60 43 Ventral 33 29 24 11 10 8 Pectoral 50 39 30 24 20 15 Depth of caudal peduncle 30 27 19 Scarus gilberti Jenkins, new species. Fig. 17. D. ix, 10. A. ii, 9 (3 anal spines show on dissection). P. 1, 12. Head 2.75 in length to base of caudal, 3.47 in total. Depth 2.44 in length to base of caudal, 3.1 in total. Scales 24; lateral line interrupted; scales slightly roughened by very fine striations of minute tubercles ; tubes but little branched. Body deep, compressed. Teeth whitish, lower jaw included. Upper lip double the whole length, covering whole of dental plate ; lower lip covering about half of lower dental plate. Two 60 BULLETIN OF THE UNITED STATES FISH COMMISSION. Scarus gilberti. No. 1. No. 2. No. 3. Scarus gilberti. No. 1. No. 2. No. 3. Total length to middle of caudal. mm. 278 mm. 251 mm. 818 Snout mm. 32 mm. 28 mm. 36 Upper lobe of caudal produced 20 15 30 Height of first dorsal spine 18 28 25 10 28 23 20 30 Length to base of caudal 2-10 208 255 Height of soft anal rays 26 Head 80 70 88 50 43 59 90 82 92 54 53 68 12 12 14 40 35 -!Li Scams paluca Jenkins, new species. Native name, Palnkaluka. Fig. 18. Head in body to base of caudal 3, in total to tips of middle caudal rays 3.68. Depth in body to base of caudal 2.8; in total to tips of middle caudal rays, 3.5. D. ix, 10. A. II, 9. P. 14. Scales 24; lateral line interrupted ; tubes mostly many-branclied, a few single. At interruption two scales with tubes stand immediately above first two of series following interruption; scales very slightly roughened by radiating lines of granulations extending to margins of scales. Body moderately deep. teeth at posterior angle of upper jaw, one at angle of lower jaw. Cheek with two rows of scales, seven in upper row and 4 to 5 in lower, the margin of one or two scales of lower projecting over anterior limb of preopercle; a series of scales along margin of opcrcle; a series of 4 scales along median line before first dorsal siiine. Dorsal spines flexible; caudal .fin middle margin straight, the lobes pro- duced nearly the length of middle portion of fin in longest specimen, less so in smaller ones. Caudal peduucle 2 in head. Pectoral 1.3 in head, its breadth about two-thirds its own length; ventral 1,6 in head, not reaching vent by one-half to one-fourth its length, inserted but slightly back of a vertical from anterior insertion of pectoral. Color notes from freshly collected specimens not obtained. Specimens in alcohol (evidently much faded) are pale, with upper parts of body somewhat dusky ; upper lip greenish-blue, margin of same color. This color from each lip unites behind angle of mouth into a band which extends to anterior margin of eye and passes along lower margin of eye, ending in a small, indistinct (in alcohol) area behind eye; transverse bar of same color across the chin. Back of this, on the throat, are markings of bluish-green; outer margins of dorsal and ventral and upper and lower margins of caudal fins bordered by same color, also anterior margins of pectoral and ventral lius; indistinct markings of same color appear on caudal. Fig. 17. — Scarus gilberti Jenkins, now species. Type, This species resembles Scants octodon Bleeker. Five specimens were obtained by me in 1889, and one by the Albatross in 1896. Type 6140, L. S. Jr. University Museum. The measurements of three are given below : FISHES FROM THE HAWAIIAN ISLANDS. 61 compressed; profile concave, teetli wliitisb, lower jaw included, no pointed tooth at angle of either jaw; upper lip double its entire length and covering little more than half of dental plate; lower lip covering half of lower plate. Cheek with 3 rows of scales, 6 in upper row, 4 to 5 in middle, 2 in lower row, which extends upon anterior limb of preopercle; posterior limb of opercle with two series of scales, anterior limb with single series ; a series of 6 scales on median line before first dorsal spine. Posterior margin of caudal lin truncate, lobes not produced. Pectoral with 14 rays, its length 1.5 in head, its breadth being more than half its length; ventral 1.85 in head, not reaching vent by two-thirds its own length, its origin but slightly behind a vertical from anterior point of the base of pectoral; dorsal spines llexible, first spine less than the succeeding ones, nearly equaling the soft rays; the soft anal one-sixth higher than soft dorsal. Coloration in life : Upper portions reddish-brown; lower parts, including ventrals and anal, a brighter red; vertical fins, and ventrals somewhat mottled, but body and head without distinct markings. Type No. 6141, L. S. Jr. University Museum. One specimen was obtained by me. This fish is similar to Scarus rubro-violaceus (Meeker), but differs in having 14 rays in pectoral, caudal lobes not produced, markings of fins, and in the absence of markings of body. The specimen gave the following measurements : Scarus paluca. Milli meters. Scarus paluca. Milli meters. Total length, to middle of margin of Height of first dorsal spine 14 184 18 150 21 88 27 50 34 53 Caudal 34 Eye 10 23 Snout 20 Scarus ahula .Jenkins, new species. Native names, U-liu-ula, Pauultumithu. Fig. 19. Head in length to base of the caudal 2.8, in total length to middle of margin of caudal 3.5. Depth in length to base of caudal 2.34; in total to middle margin of caudal 3. D. ix, 10. A. n, 10. P. 14. Scales 24, lateral line interrupted ; tubes single or once or twice branched; scales slightly roughened with very minute granulations extending their margins. Body deep compressed, profile slightly depressed. Teeth whitish, lower jaw included, no pointed teeth at angles of jaws. Upper lip double its entire length and covering about two-thirds of dental plate, lower lip covering about half of lower plate. Cheeks with one row of 4 scales, behind which are 2 scales placed one above the other; preopercle naked, posterior limb of opercle with 2 series, the anterior with 1 series of scales; 62 BULLETIN OF THE UNITED STATES FISH COMMISSION. a series of 4 scales on median line in front of first dorsal spine. Caudal fin convex, lobes rounded, length in head 1.26; height of caudal peduncle 2 in head; pectoral 1.35 in head, breadth being more than half its length; ventral 1.8 in head, not reaching vent by about half its length, inserted on a vertical from posterior limit of base of pectoral; dorsal spines flexible, but little less in height than soft portion, which equals soft anal in height. Coloration in life: Head, body, and fins uniformly brown, with reddish tinges, which are more pronounced on fins and throat. Base of caudal paler. No distinct markings on any part of body. Three specimens obtained by me of a total length of 218, 192, and 156 mm., respectively. One in Dr. Wood's collection is 237 mm. The fish was not rare, as it was frequently seen in market. It is not distinguished from Scarus paluca and brunneus by the fishermen. They are exposed for sale together and bring very high prices. Type No. 6142, L. S. Jr. University Museum. The details of the measurements of the four examined are given below : Scarus aliula. No. 1. No. 2. . No. 3. No. 4. Scarus ahula. No. 1. No. 2. No. 3. No. 4. Total to middle of caudal Length to base of caudal Distance of snout to vent mm. 224 175 110 62 73 11 27 mm. 192 152 98 55 57 10 23 mm. 156 125 83 45 47 9 18 mm. 237 193 125 70 72 11 29 Height of first dorsal spine. . . Height of soft dorsal Height of soft anal mm. 22 25 23 34 46 49 29 mm. 17 22 21 31 40 mm. 17 19 18 27 31 31 21 mm. 21 28 28 i 39 Depth 44 30 Snout Depth of caudal jjeduncle 24 Pig. 19. — Scams ahula Jenkins, new species. Type. Scarus miniatus Jenkins, now species. Fig. 20. Head in length to base of caudal 2.74, in total length 3.5. Depth in length to base of caudal, 2.6; in total, 3.3. D. ix, 10. A. ill, 9. P. 14. Scales 24. Lateral line interrupted. Body robust, deep, compressed. Teeth reddish or white; lower jaw included. A pointed tooth at angle of mouth on one side, the base of such a tooth only on other side in one specimen ; only the bases of such teeth in angles of mouth in other specimen, the teeth being worn or broken off. Upper lip double only posteriorly, covering about half of dental plate of upper jaw ; lower lip covering about a third of lower dental plate. Cheek with a single row of scales nearly imbedded, in one specimen so deeply imbedded that they are barely visible; no scales on lower preopercular limb; a series of scales on margin of opercle, those on its lower limbs nearly imbedded. Caudal fin slightly concave, lobes rounded, not produced; pectoral 1.35 in head, its breadth more than half its length; ventral half length of head, inserted just back of middle base of pectorals, and failing about half its own length of reaching vent; dorsal spines flexible; first few spines of dorsal, anal, and ventral with somewhat fleshy membranes. Coloration in life: Body, head, and fins a dull red, becoming a lighter red on lower parts and darker to a dusky reddish-brown on upper portions of body; no distinct markings, except a narrow violet line on outer margins of dorsal and anal fins ; iris brown. FISHES FROM THE HAWAIIAN ISLANDS. 63 Tlie description is based on one specimen, 500 mm. total length, obtained by me in 1887. The measurements are given below. A somewhat smaller specimen obtained by the Albatross in 1896 answers to this description in every particular. The fish is apparently not common. It brings an extravagant price among the natives in the market. Type No. 12144, L. S. Jr. University Museum. The measurements are given below : Scams miuiatus. Milli- meters. Scarus miniatus. Milli- meters. Height of longest ray of soft dorsal 390 45 142 70 152 105 Height of first dorsal spine 35 Pseudoscarus jordani Jenkins, new species. Fig. 21. Head in length to base of caudal 2.7, in total length to middle margin of caudal 3.37. Depth in length to base of caudal 2.64, in total length to middle margin of the caudal 3.3. Dorsal ix, 10. Anal ill, 9. Pectoral 14. Scales 24; lateral line interrupted. In this specimen two scales at the interrupted portion out of the series bear tubes. They are located one each over first two scales in the series following interruption. Tubes much branched; surfaces of scales, except at posterior margin of each, much roughened over the whole body by striations composed of rows of minute tubercles. Body robust and greatly compressed. An adipose lump over snout. Teeth green, lower jaw included; a strong tooth at each angle of upper jaw. Upper lip double only posteriorly, covering more than half of dental plate; lower lip covering less than half of dental plate. Cheek with 3 rows of scales, upper of 7 scales, middle row of 7, the lower row of 2 scales, which extend on lower preoper- cular limb; a series of scales along entire margin of opercle; a series of six scales on median line before first dorsal spine. Lobes of caudal fin much produced, being longer than body of the fin; height of caudal jieduncle in head 2.54. Pectoral 1.5 in head, its breadth being less than half of its own length ; ventral 1.75 in head, not reaching vent by one-half its own length, inserted on a vertical from about middle of base of pectoral; dorsal spines flexible; membrane of first few spines of dorsal, anal, and ventral somewhat fleshy on outer margin. Coloration in life: General color blue, the sides of body and head rosy or pink; region of body just below posterior two-thirds of dorsal and the caudal peduncle green; iris orange; margin of upper lip orange, above which it is bright blue; margin of lower lip blue, below which is an orange area, below this again it is bright blue which changes to a lighter blue; posterior to this a pink area; a pink wavy line from angle of mouth to eye; dorsal fin pink with upper and anterior border blue, and with a blue bar along each spine; anal yellowish with bright blue anterior outer and posterior margins; ventral blue anteriorly and pink posteriorly; caudal, upper and lower borders bright blue, interior portions with reticulations of pink and blue, colors almost wholly disappearing in alcohol; 64 BULLETIN OF THE UNITED STATES FISH COMMISSION. there remains the green below the posterior two-thirds of dorsal and on caudal peduncle as a pigment on the scales. But one specimen of this brilliant fish was obtained by me, 700 mm. in length, including lobes of caudal. A high price was asked for it indicating that the natives esteem it as food. Type No. 12143, L. S. Jr. University Museum. The measurements are given below: Pseudosearus jordani. Milli- meters. Pseudoscarus jordani. Milli- meters. Total length, to middle of margin of Snout 85 600 30 100 53 480 100 178 120 Depth 182 Breadth of narrowest jiortion of caudal Eye 20 70 Fig. 21.— Pseucloscarus jordani Jenkins, new species. Type. Fig. 22. — Pscudocheilinus octotcenia Jenkins, new species. Type. Pseudocheilinus octotcenia Jenkins, new species. Fig. 22. Head in length to middle margin of caudal 3.1, to base of caudal 2.5. Depth 3.7 (3). D. ix, 11. A. in, 9. Scales 24; lateral line interrupted; tubes single. Body moderately slender, head and snout moderately long, profile slightly concave. Teeth in a single series, no posterior canine ; three pairs of large canines in anterior of upper jaw, the posterior pair of which bends outward and backward; one pair of strong canines in anterior of lower ja w. Check with mainly two rows of scales ; anteriorly the scales are confused, hut apparently of three series, encroaching somewhat on preopercle; opercle FISHES FROM THE HAWAIIAN ISLANDS. 65 with a single series of scales; a series of 5 scales in middle line before dorsal; a series of scales largely sheaths base of dorsal; a similar series sheaths to less extent the base of anal. First dorsal spine shorter than the following, which are subequal, but of less height than soft dorsal; spines pungent, each bearing a produced filament; second anal spine slightly longer than third; soft anal about equal to soft dorsal; caudal rounded, ventral does not reach vent by about half its length; pectoral nearly half of head. Coloration in formalin: General groundwork pinkish, a conspicuous dark longitudinal band running along center of each row of scales, making eight bands on side of body, the one at base of dorsal through sheathing scales less distinct posteriorly; a similar but narrow band through middle of scales in median line before dorsal spines; faint traces of markings on head; apparently the three upper bands of body extending on to head, the cheeks and lower parts being unmarked. Spines and rays of dorsal and anal blue. Dr. Bleeker founded the genus Pseudooheilinus on a single species described by him as P. liexa- tcenia, from Amboyna. This species is included under this genus from its closer relation to P. liexatcenia, its sole representative, than to Cheilinus, as shown in dentition, there being three pairs of strong anterior canines in upper jaw, the third being largest and turning outward and backward. The second anal spine is, however, only slightly longer than third, and the squauiation of the cheeks shows anteriorly more than two series. This description is based on a single specimen, 115 mm. total length. It was obtained by Mr. Keleipio, of Honolulu, for the collection of Dr. Wood in the winter of 1899. Type No. 6122 L. S. Jr. University Museum. The measurements follow: Pseudooheilinus octotamia. Milli- meters. Total length to middle margin of caudal 115 Length to base of caudal 92 37 31 7 15 60 Depth of caudal peduncle 15 Height of first dorsal spine 5 Pseudoekeilinus octotoenia. Milli- meters. Height of ninth dorsal spine 8 Height of soft dorsal 12 First anal spine 7 Second anal spine 9 Third anal spine 8 Soft anal 13 Ventral 16 Pectoral 20 Leland Stanford Jr. University, May 26, 1899. F. C. B. 1899—5 CONTRIBUTIONS TO THE BIOLOGY OF THE GREAT LAKES. ROTATORIA OF THE UNITED STATES, WITH ESPECIAL REFERENCE TO THOSE OF THE GREAT LAKES.* By H. S. JENNINGS, Instructor in Zoology, University of Michigan, INTRODUCTION. Among the objects to be attained in the biological investigations of the Great Lakes, inaugurated by the United States Fish Commission, a preliminary one is the collection and determination of the animals and plants found in the lakes and the placing of these on record. A portion of the first summer in the field (1898) was spent by the writer in a study of the Rotatoria of the region of Lake Erie about South Bass Island, where the summer laboratory was situated; the observations and collections there made form the basis of the present paper. Three summers had been spent previ- ously by the writer in study of the Rotatoria of other parts of the Great Lakes and of some of the inland lakes of Michigan; other observations in the Great Lakes have been made by Kellicott (’96 and ’97) on the Rotatoria of Lake Erie. These researches, taken together, make possible an extended, though of course incomplete, list of the Rotatoria of the Great Lakes. As the work on the Great Lakes has included the larger number of species observed in the United States, it has been deemed advisable, in this first report of work done for the United States Fish Commission, to include a record of all the Rotifera thus far observed in the United States, together with all the localities in which they have been observed, so far as possible. Notices of the Rotifera are scattered through many publications, and it is believed that nothing will serve better as a basis for future work than to bring these scattered notices together. In beginning a study of a circumscribed group of animals such as the Rotatoria, in connection with a general biological survey of the Great Lakes, it is well to have clearly in mind an outline of the problems upon which work is to be done. The lake is to be looked upon as an organism, the various groups of animals and plants in the lake, as well as the chemical and physical conditions and forces there present, being the organs which make up the whole. These organs are necessarily as closely corre- * The papers in this series are based on investigations of the U. $. Fish Commission, under the direction of Prof. Jacob Keighard, of the University of Michigan. Three other articles of the series, by the same author, have been published in the American Journal of Physiology, vol. ii, 1898 (pp. 311-341, and 355-379), and vol. in, 1900 (pp. 229-260), as follows: (1) The Mechanism of the Motor Reactions of Paramecium; (2) Laws of Chemotaxis in Paramecium; (3) On the Movements and Motor Reflexes of the Flagellata and Ciliata. 67 68 BULLETIN OF THE UNITED STATES FISH COMMISSION. lated and adapted to each other as the organs of an animal, and the network of interrelations forms the chief object of study for such a biological survey. Our special problem is to determine just what place the Rotatoria occupy as members of this organic whole. For this a study of the group in all its relations will be necessary. Such a study will follow some such lines as the following: First, the objects of study must be known and described. To this end there must be on record somewhere full and complete descriptions and figures of all the Rotifera found in the lake, with their correct names. An important preliminary feature of the work consists, therefore, in figuring and describing such forms as have been incor- rectly or incompletely described or not described at all. On account of the large number of minute species found among the Rotifera, and the unsatisfactory work that has been done upon many of them, much remains to be accomplished in this line. The best method of carrying on this work will consist in studying carefully circum- scribed groups — the species of a single genus, perhaps — describing and figuring all the species, and going critically over the literature of the group in such a way as to set the synonymy in order. Studies of this sort on the genus Monostyla , on Cathypna and Distyla, on Metopidia , on Colurus , on the Rattulidce , or, for a much more extensive problem, on the Notommatadw, would be exceedingly valuable. Lists of all the species found in the given body of water must also be placed upon record, in order that the investigator may know with what material he has to deal. Such lists, carefully prepared, are also of much interest for a study of the general problem of the geographical distribution of the group with which we are dealing. The work above characterized must be considered as purely preliminary to the main object of study ; the present paper includes only such preliminary matter. The distribution of the animals within the lake and the study of the conditions under which they live constitute a problem of much greater interest. Most of the species of rotifers seem fitted to some special environmental conditions; what these conditions are may be determined by observation and by experiment. From the side of observa- tion, lists should be kept of Rotifera of the different regions into which a body of water maybe divided — distinguishing thus, as is usually done, limnetic species, littoral species, bottom species, and swamp species. This classification is, however, much too general to give precise results. It should be supplemented by a careful study of the Rotifera of regions of different character, with regard to the depth of the water, the character of the bottom, and especially with relation to the plant life of the water. The rotifers from the Chara beds, Ceratophyllum beds, Naias beds, Potamogeton beds, the Cladophora-covered surfaces, should be studied and listed, to see what relation there is between the animal and plant life of the lake. A beginning of this work was made during the summer of 1898, but in the two months during which the party was in the field not enough could be done, in addition to other work, to permit the drawing of any conclusions, so that the results will not be presented till further study is made. The changes in the character of the fauna with different seasons of the year and with variations in climate must be studied as an important feature of the conditions of existence. The problem may be attacked more directly under experimental conditions. The animals may be kept in laboratory cultures, and suitable changes produced artifi- cially in the conditions within these cultures. The results of these changes on the life of the animals may thus be observed. Proper control cultures will always be necessary in such work. The effects of varying conditions in modifying a given single species is a problem ROTATORIA OF THE UNITED STATES. 69 of the greatest interest, leading to the general question of the origin and nature of species and to the deepest questions of morphogenesis. The variations of a given species should be recorded and studied; from a sufficient number of records the laws of the variations may be deduced. More important still is a study of the causes of the variations. The variations occurring with a change of conditions, or with a change of season, are worthy of careful study, such a study as has recently been made by Lauter- born (’98) of Anurcea cochlearis. An experimental study of the production of variations promises most for this problem; to this end the very variable and hardy species Brachionus bakeri would probably be a most useful form. The life-history and reproductive habits of the animals furnish a further important field for study. Many features in the life-history of the Rotifera are of much interest and need such study as has been carried on thus far only for Eydatina senta, by Maupas, Tsnssbaum, and others. The striking sexual dimorphism ; the conditions of the occur- rence of males at certain periods; the determination of the different kinds of eggs pro- duced at different times; the sex of the offspring — all present problems of the greatest interest for an experimental study. A most important and neglected field lies in a study of the activities by which the animals respond to their environment. Proper conditions of existence are neces- sary, but these are not sufficient alone to preserve the life of the organisms ; the animals must respond to their environment by appropriate activities. We must know what these activities are and the laws which govern them. In other words, a study of the physiology and especially the psychology of the animals is necessary before we can understand the interaction of organism and environment; the functions and move- ments of the organisms constitute one of the chief factors in the network of interrela- tions of the “ organs ” of the lake. Having determined the general laws according to which the organisms respond to their environment — in other words, having made a study of the psychology of the animals — it will then be possible to determine by obser- vation and experiment the specific factors which cause migrations, the sudden appear- ance of the animals in a given locality, their quick disappearance from another locality, and the like. In any group of animals the investigation must follow some such line as marked out above, in each case a study of the normal psychology of the animals being a prerequisite to an understanding of the laws of their migrations and other striking activities. Commencing at the beginning, a study of this nature was made on the Protozoa during the summer of 1898, the results of which are resumed in another paper, which deals with the Protozoa; it is hoped soon to extend this study to the Rotatoria. The psychology of the Rotatoria has been studied scarcely at all; notes as to the nature of the food and the method of taking it, together with descrip- tions of the method of forming the tube in some tube-dwelling species, being the chief matters that can be gleaned from the literature. The only way in which the problems above characterized can be solved, the relations of the Rotifers (or of any other group) to their environment worked out, is for investigators to choose definite limited problems for solution and devote time and energy to observation and experimentation till the questions proposed are answered. Mere isolated observations, collected during a systematic study of the group, can do little; investigators must take up the work in the same spirit in which morphological jjrobleins are attacked, concentrating all efforts upon a given point until that is settled. The activities of animals are as worthy of such study as are their structures. Until a large amount of investigation has been done it will not be possible to give any 70 BULLETIN OF THE UNITED STATES FISH COMMISSION. satisfactory discussion of the place occupied by the Botatoria in the life system of the lake. In this paper it is only attempted to point out here and tliei'e problems that await settlement at the hands of careful investigators. The paper is, therefore, purely preliminary in character, aiming to show merely what species have thus far been recognized in the United States and where they are found, as well as giving descriptions and figures of some species that are in need of study. Future reports need not take the shape of a formal list, but will give accounts of special studies in any line or record additions to the fauna. Formal lists of species are perhaps the most uninteresting of scientific writings, yet they form a disagreeable necessity as a basis for further work. An ideal list is such a one as that given by Weber (’98) of the Botatoria of the basin of Lake Geneva, every name accompanied by a beautiful figure of the animal. The short time spent on the work thus far has rendered this impossible for the Botatoria of the Great Lakes. Most of the figures must be reserved for the future development of the work along this line. In the case of new species, or where there are other causes for special interest, figures are given in the present paper. These figures were drawn by Mrs. Louise Jennings from camera sketches made by the author. The author has endeavored to avoid, as far as possible, the naming of new species. Since the publication of Hudson and Gosse’s Monograph of the Botifera, about ten years ago, study of this group has been very active, resulting in the multiplication of papers on the subject, often without relation to one another, and describing the same forms under different names. A certain amount of this was perhaps inevitable at first, but heedless work has multiplied the resultant confusion many fold. No one has a right to cumber scientific literature with the names of species “presumably uudescribed,” as a recent paper naively puts it, without recognizing the fact that a vast volume of literature has appeared on the group since the publication of Hudson and Gosse’s Monograph, including descriptions of many new species (295 up to 1897, according to Bousselet, ’97). The recognition and description of a new species must therefore be regarded as a most laborious piece of work, involving a careful examina- tion of large numbers of papers in various languages, besides a consultation of Hudson and Gosse. There is no excuse for omitting such a study before publishing- descriptions of species as new, in view of the full lists of -new rotifers published at intervals by Mr. Gliarles Bousselet (’93 and ’97), with the titles of the papers in which the descriptions are published. If a student finds himself unable to see a large share of these papers it is his duty to recognize the fact that he is not in a position to publish names of new species. If he wishes to publish his notes and drawings, these may be of great use to other workers, but if he proceeds to append new names to his descriptions, increasing the already heavy burden of synonymy, his work becomes a positive injury to science and a nuisance to all careful scientific students. The record of American workers on the Botifera has not always been so good in this matter as might be wished. In the American paper above referred to as giving names to “ presumably undescribed ” species, six so-called new species are figured on the plate, with new specific names in the descriptions. Of these six, four are easily recognizable as old friends by anyone familiar with the recent literature of the subject, while the other two are thought by another reviewer to be old species. This illustrates the value of the description of “ presumably new ” species without comparison with those described in recent papers. Science is burdened with four, perhaps six, new synonyms. Another mistake to be avoided, as has been emphasized by Bousselet (j96), is the ROTATORIA OF THE UNITED STATES. 71 making of new species because the observer finds some structure not previously figured or described in wTiat would otherwise be considered an old species. It must be recognized that very few of the older figures and descriptions are in any sense complete; it was the purpose of many of the older authors to give merely such a figure and description as would lead to ready recognition of the animal, not to give an exhaustive anatomical account. Moreover, the improvements in optical instruments and in technique have been such as to enable even the amateur to see much that formerly remained hidden to the best investigators. A third opportunity for the introduction of confusion into the study of the Rotatoria lies in the great variability of many forms. There are few species that are not sufficiently variable to permit an observer to find specimens that differ from the type enough to allow him to immortalize his name by appending it to a synonym, if his ambition runs in that direction. Rousselet (’97) has already pointed out that many of the recent new species are but slight variations of well-known forms. It must be recognized, however, that the limits of variability are not easily defined, and that it is often very difficult to say whether a given specimen should be considered a new species or a variation of an old one. Mistakes from this source are therefore to a cer- tain extent excusable, while those resulting from the first two causes above mentioned are usually due to carelessness or ignorance. To describe a new species, one should therefore have access to all the papers in which new species have been described since the publication of Hudson and Gosse’s Monograph, or at least to all papers describing any species belonging to the genus under consideration. The titles of the papers bearing directly on the genus of which it is proposed to describe a new species may be determined — up to 1897, at least — from Mr. Charles Rousselet’s lists of new species of Rotifera (Rousselet ’93 and ’97). As a further precaution against error, it would be well to submit either mounted speci- mens or drawings and notes on proposed new species before publication to someone thoroughly competent to judge as to their claims. Mr. Charles Rousselet, 2 Pembridge Crescent, Bayswater, London W., England, is doubtless as well acquainted with the Rotifera as anyone in the world, and is always willing, with uniform courtesy and kindness, to give expert advice as to the publication of what seem to be new species. On account of the recent great multiplication of new species, a description of a new rotifer should be accompanied by a careful comparison with any other species of the same genus that at all nearly resemble it and the points of difference brought out clearly. In a number of recent cases the lack of ground for giving a new name would at once have been evident if this had been done. For example, Stenroos (’98) in his recent valuable paper on the animal life of Lake Rurmijarvi, in Finland, after describ- ing as new Cathypna magna n. sp., gives a list of the known species of Cathypna, among them Cathypna ungulata Gosse. A careful comparison of Cathypna magna Stenroos with Cathypna ungulata Gosse would have disclosed their identity. The publication of a new species without a figure, which has been practiced by some American authors, as well as by some of those of Europe, is greatly to be deprecated. Usually the figure is the most important part of the account of a rotifer, and a description could, as a rule, be much better dispensed with than a good figure. To sum up, therefore, anyone who proposes to publish a description of a rotifer as new should fulfill the following conditions: 1. Rot only Hudson and Gosse’s Monograph, but all subsequent papers contain- ing descriptions of rotifers in any way related to the one in hand, should be consulted. 72 BULLETIN OF THE UNITED STATES FISH COMMISSION. 2. New species should not be described as a result of the discovery of some hitherto unmentioned anatomical detail in an otherwise known species. 3. Great care should be exercised not to describe as new species mere variations of an old species. 4. If any doubt can possibly exist, the figures and descriptions should be submitted, before publishing, to some expert who has all the literature at hand. 5. A description of a new species should be accompanied by a detailed comparison with any very closely related species that may exist, to show wherein this one differs and why it is considered new. 6. Every description of a new species should be accompanied by a good figure or figures. For the two cases in which it has seemed necessary to describe certain forms as new in the present paper, an attempt has been made to fulfill these conditions. The subjoined list contains not only the species found by the author in Lake Erie, in the region of South Bass Island in the summer of 1898, but also, so far as known to the writer, all the species that have been found in the United States, together with the localities from which each species has been recorded. An attempt has been /made to make this list as complete as possible, but the references to the Rotatoria are exceed- ingly scattered, so that I can not hope that none have been overlooked. Nevertheless it is believed that the omitted references are v.ery few. A brief review of the history of the study of the Rotatoria in this country may be of interest in this connection. The first recorded observation of any member of the group in America seems to be that of Bose (1802), who observed some rotifer belonging to the Philodinidce in Carolina. Ehrenberg in his great work (’38) held Bose’s animal to have been Rotifer vulgaris , while in a later paper (Ehrenberg, ’43) he considers it to have been probably Callidina rediviva Ehr. The next notice of American Rotatoria that I have been able to find is that by Ehrenberg (’43). He lists a few rotifers observed by him in material sent to him from this country by various men of science. In 1S55 Bailey (’55) described Limnias annulatus Bailey. Schmarda (’59) in his trip around the world, 1853 to 1857, observed two rotifers “in brackish water near New Orleans.” From this time on, up to 1879, little notice of the Rotifera is to be found in American journals, save a few notes by Leidy (’51, ’57, ’74, and ’74^) and one by Peirce (’75), in the Proceedings of the Academy of Natural Sciences of Philadelphia. In 1879 the late Dr. D. S. Kellicott published his first note on the Rotifera, a description of Notliolca longispina Kellicott. This was followed by many other papers on the group, up to the year before the death of this author iu 1898. The decade from 1880 to 1890 was marked by numerous brief papers and notes on the group, by Kelli- cott, Herrick, Leidy, Attwood, Yorce, Forbes, Foulke, Stokes, Up de Graff, and others. The first formal list of American species was that of Herrick (’85) of rotifers found in Ohio and Minnesota, followed with one by Kellicott (’88) of rotifers found at Corunna, Micti. In the decade now coming to an end, work on the group has been much increased, especially in connection with the founding of fresh- water biological stations. Extended local lists of species have been published by Turner (’92), the present writer (’94 and ’96), Kellicott (’96 and ’97), and Hempel (’98). ROTATORIA OF THE UNITED STATES. 73 In the following list those species representing the autnor’s investigations of Lake Erie during the summer of 1898 are numbered consecutively; those not observed here are not numbered. After each locality is given the name of the investigator who has recorded it, followed by numerals showing the year in which the publication took place. The exact reference may then be determined by turning to the list of litera- ture at the end of the paper, where the authors’ names are arranged alphabetically, and the papers of a given author are distinguished by prefixing to each the number of the year (in the century) in which it was published. In certain cases species are recorded in proceedings of societies as having been exhibited by some member of the society; in every such case the citation is given under the name of the member who made the exhibit. In some cases I have recorded here for the first time localities other than Lake Erie in which I have at some time observed a species; these localities are signed with my own initials (H. S. J.). The region studied by the author during the summer of 1898 consisted of the waters about South Bass Island, especially the waters of the lake along the shore of the island. Naturally the waters in the immediate neighborhood of Put-in Bay were most carefully examined, since the laboratory was situated on the shore of this bay. Many excursions, however, were made to more distant regions. East Harbor, south of the island, on the northern shore of Ohio, furnished many of the rotifers. Others came from towings made in Lake Erie- at a distance from shore. Two swamps on South Bass Island were carefully examined; one lies close to the United States fish- hatchery, while the other lies on the east shore of the island, just east of the village of Put-in Bay. The latter is referred to in the list as “East Swamp.” The swamp near the fisli-liatchery is connected with the lake by a channel about 50 feet long, and is situated at such a level that at times water flows from the lake into the swamp, while again it flows from the swamp into the lake; therefore, as might be expected, the limnetic rotifers of the lake sometimes occur in the swamp, while at other times the fauna of the swamp is of the most pronounced stagnant- water type. East Swamp has no connection with the lake. The proper classification of the Botatoria presents great difficulties. The system most in use is that of Hudson and Gosse, as given in their Monograph of the Rotifera. This classification is unsatisfactory in many ways, and what I consider a better one in many respects has recently been proposed by Lund (’99). After consideration it was decided, however, not to introduce this new classification into the present paper, as most workers on the group are now better acquainted with the classification given by Hudson and Gosse, so that the use of their system will best facilitate reference to the list. The sequence of orders, families, and genera adopted is therefore that of Hudson and Gosse, in the Monograph of the Rotifera published in 1889, with some modifications rendered absolutely necessary by more recent investigations. 74 BULLETIN OF THE UNITED STATES FISH COMMISSION. A LIST OF ALL ROTATORIA HITHERTO FOUND IN THE UNITED STATES, AND THE LOCALITIES WHERE THEY HAVE BEEN OBSERVED, WITH DESCRIPTIONS OF TWO NEW SPECIES. [Accompanied by plates 14-22, figs. 1-46.] Order I. RHIZOTA. Much less study has heeu given to the Rhizota, or attached Rotifera, by the present writer, than to the free-swimming forms, and the same seems to' he true of other students of American Rotatoria. Doubtless much remains to lie done before even an approximately full list of the American Rhizota can he given, and some of the forms already listed are sadly in need of careful study. There is much room for work in preparing full and accurate descriptions and figures of the American Rotifera of this group. lily 1. FLOSCULARIADiE. FLOSCULARIA Oken. 1. F. cornuta Dobie. On FJodea from East Harbor, Lake Erie. Pond near Bangor, Me. (J. C. S., ’83). Nigger Creek, Grand Island, Niagara River (Kellicott, ’87). Shiawassee River at Corunna, Mich. (Kellicott, ’88). Chippewa Lake, Mecosta County, Mich., and Lake St. Clair (Jennings, ’94). Sandusky Bay, Lake Erie, and Buffalo, N. Y. (Kellicott, ’96). 2. F. algicola Hudson. On Cladophora and Cliara, in East Swamp, South Bass Island. The specimens found agreed perfectly with the description given by Hudson in the small size (210//= 0.01 inch in length), in the dots on the corona arranged in symmetrical patterns, and in other respects; but they inhabited evident tubes, while Hudson was unable to find a tube. 3. F. mutabilis Bolton. This free-swimming form seems widely distributed in waters of the Great Lakes, although it is never present in large numbers. It was frequently taken with the tow net and plankton net in the open lake about South Bass Island, in Lake Erie. Also found in a swamp close to United States fish-hatchery on South Bass Island, at times when the water of the lake flowed into the swamp. Lake St.Clair (Jennings, ’94). Lake Michigan, Pine Lake, Round Lake, and West Twin Lake, near Charlevoix, Mich. (Jennings, ’96). Sandusky Bay, Lake Erie (Kellicott, ’96). 4. F. pelaeica Rousselet. The distribution of F. pelagica is about the same as that of F. mutabilis. I found it in collections made with the tow net and plankton net in Lake Erie at various places near South Bass Island. It is noteworthy that neither of these limnetic Floscularias have been reported from the carefully studied waters of the Illinois River. Lake St.Clair (Jennings, ’94). Lake Michigan, Round Lake, and Pine Lake, near Charlevoix, Mich. (Jennings, ’96). 5. F. millsii Kellicott. A single specimen attached to a Difflugia shell, from the Portage River, Ohio, about a mile from Lake Erie. Black Creek, Ontario, Canada (Kellicott, ’85). Nigger Creek, Grand Island, Niagara River (Kellicott, ’87). Sandusky Bay, Lake Erie (Kellicott, ’97). Also found by Mr. J. B. Shearer at Bay City, Mich, (according to Kellicott, ’97). F. ornata Ehrenberg. — Pond near Bangor, Me. ( J. C. S., ’83). “American species” (Kellicott, ’84). Minnesota (Herrick, ’85). Nigger Creek, Grand Island, Niagara River (Kellicott, ’87). Shiawassee River at Corunna, Mich. (Kellicott, ’88). Exhibited in New York (Helm, ’89 and ’91). In the neigh- borhood of Cincinnati, Ohio (Turner, ’92). Chippewa Lake, Mecosta County, Mich., and Lake St. Clair (Jennings, ’94). Sandusky Bay, Lake Erie (Kellicott, ’96). Waters connected with the Illinois River at Havana, 111. (Hempel, ’98). F. campanulata Dobie. — Black Creek, Ontario, Canada (Kellicott, ’85). Nigger Creek on Grand Island, Niagara River (Kellicott, ’87). Shiawassee River at Corunna, Mich. (Kellicott, ’88). West Twin Lake, Muskegon County, Mich.; McLaren Lake, Oceana County, Mich. (-Jennings, ’94). West Twin Lake, near Charlevoix, Mich. (Jennings, ’96). Sandusky Bay, Lake Erie (Kellicott, ’96). ROTATORIA OF THE UNITED STATES. 75 F. cilent ata Collins. — McLaren Lake, Oceana County, Midi. ; Chippewa Lake, Mecosta County, Mich.; Crooked Lake, Newaygo County, Mich. (Jennings, ’94). Small pond at Cedar Point, San- dusky, Ohio (Kellicott, ’97). Swamp near Norwich, Vt. (H. S. J.). F. ambiijua Hudson. — Shiawassee River, Corunna, Mich. (Kellicott, ’84 and ’88). Sandusky Bay, Lake Erie (Kellicott, ’96). F. coronetta Cuhitt. — American species” (Kellicott, ’84). Shiawassee River at Corunna, Mich. (Kellicott, ’88). F. trilobata Collins. — Pond near Bangor, Me. (J. C. S., ’83, under the name F. trifolium). Corunna, Mich. (Kellicott, ’84, as F. trifolium). F. regalis Hudson. — Found by Up de Graff at Elmira, N. Y. (according to Kellicott, ’84). Acyclus inquietus Lcidy.— Schuylkill River, near Philadelphia (Leidy, ’82). APSILUS Metschnikoff. (Plate 14, figs. 1 and 2.) What seems to he the young free-swimming form of a species of this genus was found sparingly amid Lenina, Spirodela, and TYolffia in the swamp near the United States fish-hatchery, on South Bass Island, in July, 1898. It is not possible to determine the species of young specimens from the pub- lished descriptions of the members of this genus, because they deal entirely with the adult. In my paper of 1894 I assumed that this was A. lentiformis Metsch., that being the only species recognized as certainly well established by Hudson and Gosse. But if Stokes (’96c) is correct in his contention that four species of the genus are to be recognized, then this may belong to any of them. A thorough study of the animal was undertaken, but they disappeared before this was completed; I give my notes and drawings, however, so far as they go. A side view, accurate as far as it goes, but not showing all details, is given in fig. 1. The body is thick and clumsy and slightly curved, so as to be convex dorsally and concave ven- trally. It tapers slightly toward both ends, ending in a truncate fashion, at the anterior end in the head, at the posterior end in the broad disk which serves as a foot. The head is separated from the body dorsally by a slight depression, while ventrally the outline of the body continues uninterruptedly into the head. The posterior disk is marked off from the body by a broad shallow constriction. The animal is very transparent and entirely colorless, except for the two red eyes. The ciliated face at the anterior end is slightly oblique, a non-ciliated part projecting above the ciliated portion. The cilia are rather long and seem to form a simple circle, but observations on this point are not complete. A large lobed brain lies behind the corona, bearing in front the two red eyes, in which crystalline lenses are clearly visible. The head may be extended considerably farther than is shown in the figure, or may be entirely retracted. The body of the animal is filled with a bewildering confusion of glands, digestive organs, muscles, and nerve cords. Much of interest might be brought out by a minute study of these parts; 1 have seen nn rotifer that appeared so favorable for a study especially of the muscular and nervous systems. The figure, accurate so far as it goes, gives an idea of the complex of details awaiting disentanglement. I shall not, in the absence of minute study, attempt to interpret the structures shown. The prominent trophi lie, as in other members of this genus, at the bottom of a large sac; they are of the peculiar form characteristic of the genus. They are shown in fig. 2, plate 14. The body wall seems much more complex than is usually the case in the Rotifera. It appears to be possible to distinguish four layers, beginning with the outside: (1) A thin cuticula; (2) a thick, gelatinous cellular layer; (3) a layer of transverse muscles; (4) a layer of longitudinal muscles. The two muscular layers are not completely separated, and some of the longitudinal muscles traverse the body cavity, but in a general way the distinction into two layers is evident. The posterior disk is retractile into a sort of mantle which partly covers it, and the whole, mantle and all, may be drawn within the body. The disk is concave, with lines radiating from the deeper central portion, and is ciliated. The movements of the animal much resemble those of Asplanchna. Lund (’99) holds that Apsilus should be removed from the Flosculariadce and placed in a separate family near the Asplanclinadai. The present author (’94) recorded this same form from Lake St. Clair as J. lentiformis Metsch. A. vorax Leidy. — Schuylkill River and Fairmount Park, Philadelphia (Leidy, ’57, ’82, and ’84). A. bipera Foulke. — Fairmount Park, Philadelphia (Foulke, ’84). Pool at Trenton, N. J. (Stokes, ’96c). Sandusky Bay, Lake Erie (Kellicott, ’97). A. bucinedax Forbes. — In an aquarium at Normal, 111. (Forbes, ’82). Pool at Trenton, N. J. (Stokes, ’96c). 76 BULLETIN OF THE UNITED STATES FISH COMMISSION. A. lentiformis Metsch. — In Phipps Conservatory tanks at Allegheny, Pa. (Logan, ’95). Stephanoceros eichhornii Ehrenberg. — Philadelphia ? (Peirce, ’75). Bangor, Me. (J. C. S., ’83). Ponds in New Jersey (Balen, ’83 and ’85). Found by E. B. Grove in Rogers Glen, Oneida, N. Y., by W. R. Cross at Camden, Me., and by C. F. Park near Poughkeepsie, N. Y., according to Balen (’83). In Can- ada, across from Buffalo, N. Y. (Kellicott, ’84). Pittsburg, Pa. (Mellor, ’89). Exhibited in New York (Cox, ’89, and Helm, ’97). McLaren Lake, Oceana County, Mich., and Horsehead Lake, Mecosta County, Mich. (Jennings, ’94). Sandusky Bay, Lake Erie (Kellicott, ’97). Family 2. MELICERTADiE. MELICERTA Schrank. 6. M. conifera Hudson. East Swamp, South Bass Island, on Chardceai. Shiawassee River at Corunna, Mich. (Kellicott, ’88). Sandusky Bay, Lake Erie (Kellicott, ’96). M. rinc/ens Schrank. — Niagara River (Fell, ’82). Baugor, Me. (.J. C. S., ’83). New Jersey (Balen, ’85). Shiawassee River at Corunna, Mich. (Kellicott, ’88). Exhibited in New York (Helm, ’89, and Walker, ’94). Exhibited in San Francisco (Breckenfeld, ’89). Lake St.Clair and West Twin Lake, Muskegon County, Mich. (Jennings, ’94). Phipps Conservatory tanks at Allegheny, Pa. (Logan, ’95). Susan Lake, North Michigan (Jennings, ’96). Black Channel, Sandusky Bay, Lake Erie (Kellicott, ’97). M.janus Hudson. — In the following inland lakes of Michigan: McLaren Lake, Oceana County ; Crooked Lake, Newaygo County ; Horsehead Lake, Mecosta County ; Chippewa Lake, Mecosta County (Jennings, ’94). Mr. John Hood has recently called my attention to the fact that the Melicerta floccu- losa, described (without a figure) by Kellicott (’96), from Sandusky Bay, resembles in almost, if not quite, every respect M.janus, except in the possession of a tube without pellets. Now, Mr. Hood finds M.janus in Scotland inhabiting tubes without pellets, so that there is a strong probability that he is right in believing Kellicott’s M.flocculosa to be really M. janus. Kellicott’s description of M.flocculosa is so meager that it is difficult to find facts upon which to base a positive opinion; the only point in the description of M.flocculosa which might be held not to coincide with thatof M.janus is that Kelli- cott says the chin is spatulate, whereas in Hudson’s description the chin is said to be two-pointed. It will probably be best to consider M.flocculosa, at least provisionally, as a synonym of M.janus. M. tubicolaria Ebr. — Exhibited in New York (Helm, ’89). Sandusky Bay, Lake Erie (Kellicott, ’97). M. flocculosa Kellicott. — See M.janus. Limnias ceratophylli Schrank. — “Abundant in our rivers ” (Leidy, ’74). Pond near Bangor, Me. (J. C. S., ’83). Shiawassee River at Corunna, Mich. (Kellicott, ’88.) Sandusky Bay, Lake Erie, and Niagara River (Kellicott, ’96). Exhibited in New York (Helm, ’97). Waters connected with Illinois River at Havana, 111. (Hempel, ’98). L. shiaivasseensis Kellicott. — Shiawassee River at Corunna, Mich. (Kellicott, ’88). Sandusky Bay, Lake Erie (Kellicott, ’96). L. annulatus Bailey. — West Point, N. Y. (Bailey, ’55). Shiawassee River at Corunna, Mich. (Kellicott, ’88). Sandusky Bay, Lake Erie (Kellicott, ’96). L. socialis Leidy = L. ceratophylli Schrank. Cephalo siphon limnias Ehrenberg. — Along shore of Niagara River (Mills, ’81). Buffalo, N. Y., and Shiawassee River Michigan (Kellicott, ’87). Corunna, Mich. (Kellicott, ’88). Olentangy Creek at Columbus, Ohio (Kellicott, ’89). Lake St. Clair (Jennings, ’94). Sandusky Bay, Lake Erie (Kellicott, ’96j. Waters connected with Illinois River at Havana, 111. (Hempel, ’98). C. candidus Hudson. — Olentangy Creek, Columbus, Ohio (Kellicott, ’89). C. furcillatus Kellicott = CEcistes melicerta Ehrenberg. GZcistes melicerta Ehrenberg. — Olentaugy Creek at Columbus, Obio (Kellicott, ’89, under the name Ceplmlosiphon furcillatus). Swamp on the shore of Lake St. Clair (Jennings, ’94). This species is represented by Stokes (’81, fig. 2) without a name, probably from New Jersey. The two dorsal hooks are developed in many specimens into two great branched antler-like structures, which are shown in plate 14, fig. 3. There seems to be no justification for the change of the specific name from melicerta, as given by Ehrenberg, to ptygura, as given by Hudson and Gosse. 0. longicornis Davis. — Shiawassee River at Corunna, Mich. (Kellicott, ’88). Lake St. Clair and McLaren Lake, Oceana County, Mich. (Jennings, ’94). Sandusky Bay, Lake Erie (Kellicott, ’96). Exhibited in New York (Helm, ’97). 0. mucicola Kellicott. — Shiawassee River at Corunna, Mich. (Kellicott, ’88). Sandusky Bay, Lake Erie (Kellicott, ’96). West Twin Lake, near Charlevoix, Mich. (Jennings, ’96). Waters con- nected with the Illinois River at Havana, 111. (Hempel, ’98). ROTATORIA OF THE UNITED STATES. 77 O. crystallinus Ehrenberg. — Shiawassee River at Corunna, Mich. (Kellicott, ’88). Sandusky Bay, Lake Erie (Kellicott, ’96). O.umbella Hudson. — Sandusky Bay, Lake Erie (Kellicott, ’97). 0. intermedins Davis. — Waters connected with the Illinois River at Havana, 111. (Hempel, ’98). Lacinularia socialis Ehr. — Exhibited in New York (Balen, ’85; Damon, ’86 and ’88). Shiawassee River at Corunna, Mich. (Kellicott, ’88). Pond near Norris Pass, on the Shoshone Trail, Yellowstone Park (Forbes, ’93). Saudusky Bay, Lake Erie (Kellicott, ’96). West Twin Lake near Charlevoix, Mich. (Jennings, ’96). Much less common than Megalotroclia alboflaricans. It seems probable that the latter is often mistaken for it. MEGALOTROCHA Ehrenberg. 7. M. alboflavicans Ehrenberg. Very abundant on Chara in East Swamp, South Bass Island. Along shore of Niagara River (Mills, ’81, under name M.flavicans). Schuylkill River, Pennsylvania (Leidy, ’82, under name M. alba). Pond near Bangor, Me. (J. C. S., ’83). Exhibited in New York (Helm, ’94). Lake St. Clair; Mona Labe, Muskegon County, Mich., and Horsehead Lake, Mecosta County, Mich. (Jennings, ’94). Sandusky Bay, Lake Erie (Kellicott, ’96). Waters connected with the Illinois River at Havana, 111. (Hempel, ’98). M . semibullata Hudson. — Waters connected with Illinois River at Havana, 111. (Hempel, ’98). TROCHOSPHJERA Semper. 8. T. solstitialis Thorpe. Swamp near United States fish-hatchery, South Bass Island, in August, 1898. The genus Troclwsphcera was instituted by Semper in 1872 (Zeitschr. f. wiss. Zool., Bd. 22, p.311) for the remarkable rotifer Troclwsphcera equator! alls Semper, found by that author in the Philippine Islands. Trochosphcera equatorialis, as is well known, is a spherical rotifer, with a girdle of cilia divid- ing the surface of the sphere into two hemispheres. It approaches closely, in many respects, to the structure of the Trochophora larva of annelids and mollusks, and great importance has been attached to it as the nearest representative of the hypothetical ancestor of those groups. Trocliosplutra seems not to have been seen again until found by Surgeon V. Gunson Thorpe, of the English Navy, in 1889, at Brisbane, Australia. In 1892 the same investigator discovered in China a second species of the same genus, differing from T. equatorialis in that the ciliary girdle passes not around the middle of the sphere, but nearer one pole, like the tropic of Cancer around the earth. To this species Thorpe (’93) gave the appropriate specific name solstitialis. Much interest was aroused when in 1896 (Science, Dec. 25, 1896) Ivofoid announced the discovery of T. solstitialis Thorpe in the Illinois River and waters connected therewith near Havana, 111. Kofoid raised the question whether its presence in America was due to recent importation from China, or whether it is to be considered a native American form. Its occurrence at a station so distant from that recorded by Kofoid, on a small island in Lake Erie, seems to indicate that the latter alternative is probably correct. It is not unlikely that Trocliospliasra will be found to be widely distributed in America when proper search for it is made. The swanrp in which Troclwsphcera occurred at Put-in Bay has over its bottom a dense growth of Ceratophyllum, while the surface is completely covered with a mantle of plant material consisting of Lenina, Spirodela, and Wolffia intermixed. It is connected with the lake by a narrow short channel, and is situated at such a level that when the lake is high it receives water from the lake, while under the usual conditions water flows out of the swamp into the lake. Many of the Rotifera in the swamp are common to it and to the lake, while a number were found in the swamp alone; among the latter was Trochosphcera. The animal was never abundant, only a few individuals being obtained, and it was found for only a few days in August. Waters connected with the Illinois River at Havana, 111. (Kofoid, ’96, and Hempel, ’98). CONOCHILUS Ehrenberg. 9. C. unicornis Rousselet. Common in surface towings and plankton hauls from Lake Erie in the region of South Bass Island and from East Harbor. Lewis Lake and Yellowstone Lake in the Yellowstone Park (Forbes, ’93, under the name C. leptopus Forbes). Lake St. Clair (Jennings, ’94). Sandusky Bay, Lake Erie (Kellicott, ’97). Waters connected with the Illinois River at Havana, 111. (Hempel, ’98). C. volvox Ehrenberg.— Water from Plainfield, N. J. (Hitchcock, ’81 b). Bangor, Me. (J. C. S., ’83). Exhibited in New York (Balen, ’85, Helm, ’89, and Walker, ’94). Quincy Bay, Mississippi River, Illi- 78 BULLETIN OF THE UNITED STATES FISH COMMISSION. nois (Garman, ’90, p. 182). Lake St. Clair and Chippewa Lake, Mecosta County, Mich. (Jennings, ’94). Sandusky Bay, Lake Erie (Kellicott, ’96). West Twin Lake, near Charlevoix, Mich. (Jennings, ’96). Van Cortlandt Lake, New York City (Helm, ’97). C. dossuarms Hudson. — Sandusky Bay, Lake Erie (Kellicott, ’97). Waters connected with the Illinois River at Havana, 111. (Hempel, ’98). C. leptopus Forbes=C. unicornis Rousselet. “Apoecia amelia.” — In the Journal of the New York Microscopical Society, vol. 13, 1897, on p. 15, among the “objects exhibited” occurs the following : “Rotifer, Apoecia amelia, living, from New Jersey, by Frederick Kato.” Further on occurs the following: “Mr. Walker said of Mr. Kato’s rotifer, that it is mentioned in Trans. Acad. Nat. Sci.of Phil., and is the same as Mr. A. D. Balen’s rotifer of “pond 61,” 2^ miles from Westfield, N. J.” I have not been able to discover any reference to a rotifer with this generic or specific name elsewhere, though I myself and others have searched carefully through the publications of the Philadelphia Academy for an account of the animal. Through the kindness of Dr. Emily G. LIunt, however, I am able to present the following facts: The rotifer was discovered by Dr. W. Gibbons Hunt, of Philadelphia, about twenty years ago, on the under side of partially decayed water-lily leaves, in a lake in New Jersey. “ It was a very large and unusually beautiful rotifer, existing in colonies, the individuals of which had the tails all attached to one com- mon point and radiated out on all sides in a sphere, the whole embedded in a clear jelly.” Dr. Hunt named the rotifer Apoecia aemilia, and is supposed to have published somewhere a description of the new genus and species thus founded ; this description, if it exists, it seems impossible now to trace. From the above description it is apparent that the animal belonged to the Rhizota. Order II. BDELLOIDA. Family 3. PHILODINADiE. PHILODINA Ehrenberg. 10. P. roseola Ehrenberg. One of the commonest of the Rotifera; abundant among the plants of the bottom of Lake Erie in the region studied. This rotifer was one of the few species found in the small landlocked pools on the rocky surface of Starve Island, just south of South Bass Island. Pond near Bangor, Me. (J. C. S., ’83). Shiawassee River at Corunna, Mich. (Kellicott, ’88). Pools in the neighborhood of Cincinnati, Ohio (Turner, ’92). Lake St. Clair and the following inland lakes of Michigan: White Lake, Muskegon County; Crooked Lake, Newaygo County; Chippewa Lake, Mecosta County (Jennings, ’94). Sandusky Bay, Lake Erie (Kellicott, ’96). In old channel connecting Round Lake and Pine Lake, near Charlevoix, Mich. (Jennings, ’96). 11. P. citrina Ehrenberg. Bottom of Put-in Bay Harbor and East Harbor, Lake Erie; also in the swamp near the fish- hatchery on South Bass Island. Lake St. Clair and the following inland lakes of Michigan: White Lake, Muskegon County; McLaren Lake, Oceana County, and Crooked Lake, Newaygo County (Jennings, ’94). Sandusky Bay, Lake Erie (Kellicott, ’96). Trenton, N. J.? (Stokes, ’96«). Round Lake and swamp on the shore of Pine Lake, near Charlevoix, Mich. (Jennings, ’96). Common in pools, Hanover, N. H. (H. S. J.). 12. P. megalotrocha Ehrenberg. Common in bottom vegetation of Lake Erie in the region of South Bass Island. Also from East Harbor and the small pools on Starve Island. Pool in the neighborhood of Cincinnati, Ohio (Turner, ’92). Lake St. Clair (Jennings, ’94). Sandusky Bay, Lake Erie (Kellicott, ’96). Pine Lake, near Charlevoix, Mich. (Jennings, ’96). Waters connected with the Illinois River at Havana, 111. (Hempel, ’98). 13. P. aculeata Ehrenberg. Swamp near United States fish-hatchery, South Bass Island. Shiawassee River at Corunna, Mich. (Kellicott, ’88). Lake St. Clair and McLaren Lake, Oceana County, Mich. (Jennings, ’94). Sandusky Bay, Lake Erie (Kellicott, ’96). Trenton, N. J. (Stokes, ’96«). Tamarack swamp on shore of Pine Lake, near Charlevoix, Mich. (Jennings, ’96). Apparently the species described by Stokes (’81), as “ Philodina n. sp.?.” P. macrostijla Ehrenberg. — Lake St. Clair (Jennings, ’94). Bottom of Lake Michigan, and pool on the shore of Pine Lake at Charlevoix, Mich. (Jennings, ’96). I have lately found it in some Utri- cularia sent from Norfolk, Ya. Waters connected with the Illinois River at Havana, 111. (Hempel, ’98). ROTATORIA OF THE UNITED STATES. 79 ROTIFER Schrank. 14. R. vulgaris Schi'auk. Common among the plants of the bottom of Lake Erie about South Bass Island and in East Harbor; also from the two swamps on the island. In dirt from the crevices of pavements of Philadelphia (Leidy, ’74 &)• Summit of Roan Mountain, North Carolina (Leidy, ’80). Lake Erie (Vorce, ’82). Pond near Bangor, Me. (J. C. S., ’83). Exhib- ited in New York (Mitchell, ’86, Helm, ’89). Shiawassee River at Corunna, Mich. (Kellicott, ’88). Pools near Cincinnati, Ohio (Turner, ’92). Lake St. Clair and various inland lakes of Michigan (Jennings, ’94). Sandusky Bay, Lake Erie (Kellicott, ’96). Pine Lake, near Charlevoix, Mich. (Jennings, ’96). Waters connected with the Illinois River at Havana, 111. (Hempel, ’98). Hanover, N. H. (H, S. J.). 15. R. tardus Ehrenberg. Abundant in the swamps on South Bass Island and in East Harbor, Lake Erie ; also in Portage River, Ohio. The specimens found in this region had the spurs much shorter and thicker than are figured by Hudson and Gosse (’89) and Jauson (’93). In every other respect, however, it was exactly Ehrenberg’s B. tardus. I have since seen at Hanover, N. H., specimens having the long narrow spurs figured by Gosse and Janson. Shiawassee River at Corunna, Mich. (Kellicott, ’88). Lake St.Clair (Jennings, ’94). Sandusky Bay, Lake Erie (Kellicott, ’96). Pool on the shore of Pine Lake at Charlevoix, Mich. (Jennings, ’96). Waters connected with the Illinois River at Havana, 111. (Hempel, ’98). It.macroceros Gosse. — Shiawassee River at Corunna, Mich. (Kellicott, ’88). Lake St.Clair (Jen- nings, ’94). Sandusky Bay, Lake Erie (Kellicott, ’96). li. elongatus Weber. — Bogs, Corunna, Mich. (Kellicott, ’92). Bottom of Lake Michigan in the neighborhood of Charlevoix, Mich. (Jennings, ’96). B. trisecatus Weber.— Lake St. Clair and pools on shore of Pine Lake near Charlevoix, Mich. (Jennings, ’94 and ’96). Ii. mento Anderson.— Lake St. Clair (Jennings, ’94). B. neptunius Ehrenberg. — Pittsburg, Pa. (Mellor, ’88). Lake St. Clair (Jennings, ’94). W^aters connected with the Illinois River at Havana, 111. (Hempel, ’98). B. macrurus Schrank. — Shiawassee River "at Corunna, Mich. (Kellicott, ’88). Sandusky Bay, Lake Erie (Kellicott, ’96). Waters connected with Illinois River at Havana, 111. (Hempel, ’98). Very abundant in water from a ditch at Hanover, N. H. (H. S. J.). Callidina papillosa Thompson. Swamp on shore of Pine Lake near Charlevoix, Mich. (Jennings, ’96). C. constricta Duj. — With the last (Jennings, ’96). C. magna Plate. — With the last (Jennings, ’96). C. musculosa Milne. — Bottom of Lake Michigan near Charlevoix, Mich. (Jennings, ’96). C. eremita Bryce. — Swamp 5 miles from Norwich, Yt. (H. S. J.). C. elegans Ehrenberg. — Doubtfully reported by Kellicott (’96) from Sandusky Bay, Lake Erie. Waters connected with Illinois River at Havana, 111. (Hempel, ’98). C. socialis Kellicott. — On the larva of Pseplienus lecontei, Shiawassee River at Corunna, Mich. (Kellicott, ’88). C. rediviva Ehrenberg? — North Carolina, Bose. (See Ehrenberg, ’43.) Family 4. ADINETADiE. Adineta vaga Davis. — Tamarack swamp, on shore of Pine Lake, near Charlevoix, Mich. (Jen- nings, ’96). Order III. PLOIMA. Sub-Order ILLORICATA. Family 5. MICROCODONTIDiE. Microcodon clavus Ehrenberg.— Pond near Bangor, Me. (J. C. S., ’83). Crooked Lake, Newaygo County, Mich. (Jennings, '94). West Twin Lake, near Charlevoix, Mich. (Jennings, ’96). Microcodules orbicnlodisous Thorpe. — Lake St. Clair (Jennings, ’94). Pool on shore of Pine Lake, near Charlevoix, Mich. (Jennings, ’96). Pond in sand on shore of Sandusky Bay, Sandusky, Ohio (Kellicott, ’97, under the name M. dubius Bergendal). 80 BULLETIN OF THE UNITED STATES FISH COMMISSION. Family 6. ASFLAN CHNAD5!. ASPLANCHNA Gosse. 16. A. priodonta Gosse. In towings from Lake Erie. Not abundant. Pond in Buffalo City Park (Kellicott, ’87). Abundant in Lake St. Clair (Jennings, ’94). Lake Michigan, Iiound Lake, and Pine Lake, near Charlevoix, Mich. (Jennings, ’96). Sandusky Bay, Lake Erie (Kellicott, ’97). Waters connected with the Illinois River at Havana, 111. (Hempel, ’98). A. lierriclcii de Guerne. — It is characteristic of the poorness of the plankton in Eotifera, in Lake Erie, about South Bass Island, that Asplanchna lierriclcii was not found there at all in the summer of 1898, and that A. priodonta was not abundant. In previous examinations of Lake St. Clair and Lake Michigan both had been found very abundant. This species was first figured by Herrick (’84, plate v, fig. 8) from Minnesota, under the title “flask-shaped rotifer, hermaphrodite, with eggs and sperm.” Other localities where it has since been found in America are as follows: Lake St. Clair (Jennings, ’94); Lake Michigan, Round Lake, Pine Lake, and Susan Lake, in north Michigan (Jennings, ’96); waters connected with the Illinois River at Havana, 111. (Hempel, ’98). A. brightwellii Gosse. — Neighborhood of Cincinnati, Ohio (Turner, ’92, under the name A. cincin- natiensis Turner). Phipps Conservatory tanks at Allegheny, Pa. (Smiley, ’95). Waters connected with the Illinois River at Havana, 111. (Hempel, ’98). A. ebbesbornii Hudson. — Pond near Philadelphia (Leidy, ’87). Waters connected with the Illinois River, at Havana, 111. (Hempel, ’98). A. girodi de Guerne. — Waters connected with the Illinois River at Havana, 111. (Hempel, ’98). A. amphora Hudson. — Found at Philadelphia by Leidy, according to Hudson and Gosse, ’89 (Supplement, p. 13). A. cincinnatiensis Turner = A. brightwellii Gosse. A. magnificus Herrick = Asplanchnopus myrmeleo Ehr. ASPLANCHNOPUS De Guerne. 17. A. myrmeleo Ehrenberg. East Swamp, South Bass Island. Many. Minnesota (Herrick, ’84, under the title “ deadly enemy to Chydorus,” and ’85, under the name Asplanchna magnificus n. sp.). Pine Lake and West Twin Lake, near Charlevoix, Mich. (Jennings, '96). Marshes in the region of Sandusky, Ohio (Kellicott, ’96). Waters connected with the Illinois River at Havana, 111. (Hempel, ’98). Ascomorpha ecaudis Perty (Sacculus viridis Gosse). — Shiawassee River at Corunna, Mich. (Kelli- cott, ’88). Lake St. Clair and Whitmore Lake, near Ann Arbor, Mich. (Jennings, ’94). Round Lake, near Charlevoix, Mich. (Jennings, ’96). Waters connected with the Illinois River at Havana, 111. (Hempel, ’98). There seems to be some question as to the proper specific name of this animal. According to de Guerne (’88) Perty’s name ecaudis has the priority, dating from 1850. But Weber, ’98, in his recent very careful paper, uses the name helvetica, likewise credited to Perty, without giving the date of this name, though he cites also the name ecaudis as a synonym. Perty’s papers have not been at my com- mand in order to settle the uncertainty. A. hyalina Kellicott. — Pool at Corunna, Mich. (Kellicott, ’88). Lake St. Clair and Whitmore Lake, Mich. (Jennings, ’94). West Twin Lake, near Charlevoix, Mich. (Jennings, ’96). A. orbicularis Kellicott. — “Biemullers Cove,” Sandusky Bay, Lake Erie (Kellicott, ’97). Weber (’98) holds that this species was described from contracted examples of Gastropus stylifer Imhof ( Notops pygmceus Caiman). This appears not improbable. HERTWIGIA Plate. 18. H. parasita Ehrenberg. In Yolvox from East Swamp, South Bass Island. A rotifer parasitic in Yolvox, and therefore doubtless this species, has been recordod from Paterson, N. J. (N. N., ’75), and from Hyde Park, Chicago, 111. (Attwood, ’78). It has also been recorded by name from Sandusky Bay, Lake Erie, and Minerva Park, Columbus, Ohio (Kellicott, ’97). I have also found it in the reservoir of the town water supply of Hanover, N. H. (H. S. J.). Though often placed with the Notommatadce, the opinion expressed by many authors that this creature is more nearly related to Ascomorpha is probably correct, so that it seems best to place it here in close juxtaposition with that genus. ROTATORIA OF THE UNITED STATES. 81 Family 7. SYNCHAlTADiE. SYNCHiETA Ehrenberg. 19. S. stylata Wierzejski. Rather rare in the harbor of Put-in Bay, Lake Erie. Few in the swamp near the fish-hatchery on South Bass Island at times when the lake water has poured into the swamp. Lake St. Clair (Jennings, ’94). Lake Michigan, Round Lake, Pine Lake, and West Twin Lake, near Charlevoix, Mich. (Jennings, ’96). Sandusky Bay, Lake Erie (Kellicott, 97). Waters connected with the Illinois River at Havana, 111. (Hempel, ’98). S. pectinata Ehrenberg.— Pond near Bangor, Me. (J. C. S., ’83). Whitmore Lake, Washtenaw County, Mich. (Jennings, ’94). Sandusky Bay, Lake Erie (Kellicott, ’97). Waters connected with the Illinois River at Havana, 111. (Hempel, ’98). S. tremula Ehrenberg?— Doubtfully reported by Kellicott (’97) in marsh water in the neighbor- hood of Sandusky, Ohio. Family 8. TRIARTHRADiE. POLYARTHRA Ehrenberg. 20. P. platyptera Ehrenberg. One of the commonest of the Rotatoria. Abundant in surface and bottom towiugs and in collec- tions of plants from the bottom of Lake Erie in the region of South Bass Island, and from East Harbor. Also in the two swamps on South Bass Island. Lake Erie (Vorce, ’82). Near Minneapolis, Minn. (Herrick, ’85). Shiawassee River at Corunna, Mich. (Kellicott, ’88). Lake St.Clair; Chippewa Lake, Mecosta County, Mich.; Whitmore Lake, Washtenaw County, Mich. (Jennings, ’94). Sandusky Bay, Lake Erie (Kellicott, ’96). Lake Michi- gan, Round Lake, Pine Lake, and West Twin Lake, near Charlevoix, Mich. (Jennings, ’96). Waters connected with the Illinois River at Havana, 111. (Hempel, ’98). P. platyptera var. euryptera Wierzejski. — Waters connected with the Illinois River at Havana, 111. (Hempel, ’98). Anartlira aptera Hood.— Waters connected with the Illinois River at Havana, 111. (Hempel, ’98). TRIARTHRA Ehrenberg. 21. T. longiseta Ehrenberg. Swamp near United States fish-liatchery on South Bass Island. Few. Water from Lake Erie at Sandusky, Ohio (Kellicott, ’96). Waters connected with the Illinois River at Havana, 111. (Hempel, ’98). Pedetes saltator Gosse. — Waters connected with the Illinois River at Havana, 111. (Hempel, ’98). Family 9. HYDATINADiE. Hydatina senta Ehrenberg. — This large and interesting rotifer is said to lie common in Europe, but in all the waters which I have examined for rotifers in this country Hydatina has been conspicuously absent. But it has been reported by Kellicott (’88) from the Shiawassee River at Corunna, Mich., and by Hempel (’98) from waters connected with the Illinois River at Havana, 111. NOTOPS Hudson. 22. N. clavulatus Ehrenberg. Numerous at tiifles in the swamp near the United States fish-hatchery on South Bass Island, in company with Trochosphcera solstitialis. Also in East Swamp. There seem to be no very exact figures of the ciliary apparatus of this species; for the sake of comparison with Notops pelagicus n. sp. (see the next) I have made a study of it and present herewith a figure (plate 15, fig. 10). The corona of Notops clavulatus consists of the following parts: A circle of large cilia extends around the circumference of the head, interrupted (if at all) only at the mouth on the ventral side. The ventral half of this crown of cilia approaches a semicircle in form, but laterally there is a notch on each side, and the dorsal half is much less regular. Within this outer coronal wreath are three large styligerous prominences surrounding the buccal funnel, corresponding with the three prominences of N. pelagicus, and doubtless also with those of N. brachionus. The dorsal one of the three prominences is smaller than in N. pelagicus, and bears six long styles sheathed at the base, of the sort described and figured by Hudson and Gosse as occurring in N. brachionus. The two F. C. 13. 1899—6 82 BULLETIN OF THE UNITED STATES FISH COMMISSION. lateral prominences bear ten (or sometimes nine?) similar styles. To right and left of the dorsal median prominence are two small bundles of very slender bristles. These take the place of the two large single styles in N. pelagicus. As is well known, such large single styles are often formed of cilia or small styles united in bundles; this is probably the case in N. pelagicus, so that it is not surprising to find a large style in the one species corresponding to a bundle of small styles in the other. The buccal funnel descends directly from the lateral styligerous prominences; on its lateral walls are two elevations, each bearing five styles that project horizontally across the furrow through which the food passes. These correspond exactly with the two elevations in the buccal funnel of N. pelagicus (fig. 8) and with those described by Hudson and Gosse for N. brachionus. Thorpe (’93) has described as a new species Notops lotos, from China; the only difference between this and N. clavulatus is that the former has but three styligerous prominences on the corona, whereas Hudson says that “ N. clavulatus has a greater number of styligerous lobes” than N. brachionus, which he says has three (Hudson and Gosse, ’89, vol. II, p. 12). ' Hudson’s figure shows in N. clavulatus some six or seven or more of such lobes (vol. I, plate xv, tig. 3). But it is difficult to say how exact Hud- son’s figure was meant to be in this respect. From the figure it is not possible to say which structures represent styligerous prominences and which parts of the outer ciliary wreath, and the impression is given that the exact number and position of the prominences was not clear in the mind of the author. Such being the case, it has seemed most probable to me that my specimens are N. clavulatus, though but three large and two small ‘c styligerous prominences” exist. Moreover, the two small prominences to right and left of the median one are very easily overlooked, and it seems to me possible that they were thus overlooked by Thorpe, and that his specimens were also N. clavulatus. Some exact informa- tion as to the corona of the typical N. clavulatus of Europe would be a valuable contribution from some of the European workers in this field. It is possible that such information may show that our form should be given Thorpe’s new name Notops lotos, but I think this highly improbable. In case this should turn out to be N. lotos Thorpe, it is worthy of remark that it was found here, as Thorpe found it in China, along with Trochosphaira solstitialis. 23. N. pelagicus n. sp. (Plate 15, figs. 7, 8, and 9.) This interesting new member of the limnetic fauna of Lake Erie occurred rather sparingly in surface and bottom towings and plankton hauls from parts of Lake Erie in the neighborhood of the group of islands about Put-in Bay. On the first examination of this rotifer the generic affinities are puzzling. It has in many respects the general aspect of a Brachionus, seeming, like N. brachionus, to form a connecting link between the genera Notops and Brachionus. Its closest affinities are apparently with Notops brachionus Ehr., but with its partially loricated body it seems to resemble very closely also the Brachionus mollis of llempel(’96). Through the kindness of Dr. C. A. Kofoid, superintendent of the Illinois Biological Station, I have been able to examine the type specimens of Brachionus mollis. While the resemblance between the two is striking, Hempel’s species is clearly a Brachionus, while this is as evidently a Notops. The body is thick and clumsy, the dorsal surface rising in a regular arch from both ends to the middle (as seen in side view), the ventral surface nearly flat, but its posterior third sloping upward to join the dorsal surface (fig. 7). A ventral view shows a broad surface, widest some distance from the rear, thence narrowing suddenly backward to a blunt point (fig. 8). The coronal surface is so prone as to seem to form an almost direct continuation of the surface of the body. A short unringed foot with two inconspicuous toes completes the animal posteriorly. The integument is thickened to form a partial lorica, much as in N. hyptopus Ehr., to judge from the account given by Hudson and Gosse. In front the loricate nature of the integument is extremely evident, the dorsal edge having even four short teeth, as in species of Brachionus or Anurcea, while at the junction of the dorsal and ventral parts of the lorica there are in front two marked teeth or angles. The anterior ventral edge is nearly smooth, there being merely a rounded notch at its middle point. The corona can be partly withdrawn within the lorica, giving exactly the appearance of an Anurcea or Brachionus, with partly retracted corona — the edges of the lorica with its teeth standing out sharp and clear. Over the remainder of the animal the integument is merely stiffened, much as in some of the large species of Diglena, forming thus certain permanent folds. A pair of such folds extends backward from the head on either side, separating the lorica into dorsal and ventral portions (fig. 7). Just in front of the base of the foot a transverse fold passes across from one lateral fold to the other (fig. 8), seeming to set a posterior limit to the ventral plate. Yet the entire lorica, if it is to be so called, is pliable, not forming an unyielding shell as in the typically loricate Rotifera, and especially is the posterior region soft and yielding, so that there is no sheath or any indication of lorica about the place of attachment of the foot. Above the foot the body projects backward in a ROTATORIA OF THE UNITED STATES. 83 thick point (fig. 7), yet this is comparable in texture merely to the “'tail” of Copeus pachyurus, rather than to a projection of the lorica, such as occurs in a Brachionus. A little above the two lateral longitudinal folds above mentioned there is a broad longitudinal depression, above which the arched dorsal part of the body is much less in width than the ventral part. This depression is indicated by a strip of deeper shading along the side in fig. 7. The corona (fig. 8), is of the typical Notops character, resembling in all essential details that of N. clavulatus, just described, and in many respects very closely that of N. brachionus, as figured by Western (’90). A nearly circular outer ciliary wreath is interrupted on each side by an ear-like unciliated projection, with a deep notch in front of it; there is also a short ventral uuciliated region. Within this wreath are three curved styligerous prominences about the buccal funnel; these corre- spond in position to the three main prominences in N. clavulatus and, I should judge, to the middle dorsal and the two ventral prominences shown by Western (’90) in N. brachionus. The exact number of styles on each of these prominences was not noted, so that the figure does not attempt to be accurate on that poiut. At the side of the middle prominence, between it and the lateral prominences, are two thick styles or antennae, taking the place of the two bunches of small cilia to right and left of the middle prominence in N. clavulatus. The ciliated buccal funnel descends from within the three prominences; some distance within there are on the sides two small elevations, each with a number of stiff setae extending transversely across the buccal groove, exactly as in N. clavulatus. As previously mentioned, the coronal surface is very nearly a direct continuation of the ventral surface of the animal, so that a ventral view permits a thorough study of the corona. A thick dorsal antenna projects from a notch in the anterior dorsal margin of the lorica, exactly as in Brachionus. Lateral antennae were not observed. The foot is short and thick, and is quite without annulations. It is scarcely at all extensible^ varying little in length, so far as observed. The two toes are very inconspicuous, at times retracted, so as to be quite invisible. Each ends in a minute tube, through which at times a thick mucus is exuded, by means of which the animal adheres to objects with which it comes in contact. A broad canal can be traced from each toe to a group of small glands at the base of the foot. The trophi (fig. 9) are malleate, agreeing in all essentials with those of K. clavulatus as figured by Wierzejski (’93) and Gosse (’56). Each uncus contains five broad blunt teeth. On each side of the mastax, situated apparently iu some portion of the alimentary canal, there is — in many specimens at least — a bright red spot, the two making almost the appearance of eyes. The large brain, triangular in side view, carries at its posterior dorsal poiut the single large red eye. The other internal organs were not studied. The egg is carried by the mother, attached just above the base of the foot, in exactly the position in which a Brachionus carries its eggs. Notops pelagicus feeds upon the unicellular algae which float in the clear waters of the lake and form the primary food supply of almost all the water organisms. Thus, if we consider the organisms of the lake as forming a chain, of which these unicellular algae, deriving their sustenance directly from the inorganic constituents of the water, are the first link, while the highest carnivorous fish are the last, this rotifer forms a part of the second link, standing in relations of dependency only to the primal source of food supply. Notops pelagicus is noteworthy for its bearing upon the classification of the Rotatoria. It seems to belong unquestionably to Notops, and to be more closely related to the soft-bodied members of that genus; yet it has an evident partial lorica. In spite of this lorica, it clearly does not belong at all with those loricate members of the (former) genus Notops that have recently been separated off by Weber (’98) as Gastropus. Its relations are not with Gastropus stylifer, G. minor, and G. hyptopus, but with Notops clavulatus and N. brachionus ; at the same time, it is evidently related to the species of Brachionus. I believe with Lund (’99) that the Hydatinadre are to be grouped naturally with the Brachionidae, and that the softness or stiffness of the cuticula (upon which depends whether the animal is called loricate or illoricate) is a character of little significance in classification. It is to be noted that in the two important papers that have appeared most recently on the Rotatoria, the species of the genus Notops have been divided in the same manner, but that the name Notops has been left with a different division in each case. Both Weber (’98) and Lund (’99) separate Notops clavulatus and N. brachionus on the one hand from N. hyptopus, N. minor, and N. stylifer ( pygmceus ) on the other. But while Weber leaves the name Notops to the former group, calling the others Gas- tropus, Lund gives the name Notops to the hyptopus group, relegating the others to Hydatina. I have followed Weber, for reasons given under the discussion of the genus Gastropus. N. minor Rousselet; N. pygmceus Caiman. — See Gastropus, under Loricata. N. laurentinus Jennings. — See Proales laurentinus Jennings. 84 BULLETIN OF THE UNITED STATES FISH COMMISSION. TRIPHYLUS Hudson. 24. T. lacustris Ehrenberg. This rare and interesting rotifer occurred abundantly in East Swamp, South Bass Island, both the male and the female being found. Western (’92) gives a figure of the male of this species. CYRTONIA Rousselet 25. C. tuba Ehrenberg. East Swamp, South Bass Island, abundant. Trenton, N. J. (Stokes, ’97, under name Proales hyalina n. sp. ). Family 10. NOTOMMATADJE Albertia naidis Bousfield. Lake St. Clair (Jennings, ’91). It seems possible that the Anelcodiscus pellucidus described by Leidy (’51), from the intestine of Stxjlaria fossularis Leidy, in the neighborhood of Philadelphia, may have been a rotifer of this genus. TAPHROCAMPA Gosse. 26. T. annulosa Gosse. (Plate 14, figs. 4, 5, and 6.) Swampy parts of East Harbor, Lake Erie; common. There is so much characteristic detail about the form and structure of this animal that is not brought out in the published figures, that I have thought it worth while to give some camera figures of specimens killed in extension. Fig. 4 gives a side view. The animal is here represented as curved more than other published figures show it, but in my experience this is about the form the living specimen usually has when moving along the bottom. On account of the fact that it is so curved, the entil e body can not well be shown in a single dorsal view. Fig. 6 gives a dorsal view of the anterior three-fourths of body, while fig. 5 gives a corresponding view of the posterior three-fourths, showing the toes, with the broad tail above them. In regard to the internal anatomy, it needs to be said that the intestine does not open in the broad dorsal de2)ression near the posterior end of the body, as Weber (’98) has represented it, but the opening lies just above the toes, beneath the tail. Mr. Gosse’s statement in the monograph that the opening of the intestine is beneath the two toes is equally incorrect. This is perfectly clear in mounted specimens. Shiawassee River at Corunna, Mich. (Kellicott, ’88). Lake St. Clair, and the following inland lakes of Michigan: McLaren Lake, Oceana County; Crooked Lake, Newaygo County; Chippewa Lake, Mecosta County (Jennings, ’94). Channel between Round and Pine Lakes, near Charlevoix, Mich. (Jennings, ’96). Waters connected with the Illinois River at Havana, 111. (Hempel, ’98). 27. T. saundersias Gosse. Portage River, Ohio, among Utricularia. Shiawassee River at Corunna, Mich. (Kellicott, ’88). Lake St. Clair (Jennings, ’94). 28. T. selenura Gosse. East Harbor, Lake Erie and swamp near fish-hatchery on South Bass Island. Lake St. Clair (Jennings, ’94). Trenton, N. J. (Stokes, 96«). T. clavigera Stokes. — Trenton, N. J. (Stokes, ’96 b). PLEUROTROCHA Ehr. 29. F. parasitica n. sp. (Plate 16, figs. 13 and 14.) Parasitic on the annelid Xais lacustris, from among plants of the bottom of Lake Erie about South Bass Island. In waters connected with the Great Lakes I have several times noticed a Pleurotroclia attached by its jaws to tbe external surface of the small annelid Mats lacustris. The first one observed I thought I could identify with Ehrenberg’s P. constricta (Jennings, ’94, p. 14), but I have since been able to make a more careful study, with the result of showing that this identification is wrong, the body being much too short and broad. It resembles more nearly P. gibba Ehr., yet is clearly distin- guished from that species by the much greater size of the toes, as well as by the totally different form of the body and head. It resembles no other of the recently described species of this genus, so that it is necessary to describe it as a new species. Ventral and side views of the animal are shown in plate 16, figs. 13 and 14. ROTATORIA OF THE UNITED STATES. 85 Body very short and broad, oval in dorsal or ventral view. Head much narrower than body, tapering to the obliquely truncate corona, composed of a single wreath of cilia. Body truncate behind; from the lower side of the truncate surface rises the single joint forming the short foot. The two tapering toes are about as long as the body is thick at the posterior end; they stand some dis- tance apart at base. The internal anatomy offers nothing especially noticeable save the lack of an eye, which is of course the character that places this form in the genus Pleurotrocha. Length, 110/4. Animal ectoparasitic on the annelid Nais lacustris. Pool near Lake St. Clair (Jennings, ’94, under the name Pleurotrocha constricta Ehr. ). NOTOMMATA Gosse. 30. N. aurita Ehrenberg. Rare, among plants on bottom of Put-in Bay Harbor, Lake Erie. Lake St. Clair (Jennings, ’94). Sandusky Bay, Lake Erie (Kellicott, ’96). Waters connected with the Illinois River at Havana, 111. (Hempel, ’98). 31. N. tripus Ehrenberg. East Harbor, Lake Erie, in Utricularia. Shiawassee River at Corunna, Mich. (Kellicott, ’88). Lake St. Clair and White Lake, Muskegon County, Mich. (Jennings, ’94). Trenton, N. J. (Stokes, ’966, under name N. mirabilis n. sp.). Waters connected with the Illinois River at Havana, 111. (Hempel, ’98). Brook, Hanover, N. H. (PI. S. J.). 32. N. truncata Jennings. In Naias and Chara, bottom of Put-in Bay Harbor, Lake Erie. Lake St. Clair (Jennings, ’94). This species has recently been found by Stenroos (’98) in Finland. W. 6 racliyota Ehrenberg. — Lake St. Clair (Jennings, ’94). N. collaris Ehrenberg. — Lake St. Clair (Jennings, ’94). iV. torulosa Duj. — Lake St. Clair and Chippewa Lake, Mecosta County, Mich. (.Jennings, ’94). JV. monopus Jennings. — Lake St. Clair (Jennings, ’94). Lake Michigan, Round Lake, Pine Lake, West Twin Lake, and Susan Lake, near Charlevoix, Mich. (Jennings, ’96). N. vorax Stokes. — Trenton, N. J. (Stokes, ’97). Sandusky Bay, Lake Erie (Kellicott, ’97). W. cyrtopus Gosse. — Waters connected with the Illinois River at Havana, 111. (Hempel, ’98). N. mirabilis Stokes. —N. tripus Ehr. N. lacinulaia Ehrenberg., see Diaschiza lacinulata Ehrenberg. COPEUS Gosse. 33. C. pacliyurus Gosse. Bottom of shallow parts of Lako Erie near Put-in Bay. Tamarack swamp on the shore of Pine Lake, near Charlevoix, Mich. (Jennings, ’96). Reservoir of the water supply, Hanover, N. H. (H. S. J.). C. labiatus Gosse. — Shiawassee River at Corunna, Mich. (Kellicott, ’88). Lake St. Clair (Jennings, '94). Pine Lake, near Charlevoix, Mich. (Jennings, ’96). Brook, Hanover, N. H. (H. S. J.). C. ehrenbergii Gosse. — Sandusky Bay, Lake Erie (Kellicott, ’97). C. cerberus Gosse. — Lake St.Clair and the following inland lakes of Michigan: McLaren Lake, Oceana County; Crooked Lake, Newaygo County; Chippewa Lake, Mecosta County (Jennings, ’94). Sandusky Bay, Lake Erie (Kellicott, ’96). C. quinquelobatus Stokes. — Trenton, N. J. (Stokes, 96c). I have found a Copeus at Hanover, N. H , having five lobes to the brain, but agreeing in every other particular with C. pacliyurus ; this also seems true of Stokes’s 0. quinquelobatus. But the specimens at Hanover occurred along with typical specimens of 0. pacliyurus, having but three lobes to the brain ; moreover, the specimens with five lobes were a little larger than those with three lobes. I can not but think it probable that C. pacliyurus develops another pair of lateral lobes on the brain as it becomes larger, and that the species quinquelo- batus is founded on such specimens. This is rendered the more probable by the fact that in the speci- mens seen the development of the two lateral lobes varied greatly. C. americanus Pell. — Locality not given by describer, but probably Highland Falls, N. Y. (Pell, ’90). PROALES Gosse. 34. P. sordida Gosse. Bottom of Put-in Bay Harbor, Lake Erie. Lake St. Clair (Jennings, ’94). Sandusky Bay, Lake Erie (Kellicott, ’96). P. fells Ehrenberg. — Lake St. Clair (Jennings, ’94). 86 BULLETIN OF THE UNITED STATES FISH COMMISSION. P. werneckii Ehrenberg. — In Vaucheria from ponds along Paxton Creek, Harrisburg, Pa. (Wolle, ’82 and ’87). Lake St. Clair; rivulet at Ann Arbor, Mich. (Jennings, ’94). P. laurentinus Jennings. — Lake St. Clair (Jennings, ’94, as Notops Jaurentinus). Channel between Round Lake and Pine Lake, and in West Twin Lake, near Charlevoix, Mich. (Jennings, ’96). P. decipiens Elirenberg. — Sandusky Bay, Lake Erie (Kellicott, ’96.) P. gibba Elirenberg. — Sandusky Bay, Lake Erie (Kellicott, ’96). P. algicola Kellicott. — Sandusky Bay, Lake Erie (Kellicott, ’97). P. hyalina Stokes = Cyrtonia tuba Ehr. FURCULARIA Ehrenberg. 35. F. forficula Ehrenberg. ( F . trihamata Stenroos, ’98). Very abundant in bottom and littoral vegetation of shallow parts of Lake Erie about South Bass Island. Shiawassee River at Corunna, Mich. (Kellicott, ’88). Lake St. Clair and the following inland lakes of Michigan : McLaren Lake, Oceana County; Chippewa Lake, Mecosta County ; Round Lake, Mecosta County (Jennings, ’94). Pine Lake, near Charlevoix, Mich. (Jennings, ’96). Waters con- nected with the Illinois River at Havana, 111. (Hempel, ’98). 36. F. longiseta Ehrenberg. Common among water plants in East Harbor, Lake Erie. Shiawassee River at Corunna, Mich. (Kellicott, ’88). Lake St. Clair and the following inland lakes of Michigan: White Lake, Muskegon County; McLaren Lake, Oceana County; Crooked Lake, Newaygo County; Chippewa Lake, Mecosta County (Jennings, ’94). Sandusky Bay, Lake Erie (Kel- licott, ’96). Pool on the shore of Pine Lake ; West Twin Lake, near Charlevoix, Mich. (Jennings, ’96). Waters connected with the Illinois River at Havana, 111. (Hempel, ’98). Swamp near Norwich, Vt. ; pond, Hanover, N. H. (II. S. J.). 37. F. senrisetifera Glasscott. In Cliara from East Swamp, South Bass Island. Hood (’95 ) holds that this species is identical with Furcularia eva of Gosse. My specimens did not have the large anterior dorsal hump which Gosse mentions in his description and figures prominently, so that I feel it necessary to accept Miss Glasscott’s name, the specimens agreeing with her figures. Pool on the shore of Pine Lake near Charlevoix, Mich. (Jennings, ’96). F. gracilis Ehrenberg. — Shiawassee River at Corunna, Mich. (Kellicott, ’88). Lake St. Clair (Jen- nings, ’94). Pool on the shore of Pine Lake near Charlevoix, Mich. (Jennings, ’96). F. gibba Ehrenberg. — Lake St. Clair; Chippewa Lake, Mecosta County, Mich. (Jennings, ’94). F. micropus Gosse. — Pool on the shore of Pine Lake, near Charlevoix, Mich. (Jennings, ’96). Trioplithalmus dorsualis Ehrenberg. — Round Lake and Pine Lake, near Charlevoix, Mich. (Jen- nings, ’96). EOSPHORA Ehr. < 38. E. aurita Ehrenberg. East Swamp, South Bass Island; Portage River, Ohio. Lake St. Clair (Jennings, ’94). Round Lake, Charlevoix, Mich. (Jennings, ’96). Sandusky Bay, Lake Erie (Kellicott, ’97). Waters connected with the Illinois River at Havana, 111. (Hempel, ’98). DIGLENA Ehr. 39. D. grandis Gosse. Bottom vegetation of Lake Erie about South Bass Island and East Harbor. Lake St. Clair (Jennings, ’94). Old Channel and West Twin Lake, near Charlevoix, Mich. (Jen- nings, ’96). Waters connected with the Illinois River at Havana, 111. (Hempel, ’98). 40. D. forcipata Ehrenberg. In Naias from bottom of Put-in Bay Harbor and East Harbor, Lake Erie. Lake St. Clair, and Chippewa Lake, Mecosta County, Mich. ; Crooked Lake, Newaygo County, Mich. (Jennings, ’94). Old Channel, Charlevoix, Mich. (Jennings, ’96). Sandusky,Bay, Lake Erie (Kellicott, ’97). 41. D. catellina Ehrenberg. Bottom of Put-in Bay Harbor, Lake Erie; land-locked pools on Starve Island. Round Lake, Charlevoix, Mich. (Jennings, ’96). Waters connected with the Illinois River at Havana, 111. (Hempel, ’98). ROTATORIA OF THE UNITED STATES. 87 42. D. biraphis Gosse. Swamp near fisli-hatcliery, South Bass Island. Lake St. Clair, and Chippewa Lake, Mecosta County, Mich. (Jennings, ’94). Waters connected with the Illinois River at Havana, 111. (Hempel, ’98). D.circinator Gosse. — Chippewa Lake, Mecosta County, Mich. (Jennings, ’94). D. caudata Ehrenberg. — McLaren Lake, Oceana County, Mich. (Jennings, ’94). D. contorta Stokes. — Trenton, N. J. (Stokes, ’97). Distemma forficula Ehrenberg. — Pond near Bangor, Me. (J. C. S., ’83). Suborder LORICATA. Family 11. RATTULIDiE. There is much confusion in regard to the identification of the species belonging to this family, so that I have thought it best to give figures of the species listed, so far as possible. The group is badly in need of a thorough revision. MASTIGOCERCA Ehrenberg. 43. M. bicornis Ehrenberg. (Plate 17, fig. 15). East Harbor, Lake Erie. Pond near Bangor, Me. (,T. C. S., ’83). Lake St. Clair, and Chippewa Lake, Mecosta County, Mich. (Jennings, ’94). Round Lake, and Pine Lake. Charlevoix, Mich. (Jennings, ’96). Sandusky Bay, Lake Erie (Kellicott, ’96). Waters connected with the Illinois River at Havana, 111. (Hempel, ’98). Pools about Hanover, N. H. (H. S. J.). 44. M. carinata Ehrenberg. (Plate 18, fig. 19.) East Harbor, Lake Erie; swamp near fish-hatchery on South Bass Island; Portage River, Ohio. Pond near Bangor, Me. (J. C. S., ’83). Shiawassee River, at Corunna, Mich. (Kellicott, ’88). Lake St. Clair and the following inland lakes of Michigan : West Twin Lake, Muskegon County; Crooked Lake, Newaygo .County, and Chippewa Lake, Mecosta County (Jennings, ’94). Lake Michigan, Round Lake, and Pine Lake, near Charlevoix, Mich. (Jennings, ’96). Sandusky Bay, Lake Erie (Kellicott, ’96). Waters connected with the Illinois River at Havana, 111. (Hempel, ’98). Common at Hanover, N. II. (H. S. J.). 45. M. elongata Gosse. (Plate 17, fig. 16.) In Utricularia from Portage River, Ohio. Sandusky Bay, Lake Erie (Kellicott, ’97). Waters connected with the Illinois River at Havana, 111. (Hempel, ’98). Pools, Hanover, N. H. (H. S. J.). 46. M. bicuspes Pell. (Plate 16, figs. 11 and 12.) Iu Utricularia from East Harbor, Lake Erie. This form has recently been redescribed by Stokes (’97) as M. spinigera n. sp. As the description and figure of Pell (’90) seem not well known, I give figures of dorsal and lateral views. Like M. lata, this species has five sensory projections on the corona, as shown in the figures. Highland Falls, N. Y. ? (Pell, '90, locality not stated). Trenton, N. J. (Stokes, ’97, under name M. spinigera n. sp.). 47. M. mucosa Stokes. (Plate 17, fig. 18.) One of the most abundant of the Rotifera among the vegetation of the shallow parts of Lake Erie about South Bass Island. This is the two-keeled species mentioned without identification in my first paper on the Rotifera (’94), as being abundant in various lakes ; it has since been described by Stokes under the above name. It differs from M. bicristata Gosse (fig. 17) in its shorter thicker body, and in the fact that the two keels extend only about one-half the length of the body. Lake St. Clair, Chippewa Lake, Mecosta County, Mich., and Crooked Lake, Newaygo County, Mich. (Jennings, ’94, as “form with two large dorsal keels”). Round Lake and Old Channel, Charlevoix, Mich. (Jennings, ’96, unnamed, p. 91). Trenton, N. J. (Stokes, ’96 b). Pond, Hanover, N. H. (H. S. J.). M. bicristata Gosse. (Plate 17, fig. 17.) West Twin Lake near Charlevoix, Mich. (Jennings, ’96) ; Sandusky Bay, Lake Erie (Kellicott, ’97 ; possibly this was M. mucosa Stokes). Waters connected with the Illinois River near Havana, 111. (Hempel, ’98). M. capucina Wierz. and Zach. — Lake St. Clair (Jennings, ’94). West Twin Lake near Charlevoix, Mich. (Jennings, ’96). 83 BULLETIN OF THE UNITED STATES FISH COMMISSION. M. lata Jennings. — Lake St. Clair ( Jennings ’94). West Twin Lake near Charlevoix, Mich. (Jennings, ’96). Sandusky Bay, Lake Erie (Kellicott, ’96). Waters connected with the Illinois Kiver at Havana, 111. (Kerapel, ’98). This species has recently been found also by Stenroos (’98) in Finland. M. rattus Ehr. — New York (Ehrenberg, ’43). Near Minneapolis, Minn. (J. W., ’83). Near Cincinnati, Ohio (Turner, ’92). (Possibly the same thing was seen by Herrick (’85), who speaks of a rotifer resembling Monocerca rattus.) Sandusky Bay, Lake Erie (Kellicott, ’97). M. multicrinis Kellicott. — Sandusky Bay, Lake Erie (Kellicott, ’97). M. spinigera Stokes = M. bicuspes Pell. KATTULUS Ehrenberg. 48. R. tigris Muller. (Plate 18, figs. 20 and 21.) In Naias, Put-iu Bay Harbor, Lake Erie. The animal described and figured by Gosse in the Monograph under the above name is apparently not Ebrenberg’s species at all ; my specimens seem to agree with those of Ehrenberg. Characteristic seems to be the curved body, not enlarged in front as is figured by Gosse, hut tapering gradually from about the middle to the foot; also the single large tooth at the anterior margin of the lorica. The anterior part of the lorica has about nine longitudinal folds, extending from the anterior margin to the constriction separating that part of the lorica covering the head from that covering the body. At the base of each of the two main toes are four minute substyles (fig. 21). Pond near Bangor, Me. (J. C. S., ’83). Turner (’92) records “liattulus tigris ” from the neighbor- hood of Cincinnati, Ohio, citing liattulus tigris of Hudson and Gosse and Diurella tigris of Herrick (’85) for accounts of the animal. Now, these two latter represent two entirely different auimals, Herrick’s animal being Ccelopus porcellus, while, as noted above, the Rattulus described by Gosse is not the real liattulus tigris. It is therefore impossible to say what the animal observed by Turner was. 49. R. sulcatus Jennings. Not uncommon in shallow parts of Lake Erie about South Bass Island. Lake St. Clair (Jennings, ’94). Old Channel and West Twin Lake near Charlevoix, Mich. (Jennings, ’96). Sandusky Bay, Lake Erie (Kellicott, ’96). li. palpitatus Stokes (— Ccelopus bracliyurus Gossef). — Trenton, N. J. (Stokes, '96b). “ Diurella tigris Bory,” Herrick (’85) = Coelopus porcellus Gosse. Diurella insignis Herrick (’85). See Coelopus tenuior. CCELOPUS Gosse. As has been several times pointed out of late, this is a genus which was founded on an incorrect interpretation of the structure of the toes. When the Rattulid® are subjected to the revision which they so much need, probably the name Coelopus will disappear; until that is done it will be best to retain the names commonly used. 50. C. porcellus Gosse. (Plate 18, figs. 22 and 23.) Not uncommon in the vegetation of shallow parts of Lake Erie about South Bass Island. Ohio and Minnesota (Herrick, ’85, under the name Diurella tigris Bory). Shiawassee River at Corunna, Mich. (Kellicott, ’88). Lake St. Clair, Crooked Lake, Newaygo County, Mich. ( Jennings, ’94). Old Channel, Charlevoix, Mich. (Jenuings, ’96). Sandusky Bay, Lake Erie (Kellicott, ’96). Waters connected with the Illinois River at Havana, 111. (Hempel, ’98). Pools, Hanover, N. H. (H. S. J.). 51. C. bracliyurus Gosse. (Plate 18, fig. 24.) East Harbor, Lake Erie; swamp near fish-hatchery on South Bass Island. Shiawassee River at Corunna, Mich. (Kellicott, ’88). Pools, Hanover, N. H. (H. S. J.). C. tenuior Gosse. Doubtfully reported by Kellicott (’88) from the Shiawassee River at Corunna, Mich. Old Channel, Charlevoix, Mich. (Jennings ’96). Sandusky Bay, Lake Erie (Kellicott, ’96). Waters connected with the Illinois River at Havana, 111. (Hempel ’98). Diurella insignis Herrick (’85) apparently should he referred to this species. Weber (’98) refers it to Coelopus porcellus, yet an inspection of Herrick’s figure shows that the proportions are totally different from those of the latter species, while they agree fairly well with those of C. tenuior; moreover, Herrick had already described C. porcellus on the preceding page of his paper, under the name Diurella tigris. Diurella insignis ( C . tenuior) was found in Minnesota. Helerognatlius nolommata Schmarda (— Coelopus tenuior?). Brackish water near New Orleans (Schmarda, ’59), ROTATORIA OF THE UNITED STATES. 89 Family 12. DINOCHARIDJE. DINOCHARIS Ehrenberg. 52. D. pocillum Ehrenberg. East Harbor, Lake Erie, in bottom vegetation. Minneapolis, Minn. (J. W., ’83). Pond near Bangor, Me. (J. C. S., ’83). Minnesota (Herrick, ’85). Shiawassee River, at Corunna, Mich. (Kellicott, ’88). Lake St. Clair and the following inland lakes of Michigan : McLaren Lake, Oceana County ; Crooked Lake, Newaygo County ; Chippewa Lake, Mecosta County (Jennings, ’94). Round Lake, Charlevoix, Mich. (Jennings, ’96). Sandusky Bay, Lake Erie (Kellicott, ’97). Waters connected with the Illinois River at Havana, 111. (Hempel, ’98). 53. D. tetractis Ehrenberg. Bottom vegetation of Put-in Bay Harbor and East Harbor, Lake Erie; more numerous than the last. Herrick (’85, p. 52) mentions as occurring in Minnesota a species of Dinocharis resembling I). pocillum, but lacking the spine on last joint of the foot ; this was evidently D. tetractis. Shiawassee River, at Corunna, Mich. (Kellicott, ’88). Lake St. Clair and the following inland lakes of Michigan : Crooked Lake, Newaygo County ; Chippewa Lake, Mecosta County (Jennings, ’94). West Twin Lake, and pool on the shore of Pine Lake, near Charlevoix, Mich. (Jennings, ’96). Hanover, N. II. (II. S. .J.). POLYCHIETUS Perty. 54. P. subquadratus Perty. Bottom vegetation, Put-in Bay Harbor and East Harbor, Lake Erie. Lake St. Clair (Jennings, ’94). Old Channel, Charlevoix, Mich. (Jennings, ’96). Sandusky Bay, Lake Erie (Kellicott, ’97). 55. P. collinsii Gosse. In Myriophyllum from East Harbor, Lake Erie; in TJtricularia from Portage River, Ohio, not far from the mouth. Pools and Old Channel, Charlevoix, Mich. (Jennings, ’96). P. serica Thorpe. — Sandusky Bay, Lake Erie (Kellicott, ’97). SCARIDIUM Ehrenberg. 56. S. longicaudatum Ehrenberg. Very abundant in shallow parts of Lake Erie about South Bass Island and in East Harbor. Shiawassee River, at Corunna, Mich. (Kellicott, ’88). Near Cincinnati, Ohio (Turner, ’92). Lake St. Clair and the following inland lakes of Michigan: McLaren Lake, Oceana County; Crooked Lake, Newaygo County; Chippewa Lake, Mecosta County (Jennings, ’94). Pine Lake and Old Channel, Charlevoix, Mick. (Jennings, ’96). Sandusky Bay, LakeErie (Kellicott, ’96). Trenton, N. J. ( ?) (Stokes, ’96«). Waters connected with the Illinois River at Havana, 111. (Hempel, ’98). S. eudactylotum Gosse. — This animal is represented in fig. 1, plate IV, of Herrick, ’85, under the title “undetermined.” It was thus evidently found by Herrick somewhere in America; no locality is given. Chippewa Lake, Mecosta County, Mich. (Jennings, ’94). - Stephanops muticus Ehrenberg. — Said by Herrick (’85) to occur somewhere in America. Lake St. Clair; Chippewa Lake, Mecosta County, Mich. (Jennings, ’94). Sandusky Bay, Lake Erie (Kellicott, ’97). S. lamellaris Ehrenberg. — Minneapolis, Minn. (J. W. ’83). Shiawassee River, at Corunna, Mich. (Kellicott, ’88). Sandusky Bay, Lake Erie (Kellicott, ’96). S. (Mama Gosse. — Sandusky Bay, Lake Erie (Kellicott, ’96). Family 13. SALPINADiE. DIASCHIZA Gosse. 57. D. semiaperta Gosse. Abundant in bottom vegetation of Put- in Bay Harbor and East Harbor, Lake Erie. Lake St. Clair (Jennings, ’94). Round Lake and pools, Charlevoix, Mich. (Jennings, ’96). 58. D. lacinulata Ehrenberg. ( Notommata lacinulata.) Common in vegetation of bottom of shallow parts of Lake Erie about South Bass Island. Shiawassee River, at Corunna, Mich. (Kellicott, ’88). Abundant in Michigan lakes (Jennings, ’94). 90 BULLETIN OF THE UNITED STATES FISH COMMISSION. Round Lake and West Twin Lake, Charlevoix, Mich. (Jennings, ’96). Sandusky Bay, Lake Erie (Kellicott, ’96). Waters connected with the Illinois River at Havana, 111. (Hempel, ’98). Pond, Han- over, N. II. (H.S.J.). DIPLAX Gosse. 59. D. trigona Gosse. Rather common at times in the swamp near the fish-hatchery on South Bass Island. SALPINA Ehrenberg. 60. S. brevispina Ehrenberg. Bottom vegetation, East Harbor, Lake Erie; swamp near fish-hatchery, South Bass Island. Shiawassee River, at Corunna, Mich. (Kellicott, ’88). Near Cincinnati, Ohio (Turner, ’92). Lake St. Clair and the following inland lakes of Michigan: McLaren Lake, Oceana County; Crooked Lake, Newaygo County; Chippewa Lake, Mecosta County (Jennings, ’94). Sandusky Bay, Lake Erie (Kellicott, ’96). 61. S. macracantha Gosse. In swamps on South Bass Island; in Utricularia from Portage River, Ohio. S.ventralis Ehr. — Lake St. Clair; Chippewa Lake, Mecosta County, Mich. (Jennings, ’94). San- dusky Bay, Lake Erie (Kellicott, ’96). 5. mucronata Ehr. — Near Cincinnati, Ohio (Turner, ’92). S. eustala Gosse. — Waters connected with the Illinois River at Havana, 111. (Hempel, ’98). S. macrocera Jennings. — Chippewa Lake, Mecosta County, Mich. (Jennings, ’94). S. similis Stokes ( =S . macracantha Gossef). — Trenton, N. J. (Stokes, ’966). S. affinis Herrick ( =