ve Cee ee pw vr fy mi pee a inet base rence a rn meee eat Ag Ma ititest : nh vena ; Seer eek neta ete pea ae * ¢ * - nw 4 : ae yeapty Se i ee ; if ‘ a eb AN Wad aie ed pris tes t } at ee i fit oles f 3 5 aot he vaskats RNR tS Neen) ti pai naitetsshge wa tetd a, tie vi . bnernt Ml Ath wh i ne att A i nt sty Fs nies | iM oS Mae i: ee: ee S “np, Oy i" “pi 2, UA iil vo : i W oN Ay es eo K ale ‘ MW SS eis mE? Ons” ae Ke wr, a : ’ i" y . on i" Pa : tb, i, J ay < ¥ 4 ae ; e a a be > o, oy ; “ = ase ES | ao wr. “an” i: os “th G Pee ib Lene, ria Nts, Sf, WER Pa vanes Hon The yo NEG SY fl ‘ % ae y eS Oh We, Wr fy SA , e SE D> oe ae : ies Sy, % S y te hes wy in of S, 1 | | | 4 Me ab > “ay & p27 ANH REN Sor i Gp cS y, SY | Hy Hi y i f 3 | , | / Gi : Ze ——— O, gM: q ia | i iD ao. i cs ? > a seit HH ray all i | Y 4 | ye %, LW Scat el Oo LES ° As if a 1 i M t iron XY eee & EE © — : : 4 pene | | | t S lt alll ih lll Zee | Ht | Kerr | is ei ; Wf ‘Ss = mit / en Whhy === -_ ay SS 2 —————— == < “ we I ao S S yh is es ie 5 if ' ald 7 vopGt Dy U. S. NATIONAL MUSEUM LIBRARY OF Henry Guernsey Hubbard | AND Eugene Amandus Schwarz DONATED IN 1902 ACCESSION NO.. NATIONAL AGADEMY OF SCIENCES. NBO LV FIRST MEMOIR. 3 - CAVE FAUNA OF NORTH AMERICA, WITH REMARKS ON THE ANAT- U. S. NATIONAL MUSEUM LIBRARY OF Henry Guernsey Hubbard Eugene Amandus Schwarz 5 4 DONATED IN 1902 ACCESSION NO... on, THE GAVE FAUNA OF NORTH AMERICA, WITH REMARKS ON THE ANAT- OMY OF THE BRAIN AND ORIGIN OF THE BLIND SPECIES, - Lew 2 ln a tp args 0 By A. SJ Packarp.* CONTENTS. I. Description of the caves, with notes on their hydrog- Y. Geographical distribution of the cave species. raphy, temperature, their origin, and geological VI. List of American and European cave animals. age; the source of the food supply of their inhabit- VII. List of blind non-cavernicolous animals, including ants; the probable mode of colonization; with lists deep-sea forms. : of the species inhabiting the better-known caves. | VIII. Anatomy of the brain and rudimentary eyes (when Il. The vegetable life of the caves. present) of certain blind Arthropoda. Ill. Systematic description of the invertebrate animals. IX. The origin.of the cave species and genera. IV. Systematic list of the cave animals of North America. X. Bibliography relating to caves and to cave life. PREFACE. Most of the materials on which this essay is based were obtained while connected temporarily with the geological survey of Kentucky, in charge of Prof. N. S. Shaler. During portions of the months of April and May, 1874, I examined Mammoth Cave and several adjoining, i. ¢., White’s Cave, Dixon’s Cave, Diamond Cave, and Proctor’s Cave, in company with. Professor Shaler and Mr. F. G. Sanborn, assistant on the survey, and subsequently Mr. Sanborn éxplored these, Carter, and many smaller caves. In company with Professor Shaler, I also made a slight examination of the four Cartercaves. Fully appreciating the importance of the subject of cavern life and of comparing the fauna of different caves, Professor Shaler invited me to visit Wyan- dotte Cave, and the Bradford caves in Indiana. The Bradford caves I visited in company with Dr. John Sloan, of New Albany, Indiana, who had already examined with much success many of the small caverns in southern Indiana. The collections made by him, and contained in the Museum of Natural History of New Albany, were also examined, and he has kindly sent me other material. I have also received specimens and notes from Dr. Moses.N. Elrod of Orleans, Orange County, Indiana, a region abounding in small caves. On my. return I examined Weyer’s Cave and adjoin- ing Madison’s Cave near Staunton, Virginia, and discovered about twelve forms, where before none were known to inhabit those caves. In the autumn Professor Putnam made a thorough explora- tion of Mammoth Cave. This paper is accordingly based on material collected by Mr. Sanborn, Professor Shaler, Mr. Cooke, Professor Putnam, Dr. Sloan, and myself. Mr. Emerton kindly identified and described the spiders of the caves, and his descriptions and drawings accompany this article. The Coleoptera have been identified by Dr. Le Conte, the Dip- * Read November 9, 1256, 4 MEMOIRS OF THE NATIONAL ACADEMY OF SCIENCES. tera by Baron Osten Sacken, and the only Psocide found have been described, so far as they could be, by Dr. Hagen. Tn 1880 I visited the New Market and Luray caverns. Mr. H. G. Hubbard has kindly loaned me many of the specimens which he collected in Mammoth Cave. The entire collection of cave insects, excepting some of the duplicates, embracing the types obtained by Mr. Sanborn and myself while attached to the Kentucky geological survey, have been placed in the Museum of Comparative Zoology at Cambridge, Massachusetts. I. DESCRIPTION OF THE BETTER-KNOWN CAVES. MAMMOTH CAVE. As Mammoth Cave is the largest and most frequented, and was the first known to contain eye- less animals, we will first briefly describe this great cavern, simply dwelling on those points which are of interest from a biological point of view. One can form little idea of the general geological relations of this cave from a few visits, especially when busied with the search for cave animals; and we are indebted to the Kentucky geological reports, containing accounts by Professors Owen and Shaler, also to an excellent paper by Mr. W. Le Conte-Stevens,! and the carefully-prepared work of Rev. H.C. Hovey on “Celebrated American Caverns,” who has given the results of much time spent in exploration, and has taken, with more care than any cne else, the temperatures of this! and other caves. By consulting Mr. Hovey’s map of the cave? and reading Mr. Stevens’s condensed account, aided by his sketches, we can obtain a fair idea of the topography of the cave, embracing the dry and damp portions; 7. ¢., those portions not deserted, and those most frequented by the animals of the caves. Mammoth Cave is situated in latitude 37° 14’ N., and longitude 86° 12’ W., in Edmondson county, Kentucky. It is the largest out of five hundred caverns estimated to exist in this county. These caves are excavated in the subcarboniferous limestone, covering a more or less elevated area, estimated to be 8,000 square miles in extent, and varying in thickness from 10 to 300 or 400 feet. This plateau is so honey-combed, that the drainage is almost entirely subterranean. The general features of this limestone table-land are paralleled by those of the less extensive Carniolan caves, described as follows, in Geikie’s Hlementary Lessons in Physical Geography (p. 246-247).. One of the most remarkable examples of this kind of scenery is that of the Karst, in Carniola, on the flanks of the Julian Alps. It is a table-land of limestone, so full of holes as to resemble a sponge. All the rain which falls upon it is at once swallowed up and disappears in underground channels, where, as it rushes among the rocks, it can be heard even from the surface. Some of the holes which open upon the surface lead downward for several hundred feet. Some turn aside and pass into tunnels, in which the collected waters move along as large and rapid subter- ranean rivers, either gushing out like the Timao at the outer edge of the table-land, or actually passing for some distance beyond the shore, and finding an outlet below the sea. Here and there the labyrinths of the honey-combed rock expand into a vast chamber with stalactites of snowy crystalline lime hanging from the roof or connecting it by massive pillars and partitions with the floor. Such is the famous grotto of Adelsberg near Trieste—a series of caverns and passages with a river running across them.” Stevens states that the subcarboniferous limestone in which the Mammoth Cave is situated is overlaid with a thin stratum, mostly of sandstone, which is pierced by thousands of sink-noles, through which the surface drainage is carried down into limestone fissures and thus to the general drainage level of the Green River. ‘This stream passes at the distance of less than a mile from the Cave Hotel, the floor of the latter being 312 feet above the water and 118 feet above the mouth of the cave.” He adds: “The rate of erosionin the Mammoth Cave has been variable. The older parts are perfectly dry, and entirely free from stalagmitic deposits, indicating rapid erosion, fol- < lowed by elevation, so as to deviate the water completely into other channels. In the newer parts the water is still dripping from the surface above, and depositing stalactites and stalagmites.” It is in the newer damper parts, as well as in or near the subterranean streams and pools of this and most if not all the other caves that the animal life mostly congregates. It will be seen that the caves have frequent passages communicating with the upper world, and it will also be seen how 1 Yor the titles of these articles see Chapter X, Bibliography. 2 Kindly loaned by the publishers, Robert Clarke & Co., Cincinnati. eee | | | ON THE LONG ROU/TE. 39, Scotchman’s Trap. 40. at Man’s Misery. 41. Great Relief. 42. The Dead Sea. 43. Cascade at the Styx. 44, Cascade Hall. 45. Serpent Hall. 46. Valley-Way Side-Cut. 47. Great Western. 48, Valley of Flowers. 49. Lucy’s Dome. 50, Ole Bull’s Concert Hall 61. Fly Chamber. 62, Sheep Shelter. CROGHAN’S HALL» ST-CECILIA’s Bogart, Cin.o- 53. Corinna’s Dome. 54, Black Hole of Calcutta. 55, Washington Hall. 56. Snow Ball Room. 57. Floral Cross and Last Rose of Summer. 58. Paradise. 59, Zoe’s Grotto. 60. Tlora’s Garden. 61. Vale of Diamonds. 62. Charlotte’s Grotto, 63. Serena’s Arbor, 64, The Maelstrom. SVE ANS OF THE MAMMOTH CAVE, PREPARED FOR ~~ H. C. HOVEY, 1882. IAS SRO, CEEVECAND’S @ ST CeEcirs," Ss CABYRINTE fo. SO, “| ROGHAN’@ HALL. — MAIN CAVE. i f This Lernzan is said by Cope to be a parasite of the blind-fish, and to live attached by the dise to the inuer edge of the upper lip. “ This position being maintained, it becomes a favorable one for the sustenance of the parasite, which is not a sucker or devourer of its host, but must feed on the substances which are caught by the blind-fish and crushed between its teeth. The frag- ments and juices expressed into the water must suffice for the small wants of this Crustacean.” Cope describes and figures the two egg-pouches, but these were not present in the specimen we had for examination. Whether Mr. Cope’s genus Cauloxenus really differs from Achtheres or not, we are not prepared to say. “The character,” he remarks, “ which distinguishes it from its allies is one which especially adapts it for maintaining a firm hold on its host, 7. e., the fusion of its jaw- arms into a single stem.” In the specimen we figure, however, the arms are much as in Achtheres carpenteri, separate for a part of the way, but not so widely so as in A. carpentert. The body is short and thick, the head oval-rounded, the “arms” short and thick, and of the same length as the head. The shape of the rest of the body in my specimen is irregular. The reader is, for further details, referred to our camera figures. COPEPODA. CANTHOCAMPTUS CAVERNARUM Pack. PI. I, figs. 2, 2a, 2b. Canthocamptus cavernarum Pack., Zoology, 297, Fig. 238, 1879. Body slender, cylindrical, tapering slowly to the end of the body. Carapace not wider than the segment next behind it, but about three and a half times longer, and equal to the combined length of the succeeding five segments; sixth and seventh segments of the body the longest, the eighth being but little shorter than the seventh. The hinder edge of each segment except the last =- MEMOIRS OF THE NATIONAL ACADEMY OF SCIENCKS. 29 is serrate, the teeth being like those of a saw, conical, with the bases approximate. Hyes distinct, broader than long, blackish, situated directly in front of the stomach; the latter being short, pyriform. The intestine, as usual, long and straight. The first antennz of the male composed of eight very unequal joints, the two basal shorter than thick, the second with three spines; third joint very short, less than half as long as the first, and with the distal end oblique, and with a stout spine; fourth joint large, much swollen, as long as the second and third together, and with a row of slender sete on the inner side; fifth joint small, two-thirds as thick and as long as the fourth; sixth long and slender, one-half as thick as the fourth; the seventh and eighth much slenderer than the sixth; the eighth (termina! one) considerably shorter than the seventh, and tipped with three hairs. Second antennz three-jointed, basal joint short, no longer than broad ; second and third joints subequal, of about the same length and thick- ness; from the base of the second a slender curved joint arises, which is about one-third as thick as the second joint, and gives origin to four large sete; third joint subclavate, with about six hairs, besides four Jong terminal setz, which are flattened and bent in the middle, with the ends hair-like. In the first pair of feet the outer branch consists of rather short joints, shorter than those of the inner ramus, which is 3-jointed and nearly twice as long as the outer. The feet in general are rather long, with large, long spines. The fourth arthromere behind the carapace bears a pair of tapering triarticulate rudimentary feet as long as the segment itself, and ending in hooks. The succeeding segment only bears a pair of two unequal stout sete. Caudal setz not more than one-third longer than the appendages themselves, which are conical -and of the usual form; the inner pair are twice as long as the outer; the sete are stiff and stout, simple, with no hairs. No spines on the terminal segment between the caudal appendages; only a few fine minute hairs. Color of the body and appendages, snow white. Length, .03™™ (or ;; inch.) This species differs from C. cryptorum and C. staphylinus, minutus, and rostratus of Europe in the simple caudal setz and the large coarse spines on the edges of the arthromeres. In its general appearance, and especially the form of the male antenne, the present species is much nearer allied to C. eryptorum Brady than to the other European species figured. Several specimens occurred in water taken from Wandering Willie’s Spring. The intestines were filled with a dark mass of food, and the dejections were elongated pyriform. This spring is probably fed from pools above, or the water percolates from above down into the spring. Whether this species of Copepod is peculiar to the cave or earried down into the cavern each year from out-of-doors remains to be seen when the Copepod fauna of the water in the neighborhood of Mammoth Cave has been thoroughly investigated. Family ASELLIDZ. CHCIDOTHA STYGIA Packard. PI. LI, IV, figs. 1, la—1m. Cecidotaa stygia Packard, Amer. Naturalist, v. 752, Figs. 132, 133, 1871. Cecidotea microcephala Cope, Amer. Naturalist, vi. 411, 419, Figs. 109, 110, July, 1872; ‘‘ Mammoth Cave and its Inhabitants”, etc., 19, 1872; Annals and Mag. Nat. Hist., Nov., 1871 (no name). Cacidotea stygia Pack., 5th Rep. Peab. Acad. Sc., Salem, 95, 1873; Smith, U. S. Com. Fish and Fisheries, Ii. Report for 1872 and 1873, 661, 1874. Asellus stygius Forbes, Bull. Illinois Mus, Nat. Hist., I, 11, 1876. Cacidotea stygia Hubbard, Amer. Ent., Il; Fig. 10, p. 79, March, 1880. Generic characters.—Differs from Asellus in the much longer body, the absence of eyes, and in the longer and narrower head. The second (larger) anteune are almost as long as the body; in Asellus only one-half or a little more than one-half as long as the body; the fifth joint nearly twice as long as the fourth; in Asellus it is but slightly longer than the fourth. The legs are also slenderer, the terminal joint much narrower, aud the claws much smaller. The telson (abdomen) is very long and narrow, oblong, two-thirds as broad as long; whereas in Asellus it is not so long as broad, and is much rounded. The mouth parts also present some differences ; the lobes of the maxilla are narrower; the penultimate joint of the palpus of the maxillipedes are one-half as thick 30 MEMOIRS OF THE NATIONAL ACADEMY OF SCIENCES. a8 in Asellus, while the middle joint of the mandibular palpus is not so large and thick, and there are other differences, though of less importance. In our first notice of this genus (Amer. Naturalist, v, 752) it was stated that it was Sosy allied to Idotea. This was a most unfortunate comparison, my observations having been based on imper- fect specimens, which lacked the cercopods or anal stylets, and also the larger antennz. After comparing it with Asellus communis the error was noted.* Ubpder these circumstances I earnestly desire to change the name Cecidotea to Cecasellus, but defer to the present rule of nomenclature, that one author can not change a name based even on an unfortunate error. The present genus has not been regarded as a valid one by Mr. Forbes, who thus speaks of it in his *‘ List of Hlinois Crustacea”: This species has been peculiarly unfortunate. Described originally from an injured specimen, its structure and relations were misunderstood, and it was made the type of a new genus (Cacidetwa Packard). It was soon redescribed by Professor Cope, under the specific name microcephalus; and these imperfect descriptions have since been supple- mented by several fragmentary notices in various papers by Packard and Smith. * * * A detailed comparison of this species with undoubted Asellus, especially with the admirable plates of 4. aquaticus in the Crustacés d’eau donce de Norvége, has failed to reveal any structural peculiarities which could positively serve as the characters of a distinct geaus, and I have therefore united it to Asellus (p. 11). It remains to be seen, however, whether Mr. Forbes has not somewhat overstated the case, and whether there are not a number of structural peculiarities which forbid our placing the two known species in the genus Asellus. The more obvious and important of these have been already noticed in the foregoing diagnosis of the genus. It should be observed that not only are Cwei- dotca stygia and Cecidotea nickajackensis without eyes, but that the body and appendages also differ a good deal from any of the known species of Asellus. The genus seems as well founded as many others in the Isopoda and other groups of Crustacea. We have little doubt but that Cecidotea has by modification and heredity been derived from Asellus, but because this is most probable it is no reason why, from a systematic point of view, we sboula disregard its evident _generic characters; for it is now generally believed that somehow all the genera of Isopoda have descended from some primitive form or genus. Because, then, we do know with some degree of certainty that Cecidotza has recently diverged from Asellus, and can see that the generic charac- ters if possesses have been the result of its under-ground life, we should yet, from a purely taxo- nomical point of view, regard it as a good genus. Of the genus Crangonyx some species are blind and others are not, but the blind species do not present other important differences. It is so with the species of Phalangodes, where the loss of eyes is not always acccompanied by other changes in form and structure; and so with other cases. If we turn to the European Asellus forelii Blanc, a blind species from the abysses of Lake Leman (for specimens of which we are indebted to the kindness of Professor Forel), we see that it does not belong to our genus Oxcidotea; although it has been referred to Cecidotea by Fuchs in his paper on the fauna of the deep sea. Asellus forelii, compared with specimens of Asellus aquaticus from Belgium (obligingly sent us by Prof. E. Van Beneden), is about half as long and broad as A. aquaticus ; the body has retained about the same proportions; the telson (abdomen) is little if any narrower or more elongated. Both branches of the caudal stylets are of about the same length as in A. aquaticus. Asellus forelii, then, appears to us to be evidently a depauperated species, closely allied to A. aquaticus, which has lost its eyes by its life in supposed perpetual darkness at or near the bottom of Lake Geneva. Its generic characters are identical with those of its parent form, A. aquaticus. So also are those of A. cavaticus Schiddte, found in wells im Germany, and which closely resembles A. forelit, only differing in slight specific characters. It is evident that these two blind species were originally derived from A. aquaticus, and hence have retained the generic characters and specific marks of that Huropean species as ebpupaned with our American A. communis. ~ When, however, we turn to our Cecidotea stygia and nickajackensis, we find that they are not only not congeners of the blind European Aselli, but that they are also not congeneric with the American Asellus communis, and that there are no intermediate forms connecting them, although- the eyed species of Asellus are somewhat variable. Hence we feel warranted, on taxonomic grounds, whatever may be our theory about their origin, to retain the genus Cecidotza. * See 5th Rep., Peab, Acad. Sc., Salem, 95, 1873, MEMOIRS OF THE NATIONAL ACADEMY OF SCIENCES. 31 Another point of interest is, that we should naturally expect that the abdomen (telson) in the ‘blind European forms would be longer than in our American Cecidotza, as the abdomen in the European Asellus aquaticus has a telson slightly longer than in the American Asellus communis ; moreover, the legs of the former are longer; the antennz much longer, as well as the caudal stylets, and yet the European blind forms do not exaggerate any of these features of the eyed Asellus, from which they have originated, while the American species do. Specifie characters.—Normal and well-developed male and female from Mammoth Cave. Body long and narrow; head about as long as broad or slightly narrower than long, much narrower than the first thoracic segment, the latter not so wide and one-fourth longer than the second seg- ment, which is erescentiform ; third slightly shorter but like second segment in shape; fourth a little shorter than third segment, and fifth shorter than fourth and considerably shorter than the sixth; the seventh a little longer than first and considerably longer than sixth; the spaces between the thoracic segments large and deep. Abdomen nearly as long as the fifth to seventh, or last three thoracic segments; not quite but almost as wide as the thoracic segments; end of telson a little full in the middle. First (smaller) antennz; in large slender specimens total number of joints, twelve; in shorter less-attenuated females there are eleven to twelve joints. The proportions of the different joints, their relative length and size, do not essentially differ from those of A. communis, except that the two terminal joints are apt to be smaller, though the individual differences are marked, the joints being sLorter in short-bodied than long-bodied individuals; so also the sete vary in number and size : they are arranged much as in A. communis. Olfactory rods, usually three in number, situated on ‘the three penultimate joints, like a cricket-bat in shape, rather more contracted at base, forming a eylindrical handle, than in A. Gommunis; of the same size as 1n A. communis—certainly not smaller Owing to the usually much smaller size of the terminal joint the terminal olfactory rod is two to three times as long as the last joint, and the second rod is sometimes longer than the two terminal joints of the antenna collectively. Sometimes there are five olfactery rods (Fig. 8a). The auditory bristles (Sete auditorie, Fig. 2c, a.s) do not present avy specific characters either ia form or arrange- _ ment; there is a short one on the outer side at the end of the basal joint and three longer ones on the end of the second joint, one external, the other twointernal. In A. communis one ocenrs at the end of the short fourth antennal joint (Fig. 3b). This has not been yet observed in Cecidotwa stygia. (The variations in the shape and number of joints are noticed further on.) Second antennz with the four basal joints short, third longest of the four and scarcely longer than thick; the fifth and sixth very long, the sixth over cne-third longer than the fifth; the remain- ing joints of the flabellum, eighty-five or less, the number variable; the joints frequently bearing single sete. The mandibles differ decidedly from those of Asellus communis; they are more triangular or pointed in the right (Pl. IV, fig. 1a); the teeth of the cutting-edge are very unequal, while the setose edge within the teeth is much more parallel with the outer edge of the mandible; the 2-jointed inner lobe is broader and with stouter longer sete; the left mandible is longer and narrower. The palpi are three-jointed; second joint about half or one-third as thick as in A. communis, but the arrangement of the set nearly the same; third joint much as in A. communis, equaling in iength the basal joint of the palpus, and with two terminal claws. The lobes in front of first maxille, forming the so-called under lip (PI. III, fig. 2h), are provided with narrower lobes than in A. communis. The first maxille (PI. IT], figs. 2e, 2), with the outer lobe a little narrower at base but broader at the end than in A. communis; the terminal spines of the outer lobe much slenderer, more hair-like than those of the inner lobe (Fig. 2f), which are spinulate, and much as in A. com- munis. Second maxille (PI. III, figs. 2g, 2h,) much as in A. communis, the two outer lobes with the inner stout setz curiously denticulated on the inner edge; the hairs on the innermost lobe much longer and denser than in A. communis. The maxillipedes (Pl. IV, figs. 1b, 1c) differ from those of A. communis mainly in the two terminal joints of the palpus; the fourth or penultimate joint is in Cecidotwa stygia slender and about half as thick as in Asellus, but the terminal joint is a little shorter; the lateral lobes are much more full and rounded where the sete are longest than in Asellus communis. 32 MEMOIRS OF THE NATIONAL ACADEMY OF SCIENCES. First pair of legs in the male a little slenderer than in A. communis, the second joint especially much slenderer, but the hand is but slightly narrower,.and the finger is nearly the same in the female. First pair of male uropoda with the basal joint as long as broad, much shorter than in A. com- munis; second joint not quite twice as long as broad, obliquely truncated at the end (regularly oval- rounded in A. communis), with a few minute sete, while on the outside are nine sete; this group not present in A. communis. Second pair minute and short compared with those of A. communis, and very different in shape (Pl. IV, fig. 12). In the female the basal lamellate uropods oval, one- third as large as the second pair, which are also oval and very thin. Caudal stylets (last pair of uropoda) remarkably long and slender, between one-third and one-half the length of the body ; basal joint linear; the second joint about half as long as the first and slightly more than half as wide, with 8 to 10 unequal terminal bristles, of which a single external one is stouter than the others and bulbous at the end; outer branch one third to a little over one-half as long as second joint, and bearing at the end three sete, flattened at the base. The great differences from the last uropoda of A. communis may be seen by reference to Pl. III, figs. 1b, le, and Pl. IV, fig. 3g. There are no auditory sets on the end of the last pair of uropoda. Length of body, male, 12™™; width, 2.5™™s of caudal stylets, 4™™; the males a littie smaller than the females. This crustacean is snendlen? in the subterranean streamlets of Mammoth Cave, especially in one flowing through the Labyrinth, which we have named Shaler’s Brook. The females May 1, 1874, had eggs in the brood-pouch. It also occurred in the River Styx, in the water of River Hall, and in the Dead Sea. Also in the pools in White’s Cave and in Diamond Cave, as well as Salt Cave. It also occurs in Wyandotte Cave, where it was first discovered by Professor Cope, who named it Cecidotea microcephala. Having, however, received Professor Cope’s type-specimens from Wyandotte Cave, we have been unable to find any specific differences.* A variety (P1. III, fig. 5a—5d) occurred in Long Cave, 2 miles from Glasgow Junction, Kentucky, in a water-trough nearly 1 mile from the mouth, May 11, 1874 (Sanborn). In the female, which is blind, the body is considerably broader, the head broader if not shorter, and the abdomen is shorter than in the Mammoth Cave form. Fig. 5a represents the first antenna, which has but 8 to 9 joints, and there are slight differences in the anterior feet (Fig. 5d). From Walnut Hill Spring Cave, near Glasgow Junction, at a point 300 yards from day- light, May 16, specimens were collected by Mr. Sanborn; others collected by Mr. Sanborn in day- light, 50 feet from the entrance of Walnut Hill Spring Cave, Glasgow Junction, May 14, proved to exhibit no differences from Mammoth Cave forms. The first antenne (Figs.-6, 6a) have 10 joints. Those collected by us in Bradford Cave, Indiana, were exactly like the longer slenderer forms from Mammoth Cave. Fig. 7 represents the ten-jointed first antenna. ~ The Carter Cave specimens agree with those from Mammoth Cave, except that the first antennz are much better developed, being larger and with several more joints (15 in all); moreover, there © are as many as 5 olfactory rods, the first one on the seventh joint from the base (Figs. 8, 8a). The first pair of feet (Pl. IV, fig. 2) do not differ from those from Mammoth Cave. We have received numerous specimens from wells in Illinois through the kindness of Mr. 8. A. Forbes. They have narrower, more linear bodies than those in Mammoth Cave. They were 11™™ in length and 2™™ in width. According to Forbes’s sketches (his Pl. I, figs. 19, 20), the two pairs of male abdominal append- ages are somewhat slenderer than in our Mammoth Cave specimens. Forbes also states that the first (upper) have 10 to 12 jointed flagella, “‘ having a slender olfactory club at tip of each of the four or five joints preceding the last. * * * The lower antennz are about two-thirds as long as the body in the female, in the male somewhat longer,” the flagellum containing 75 to 80 joints. Forbes - also remarks that the length of the rami of the caudal stylets varies greatly with age and sex. *In many old males the inner is very long and the outer minute.” Mr. Forbes adds to his description that it “is found quite frequently in deep wells of central Illinois in company with, but much more abundant than, Crangonyx mucronatus. After a long period of heavy rains during the last summer had greatly swelled the subterranean streams which * See Fifth Rep., Peab, Acad. Sci., Salem, 95, 1873, : : f + 4 f . ; 1 ee ee ee a en ee Oe ee A ee ee a ee ee ee ee ee Ree | Ee ee ee ee a ee 4 MEMOIRS OF THE NATION AL ACADEMY OF SCIENCES. 33 these species inhabit, they appeared at the surface in springs, and even at the mouths of tile drains, in such numbers that a hundred could be taken in an hour. A few females were observed with eggs at this time (July).” Upon the whole, the Illinois specimens are the most aberrant, 7. e., the farthest removed from Asellus communis, of any forms found in Mammoth or other caves. They are more linear than some specimens taken from wells in Annville, Lebanon county, Pennsylvania, kindly sent us by Mr. L. H. McFadden, who writes me that they also occur in the springs and welis in the limestone rocks of Cumberland and York counties. The Pennsylvanian examples were large and well developed; the females were 16™™ in length; and the greatest width, 4™™; length of first antenne, 3™™; of second antenne, 15™™, or almost as long as the body itself. The first antennez were large, well developed, and sixteen to seventeen jointed. PI. III, figs.4; 4a represents the end of the antennule, showing three olfactory rods arising from the sixth joint from the end, this being probably an anomaly.* Ca@cIpoT@A NICKAJACKENSIS Packard. PI. III, figs. 9, 9a. Cacidotea nickajackensis Pack., Amer. Naturalist, xv., 879, November, 1881. Body longer, narrower, and slenderer than in C. stygia. The antenna are sometimes very . long, and reach to the end of the third joint of the second antenne; they are sometimes nearly twice as long as in C. stygia, and are purplish-white, while the flagellum is provided with long hairs. Figs. 9, 9a, however, represents a short first antenna composed of only 8 to 9 joints with a single long olfactory rod on the sixth joint. The second antennez are as long as the head and extend back- wards as far as the base of the abdomen. The legs are much longer and slenderer than in C. stygia. The abdomen is long and narrow, and the caudal appendages are moderately long in one specimen and short in another; in one individual the outer branch is much shorter and smaller than in the others, and in most it is as long as the basal joint. On the whole, the caudal appendages are no longer than the telson or terminal segment of the abdomen, while in C. stygia they are half as long as the entire body. This species forms, in the antenne and slightly purplish color and the proportions of the leg-joints, perhaps a nearer approach to the _ genus Asellus than that of Mammoth and Wyandotte caves; on the other hand, C. stygia approaches Asellus more in its shorter, broader body, with the shorter, broader abdomen. It seems quite evident that the two species must have descended from different species of Asellus. Whether there is an additional species in the Southern States from ‘Fic. 7.— Cecidotea nickajackensis, which the present species may have been derived remains to be seen. ec Oe Dy seustal ehylatss The two specimens from Lost River, found by Dr. Sloan in that subterranean abode, were of the normal form and size of A. communis, but bleached as white as C. stygia; the eyes are black and distinct. This variety may be called pallida. It is interesting to note the occurrence of this bleached variety, which may have become thus modified after but a few generations; perhaps but one or two. Besides Asellus communis, which is widely diffused throughout the Eastern and Central States, Mr. Forbes has described two species which occur in southern Illinois, neither having been detected in central or northern Illinois, although the most varied situations were carefully searched. Our sketches (Pl. IV, figs. 3, 3a-39) will give some of the details of structure of A. communis, which will serve as a basis of comparison. Asellus intermedius Forbes, Bull. Ill. Mus. Nat, Hist., No. 1, 10. This species, as remarked by its describer, is intermediate between Asellus communis and A. brevicauda. As will be seen by reference to Forbes’ Fig. 14, the caudal stylets are like those of *The anditory bristles of this species were mistaken by Mr. Hubbard for ‘an unknown ciliate Infusorian?” See his Fig. 10b, and Amer. Ent., iii, p. 80. S. Mis. 30, pt. 2——2 34 MEMOIRS OF THE NATIONAL ACADEMY OF SCIENCES. A. communis, only narrower, and with longer sete, especially the terminal ones. The first antennz are fourteen-jointed, while the olfactory rods and auditory bristles are essentially asin A. communis. Mr. Forbes writes me that “there is some room for supposing that my A. intermedius is a dwarf form of A. communis; but I think no one could suppose that A. brevicauda was other than a distinct species. It occurs in the hill country of southern [inois, and probably also farther south.” Asellus brevicauda Forbes, l. c., p. 8. In this form, which lives in clear rocky rills in Illinois, which is described in detail by Mr. Forbes, the first antenne are fifteen-jointed. The auditory bristles are well developed, and there are five on the end of the second joint, where three are usually observed. ‘The last olfactory rod is twice as long as the terminal joint of the antenna. ‘The first pair of legs are shorter than in A. communis, while the head is much smaller, and of an entirely different shape. The caudal uropoda are dwarfed, very short, but the sete: much as in A. communis. Remarks.—It follows from the foregoing statements that the geographical range of Cecidotwa stygia is as great or greater than that of any other cave animal, unless Pseudotremia cavernarum be excepted. it also appears that upon the whole it does not vary much, being invariably white in color, blind, usually with no traces of eyes, and of a narrow, elongated shape. The parts which vary most are the organs of sense—é. e., the first and second antennsz, especially the first pair. Its parent form is evidently one of the species described from Illinois, and as Asellus communis is widely distributed over the Mississippi valley and the Atlantic States, we are justified in regarding this as the parent form. It seems to be more abundant than any other species of the genus. This is evidently due to its immunity in its subterranean retreats from the attacks of the host of enemies—insect, crustacean, and fish—which prey upon the eyed out-of-door forms. Although blind, its loss of eye-sight is made up to i by its greater development of sense- appendages (antennze), though after all the loss of sight is perhaps of little moment, since it is not exposed to the attacks of stronger animals. It breeds from April to May in Mammoth Cave, and probably all summer, since it was found by Mr. Hubbard with eggs in Cave City Cave, July 29, 1881, and in wells with eggs in Illinois by Mr. Forbes. The number of eggs produced and kept within the incubatory pouch appears to be no greater than in the out-of-door forms. We think we have shown that on taxonomic grounds the genus Cecidotea is as well founded as many other genera which are accepted by carcinologists, It presents, at any rate, certain constant differences from the blind species of European wells and caves, as well as the dark abysses of Lake Geneva, and though exposed to the same general surroundings, has developed in different directions. It affords an interesting example of the origin of generic characters by changes in an environment the nature of which we can easily estimate. CRANGONYX VITREUS Smith. PI. V. figs. 1 to 4. Stygobromus vitreus Cope, Amer. Naturalist, vi, 422, July, 1872. _ Not Crangonyx vitreus Packard, Fifth Ann. Rep. Peab. Acad. Sci., Salem, 95, 1873. The following description and accompanying sketches on Plate V have been kindly prepared by Prof. 8. I. Smith, of Yale University: All the Amphipods which I have seen from the Mammoth Cave belong to a single species, undoubtedly the same as the one badly described from the same locality by Professor Cope. In all I have examined five specimens, collected by Professor Packard in Shaler’s Brook, the Laby- rinth, and Willie’s Spring. The largest specimen, the one figured, is from Shaler’s Brook, and was found under a stone. This specimen is a female, 5.2"™ in length (from the front of the head to the tip of the telson). The secondary flagellum of the antennulz is minute, scarcely larger than the first segment of the flagellum, very slender, and composed of two segments, of which the terminal one is very minute, and about one-third as long as the first segment. The caudal stylets are all short and stout, the first and second pairs, with the outer rami, a little shorter than the inner, and both armed with spines which increase in length distally and at the tips are very long and slender; the third, or posterior pair, are almost rudimentary, being much shorter than the telson. The basal portion eee es ee ee ee ee ee ee ee MEMOIRS OF THE NATIONAL ACADEMY OF SCIENCES. 35 is very short but quite stout and wholly unarmed, while the single terminal segment or ramus is © minute, scarcely longer than broad, and tipped with three spinules. The telson, as seen from above, is rectangular, nearly as broad as long; the posterior margin has a very shallow sinus in the middle, each side of which is armed with several slender spines. The other appendages are suf- ficiently well shown in the figures. A smaller female, about 3.5™™ long, differs very slightly from the last. The flagella of the antennule are each composed of ten segments and those of the antenne of four. In the second pair of legs the propodus is relatively not quite as large, is a very little narrower, and the palmary margin has one or two less spines on the outer side; all characters of a slightly less mature specimen. The three other specimens are very small, and I have not been able to determine the sex of any of them. They may be either young males or immature females. One of these, about 2.7™™ long, differs considerably from the larger specimens, but only in such characters as immature specimens usually differ from the adults. There are only eight segments in the flagellum of the antennula and four in that of the antenne. Im the first pair of legs the propodus is a little more slender than in the adult, the palmary margin is not quite as oblique, and is armed with one or twoless spines on each side. The second pair are only very slightly larger than the first, and of course very much more slender than in the adult. The propodus is narrower in proportion and scarcely wider thanthecarpus. The palmary margin is less oblique, not longer than the posterior margin, and is armed with fewer spines on each side. The first and second pairs of caudal stylets and the telson are armed with a few less spines. I have been unable to discover even rudimentary eyes in any of the specimens. This species agrees with Bates’s description of the typical species of the genus in having the posterior caudal stylets ‘‘unibranched,” and thus differs from the following species which we have referred to the genus, although in Bates’s species the terminal segment or ramus of the stylet is as elongated as the outer ramus in C. gracilis and C. packardii. In the structure and size of the posterior caudal stylets, in the stoutness of the second pair of legs, and in wanting eyes, this species approaches C. tenuis, from wells at Middle- town, Connecticut, to which it is apparently more nearly allied than to any of the described American species. The C. tenuis is, however, a wholly distinct and quite different species. I know of no species with which this is closely enough allied to make its affinities of any value on the question of the origin of the cave fauna (S. I. Smith). Figs. 1 to 4.—Crangonyx vitreus, female, 5.2™™ long: 1, lateral view, enlarged 20 diameters; 2, one of the first pair of legs seen from the outside, enlarged 48 diameters; 3, one of the second pair of legs, enlarged 48 diameters; 4, terminal portion of the abdomen, side view, enlarged 48 diameters; a, telson; b, posterior caudal stylet; c, second caudal stylet; d, first caudal stylet. This species was not uncommon in the pools of Mammoth Cave, occurring in Richardson’s Spring and Wandering Willie’s Spring. It has the habits of Gammarus, scooping a furrow in the mud of the bottom of the pools in which it lives. CRANGONYX PACKARDII Smith. Pl. V, figs. 5 to 11. Crangonya vitreus Packard, Fifth Ann. Rep. Peab. Acad. Sci., Salem, 95. July, 1873. Not Siygobromus vitreus Cope, Amer. Naturalist, vi, 422, 1872. This species is so closely allied to Crangonyx gracilis that it might readily be mistaken for it were it not for the peculiar structure of the eyes. The eyes of C. gracilis are composed of a few facets, and are abundantly supplied with black pigment. In all the specimens of C. packardii which I have seen the eyes are observable with difficulty, the black pigment being wholly wanting. The specimens received at first were very badly preserved, and I then thought the absence of the pigment might be due to this fact; but subsequent examination of more perfect specimens shows that this cannot be the éase, and that the eyes are, in life, undoubtedly wholly without black pigment. The eyes are scarcely, if at all, observable in the ordinary alcoholic specimens, but when rendered translucent by immersion in glycerine the structure of the facets is distinctly observable, as shown in Fig. 5. As observed by Dr. Packard, the flagella of the antennule of O. packardii are a little shorter, and usually contain four or five segments less than in C. gracilis, but this is an uncertain - 36 MEMOIRS OF THE NATIONAL ACADEMY OF SCIENCES. character, and some specimens of C. gracilis from Lake Superior actually have only one or two more segments than the subterranean species. In the antenne there are no constant differences. There are some very slight differences in the first and second pairs of legs, especially in the females, but not greater than usually exist in the individuals of a single species in allied genera, and any large series of specimens would undoubtedly show all the intermediate forms. In the third to the seventh pairs of legs there is a constant difference in allthe specimens examined, the Spines being more numerous, longer, and more slender in C. gracilis. The spines upon the first and second pairs of caudal stylets are a little shorter and more obtuse in C. packardi than they usually are in C. gracilis; otherwise there is no difference in the caudal stylets and telson. These differences are all such as very naturally lead to the supposition that this subterranean form has been derived from the C. gracilis at no very remote period, although this supposition may well be held in reserve until we have a more complete series of the subterranean species for comparison. C. gracilis occurs as far south as Grand Rapids, Michigan, whence we have received specimens from Mr. N. Coleman, and it very likely occurs in the same region as C. packardii. The figures are all from Professor Packard’s original specimens, collected from wells in Orleans, Indiana, by Dr. Moses N. Elrod. Only one of these has the body entire; this is a female, 5.5™™ Jong, from which Figs. 5 to 8 were made. A larger specimen, a female about 7.5™™ long, unfortunately wanting most of the antennule and antenne, collected from a well at New Albany, Indiana, by Dr. John Sloan, was sent to Dr. Packard for examination. Figs. 5 to 11.—Crangonyx packardit, details all enlarged 48 diameters; 5 to 8, female, 5.5™™ long; 9 to 11, female, about 7.5™™ long: 5, lateral view of head; 6, terminal portion of one of the first pair of legs, outside; 7, same of second pair; 8, terminal portion of abdomen, lateral view; 9, one of the first pair of legs, outside; 10, one of second pair of legs, outside; 11, antennula and antenna, side view. [All the figures, 1 to 11, drawn by Prof. S. I. Smith.]} CRANGONYX ANTENNATUS Pack. Crangonyx antennatus Pack., Amer. Natuvalist, xv, 880, 1881. The second crustacean discovered swimming about in the subterranean stream of Nickajack Cave was a species of Amphipod belonging to the genus Crangonyx, and has been described as Fig. 5.—Orangonyx antennatus Packard. a, end ef abdomen and appendages; 6b, head, with base of upper and entire lower antenne and 5 eyes; ¢, manus of second pair of feet; all enlarged. . Crangonyx antennatus Packard.* It is a large purplish species, with very long antenpe, and dis- tinet, well-developed black eyes. * Tt is a large and purplish species; the first antenne very long; the flagellum with 20 to 24 joints; the entire antenne being over one-half and nearly two-thirds as long as the body; the last joint of the peduncke being slightly more than half as long as the penultimate joint. Compared with C. gracilis Smith, from Lake Superior, it differs in the form of the eyes, the longer and stouter first antenne, the flagellum having a greater number of joints, and in the different proportions of the joints of the peduncle; the second joint of the latter being much longer than in C. gracilis, while the first joint of the scape is much longer, and the second and third joints one-third longer in proportion than in C. gracilis. The-fourth pair of epimera are unusually large and nearly square. The telson, together with the caudal stylets, is much as in C. gracilis, but the rami are slightly stouter and more polished, and the spinules a little stouter. It probably is a little larger species than C. gracilis, the specimens being 6 to7™™ in length; the eyes are not so distinct, and are only one-fourth as large asin C. gracilis. It is very different from ©. vitreus (Cope), of Mammoth Cave, and from C. packardit Smith, differing in its distinct eyes, and larger, more numerously-jointed antenna. MEMOIRS OF THE NATIONAL ACADEMY OF SCIENCES. 37 CRANGONYX MUCRONATUS Forbes. ' Crangonyx mucronatus Forbes, Bull. Illinois Mus. Nat. Hist., i, 6, Dec., 1876. Figs. 1 to 7. This remarkable species, remarks Mr, Forbes,* is perhaps entitled to rank as the type of a new genus; but, until I have the material for a more general study of its relations than I am able to make at present, I prefer to place it with its nearest allies in the genus Crangonyx. Colorless; blind; length, 9 to10™™; width, i™™. The head is a little longer than the first tho- racic segment, its anterior margin concave at the bases of the upper antennz, convex between them; the posterior margin straight in the middle and curving forward on the sides. The front angles of the first thoracic segment are uncovered and produced a little forward; the hind angles of the first five segments are rounded and produced strongly backward. The first three abdominal segments have the lateral margins and all the angles broadly rounded, and the posterior angles, as well as the posterior margin of the seventh epimeron, are sligitly notched and bristled. The upper anteunz of the male are two-thirds to four-fifths as long asthe body. The first and second joints of the pedicel are subequal, each about as long as the four basal joints of the flagellum; the third is one-third as long as the second. The flagellum is about five times the length of the pedicel, and is composed of 30 to 35 joints, each with a few short hairs at tip, and all except the seven or eight basal joints and the last with a slender olfactory club. The secondary flagellum contains two bristled joints, together a little longer than the first of the primary flagellum. Pedicel of lower antennz longer than that of upper, the last two joints equal, each a little longer than basal joints of upper antenna. Flagellum nine or ten jointed, with olfactory clubs. Right man- dible with dental laminz; each with five conical, obtuse, subequal teeth. The anterior lamina of the left mandible is much the larger and stronger, with three very strong, blunt teeth; posterior lamina with three slender and acute teeth. Paipus, three-jointed; basal quadrate about half as long as second, which is clavate and nearly twice as wide as long, with about ten long hairs on its rounded hind margin, which are longest and closest distally. Last joint a little longer and nar- rower than second, regularly convex in front, straight on proximal half of hind margin, slightly concave on distal half, and fringed here with about twenty-four slender hairs, the three or four at tip becoming suddenly very much longer. A few scattered hairs on front margin of this joint. Inner plate of anterior maxilla is nearly hemispherical, about half as long as outer, with four plumose hairs on the rounded margin, which are about as long as the plate itself. Palpus two- jointed ; first quadrate, one-third as long as second, which is oval, pointed, tipped with two claws and some smailer spines. Lamine of basal joints of maxillipeds short, neither hair extending beyond tips of succeeding joints. First two pairs of feet equal. Dactyl of first pair in male curved, two-thirds as long as hand. The latter is broad-ovate, two-thirds as wide as long, the palmar and posterior margins forming a wide angle. Long hairs on posterior surface in transverse rows. Palm with about fifteen short, -notehed spines, each with a hair arising from the notch. Carpus subtriangular, tliree-fourths as wide as propodus, hind margin very short, with one or two pectinate spines and a few long hairs; second pair similar; propodus a little longer and narrower; carpus as wide as propodus, posterior margin longer, with about five transverse rows of long bristles, of which the distal row are doubly * pectinate on terminal third. The three posterior pairs of thcracic legs increase in size backward, the first of these being not quite two-thirds as long as the last. The seventh epimeron is narrow, with the lower margin regularly arcuate. The tips of the first pair of anal legs extend beyond the tips of the second, and these beyond the tips of the third. The latter are therefore very short, about as long as the pedicel of the second pair. The outer ramus is ovate, truncate, half as long as the pedicel, and hairy at tip; the inner is an unarmed rudiment, one-fourth or one-fifth the length of the outer. The telson of the male is a smooth cylindrical appendage, usually about as jong as the first three abdominal segments, and as large as the last joint of the pedicel of the lower antenna. It presents a very slight double curve, iS obliquely rounded at the end, and tipped by a cluster of short hairs. In some eases this appendage is half as long as the body. The female differs in the following particulars: The upper antenne are only about half the length of the body, the flagellum not more than three times as long as the pedicel, and the secondary j * The following description is copied from Mr. S. A. Vorbes’ article. 38 MEMOIRS OF THE NATIONAL ACADEMY OF SCIENCHS. flagellum is usually a little shorter. The propodus of the first pair of feet is similar in outline, but the palmar margin and dactyl are shorter, and the posterior margin longer. The second pair is — extremely like the second of the male, but is decidedly smaller than the first. The telson affords a difference so remarkable, that the two sexes, at first sight, would hardly be referred to the same genus. In the female this is very similar to the telson of C. gracilis Smith. It is flattened and slightly emarginate, a little longer than broad, extending to the tips of the second pair of anal legs, and bears two terminal clusters of spines of four or five each. This species was first discovered by me in a well at Normal, Illinois, during the summer of 1875. It was subsequently found by Mr. Harry Garman in great numbers in springs, and even at the mouths of drains, after a long period of heavy rains. With the advent of dry weather it entirely disappeared from these, but still occurs sparingly in wells. (S. A. Forbes.) CRANGONYX LUCIFUGUS Hay. Crangonyx tucifugus Hay, Amer. Naturalist, xvi., 144-5. 1882. The following deseription is copied from Mr. Hay: This is a small, rather elongated species, that was obtained from a well in Abingdon, Knox county, Illinois. As befits its subterranean mode of life, it is blind, and of a pale color. Inlength the largest specimens measure about 6™™. Male.—Antennule scarcely one-half as long as the body. The third segment of the peduncle two-thirds as long as the second; this, two-thirds the length of the first. Flagellum consisting of about 14 segments. The secondary flagellum very short, and with but 2 segments. Antenne short, only half as long as the antennule. Last 2 segments of its peduncle elongated. Flagellum consisting of but about 5 segments, and shorter than the last two segments of the peduncle taken together. Second pair of thoracic legs stouter than the first. Propodite of first pair quadrate, with nearly a right angle between the palmar and posterior margins. Palmar surface on each side of the cutting edge, with a row of about 6 notched and ciliated spines, one or two of which at the posterior angle are larger than the others. The cutting ecge is entire. Dactylopodite as long as the palmar margin, and furnished along the concave edge with a few hairs. Propodite of the second pair of legs ovate in outline, twice as long as broad. The palmar margin curving grad- ually into the posterior margin. The cutting edge of the palmar surface uneven, and having, near the insertion of the dactyl, a square projection. The palmar surface also armed with two rows of notched and ciliated spines, five in the inner row, seven in the onter. Dactyl short and stout. Two posterior pairs of thoracic legs longest of all and about equal to each other. All the legs are stout and their basal segments squamiform. Postero-lateral angle of first ab dominal segment rounded; of second and third, from obtuse-angled to right-angled. First pair of caudal stylets extending a little fart her back than the second; these slightly exceeding the third. The peduncle of the first pair somewhat curved, with the concavity above, the rami equal and two-thirds as long as the peduncle. The peduncle of the second pair little longer than the outerramus. Inner ramus nearly twice as long as the outer. Third pair of caudal stylets rudimentary, consisting of but a single segment. This somewhat longer than the telson, broadly ovate, two-thirds as broad as long, and furnished at the tip with two short spines. Telson a little longer than wide, narrowing a little to the truncated tip, which is provided at each postero- lateral angle with a couple of stout spines. j Female.—In the female the propodite of the anterior pair of feet resembles closely that of the corresponding foot of the male. The palmar margin of the second propodite is less oblique than in the second foot of the male, and does not pass so gradually into the posterior margin. It is also destitute of the jagged edge and the square process of the male foot. There are fewer spines along the margin. One of the spines at the posterior angle is very long and stout. This species appears to resemble C. tenuis Smith, but is evidently different. In that species, as described by Prof. 8. I. Smith, the first pair of feet is stouter than the second, and has the palmar margin of the propodite much more oblique. The reverse is true of the species I describe. Nor dol understand from the description of C. tenwis that the posterior caudal sty lets each consist of a single segment. There are some minor differences. From C. vitreus, judging from Professor Cope’s description in American Naturalist, volume vi., page 422, it must differ in the caudai stylets. ‘“‘Penultimate segment with a stout limb, with two equal styles,” is a statement that will not apply to my species, whichever the ‘‘ penultimate” segment may be. CAMBARUS PELLUCIDUS (Tellkampf). Astacus pellucidus Tellkampf, Arch. f. Anat., Phys. u. Wissensch. Med., 383, 1844. Cambarus pellucidus Erichson, Archiv f. Naturgesch., xii, Jahrg. I, 95, 1846. Orconectes pellucidus Cope, and Orconectes inermis Cope, Amer. Naturalist, vi, 410, 419, 1872. Canbarus pellucidus Smith, Rep. U. S. Com. Fish., 1872 and 1873, 639, 1874. ; Cambarus pellucidus Faxon, Proc. Amer. Acad. Arts and Sci., xx, 139, 1884; revision, 90, 1885. Next to the blind-fish of Mammoth and other caves, the blind crayfish first discovered in 1842 by Dr. W. T. Craige has, from its size, attracted the most general attention from the public. We ee ee ee TO ae MEMOIRS OF THE NATIONAL ACADEMY OF SCIENCES. 39 will extract the following remarks as to its specific character and affinities from the first mono- grapher of the genus, Dr. Hagen: . The shape of the rostrum is somewhat analogous to that of C. affinis; the margins are more paralle! at the base. The lamina of the antenns is long, but strongly dilated nearer to the tip; the epistoma is shorter and broader than in the other species; the basal joint of the inner antenne has a spine at the tip which in the other species is always nearer to the base. The fore border of the cephalothorax is not angulated behind the antenn~, as in all other species. Nevertheless the number of the hooked legs, the form of the abdominal legs, and the elongated body and hands, exclude C. pellucidus from the other groups. Some, no doubt, will prefer to regard C. pellucidus as a distinct group or genus, still, as Tam convinced, without foundation. The most striking differences consist in the aberrations in the shape of ths foreparts and of the limbs of the head (pp. 33, 34). In speaking of the eyes, Dr. Hagen appears to be in error in stating that the “‘ optic fibers” are not developed. If reference is here made to the optic nerve, it is, as we shall see further on, well developed, while, as he truly says, the “‘ dark-colored pigments” are not developed. The specimens from Wyandotte Cave described by Cope as Orconectes inermis are scarcely a variety of C. pellucidus, as originally stated by us in our paper ‘On the Cave Fauna of Indiana,”* Professor Faxon also remarks: C. pellucidus is subject to considerable variation. In some specimens the rostrum is shorter than in typical specimens, and contracts more from the base to the lateral teeth, which are much less prominent. The spines of the postorbital ridge and sides of the carapace are slightly developed. This is the form described as a new species (Orconectes inermis) from Wyandotte Caye, Indiana, by Professor Cope, in 1872. LIowe to Prof, A. S. Pagkard an oppor- tunity to examine Cope’s type. Itis a male, Form II, with the first pair of abdominal appendages not articulated, a condition often found in the second form males of this species. After an examination of this specimen I can indorse the opinion of Hagen (Amer. Naturalist, Aug., 1872), and Packard (Fifth Ann. Rep. Peab. Acad. Sci. for 1872), ex- pressed before seeing the specimen, that the variation is not of specific value. All the specimens which I have seen from the Indiana caves, amounting to six in number, belong to this form. But the same form also comes from the Mammoth and neighboring cavesin Kentucky. Ina gigantic female in the Museum of Comparative Zoology (No. 3417, collected in Mammoth Cave by F. W. Putnam) the peculiarities of Cope’s form are intensified. The point of the rostrum does not reach the distal end of the peduncle of the antennule, and hardly attains the proximal end of the distal segment of the peduncle of the antenna.t The lateral rostral spines are reduced to salient angles. The postorbital ridges are destitute of spines, as in C. bartonit. The antennal scales reach but to the proximal end of the terminal segment of the peduncle of the antenna. The lateral spinules of the carapace are represented by granular tubercles. The spines of the meros of the cheliped are short and tooth-like, those on the upper surface are blunt, those beneath are irregularly disposed, without the clear biserial order seen in the typical form, and also in Cope’s type of O. inermis. The hands are broad, flattened, and tuberculate. This species is more widely diffused throughout the cavernous region of Kentucky and Indiana than is generally supposed. A male collected by us from Bradford Cave, Indiana, does not differ from a male from Mam- moth Cave. In a male from one of the Indiana caves (which cave is not indicated) the cephalothoracic suture is much more acutely produced posteriorly than in a male from Mammoth Cave. In the male the rostrum is narrower and its spines, both frontal and lateral, are longer and slenderer than in the female. ¥ Comparing two males of the same form from Mammoth and Wyandotte caves, the former has the right hand the larger, and the Wyandotte one the left hand the larger. Although the Wyan- dotte male is a little the smaller, the large hand is about one-fourth larger and is broader than in the Mammoth Cave one. The rostrum of the Mammoth Cave example is broader and the sides less raised and thickened. In the Mammoth Cave specimen the inner edge of the end of the antennal scale reaches as far as the lateral spines of the rostrum; in the Wyandotte male the scale reaches far be- youd the lateral spines, more than half way between the lateral spine and the end of the median spine. These Lregard as simply individual differences, as in another male from Wyandotte or Bradford Cave the large hand is of the same relative size and on the same side as the Mammoth Cave male. In the females from different caves in Indiana (received from Dr. John Sloan), one has a slightly narrower rostrum than the other, which is a larger individual. In a female from Diamond Cave, which we collected, the larger hand is the left one, the rostrum and antennal scale are shorter than in a large female from Mammoth Cave; otherwise it does not differ essentially. * Fifth Rep. Peab. Acad. Sci., Salem, 94, 1873. t In the typical form of C. pellucidus the rostrum equals or exceeds in length the pedancle of the antenna. 40 MEMOIRS OF THE NATIONAL ACADEMY OF SOIENCES. This species was also found by Mr. J. B. Proctor to occur in Bat Cave, Edmonson county, Kentucky, June 13, 1874. It does not differ from Mammoth Cave examples. ‘he larger hand is on the left side. One female collected by Mr. Sanborn in a cave near Haunted Cave, same county and State, was rather small, but presented no differences from Mammoth Cave females. In none of the specimens examined did the rudimentary eyes seem to vary. Relation of O. pellucidus to out-of door species of Cambarus.—In comparing pellucidus with C. bartonit, a young male of Form II, 1.30 inches long, from Mammoth Cave, it was found to differ from O. bartonii in the spinules on each side of the apex of the rostrum. It seems to us that O. pellucidus is, in the proportions of the body and particularly the shape of the rostruin more like C. bartonit than C. affinis. C. pellucidus differs from C. bartonii in the longer hand and the fourth joint of the limb, while the thorax and rostrum are much longer. The antenne are rather the stouter and shorter in C. pellucidus. The elongated body, long hands, and the limbs bearing them are changes such as we would expect to meet with in cave animals. The C. bartonit, which is a good deal bleached, is as white and as pale as the pellucidus, except that the head and first pair of limbs and hands have scattered blackish speckles. C. affinis is evidently, however, the parent form of C. pellucidus. On comparing males of Form II of the two species, C. pellucidus has stouter and larger second antenne; the antennal scale is broader at the . end, the rostrum is wider, the head is rather wider and shorter, the hinder edge is less convex, the thorax is a third longer, the abdomen but slightly longer, the difference being in the cephalo- thorax. The ischium of first pair of legs is one-third as thick and about one-third as long, the meros one-half as thick and one-third longer; the carpus is of about the same length, but the hand is one-half as wide and a little longer than in C. affinis. Of the four succeeding pairs of feet the ischia are about the same length, the meros somewhat longer. The first antennz are longer and slenderer. ; The gonopods in Form II are very distinct from the out of-door species, being nearly one-halt shorter. Cambarus rusticus, which is closely related to C. affinis, was found by us in abundance at a point only about 20 feet from the mouth of the cave in the brooks which flow out of Bradford Cave; inside of the cave pellucidus is not uncommon. Remarks.—Two alternatives present themselves in considering the origin of the form pellw- cidus. First, it either is derived, with C. affinis, from a common ancester; or second, and what seems more probable, it is a modification of C. affinis or an allied enc é. g., rusticus. The characteristics which separate C. pellucidus from C. affinis or C. bartonii or any out-of-door species are those which have been induced by its life in total darkness and the diminution in its food- supply. The close neighborhood of the habitat of the two forms at the Bradford Cave, the blind one living only a few yards away and in the upper part of the same brook as C. rusticus, is very significant, and this affords us the best means of ascertaining the origin of this form. It is paral- leled by the probable origin of the Myriopod Seoterpes from Trichopetalum and of Pseudotremia from Lysiopetalum. CAMBARUS HAMULATUS (Cope and Packard). Orconectes hamulatus Cope and Packard, Amer. Naturalist, xv, 881, Pl. vii, figs. 1, 1a, 16, Nov., 1881. Cambarus hamulatus Faxon, Proc. Amer, Acad. Arts and Sci., xx, 145, 1884; Revision of the Meneices Mem. Mus. Comp. Zool., x, 4, 81. Pl. iv, fig. 6, ix, figs. la, tas ies 1885. In this species the epistoma is much as that of C. bartonii, but shorter and broader ; while ihe median terminal tooth is less marked than in C. latimanus, and the sides fall away panicle from the front margin. It is entirely different in shape from that of C. pellucidus. The antennal lamina is Shorter, broader, and much more rounded on the inner edge than in C. pellucidus, and in this re- spect differs from C. latimanus. The rostrum is narrower than in C. pellucidus, while the first pair of (large) claws are much slenderer, and the telson narrower than in C. pellucidus. The most ob- vious difference is seen in the modified first and second pairs of abdominal feet of the male, to which we may apply the term gonopod, for it is not properly an intromittent organ.* The first and second pair of gonopods differ decidedly from those of C. pellucidus, and closely resemble — those of Form II of Cambarus latimanus (from Athens, Géorgia, figured by Hagen), those of the MEMOIRS OF THE NATIONAL ACADEMY OF SCIENCES. 41 first pair being shorter, thizker, and the last joint being much bent, hook or sickle-shaped, whence the specific name hamulatus. The first gonopods differ in the proportion of parts from those of OC. latimanus, but the joint is much more acute than in C. latimanus. The first pair of gonopods, compared with the latimanus form of obesus from Maryland, given me by Mr. Uhler, are much like it in general form, but the sinuous branch is longer and straighter, while the hook is much slenderer. In the second pair of accessory gonopods the knob is propor- tionately smaller. In other more important characters C. hamulatus is quite unlike the latimanus form of CO. obesus, the scale of the second antenne being very different, the chele one-half as wide, and the antennz much longer, while the rostrum is much longer and more pointed. Length of the largest male, 5 centimeters. Cambarus hamulatus is quite different from C. pellucidus of Mammoth aud Wyandotte caves in the rostrum, the slender hands, the much broader antennal scale, and in the form of the gonopods, while the whole creature is slightly slenderer than 0. pellucidus, though the rudimentary eyes are of the same propor- tion to the neighboring parts as in the other species. It is obvious that the form from which C. hamulatus has been derived is quite different from that which has given origin to the blind crayfish of the Kentucky and Indiana caves. The mostcomimon species in Northern Georgiais Oambarus latimanus, which has been found at Athens and Milledgeville, Georgia, and probably is abundant in the northern limestone region of Ala- bama. At any rate, itis perhaps to Cambarus latimanus that we look for the ancestors of Cambarus hamulatus. On the other hand, in the form of the body, of the scale and rostrum, as weil as of the upper lip and the chelz (though not of the gonopods), C. hamulatus approaches Cambarus affinis. Now, of all our North American erayfishes, it would appear, as Mr. Uhler has told the writer, and as seems evident to us upon an examina- tion of several types and: the excellent figures of Dr. Hagen, that C. afinis is the more generalized form, and this is tanta- mount to saying that it is the ancestral form of our North American crayfishes. So, while our Nickajack blind cra, fish may have been an immediate derivative of C. latimanus of the Gulf States, it probably ultimately originated from OC. affinis, a wore wide-spread species. Prof. W. Faxon, in his “ Revision of the Astacidz,” remarks as follows regarding this species, based on an examination of four males, Form II, and two females, the types of Cope and Packard’s description : s In general form and appearance it bears a close resemblance to C. pellucidus, but the carapace is less spiny, and the male has hooks on the third pair of legs only, and the first pair of abdominal appendages are formed after the fashion of the C. bartonii group. The rostrum tapers towards the tip more than it does in the typical form of C. pellucidus, resembling, in this respect, the form C. pellucidus inermis. The terminal segment of the telson narrows at the hinder end more than in C. pellucidus, Ido not find the differences in the mandibles, antennal scales, aud chelee meationed by Packard. Fic. 10.—Cambarus hamulatus: a, antennal scale; b, gonopod of the first pair; alLenlarged. (Kingsley, del.) * Note on the function of the gonopods.—As stated by Milne Edwards and others, the gonopods of the crawfish are not intromittent, but simply rude gutters for the passage of the fertilizing fluid to the eges. It is obvious that in the lobster the gonopods form simply a rude tube or gutter to conduct the seminzl fluid to the eggs as they pass back- ward from the oviducts to the swimming feet of the female. During the process of fertilization of the eggs the male, without doubt, as in the crawfish, holds the female by the-claws, she resting on her back. The term gonopod is ap- plied for convenience in descriptive carcinology to the external reproductive organs of the Crustacea, since they are only modified limbs, 4? MEMOIRS OF THE NATIONAL ACADEMY OF SCIENCES. While Mr. Cope has proposed the genus Orconectes to include the two blind species of cray- fishes (0. pellucidus and C. hamulatus) on account of the absence of eyes, and we were inclined after the discovery of a second species (hamulatus) to adopt the genus, for the reason that the loss of the cornea and retinal elements, as well as the reduction in the size of the eye itself, is common to two species; yet we should hesitate to do so, from the fact that there are so many other genera of Arthropods in which there are both eyed and eyeless species—the character being one of great instability. At the same time due consideration should be given to the fact that such loss, total or partial, of the organs of vision is of profound significance, more so than the mere syste- matic zoologist is apt to recognize. Were there other good generic characters than those afforded by the reduction of the eye we should retain Cope’s genus Orconectes; meanwhile it may stand as a subgenus. ARACHNIDA. ACARINA.* Rhyncholophus cavernarum n. sp. (Pl. X, figs. 1, 1a, 1¢)—This is a minute white species, .8™™ in length, found near the end of White’s Cave; also by Mr. Sanborn in Long Cave, 2 miles from Glasgow Junction, Kentucky. Bryiobia ? (or Penthaleus ?) weyerensis n. sp. (Pl. XI, figs. 1, 1a, 1b)—Body stouter, larger, and maxille slenderer than in the other species; legs very long. Color, dull white. Length, 1.1™™., Weyer’s Cave. Mr. A. D. Michael writes me that this is probably a Labidostoma Kramer, or Nicoletia. Lelaps ? (or Holostaspis ?) wyandottensis nu. sp. (Pl. X, figs. 2, 2a, 2b)—Body thick, oval; max- ille very short, minute; maxillary palpi five-jointed, the terminal joint bearing a broad triangular plate; tarsal claws long, elbowed, and spatulate. Length, 1.8™™. Little Wyandotte Cave. Lelaps (= Iphis ?) cavernicola n. sp. (Pl. X, fig. 3)—Body oval; no eyes; pale horn color. Length, 1™™. Labyrinth, Mammoth Cave, under QOozonium auricomum on sticks. Perhaps a Hypoaspis. ; Gamasus (or Hypoaspis ?) troglodytes n. sp. (Pl. X, figs. 4, 4a, 4b, 4c)—Occurred in Mammoth Cave with the preceding species. Gamasus stygius n. sp. (Pl. X, figs. 5, 5a, 5b, 5c)—Of the same form as in the preceding species, but the beak is nearly one-half shorter. Color, pale horn. Length, 1™™. Bat. Cave, Carter Caves, Kentucky. ; Dameus (= Delba) bulbipedata n. sp. (Pl. X, figs, 7, 7a)—Head conical; abdomen orbicular; the legs long and slender, with all the joints more or less bulbous, and each bearing two or three long sete. Fig. 8, Dr. Trouessart thinks, is perhaps the nymph of this species. The legs are a little shorter, and the sete on the end of the abdomen much longer. End of Dixon’s Cave. Oribata alata n. sp. (Pl. XI, figs. 2, 2a)—This is a short, round species, with a stout conta head, and two large wing-like ‘empansianns on each side, extending in front nearly as far as the end of the head; the legs are long and slender. Collected at the end of Dixon’s Cave. Uropoda lucifugus n. sp. (Pl. X, fig. 9)—Body suborbicular, nearly as wide as long, with short legs, the longest about two-thirds as wide as the body. Found in New Wyandotte Cave, attached to Pseudotremia. Allied to U. krameri according to Trouessart (in litt.). Sejus ? sanborni u. sp. (Pl. X, figs. 6, 6a)—Body thick, spherical, white. Length, 5™™, Maxille and palpi very short, about one-third as long as the first pair of legs. Cave near Dismal Creek, Kentucky (F. G. Sanborn). Dr. Trouessart refers this form to Sejus ? or Zercon? Family CHERNETID A. OBISIUM CAVICOLA Pack. Obisiwm cavicola Pack., Amer. Naturalist, xviii, 202,203, with fig. February, 1884. + This is certainly an aberrant species of the genus, whether we regard the size of the chelicere or the shape of the cephalothorax. The latter is much longer than broad, widest just before the * The descriptions of the mites are brief and imperfect, as little is known of the genera and species in the United States, and the following notices are only preliminary. ‘The figures, it is hoped, will enable them to be rec- ognized. (See also explanation of Plates X and XI.) I am indebted to Dr. Trouessart for most of the generic names. t By an oversight the author’s name was omitted at the end of the article. MEMOIRS OF THE NATIONAL ACADEMY OF SCIENCKS. 43 middle, narrowing in front and behind, and between the chelicere deeply cleft; a feature unusual in the genus. There are no traces of eyes, either of a cornea or pigment spot. The chelicerz are rather smaller than usual and separate at base, and more conical, less pyriform than usual; the manus is shorter, and the fingers longer than usual, both finger and thumb (the fixed finger) are curved, the tips acute, and the inner edge denticulate. The pedipalps are as long as the body without the chelicerze; they are rather thick, not especially long; the first joint is stout and of uniform thickness, as is the second, which is not con- tracted at the base, being of uniform thickness; it equals in length the width of the cephalothorax ; the third is three-fourths as long as the second, is slightly contracted at the base, subconical in form; the hand is thick and heavy, it is about twice as long as the third joint, and the fingers are moderately curved. The abdomen is narrow and rather long, with the segments well marked. The body (in alcoholic specimens) is dull white, while the pedipalps are horn-colored, with a reddish tinge. Length of body, including the chelicere, 2™™. One specimen from the New Market Cave, Virginia.* CHTHONIUS PACKARDII Hagen. Plate XI, figs. 3, 3ato 37. hthonius packardi Hagen, Zool. Anzeiger, II Jahrgang, 399. July 28, 1879. ma Q eee - packard reer Amer. ee bet ae 83. March, Hi HIG SA ST aan Peeks Chthonius packardi Hubbard, Amer. Entomologist, iii, 83. March, 1880. Body rather long and narrow; thorax rather flat, considerably longer than wide, nearly one- fourth, and about (perhaps a little less than) one-half as long as the abdomen; a little broader just behind the front edge, and narrowing somewhat on the hinder edge. The abdomen is considerably contracted at the base, especially on the second segment; it is rather narrow, less swollen in the middle than usual; it consists of ten distinct segments, of which the seventh and eighth are the widest. On the thorax is a lateral row of sete, five on each side and two in front, but none in the center, and two behind. In the young there are four in front. On each side of the abdomen are two rows of setz, of the same length as those on the thorax. There are no traces of eyes in three specimens; in a fourth rudimentary ones exist. General color (in alcohol) pale horn, with a reddish tint; body and legs of the same tint; the chelicere of the same tint, but the hand of a little deeper tint, while the pedipalps are decidedly deeper in color than the rest of the appendages aud body ; beneath not paler than the upper side of the body. Chelicerz (mandibles) of moderate size, full, and swollen towards the base; two-thirds as long as the thorax, and about as long as the latter is wide behind the middle; the movable claw or finger is shorter than the chelicera is thick, and the fixed thumb is about half as long as the body of the chelicere. On the hand near the base of the finger are two hairs, and the finger itself bears two external sete within the middle; on the hand near the division into fingers is a group of three straight, stiff, barbed sete (Fig. 3e). On the inside of the middle of the long slender, regularly-curved finger are six teeth,+ the distal one the largest. The serrula has eighteen divisions, the first one free, the distal one very acute. The pedipalps are long and slender, longer than the body by nearly the length of their finger. The third joint is about one fourth as long as the second; the manus is one-half as thick as long; the fingers are long and slender, nearly but not quite as long as the second joint; both are a little curved, with the extreme tips suddenly bent in; each finger with a series of fine acute teeth on the inner edge, which end in short sete. The legs long and slender, the posterior pair about twice as long as the abdomen; the two posterior trochanters twice as large as those of the second pair of legs, and the latter much larger than * Chelifer cancroides L.—This species occurred in company with Crthonius packardii in Salt Cave in partial day- light, 50 to 100 feet from the entrance. This is a cosmopolitan species, being found everywhere in Europe and in the Atlantic as well as Pacific States. This is its first occurrence in caves; but its appearance in a single cave, may for the present, at least be regarded as accidental. t These hairs form the so-called organs of smell of Stecker. (See Hagen, Zool. Anzeiger, July 28, 1878, p. 400.) 4.4. MEMOIRS OF THE NATIONAL ACADEMY OF SCIENCES. those of the first pair of trochanters; the terminal tarsal joints of first and fourth legs of nearly the same length, and quite setose, with scattered longer stiff setz, each with a pair of slender curved hooks and a slender curved anchor-shaped plantula with a thin cylindrical shank. Length 2.5 to 33™™, Labyrinth of Mammoth Cave, under stones and fungus (Oozonium), three specimens; and ano ther from the same cave in total darkness; also Dixon’s Cave (Packard and Sanborn). One specimen with rudimentary eyes occurred in Salt Cave, 50 to 100 feet in from the mouth, in partial daylight. Of the four specimens from the Labyrinth three are totally blind, with no traces of eyes; the fourth one, of which the exact locality in the cave is unknown, has rudimentary eyes; the cornea could not be detected, and the presence of the eye was only indicated by a silvery-white pigment- spot situated on each angle of the front edge of the thorax, in front of the lateral sete. One specimen from the New Wyandotte Cave, Indiana, was totally blind, with no trace of cornea or pigment-spot, and the body was white. Our specimens agree with Hagen’s description in the American Entomologist. My description is drawn up from males. His type-specimens were from Wyandotte Cave. Dr. Hagen’s Kentucky Specimens were from the bottom of the Dome, Mammoth Cave, “‘with dead bat,” November 9, and Long Cave, near Glasgow Junction, Kentucky, 1 mile from daylight, May 11. Hagen writes as follows of the Mammoth Cave form with two eyes: It is pale yellowish; the thorax, mandibles, palpi, legs, and segments of the abdomen about the same color; the base of the mandibles a little darker; the abdomen bet ween the segments and on the sides paler. I have seen only three specimens in alcohol, ail from the Mammoth Cave region, one couple from one locality and a female from another locality. I have compared all very carefully with C. (Blothrus) packardi, from Indiana. They are a little longer, 3 to 3.2™™ long, but a little darker, or perhaps a little less white, but all three have on each side of the thorax . one eye, distant from the anterior border as far as the length of the diameter of the eye. The movable finger of the mandibles is not indented. The examination of all other details shows no difference. In his article in Zoologischer Anzeiger, July, 1879, Dr. Hagen refers to this form as follows: As the position and number of the eyes has hitherto furnished for Chelifer genera a trustworthy indication, 7 had described it as a new species.* A subsequent very close comparison with (Blothrus) packardi gave as a result that the two species appear to be identical; only the former has two eyes; the latter is blind. Further research showed that neither can be separated from the genus Chthonius, which has two eyes on each side. Consequently we have here the interesting fact that Chthonius liviug without the cave have two eyes on either side, and that within the caves live forms of this genus in which either the anterior pair of eyes is aborted or these two are wanting, and light-refracting cells (lichtbrechende Kerne) under the skin, at the base of the sensitive hairs, seem to form a partial substitute for the wanting organs of sight. l append Mr. Hubbard’s excellent description of this form. Dr. Hagen has, with the greatest liberality, placed at my disposal his unpublished descriptions and figures of new Pseudoscorpions belonging to the genus Chthonius. The following is the description of the cave species. The few changes I have taken the liberty of making in his manuscript haye been rendered necessary by new facts devel- oped in correspondence, and in an article published in advance of the descriptions by Dr. Hagen in the Zoologischer Anzeiger, Leipsig, July, 1879. 3 “Ohthonius packardi Hagen, n. sp —(Fig. 9, a male, enlarged fifteen times; b left mandible from below; c feathered bristle of the mandibles more enlarged; d movable finger of mandible, occasional in female; ¢ chela of the palpus; f termination of tarsus; g plantula from above.) “Dull whitish mandibles and palpi very light brown, segments of the abdomen yellowish white. Thorax flat, a little longer than broad, quadrangular; very little enlarged just before the anterior border, and a little narrowed behind in female; lateral borders nearly straight, a little convex just before the anterior border where the eyes should have been; angles rectangular, the hind ones scarcely rounded; noeyes; thorax smooth, with a few sensitive bristles, four on each side, two near the anterior border and two near the posterior one, two more in the middle each side nearer to the anterior border, which is a little produced in middle; mandibles large, one-third shorter than the thorax; base convex above, oblong, a little narrowed to the fingers, which are shorter than the base; the movable finger incurved, strongly pointed, with sometimes a small knob or external indentation before the tip; on the under side of the base, just before the division into fingers, internally, three long feathered bristles placed in a line, usually larger on the left mandible, sometimes wanting (or rubbed off?); abdomen less than twice as long as the thorax, scarcely broader at base, ovoid, thicker in the female; two rows of hairs on the segments 1 to 3, four rows on the three following segments, and six rows on the two following segments ;t a transversal row of *Under the manuscript name C. (Blothrus) incertus. t The number of hairs is found to vary on the abdomen. (Hub.) MEMOIRS OF THE NATIONAL ACADEMY OF SCIENCES. 45 hairs on each ventral segment. Male with the second ventral segment triangularly excised; female with two small holes; palpi thin, longer than the body by about the length of their fingers; hypopodium oblong, incurved; tro- chanter short, enlarged at tip, incurved, about half as thick as long; femur very long, straight, cylindrical, slender, a little convex above before the tip, as long as the thorax and the three basal segments of the abdomen; tibia similar to trochanter, ineurved, enlarged at tip, Jess than halfas long as the femur; chela thin, one-half longer than the femur; the hand as thick as the tibia, cylin- drical, a little enlarged below just beyond the artic- ulation, straight, shorterthan the femur; fingers as long as the femur; slender, straight, viewed in pro- file a little ineurved, the tips hooked suddenly, the movable finger a little shorter; both with a series of sharp teeth inside; legs slender, the two anterior pairs as long as the body, the two posterior pairs ex- tending beyond the body the length of the tarsus; hypopodia oblong, a little incurved, those of the first pair a little pointed before; trochanter short, a litule Jonger than thick; femur long, cylindrical, tibia half as long as the femur; first tarsal joint as long as the tibia, second as long as the femur; the two posterior pairs with the trochanter and base of the femur enlarged; femur with a spurious articula- tion before the middle; first tarsal joint shorter than the tibia; all legs with long fine hairs; two very slender and strongly-curved hooklets on tip; between them an anchor shaped plantula with a thin ecylin- dricalstem. The palpiof female are aslong as those ofthemale. Length 2.3™™—.09inch, Habitat: Wy- andotte Cave, Indiana; five males, one female, in alcohol. The female has a small external indentation of the movable finger of the mandibles; the finger of the males has no indentation, but in two specimens - ee: a aE the tip is somewhat broader, more obliquely ent, and e with a fiue engraved line where the indentation ¥ic. 12.—Chthonius packardii: a, male enlarged fifteen times; b, man- hk ” dible; c, serrula of the mandible; d, finger of the mandible; e, chela of should be. _pedipalp ; f, end of tarsus; g, plantula.—A fter Hubbard. pedipalp 39; P The discovery of this blind Pseudoscorpion in America is very interesting. It belongs without any doubt to Schiedte’s genus Blothrus, which, on careful exami- nation, proves, however, to be merely Chthonius with undeveloped eyes, and is the smallest species known. “C. (Blothrus) speleus ditters by the longer tibia of the palpi, and by the two anterior pairs of legs with a two- jointed tibia. The last statement is doubted by Mr. Simon, but Mr. Schicdte’s accuracy is so well known that his statements are to be accepted. (B.) abeillit has much longer palpi and legs, and the sexes of dissimilar development. (B.) brevimanus is only known to me by an insufficient diagnosis. (B.) cephalotes seems rather similar to (B.) packardi, only a little larger, the mandible granulated, nearly as long as the thorax; the fingers of the palpi equally longer.” Another form, with two eyes, occurs in the Mammoth Cave: “This pale yellowish; the thorax, mandibles, palpi, legs, and segments of the abdomen about the same color the base of the mandibles a little darker, the abdomen between the segments and on the sides paler. ““Thave seen only three specimens in alcohol, all from the Mammoth Cave region, one couple from one locality and a female from another locality. I have compared all very carefully with C. (Blothrus) packardi from Indiana. They are a little longer, 3 to 3.2™™ long, a little darker or perhaps a little less white, but all three have on each side of the thorax one eye, distant from the anterior border as far as the length of the diameter of the eye ; the movable finger of the mandibles is not indented. The examination ef all other details shows no difference. Habitat, from, the bottom of Dome, Mammoth Cave, with dead bat, November 9, and Long Cave, near Glasgow Junction, Kentucky, one mile from daylight, May 11.” My specimens, two males and two females, from the Rotunda in Mammoth Cave, have each two eyes, which, how- ever, vary in the convexity of the cornea and are so faint as to be easily overlooked. The males are very white; one of the females shows traces of an indentation on the mandibular finger. The male from which the figure was drawn measures 3™™ in length, or, exclusive of the mandibles, 2.3%™, The hairs upon thorax and abdomen, which are cor- rectly represented in the figure, differ slightly from the description of the blind form, but they are probably variable. To Dr. H. Hagen my grateful acknowledgments are due for invaluable aid and suggestions. I have added nothing to his observations on Pseudoscorpions; the portions indicated by quotation marks are copied almost ver- batim from his manuseript. Hagen states in his original notice in the Zoologischer Anzeiger, in which he speaks of the present species as Blothrus packardi, that “it certainly belongs to Blothrus, Schicedte, and is the smallest known species.” 46 MEMOIRS OF THE NATIONAL’ ACADEMY OF SCIENCES. Having received C. speleus from Mr. Simon, I have been able to compare the two forms. It seems to us that the present species belongs at least to a separate section of the genus from C. (Blothrus) speleus. This is seen not so much in the form of the body of C. speleus (Schicedte), though the cephalothorax is much narrower, as in the much greater length of the pedipalps. In Chthonius packurdi the base of the hand, when stretched back, reaches the end of the abdomen; in — speleeus the distal end of the second long joint reaches nearly to the end of the abdomen, hence the pedipalps are nearly as long again as in Chthonius packardii; the third joint is very slender, and three to four times as long as in C. packardii. The chelicere differ but little, as they do not vary much throughout the family, though they are smaller, and at the base are less swollen or pyri- form, while the fingers are less bent. The legs in C. spelceus are much longer, the joints being longer and slenderer. It is twice the size vf our Mammoth Cave species, and in aleohol it retains its bleached, white color. Chthonius abeillet Simon closely resembles C. speleus, agreeing with it not only in the slender- ness of the appendages, but also in the great length of the third joint of the pedipalps. Obisiwm (Blothrus) cerdberus (Simon) also agrees with the two foregoing species in the same characters. Obisium cavernarum appears to connect these three species with O. simoni Koch and the other out- of-door forms. Dr. Hagen remarks that C. packardit is nearest allied to C. cephalotes Simon, as shown by Simon’s figure 20 (Pl. XIX). This must be the case, as the third joint of the pedipaips is short and broad and of the same form as in our C. packardii. CHTHONIUS Cascus Packard. Plate XI, figs. 4, 4a, 4b, 4c. Chthonius cecus Pack.,* Amer. Naturalist, xviii, 203, Feb., 1884. Body unusually short and broad, and the limbs short and thick; thorax a little shorter than broad, the sides parallel, slightly narrower behind than in front. The setz on the cephalothorax and abdomen less distinct than in C. packardu, but arranged in nearly the same manner, though I could find none on the front of the cephalothorax; the disc is free from them, as in C. packardii. The . eyes are entirely wanting, there being no traces whatever of them, not even any pigment-spot. The general color of the body (in alcoholic specimens) the same as in C. packardvi, the body being pale horny, the abdomen whitish, the dorsal sclerites pale horn, the chelicere reddish, and the pedipalps a little paler. Chelicerze very stout and thick, considerably more so than in C. packardii, the fingers quite blunt at the end, the thumb acute, with six distinct teeth; the serrula much as in C. packardti. Pedipalps unusually short, the second joint short and nearly twice as thick as in C. packardii ; the third joint short, thick, conical, much asin C. packardii, except that it is somewhat thicker; the manus is very short and thick, really but little longer than the third joint; the mov- able finger nearly twice as long as the manus, and with long seta; it is stout and very straight, and serrulate on the inner edge. The legs are very short and thick; the second trochanters very thick; the fourth or hinder pair of legs not much longer, if any, than the abdomen, while in C. packardvi they are fully twice as long. Abdomen of 11 segments, broad and short, with the tergal sclerites rather more distinct than in C. packardit. Length of body, with the chelicer, 1.5™™. Two specimens from Weyer’s Cave, Virginia. Remarks.—This species differs remarkably from the other cave species, being very broad and short, with straighter, less curved fingers of the pedipalps, while the hind legs are scarcely longer than the abdomen; and the cephalothorax is actually shorter than broad. It is totally blind, with no traces of even a pigment-spot. It is a notable exception to the law that blind forms have usually a slenderer body and attenuated extremities. Unfortunately I have no out-of-door species to compare it with. It agrees with the generic characters given by Simon in his great work on the Arachnida of France, the cephalothorax narrowing behind, and the fingers of the pedipalps being straight. It must be confessed, however, that these are very slight generic characters, though Simon of course mentions others. It seems to us that the three species of the suppressed genus Blothrus, 7. ¢., B. speleus, B. abeillix, and B. cerberus, all of which bave remarkably long pedipalps (the iid join being very long and slender), hein * The article containing the description of this species was by enroute unsigned by the author’s name, a MEMOIRS OF THE NATIONAL ACADEMY OF SCIENCES. 47 slender legs and no eyes, in reality differ more from the other species of Obisium than does the generally-accepted genus Chthonius. As we have attempted to show however, contrary to Hagen’s opinion, the American OC. packardii is not a true Blothrus; it could not properly be separated from C. cecus. They both agree in the short, thick, conical third joint of the pedipalps. It may be interesting in this connection to learn something of the habits of the species of the two genera Obisium and Chthonius. As almost nothing is known of the habits of our American species, we are obliged to compile the following account from M. Simon’s excellent work. The species of Obisium live in moss and vegetable detritus; they are very agile, and run back- wards easily. Of seventeen species three (Blothrus) have no eyes and inhabit caves; these three have been already referred to by name. Of fourteen French species with eyes, one only (0. caver- narum) has “very small punctiform eyes.” It isacave-dweller. Of the fourteen species, only one other ‘lives in caves; this is 0. lucifugum. In this species the eyes are “ very small.” All the other species have either four or two eyes. We here see a very direct connection between cave- life and the eyeless forms in the species of southern and central Europe. Turning to the genus Chthonius, there are seven French species described by Simon. He says that ‘‘the Chthonii also seek dark and damp places; several are peculiar to grottoes, and the ordinary species are sometimes met with in wood-piles and caves. Thus at Troyes M. J. Ray has observed in caves three species of Chthonius (rayi, orthodactylus, tetrachelatus), where they run with agility on the walls, and escape by hiding in the little fissures, in which they shoot forward as previously described.” Of the seven species of Chthonius enumerated by Simon one (C.-cephalotes) has no eyes; it lives in caves. In C. microphthalmus the eyes are very small; it also usually occurs in caves. Thus the majority of the species occur in caves; three of them permanently so. It will also be seen that there is an obvious relation between those which are totally blind and: live in caves; 7. ¢., those without eyes inhabit caves exclusively, those with very small eyes are found partly or mostly in caves, and those with perfect eyes are not cave-dwellers. The relations of cause and effect in the blind species of this family are, then, very marked, the adaptation to life in partial or total darkness involving the disuse and Gonpeauent atrophy of the organs of sight. It is so also with the North American species. To return to the American cave species: It appears that in inhabitants of even one cave (Mammoth) there are individuals of Chthonius packardii existing, as regards the eyes, in three conditions, though all must be in total darkness. (1) Some have two eyes, with the cornea as usual; (2) some have no cornea, but the silvery dot indicating the retina is retained; while in others all traces of the eyes have disappeared. On the other hand, in those individuals existing in Salt Cave (a small cave not wholly dark) the two eyes are distinct. These facts would indicate that the Mammoth Cave examples must be the descendants of some out-of-door species, to which the Salt Cave individuals are nearest allied. Moreover, the difference between the eyed and eyeless forms are apparently individual rather than varietal.* The presence or absence of eyes in this case are not generic characters. The cave forms retain the generic characters of out-of. door species of Chthonius, both American and European. The characters in which individual vari- ation of C. packardit occurs are apparently the parts of the eyes alone. This loss of eyes, partial or total, seems to us to be the result of the direct influence of the surroundings upon the organism. When we take into account both the European and American species as a whole (0. cocus, with its stout body, being an as yet inexplicable exception), the eyeless species have, besides the loss of eyes, very slender bodies and remarkably attenuated pedipalps and legs, especialiy the hinder pair. We do not think that natural selection can in such a case as this be regarded as an efficient cause in producing the cave forms. The eyes are useless in total darkness; hence from disuse they gradually, after a few generations, disappear. On the other hand, the limbs tend to grow longer, to perhaps exert a tactile sense; and this trait, being favorable to the species, is gradually furtber developed, until it becomes fixed in the organism by heredity. The result is that all the * Those from the Rotunda have eyes (Hubbard); those from the Labyrinth had pigment-spots or were totally blind. The Rotunda is much nearer the mouth of the cave than the Labyrinth (see map); hence the eyed forms may have been more lately introduced into the cave. Further research should be made in this direction. Dr. Ha- gen’s specimens were mixed up, part from Mammoth and part from Long Cave. 48 . MEMOIRS OF THE NATIONAL ACADEMY OF SCIENCKS. individuals become long-limbed and blind. There is apparently no struggle for existence, but the direct influence of darkness, united with heredity, are plainly the immediate agencies in the trans- formation. Family PHALANGIDA. PHALANGODES ROBUSTA Simon. Plate XIV, figs. 2, 2a, 2b. Scotolemon robustum Pack. Bull. Hayden’s U. §. Geol. Sury. Terr., iii, 164, 1877. Phalangodes robusta Simon, Arachnides de France, 156, 1879. This species is here referred to at length, because it is the only out-of-door species of the genus yet known in the West; and by reference to the figures and description it will be seen how much the cave forms differ from it. Hight females?. Tegument deep reddish, with the hinder- segments finely bordered with brown; tarsal joints paler, with dense blackish specks; cephalothorax a little paler red, marbled with reticulated darker lines. Body pyriform, two-thirds as long as broad; cephalothorax a little more than half as long as wide, the front edge slightly rounded, with the angles well marked. Theeye-tuberele not so large and high asin S. terri- cola Simon, being of moderate size. Eyes black and large, fully _ developed, while those of S. terricola are nearly obsolete. Abdomen a little longer than broad; the first five segments well marked, the sutures being much more distinct than in S. terricola or probably any other European species, judging by Simon’s drawings. The last three segments, with the outer edge of each segment, free, not united with each other, as are the five basal joints; last segment with the ventral slightly projecting beyond the tergal portion. Beneath are seven well-marked sterna, the first and second being united without suture. Chelicere of the usual form, rather stout at base of first Fic. 13.—Phalangodes robusta Pack., (enlarged). joint, but much as in SW. terricola; second joint moderately a, pedipalp; b, c, chelicera. 2 7 & ; long; hand of the usual form, a little unequal. Pedipalps unusually short and thick, much more so than in S. terricola or any other species described by Simon; basal joint broader than long, with a pair of stout, sharp spines and four small ones; second joint nearly two thirds as broad as long, full and swollen above, beneath with four large spines; third joint much slenderer, one half as long as the second; fourth joint nearly twice as long as broad, with five stout spines, of which the fourth is much larger than the others, the fifth minute; fifth joint as long as, but slenderer than the fourth, with five stout spines, the fifth and terminal spine much larger than the others, and as long as the joint is wide. This joint is a little hairy, while the others are nearly naked. Legs stout, much more so than usual in the genus; anterior pair with three tubercles ending in hairs on the second joint; a larger tubercle on the fourth joint; the three other pairs are unarmed. Second pair of legs longer than the first by one-third of their length. The second and fourth pairs are of nearly equal length, the fourth pair differing in having the third joint consid- erably swollen; the third and first pairs of the same length. On the cox of the second pair of legs is a pair of stout conical spines, meeting over the median line of the body. The anterior tarsi are tbree-jointed, as in S. terricola of Europe, the middle one much shorter than the other two, which are of equal length; those of the second pair five-jointed; those of the third and fourth pairs four-jointed, the ends of the tibia being constituted so that the limbs appear as if they had five tarsal joints. Ungues rather long and moderately curved. The legs are stouter and shorter- than in S. terricola, and none of my specimens have the long, singular, sinuate appendage on the first joint present in S. terricola. (They are not referred to by M. Simon in his description, though my Specimens were received from him.) . Length of body, exclusive of the mandibles, 3.5™™; breadth, 2.5™™, MEMOIRS OF THE NATIONAL ACADEMY OF SCIENCES. 49 Compared with 8S. terricola Simon, from Corsica, which also lives under very large stones, and is found common at Porto-Vecchio after the heavy spring rains, but which has not yet occurred in caves, our out-of-door form is much stouter, with much shorter legs, and also differs in its well- developed eyes, dark brick-red tegument, and dark markings. [t was discovered in Colorado in 1874 by Mr. Ernest Ingersoll. while attached to Hayden’s Geological Survey of the Territories. He tells me that it did not occur in any cave, the exact locality and mode of life being forgotten. It will most probably be found under stones. Compared with Phalangodes flavescens (Erebomaster flavescens Cope, Amer. Naturalist, vi, 420, 1872, from Wyandotte Cave, Indiana), which is allied to the European S. piochardi, which inhabits caves near Orduno, it differs in the basal segments being much more distinct, where the sutures in the tergum are obsolete in S. flavescens. The eye-tubercle is aJittle smaller proportion- ately, while the eyes themselves are much larger. The mandibles and maxillz are shorter, while the legs are very much shorter and stouter. The color is deep red, the cave species being pale yellow. These are all differences such as we should expect to find between a cave dweller and one which has lived out-of doors under stones, etc. In these two species we have forcibly brought before us the great structural differences brought about by striking changes in the environment of the two species. PHALANGODES FLAVESCENS (Cope). Plate XII; Plate XIV, fig. 1. Erebomaster flavescens Cope. Amer. Naturalist, vi, 420. Figs. 114,115. July, 1872. Phalangodes flavescens Simon. Arachnides de France, vii, 156. 1879. Male and female specimens. Body broad and stout, uniformly straw yellow, including the body and appendages. Cephalothorax broad and short, but little longer than broad, the sides widening a little toward the hinder edge, being wider on the hinder édge than elsewhere; it is somewhat constricted on the anterior third just behind the eyes; the surface is considerably rounded, the posterior edge is quite free and distinct from the abdomen; the latter is unusually short aud broad, with three segments visible from above and six, in all, beneath, the basal one being the longest. The eye-tubercle is large and high, usually forming a cone slightly higher than broad at the base. The two eyes are black, distinct, and situated on each side, near the base of the conical tubercle. The cornea is underlaid by a broader dark- mass forming the retina. Cheli- cere with the first joint rather long and slightly contracted iu the middle; the second joint or hand is rather thick, not twice as long as thick; the outer.finger is much larger than the inner, much curved and pointed, with a series of ten small conical teeth on the inner edge; the inner tooth is straight, with five large blunt teeth on the inner edge. Three short sete can be seen on the outer half aud two on the inner side near the fingers. Pedipalps less than twice as long as the body, but nearly twice as long as the cephalothorax; coxal joint very broad and short; first - joint longer than broad, cylindrical, with a small setiferous external spine in the middle, and on the inside an anterior much larger spine bearing a bristle; second joint longest of all, with four sub- equal setiferous spines on the outside; on the inner side two large spines, the first half as long as the joint is thick, and bearing a stout movable spine as long as the joint is thick; the second spine is somewhat smaller; these are succeeded by four very small spines situated on the proximal half of the joint, while the edge, especially in the middle, is finely toothed; on the distal end is a group of three unequal spines, one large and long; third joint a little more than half as long as the second, with five or six unequal spines visible on each side, two of them being as long as the joint is thick ; joint four is somewhat elongated, barrel-shaped, with four stout spines on the outer and three on the inner edge, the latter being long and slender, with long setz, the basal one as long as the joint is thick in the middle; the fifth and last joint has four external stout curved spines, and three large and two minute straight inner spines; four of the sete, though differing in length, are as long as the joint is thick; the terminal spine is shorter and thicker than the others, with a movable seta, which is as long and fully twice as thick as the others. First pair of feet much smaller and shorter than the second pair, the last tarsal joint four- jointed ; second pair slenderer than the fourth pair; the third joint not so much swollen; the last tarsal joint divided into twelve joints; length of entire leg, 8™™; third pair a little longer than first S. Mis. 30, pt. 2——4 50 MEMOIRS OF THE NATIONAL ACADEMY OF SCIENCES. pair, and one-third shorter than fourth pair; the latter pair 8™™ in length, being of the same length as the second pair, but with the basal joint longer, the last tarsal joint being subdivided into but three joints. The penis is 1.25™™ in length, the basal portion not curved, the smaller distal portion a little shorter than the basal, and ending in a corneous short cultriform thin appendage with four subacute teeth on the outer edge. Length of body, not including the chelicere, 3™™; breadth, 2™™. Subvariety Weyerensis. Those from Weyer’s Cave, Virginia, all differ in habit from the Wyandotte forms, and are easily distinguishable, as seen in male and female specimens from Weyer’s Cave. They are larger and darker, with brown specks, as seen in Plate XII, fig.—. The second pair of legs, as compared with those from the Wyandotte Cave, are shorter and thicker. The penis, however, is ‘a little longer, being 1.5™™ in length, rather slenderer, and the cultriform appendage differs a little in shape, while the basal half is much more curved at base. It should be borne in mind that these differences are such as we would expect to meet with in individuals from a smaller cave and one more liable to be reached from the outside world; and the differences, moreover, are such as ally this variety to the more robust and out-of-door forms. One specimen from a cave near Dismal Creek (collected by Messrs. Sanborn and Beckham) was like Wyandotte specimens in size, and smaller than the var. weyerensis, but resembled the latter in color. Variety 2, cecum, (Pl. X11) 10 males and females from the Carter Caves (Bat Cave) are blind, the cornea being present, but with no retina. Im all the specimens the cornea is equally colorless, and the individuals must be practically blind. The eye-tubercle is smaller and blunter than in Wyandotte examples. The individuals are a little smaller and paler than the variety weyerensis; the pedipalps are the same, except that there appear to be two spines on the inside of the second joint instead of one, as in weyerensis and the typical forms from Wyandotte Cave; the length and size of the second leg are identical with those of the typical Wyandotte examples, being slenderer and slightly longer than var. weyerensis, but the third joint is setose; that of the Wyandotte specimens, at least in some cases, being without sete. The penis in cecum is but slightly over 1™™ in length; it is shorter and smaller than in the Wyandotte and Weyer specimens, the basal joint shorter as well as the second joint, and the teeth on the cultriform appendage are shorter and blunter. Remarks.—While, as observed above, the Weyer’s Cave specimens are a pretty well marked variety, aud are more like out-of-door forms than the Wyandotte Vave examples, and this would _be what we should expect to find in inhabitants of a larger and much deeper cave, it is singular to find that the individuals of var. cecum from Bat Cave, which is a smaller cave and apparently more open to daylight than Weyer’s Cave, should be a more attenuated and blind form. We should naturally expect that the Wyandotte individuals would be blind. PHALANGODES ARMATA (Tellkampf). Plate XIII, figs. 1, la-Ih. Phalangodes armata Tellkampf, Archiy fiir Naturgeschichte, X Jahrg., Bd. 1, 320, Taf. viil, figs. 7 to 10. 1844. Acanthocheir armata Lucas, Ann. Soc. Ent. France. 1860. Wood, Proc. Essex Institute, 36. 1568. Phriais longipes Cope, Amer. Naturalist, vi, 421, July. 1872. Phalangodes armata Simon, Arachnides de France, vii, 156. 1879. Phrixis longipes Hubbard, Amer. Ent., iii, 39 Feb. 1880. Maleand female. Body rather narrow and long compared with P. flavescens, being considerably longer than broad; whitish straw-yellow, including the body and appendages, the young being white. Cephalothorax considerably longer than broad; the sides widen somewhat towards the hinder edge; they are not constricted near the middle; the surface is moderately convex, and the posterior edge is nearly straight and free from the abdomen, which is broad and short, but longer and more pointed than in P. flavescens, with five segments to be seen from above and six beneath, the latter being the number of uromeres in the genus Phaiangodes; the iast segment (seen from above) is less than half as wide as the last one in P. flavescens, and is narrow and conical in shape. The eye-tubercle is about half the size of that of P. flavescens, and is conical in shape; there are no traces of the eyes, either of a cornea or dark pigment mass. MEMOIRS OF THE NATIONAL ACADHMY OF SCIENCES. 51 Chelicerze slenderer than in P. flavescens, the fingers rather longer; the movable finger with about twelve teeth on the inner edge, the thumb with six obtuse teeth. Pedipalps slender; the coxa and basal joints much as in P. flavescens, but the spine on the inner side of the first joint twice as long; second joint (female) scarcely thicker than the first, inner edge not so serrulate as in P. flavescens, the two large spines twice as long as the joint is thick; the third joint a little more than half as long as the second, with two long spines on one side and two unequal much smaller ones on the other; fourth joint but slightly longer than the fifth, with two large long . spines on each side; fifth and last joint not so thick as the fourth, only two long spines on each side, the terminal spine two-thirds as long as the joint itself. Ali the legs remarkably long and slender, eight-jointed ; the tarsus five-jointed. First pair the shortest, not quite so long as third pair; second pair about one-third longer than the first, with remarkably attenuated tarsi, with a single weak claw; length, 18™™; third pair with the coxal joint larger than in any of the other pairs (first and second, and even the fourth), and considerably shorter than the fourth; fourth pair of legs very long, but considerably shorter than the second pair, with six tarsal joints, the last somewhat swollen, considerably longer than the penultimate joint, and bearing two rather large, long, slender claws, and three long, large sete. ' ; The ovipositor* is a very large organ, being 1.5™™ in length, or more than a third as long as the body, and is as thick as the chelicere or the third pair of coxe; it projects forward when fully protruded (at least in alcoholic specimens); it is not chitinous, is muscular, not jointed, is slightly curved, and ends abruptly, with (as seen in profile, Fig. —) about six fine sete on each side, those on the extreme side being nearest together; these are probably tactile hairs (the eggs are prob- ably laid in crevices). ; The penis differs remarkably from that of P. flavescens; it is rather thick when extended, and slightly over 1™™ in length, or about one-third as long as the body; the basal segment is not much longer than broad, the second is over three times as long as the basal segment, of uniform width, and divides at the end into two lateral slightly-curved points, and a longer, straight, acute pro- jection; the sides are setose, especially towards the end, including the lateral horns. Length of body, 3 to4™™"; breadth, 2 to 2.5™™; of chelicerz, 2™™; of pedipalps,5™™. Labyrinth and Dead Sea, Mammoth Cave (Packard); Dixon’s Cave (Packard); Martha’s Vineyard, Mammoth Cave (Hubbard); White’s Cave, several (Packard); Diamond Cave, Kentucky (Packard). Tellkampf’s figure was crude and imperfect, as the second pair of legs were wanting in his specimens, which were evidently young, while the tarsal joints of the remaining limbs were not correctly drawn; hence his description was incorrect as regards the limbs, as he says the length of the limbs differs little. Professor Cope, having specimens of this species with all the legs present, and basing his comparison with Tellkampf’s genus on the latter’s figure, copied by us in the American Naturalist, founded his genus Phrixis on the character of the “ multiarticulate tarsi,” although he says the tarsi of the ‘‘longest legs” were “not counted.” His specimens were mature. Mr. H. G. Hubbard gives an excellent figure of the female, which he incorrectly considers to be a male. The following note on this species is copied from Mr. Hubbard’s paper. Phrizis longipes Cope (Fig. 2).—In Professor Cope’s description of this species (1. c. vi., 421) some confusion occurs as to the tarsi. In the longest legs the number of joints was not counted, although they are mentioned as “ multi- articulate,” and this, with the absence of eyes, is made to characterize the genus. In the specimens before me the anterior and shortest pair of legs have five-jointed tarsi, ending in a single claw, without an opposing bristle, as given by Cope; the second and longest pair have nine tarsal joints, with a single claw; the third and fonrth pairs are intermediate in length between the first and second, they have each six- jointed tarsi and a pair of claws. The first tarsal joints in all the legs equals or exceeds the femora and tibie; the second joint, though shorter than the first, is very long. The palpi (that of the left side is omitted in the figure) have five joints and a terminal spine; the basal joint bears a single spine; the second joint has five, three below the middle, springing from the outer edge, two above, springing from the inner edge; the third joint has one on the outer and two on the inner edges; the fourth joint has two external and three internal; and the fifth, two spines on * This organ was mistaken by Mr. Hubbard in his description for the penis and figured as such. Amer. Ent., hii, 79, 52 MEMOIRS OF THE NATIONAL ACADEMY OF SCIENCES. either edge; the spines are all tipped with long bristles, bent towards and crossing those of the opposite row. The male *organ is cylindrical, without joint or median swelling, as thick as the coxe, not chitinous, bearing at tip a few fine hairs; when fully protruded it equals one-third of the body in length. The abdomen shows but four narrow and one conical terminal segment beyond the cephalothoracic shield. The conical eminence at the anterior border of the dorsum, between the first pair of legs, is without trace of ocelli. Length without appendages, 2™™=.08 inch; longest leg, 18™™ = .72 inch. Two specimens in alcohol from Martha’s Vineyard, in the Mammoth Cave. Fic. 14.—PHALANGODES ARMATA: a, claw of anterior tarsus; 6b, claws of posterior tarsi; ¢c, enlarged ten times (after Hnbbard). Remarks.—This form, though living in Mammoth and adjoining smaller caves, and sometimes occurring near daylight, seems to undergo almost no variation, either as regards the absence of any traces of eyes or the remarkable length of legs. It is the extreme in a series of forms, including Phalangodes flavescens, var. ceca, weyerensis, and the terricolous P. spinifera, with the very stout short-legged P. robusta. Without much doubt P. armata has been derived from a form like P. spinifera; or from an earlier terricolous species, from which both diverged. PHALANGODES SPINIFERA, n. sp. Plate XIII, figs. 2, 2a, 2b, 2¢ One female. Body rather broad and stout; more contracted on the sides behind the inter- ocellar spine than in P. robusta. Cephalothorax widening considerably behind the hinder edge; it is two-thirds as long as the body. The eyes are considerably smaller than in P. robusta, black, and placed unusually far apart; between them is a conical projection ending in a high, sharp, prolonged spine. There are five abdominal segments seen from above, the fifth narrow, conical minute. The surface of the cephalothorax is rough, with sharp granulations, and the hind edge, with that of the succeeding segments, are adorned with a row of sharp spines, the median ones largest, those on the fourth segment being more numerous, large, and sharp, with a group of about five large ones of unequal size on each side of the body. The chelicere are 2.5" long, much longer and slenderer than in P. robusta; but regely like those of P. armata from Mammoth Cave, the first joint being similarly contracted toward the base; * This appears to be the ovipositor of the female. MEMOIRS OF THE NATIONAL ACADEMY OF SCIENCES. 53 the second joint is a little stouter, but the fingers are larger. The pedipalps are remarkably long and slender, the terminal joint very long, and ending in a short, powerful, solid spine; the first joint is scarcely longer than the second is thick, with three inner teeth and a single stout outer one; second joint nearly as long as the third and fourth together; on the inside are two large basal teeth ending in hairs, and in the middle are three small unequal ones; third joint thicker than the second, a single distal internal seta, while the outer edge is finely denticulate; fourth joint nearly twice as long as the third, with four or five fine external spines; a stout setiferous basal internal spine and a much larger one on the distal third; fifth joint about the same length as the second, being very long, the inner edge nearly straight, the outer edge regularly curved, with no spines; on the inner are four large setiferous spines, the longest seta being somewhat longer than the joint is thick; the joint ends in a large, stout, short, solid spine; length 7™™. Legs long and slender; first pair 4.5" long, with four tarsal subjoints; second pair 9™™ long, with ~ eight small tarsal subjoints, while P. robusta has but five; fourth pair with five subjoints in the tarsus, and white in color. Color of the body dull honey-yellow; edge of the abdominal segments dusky; second pair of legs dusky, especially on the distal two-thirds; second, fourth, and fifth joints of the fourth pair dusky, except at the ends. Length of the body, including the folded chelicerz, 4™™. This in some respects remarkable species differs from all the others of the genus either in Europe or this country in the spiny body, the very long pedipalps, with their long terminal joint ending in a short stout spine, also the long interocellar spine and the remote eyes. It has long legs for a terricolous species, the second pair having eight tarsal subjoints. It approaches P. armata in the shape and length of the pedipalps and chelicerz, as well as in the many-jointed second tarsi. It is possible that the Mammoth Cave species has been derived from some such form as this. The present species was collected by us either in Key West or Tortugas, Florida, proba- bly the former locality. No note was taken as to its exact habitat. The description is introduced here because it is the only out-of-door form east of Colorado as yet known. Remarks. —Of the genus Phalangodes it may be said that while it was instituted by Te!ikampf for a single species, that inhabiting Mammoth Cave (P. armata), it is evident that this is the mast aberrant species of the genus, and that the terricolous spec:es, as well as the more robust of the cavicolous forms, should more properly be regarded as the most typical species. On the other hand, as our figures and descriptions wili show, P. armata should not be regarded as generically different from the eyed species formerly by Simon and ourselves placed under Scotolemon. We see that the differential characters are elastic and only specific, since, for example, P. flavescens has a blind variety. Simon records six French species (two additional ones occur in Spain and the other in Italy) and says: “The Phalangodes are all essentially lucifugous; the most are cavernicolous, some are terricolous, others are found simply in the mosses of thick and humid woods. The Phalangodes armata Tellkf. of the United States presents no traces of eyes; in the European species these organs are, on the contrary, clearly visible, being colored black. Authors have founded on this character the genus Scotolemon; but I have recently found a species in which the eyes are extremely reduced, deprived of pigment, and even sometimes disappearing, thus compelling us to reunite the genera Scotolemon and Phalangodes. The fineness and length of the appendages are always in relation with the atrophy of the eyes; thus, in P. armata, which is blind, the limbs attain their maximum of development, while in clavigera and terricola they remain short and more robust, but other species, such as lucasi and navarica, make exactly the passage between the two extremes.” P. navarica Simon, from a cave in the Lower Pyrenees, approaches nearest to P. armata in three of five spevimens; ‘the eyes are excessively small, punctiform, and deprived of pigment; in the two others it is impossible to distinguish them; the feet are also slenderer and longer than in the other Pyrenean Phalangodes, without, however, attaining the dimensions of those of P. armata.” It appears that P. armata is in the form of its chelicere and pedipalps related to the Floridan P. spinifera, differing mainly in the slenderer body, the longer legs, and absence of eyes and spines on the body. It is from such a form as this that P. armata may have been derived. 54 MEMOIRS OF THH NATIONAL ACADEMY OF SCIENCES. PHLEGMACERA* CAVICOLENS Pack. Plate XIV, figs. 5, 5a—5g. Phlegmacera cavicolens Pack., Amer. Naturalist, xviii, 203, Feb., 1884. Generic characters.—In this genus the body is not spiny, and is slightly compressed, much less flattened than usual, no broader than high, and the tergal as well as ventral surface is unusually convex and rounded. The cephalic plate bearing the eyes is about half as long as broad; behind the cephalic plate are two very short thoracic segments, both together not so long as the cephalic plate. The abdomen forms two-thirds of the length of the body. There are nine well-marked abdominal segments seen from above, and six short well-marked urosternites, besides the basal triangular urosternite. The chelicere are three-jointed, and the hands are bent inward somewhat asin Nemastoma. Pedipalps six-jointed, considerably longer than the body, the joints simple, not spiny; the fifth joint longer than the others, much swollen; the sixth oval, simple, not spiny. Second pair of legs (probably) ending in multiarticalate whiplash-like tarsi. One pair of legs, either first or third, with undivided tarsi. This genus does not approach very near any of the European genera, such as Liobunum, Megabunus, Oligolophus, Acantholophus, ete. It approaches Prosalpia most in the form of the body, especially in the relations of the cephalic plate to the abdomen and the size of the chelicerx; but differs in the pedipalps being simple, while the first and third pair of legs are probably quite different in the undivided tarsal joint. It appears to belong to Simon’s subfamily Phalangiine, but has no very close affini ties to any of the European genera. Specific characters.—Body dark brown; appendages of a pale horn color; cephalic plate between one-third and one-half as long as broad. Hyes large, prominent, contiguous, scarcely situated on an eminence; they are black and well developed. The abdominal segments above with numerous scattered dark granulations, which become larger dorsally ; a series of large, short, but broad, dorsal transverse blackish discolorations ; a broad, dusky, lateral, diffuse band low down on the sides of the tergal sclerites next to the upper edge of the urosternites. Chelicere pale horn color, black at the tips of the fingers. Second joint moderately long, equal to the hand of the third joint in length; manus rather thick, oval in outline; the outer suriace with numerous fine sete ; the fingers very unequal, the outer or movable one about two-thirds as long as the manus, a good deal curved, with a.single tooth near the end, and a series of about twenty-three or twenty- four separate, stiff, straight, even sets, corresponding to the serrula in Phalangodes ; inner finger (thumb) straight, not much over half as long as the other finger, with two or three teeth near the tip, and along the inner edge a sinuous series of small sete of unequal length, which ends at the innermost tooth. The pedipalps are from one-fourth to one-third longer than the body ; the first joint is as thick as long; second twice as long as thick; third twice as long as second and not so thick ; fourth not so long as third but considerably thicker; fifth longer than any of the other joints and much swollen, oval in form; sixth no longer than fifth is wide, and obtuse at the tip, contracted at the base; all beyond the basal joints densely and finely setose. Of the legs, which were unfortunately detached from the specimen, two were observed; what were perhaps the first pair are five-jointed, the basal joint minute, the second and fourth of equal length, the third not being much longer than second is thick at base, while the tarsus is long and slender, tapering to the minute claw; second leg (?) very long, first joint very small and short, second shorter than fourth, the third between one-third and one half as long as the second, the fourth, with joint five, divided at the end into nine minute joints, and the last joint (joint six) iste divided into twelve joints, the last being equal in length to the four preceding, and bearing a single minute claw. Length of body, 4™™; thickness, 2.5™™; width, 2™™. Bat Cave, Carter county, Kentucky. (Packard.) Two siadnens. Family NEMASTOMATID A. NEMASTOMA TROGLODYTES Pack. Plate XIV, figs. 3, 3a, 3b, 3b’. Nemastoma troglodytes Pack., Bull. Hayden U.S. Geol. Surv. Terr., iii, 160, 1877. Ten females. Body rather long and slender compared with the European N. dentipalpis Koch, the latter being short and ovate, while our species is contracted at the base of the abdomen. The eye-tubercle is rather large and prominent; the eyes themselves well developed, black in recently- * Nov. genus; ®Aeypazos, inflated; Képas, horn or feeler. MEMOIRS OF THE NATIONAL ACADEMY OF SCIENCES. 55 molted specimens, but in others scarcely distinguishable from the dark-brown, finely-shagreened tegument. Behind the eyes the body contracts dorsally, as well as laterally. On the front edge of the cephalothorax is an acute median spine. The six basal abdominal joints are coalesced, forming a single piece, segments 3 to 6 being indicated by a pair of somewhat transverse, high, well-marked tubercles (not form- ing true spines as in NV. dentipalpe). The four terminal segments are free; the terminal one subtriangular, one fourth shorter than wide. Beneath are seven well-marked sterna, with lunate, dark spiracles on the sternum of the second segment. Chelicerze (Plate XIV, fig. 3) hairy, with the basal joint not so long as broad; second joint of the same width throughout, not swollen toward the end; third joint bent downward and inward at right angles, the hand directed a little outward; the movable finger as long as the hand is thick. Pedi- palps (Plate XIV, fig. 3a) very long and slender, hairy, nearly twice as long as the body, while in the European dentipalpe they are scarcely half as long in proportion; six-jointed (in dentipalpe five-jointed), the basal joint sub- triangular in outline, owing to the upper edge being dilated; second a little longer and much slenderer than first, and slightly curved; third a little more than twice as long as the second, very slender; fourth a little shorter than third; fifth three-fourths shorter than fourth; and sixth' slightly shorter than the second, rounded at the end, being cylindrical, ' ovate, and unarmed, though with rather stiff hairs. Legs much longer and slenderer than in J. dentipalpe, with all the coxe of nearly the same size, the hinder pair being a little shorter and broader. First pair about twice as long as the body, with eight tarsal joints ; joints four to seven, together a little longer than the terminal one; a single long, stout, curved claw. Second pair nearly three times as long as the body; length, 4™™; tarsi very long and sinuous, like a whip-lash, the last joint divided into nine subjoints; the claw rather feeble; the second joint half Sayingcl “wmoten aay as long as the first. Third pair of legs of the same length as the first pair; tarsi eight-jointed, the two terminal joints subdivided into two joints. Fourth pair nearly three times as long as the body; tarsi eight-jointed, the two last sometimes subdivided into two sub- joints (internodes) Length, 3™™. Found under stones on the bottom of Clinton’s Cave, Lake Point, Utah, in a damp place, not infrequent, July 28,1875. Quite active in its movements. Most of the specimens were apparently distended with eggs. This is the first occurrence of the genus in America. I have been able in drawing up the above description to compare our species with the European Nemastoma dentipalpe of Ausserer, a specimen of which was kindly loaned me for the purpose by Mr. J. H. Emerton. It differs from its European congener by the pedipalps being twice as long, while the tarsal joints of the three hinder pairs of feet are much fewer in number, there being twenty-four well-marked ones on the second pair of legs of N. dentipalpe, while the fifth joint of the leg (including the coxa) is subdi- vided in dentipalpe into thirteen subjoints, these divisions in WV. troglodytes not being well marked. From the European WN. bimaculatum (Fabr.), French specimens of which have been kindly loaned by Mr. J. H. Emerton, our species differs in the body being much narrower and slenderer, while the maxillze and legs are much longer, the tarsi especially being much slenderer and the joints very much less distinct. The back of N. bimaculatum is not tuberculated. The European N. dentipalpe lives in moss in woods, while N. lugubre (Muller=N. bimaculatum Fabr.) is, according to Simon, common in moss and detritus, etc. The effects of a cave life on the American species is seen in the very long palpi and legs and the indistinct sub-joints. NEMASTOMA INOPS Pack. Plate XIV, figs. 4, 4a—4c’. Nemastoma inops Pack., Amer. Naturalist, xviii, 203, February, 184. Body of the usual form, in general like the European N. lugubre and our N. troglodytes, the abdominal segments being nearly the same, but as the specimen is immature it is not chitinized ; 56 MEMOIRS OF THE NATIONAL ACADEMY OF SCIENCES. it is oval and somewhat flattened. The body is white, the appendages being slightly dusky. The eye prominence is rather large compared with that of NV. troglodytes Pack. The eyes are wanting, the pigment. being colorless, but with a dark line indicating the traces of a retina. Chelicere slender, rather long, the inner edge of each finger with short, stiff setae; om the upper side, at the base of the immovable finger, are two straight, stiff hairs; the hand is not setose, as in that of NV. troglodytes. The pedipalps are only of moderate length, being, in proportion, only about one- third as long as those of WV. troglodytes ; the second joint is not much longer than the basal, being slightly longer than thick; third joint three times as-long as the second; third and fourth of the same length, but the fourth a little thicker; fifth slightly longer and thicker than fourth, with numerous stout setz of nearly even length; sixth (terminal) two-thirds as long as the fifth and nearly as long in proportion as the terminal joint in NV. troglodytes ; it-is very setose and the tip is rounded. Legs of second pair 3™™ in length, hairy, last tarsal joint undivided; fourth pair 4™™ in length, the last tarsal joint with nine subjoints, and the ungues smaller than in the second pair. Length of the body, including the cheliceres, 1=™ Locality, Bat Cave, Carter county, Kentucky. The specimens found were immature, but the species is so characteristic that I have ventured to describe it. It differs from the Utah N. troglodytes chiefly in the much shorter pedipalps, with proportionally much shorter joints in its naked hand and much slenderer legs. The specific name is given it in allusion. to its feebly-developed, degenerate eyes. It is the first species of the genus known to occur in the eastern United States. ARANHEINA. It isin the small caverns of Carter county, Kentucky, and the two Weyer caves (Weyer’s and the adjoining Cave of the Fountains), which are often but a few (less perhaps than a hun- dred) feet below the surface, that the variation and number of species of spiders is greatest. In each set of caves there are three species to one in Mammoth and Wyandotte caves. The individual variation was the greatest in Nesticus pallidus, and, aS might be suspected, in the eyes. The degree of variation is indicated in Mr. Emerton’s description. The spiders oceurred more abundantly in all the caves than we expected. The individual abundance was greater in the smaller caverns, especially the Weyer caves, than any others. In the Mammoth Cave the Anthrobia occurred under stones in dry, but not the driest, places on the bottom at different points in the cave. Sometimes two or three cocoons would be found under a stone as large as a man’s head. ‘The cocoons were orbicular, flattened, an eighth of an inch in - diameter, and formed of fine silk, and contained from two to five eggs. They occurred with eggs in which the blastodermic cells were just formed April 25. The eggs were few in number and. seemed large for so small a spider, being 7$3>5 inch in diameter. The chorion is very thin and finely speckled. ‘The blastodermic cells seemed very large, the largest measuring nearly 73455 inch in diameter. They were round, not closely packed, and showing no indications of being polygonal. They all had a dark, very distinct nucleus. I was unable to trace the developinent of the young and ascertain if the embryos are provided with rudimentary eyes. Two young Anthrobie hatched out May 3in my room. The whole body, including the legs, is snow-white, with the legs much shorter than in the adult. The adult in life is white, tinged with a very faint flesh color, with the abdomen reddish. In some specimens the abdomen has beneath several large transverse dusky bands. The Linyphia subterranea, as observed living in Wyandotte Cave, is pale pinkish, horn- brown on the thorax and legs, while the abdomen is dull honey-yellow. What constitutes the food of these diminutive, weak, sedentary spiders I can not conjecture, unless it be certain minute delicate mites or young Podure. They spin no web, though some of the spider’s in Weyer’s Cave (Cave of the Fountains) do spin a weak, irregular web, consisting of a few threads. The Sciarz and Chironomus are too large and bulky to be captured by them. The probable insufficiency of food as well as light may account for their small size and feeble repro- ductive powers. The individuals were far less numerous than those of the Phalangodes and Chernetide. MEMOIRS OF THE NATIONAL ACADEMY OF SCIENCES. 57 NOTES ON SPIDERS FROM CAVES IN KENTUCKY, VIRGINIA, AND INDIANA. The. following remarks and descriptions, by J. H. Emerton, are copied from the American Naturalist, vol. ix, May, 1875: The collection of cave spiders contained about one hundred specimens of eleven species. Two species were found only about the mouths of caves. These are Theridion vulgare Hentz, a spider found all over the country in shady places, and Meta menardi, which has been found in similar situations in Massachusetts and New Hampshire, and resembles Epeira fusca Blackwall. One young spider allied to Tegenaria was taken in Fountain Cave, Virginia, and four specimens of a species of the same family were found in small caves in Carter county, Kentucky; all were immature except one female, and none showed any subterranean characters. The remaining six species, all belong- ing to the Theridioid, were found in considerable numbers in the larger caves, where there is little or no light and the climate is little affected by outside changes. One species of Linyphia from Weyer’s Cave, Virginia, has the eyes of the normal size and number, and the colors and markipvgs of some specimens are as bright as on spiders of the same _family living in cellars or shady woeds. The other five species are all pale in color and show some unusual condition of the eyes, three species having the front middle pair very small, one having all the eyessmall and colorless, with the front middle pair wanting in the males and some females, and one species being entirely without eyes. Follow- ing are descriptions of the last six species: Nesticus pallidus n. sp. (P1. XV, figs. 22 to 27).—Cephalothorax and legs pale orange-brown, abdomen yellowish- white, with brown hairs. Length of female 3.5™™. Cephalothorax 1.5™™ long and nearly as broad, little elevated in front; three lines of hairs from the eyes to the dorsal pit. Front middle eyes black and half as large as the others, nearly touching each other; rear middle eyes separated from each other by their diameter and from the front middle eyes by half that distance; lateral eyes in pairs, separated from the middle eyes by half their diameter. Mandibles half as long as the cephalothorax. Maxille and labium short and wide. Palpal claw long and slender, with six teeth along the middle. Legs 1, 4, 2, 3; first pair, 10™™; second, 8.25™™; third, 8.15™™; fourth, 9.6™™; thinly coy- ered with long hairs and without spines. Tarsal claws long and slender, the lower with two teeth, the upper with nine orten. Epigynum (Fig. 27); the sacs showing through the skin in some specimens. The only male taken had not finished molting and was much distorted by the alcohol. The palpus which had cast its skin is shown in Fig. 26; the penis is raised from its natural position, which is in a groove passing spirally round the end of the palpal organ to a fleshy conductor. A long process, with two teeth at the end, branches from the base of the tarsus. Fountain Cave, next to-Weyer’s, Virginia, among stalactites where there was no daylight. Several loose cocoons were found, one containing thirty or forty young just hatched (Packard). Nesticus carteri n. sp. (Pl. XV, fig. 28).—Cephalothorax and legs light yellow, hairs shorter than in N. pallidus. Abdomen in some specimens with indistinct gray markings. Eyes smaller and farther separated from each other than in N. pallidus. Epigynum (Fig.28). This species is otherwise much like N. pallidus. Bat Cave, Zwingle’s Cave, Carter county, Kentucky (Packard). A cocoon collected by Mr. Packard from Bradford Cave, Indiana, contains young which had passed their second molt, probably of this species. LTinyphia sublerranea n. sp. (Pl. XV, figs. 29 to 31).—Cepbalothorax and legs yellowish-brown ; in some specimens reddish. Abdomen white, with brown hairs; in two specimens from Zwipvgle’s Cave gray, with white spots. Eyes eight (Fig. 30), white, surrounded. by a dark border; in one specimen colorless, without dark borders. Front middle eyes very small, and in the two dark specimens from Zwingle’s Cave obscured by-dark markings on the head. Mandibles with seven teeth in front of the claw-grooves. Legs short 1, 4, 2, 3; spines on patella and tibia. Under claw of tarsus with two teeth, the upper claws with eight or nine; no claw on palpi. Epigynum external, as long as the maxillw#, extending backward along the under side of the abdomen (Figs. 29 to 31), or when the abdomen is distended projecting out from it at a right angle. Under stones in Carter and Wyandotte caves (Packard). Linyphia weyeri (P|. XV, figs. 7 to 12).—- Cep halothorax and legs yellow-brown, abdomen from dark gray to white. Length of female 2.25™". Cephalothorax wide and but little elevated in front in eithersex. Front middle eyes near each other on a black spot, rear middle eyes separated by their diameter and by the same distance from the front middle eyes, lateral eyes in pairs, eacly pair surrounded by a black area and distant twice its width from the widdle eyes. Mandibles long, spreading apart at the tips, and inclined backward toward the maxille, beyond the ends of which they extend a third of their length in the female and farther in the male; five long teeth in frout of the claw- groove. No palpal claw. Legs 1, 4, 2, 3; first pair 4™™ long in female and 4.4™™ in male, with two spines on femur, one on patella, and two on tibia. Under claw of tarsus with one tooth, upper claws with nine or ten teeth. Epigynum with an oval opening behind twice as wide as long, in front of which is a short flexible appendage (Fig. 11). Palpusof male, Figs.9and 10, The tarsal process is a small hook on the upper side; the penis is long, and passes one aud a half times around the palpal organ, supported through nearly its whole length by a wide thin conductor ending in ahard tooth. Under the end of the penis is a soft brush-like appendage, and beside it two hard processes. Weyer’s Cave, Virginia, in darkness, but not far from the entrance (Packard). Linyphia incerta n. sp. (Pl. XV, figs. 13 to 21).—Length 2™™. Cephalothorax and legs orange-brown, abdomen white, with short, fine, brown hairs. Cephalothorax 1™™ long and two-thirds as wide; in the male elevated in front (Fig. 20) and furnished with longer hairs than in the female. Eyes small and colorless, and separated far from each other (Figs. 13,and 21); the front middle pair are very small, hardly larger than the circles around the bases of the hair by which they are snrronnded, and only distingnished from them by wanting the dark rim which surrounds the hair circles. In five females from Fountain Cave all the eyes are present (Fig. 18); in one female one of the front middle eyes is wanting. In three males from the same cave both front middle eyes are wanting, asin Fig. 21; in 58 MEMOIRS OF THE NATIONAL ACADEMY OF SCIENCKS. one male one only of the front middle pair is wanting. In four females and one male from Bat Cave, Carter county, Kentucky, the front middle eyes are wanting. Mandibles long and spreading at the tips, inclined backward toward the maxillw, seven teeth in front of the claw groove, which are longer in the males. No palpal claws. Legs 1, 4, 2,3; longest 4.75™™, Tarsal claws short and slender, under claw with one tooth, upper claws with seven or eight teeth. Spines on patella and tibia. Epigynum with a small oval opening behind, with dark brown border. Palpus of male (Fig. 17), having a sharply-curved process at the base of the tarsus. The penis is supported by a stout con- ductor nearly to its end, where it passes a soft brush-like appendage. Fountain Cave, Virginia, among stalactites, in company with Nesticus pallidus (Packard); also in Bat Cave, Carter county, Kentucky (Shaler and Packard). Anthrobia mammouthia (Pl. XV, figs. 1 to 6).—In 1844, Tellkampf described and figured roughly in Wiegmann’s Archiv fur Naturgeschichte several Arthropods from the Mammoth Cave; among them an eyeless spider which he referred with doubt to the Mygalidx, apparently because he saw only four spinnerets. The eyeless spiders, found by Dr. Packard in the Mammoth Cave in 1874 agree generally with Tellkampf’s description, and his Fig. 13 represents quite well the outline of a specimen flattened by pressure between two glasses. No other eyeless spider was found, and no other which could be identified with Tellkampi’s description. There seems, therefore, little doubt that these are spiders of the same species as those described by Tellkampf. Adults, 1.5™™ long, pale brownish-yellow ; abdomen almost white, with brown hairs; ends of palpi, palpal organs, and epigynum reddish brown. Cephalothorax with scattered hairs in front. Noeyes. Mandibles with four long teeth in front of the claw groove. Maxille short and wide. Sternum wide and hairy. Legs 1, 4, 2,3; longest about 2.5™™, hairy, with spines on patella and tibia. Under tarsal claw with one tooth, the upper claws with six or more short teeth. No palpal claw. Palpus of male (Fig. 3) with a long process on the outside of the tibia, ending in a sharp hook. The tarsal process forms a small thin hook. Palpal organ very simple; the penis very short, and accompanied by a soft, thin appendage. Spinneretsshort; hypopygium one-third the length of the first pair. Mammoth Cave and Proctor’s and Diamond caves, under stones (Sanborn and Packard). Small flat cocoons were found with some specimens, containing small numbers of eggs, which were unusually large in proportion to the size of the spider. [In this connection it may be of interest to learn the opinion of Dr. T. Thorell, the accomplished arachnologist of Upsala. Upon receiving a specimen of Anthrobia mammouthia which I sent him he writes me that ‘‘ the Anthro- bia, if it really is the true A. mammouthia Tellkampf, scarcely differs from the genus Hrigone by anything more than the want of eyes; it may, however, be added as a peculiarity, that the three long and slender tarsal claws are quite smooth, neither dentated nor pectinated. The species belong most certainly to the family Theridioide.” On the other hand, on the receipt of a specimen of the same species of spider and from the same cave (Mam- moth) as that from which the specimen was taken which was sent Dr. Thorell, M. Simon, of Paris, writes me that “the Anthrobia is not allied to Mygalide, as was supposed from the imperfect description of Tellkampf, but to our Dysderidz, and the genus Leptoneta, only it is blind.”—A. S. P.] 5 The following additional species have also been described by Count Keyserling, as I am informed by Dr. G. Marx: Willibaldia cavernicola Keys (closely allied to Linyphia incerta Kmer.): Phanetta subterranea Keys (=Linyphia subterranca Emer.); Hrigone infernalis Keys. The descrip- tions of the two last species appeared in Die Spinnen Amerika’s Theridiide, 2te helfte, 1886, pp. 125 and 180. JLiocranoides unicolor Keys, from Mammoth Cave, should also be added. MYRIOPODA. In order to facilitate the identification of the cave Myriopods, all of which, except Cambala annulata, belong to the Lysiopetalide, I give the descriptions of the genera and species from my “Revision of the Lysiopetalide” in the Proceedings of the American Philosophical Society, xxi, 177, September 15, 1883. LYSIOPETALUM Brandt. Julus Say, Journ. Acad. Nat. Sci., Phil., ii, part i, 104, 1821. Lysiopetalum Brandt, Recueil, 42, 1840. Spirostrephon Brandt, Bull. Sci. Acad., 1841; St. Petersb., 1840. Recueil, p.90, 1840. Platops Newport, Ann. and Mag. Nat. Hist., xiii, 266, 1844. Lysiopeialum Gervais (in part), Aptéres, iv, 133, 1847. Cambala Gervais, Aptéres, iv, 134, 1847. Exped.a& Amer. du Sad ACastelneaw); Myriop., 17. Reasia Sager, Proc. Acad. Nat. Sci. Phil., 109, 1856. Spirostrephon Wood, Myriop. N. Amer., Trans. Amer. Phil. Soc., 192, 1865. Spirostrephon Cornet Proc. Amer. Phil. isoen 179, 1869. Spirostrephon Ryder, Proc. U. S. Nat. Mus., iii, 526, 1881. Not Cambala Gray, Griffiths, Cuvier An. King. Ins., Pl. 135, fig. 2, 1832. Not Reasia Gray. Not Reasia Jones, Todd’s Cyc. Anat. Art. Myriop., 546. Body segments numbering as many as upwards of sixty, with as many as one hundred and fifteen pairs of legs; the body unusually long and slender, tapering gradually towards the subacute tip. Head with the front flat, high, and narrow, more so than usual; the eyes in a rectangular triangle, composed of as many as forty to forty-one MEMOIRS OF THE NATIONAL ACADEMY OF SCIENCES. 59 facets, and not depressed. Antennx rather long, the joints subclavate, joint 6 not much longer than 4; joints 3 and d of the same length ; joint 6 rather thick at the end; joint 7 short, thick, and conical, much more so than usual. Body segments swollen and full, becoming saddenly depressed on the front edge; the swollen portion with numerous raised lines or ridges, with deep concave valleys between; the ridges projecting behind in an acute point. The segment next to the head rather narrower than the head, with the posterior two-thirds ridged; the sides of the segments are somewhat swollen high up on the sides, but not so conspicuously as in Pseudotremia. Legs rather stout and larger than in Pseudotremia; the first pair rather short and broad, with a regular comb of stiff sets on the inner edge of the terminal joint. The seventh and ninth pairs of legs, i. ¢., the pair immediately preceding and following the genital armor, are like the others, not being in any way modified, as in Pseudotremia, etc. The genital armature is large and better developed than in any other genus of the family; the outer lamina large, stout, spatulate-mucronate at the tip; inner lamina much shorter than the outer, and with two long acute forks; repugnatorial pores difficult to find. The genus may be recognized by the long, slender body, taperiug to a point, and by the very short conical sev- enth antennal joint; by the ribbed, swollen segments, which are very numerous; by the seventh and ninth pairs of legs being normal, like the others; and by the short, broad first pair, with the regular comb of set on the terminal joint. The genus as here defined will apply to the two southern European species Lysiopetalum carinatum Brandt and L. illyricum Latzel, except that they are setose, while our species is not. Iam indebted to Dr. Latzel for specimens for comparison. 4 In proposing the genus Spirostrephon, Brandt (Bull. Sci. Acad., St. Petersb., 1840) regarded Say’s Julus lactarius as the type species, and adding that the eyes are in a triangular area, he indicates its generic difference from Cambala annulatus, with which it has been so often confounded. Although I had originally retained Brandt’s name Spirostrephon for our species, yet upon receiving from Dr. Latzel authentic types of European Lysiopetalum, it is plain that our S. lactarius is congeneric with them. The name Spirostrephon should, then, be considered asa synonym of Lysiopetalum. It is difficult to see why Brandt should have separated lactarius from his LZ. carinatum. In his Recueil, page 42, Brandt thus characterizes his genus Lysiopetalum: ‘“‘Laminzw pedifere omnes libera, mobiles, cutis ope cum parte abdominali corporis cingulorum conujuncte. Frons ante antennas dilatata et deplanata in maribus in simul depressa.” The two species mentioned under the generic diagnosis are Lysiopetalum fatidissimum (Savi) and L, carinaium Brandt. Again, on page 90: ““Subgenus seu genus II, Spirostrephon Nob.” is thus characterized, and he apparently regards it as a subgenus of Julus: ‘‘Gnathochilarii pars media fossa haud instructa, sede jus loco aream tetragonam planam, plica seu linea derata duplici, superiore breviore et inferiore longiore, supra et infra terminatam, sed sutura longi- tudinali haud divisant offerens. Spec. 27. Julus (Spirostrephon) lactarius Nob.----- Differt habitu a Julis genuinis et Julo (Lysiopetalo) feetidissimo et plicato affinis apparet. Annuli corporis, quorum posteriores brevissimi, incluso anali53. Pedumparia95. Longitudo 10 toil; latitudo summa 3". @culi triangulares—Julum lactarium protypo generis Cambala Grayi habuissem, quum figura ab hocce zoologo sub nomine Cambale lactarii data” (Griffith Anim. Kingd. Insect., Pl. 135, fig. 2). The generic characters are not very applicable in distinguishing the genus, the men- tion of the type alone rendering it possible to understand what the genus is. The synonym will be further discussed under Cambala. In 1844 Newy:ort, having been misled by the specimen of Cambala annulata alleged to have been sent by Say as the type of his Julus lactarius, places the latter in his genus Platops, which he proposes, with a doubt, thus: ‘“‘Genus Platops? mihi.” The generic characters apply well to the present species, S. lactarius. Dr. Wood, in his Myriopoda of North America, does not attempt, for want of material, to define the genus. Professor Cope characterizes this and the next genus thus: PAU ULLEH MMVI CLOG DOMES Ere et eset caia\e eis iaie one Sie se cia kee cars als wsieisarseceeacecisce Spirostrephon. Annuli with two pores on each side the median line ..............-.-...-.---- Pseudotremia. As we have seen, there are pores in Lysiopetalum, while the ‘‘two pores” of Pseudotremia are two of the three setiferous tubercles on the side of each segment. The genus appears thus far to be represented in North America by but a single species, which ranges from Mas- sachusetts west fo lowa and south to Florida and Louisiana, while in southeastern Europe Lysiopetalum is rich in - species. LYSIOPETALUM LACTARIUM Say. Plate IX, figs. 3, 3a—3h. Julus lactarius Say, Journ. Acad. Nat. Sci., Phil., ii, pt. i, 104, 1821. Spirostrephon lactarius Brandt, Bull. Sci. St. Petersb., 1840. Recueil, 90, 1840. Platops lineata Newport, Ann. Mag. Nat. Hist., xiii, 267, April, 1844. Lysiopetalum lineatum Gervais, Aptéres, iv, 133, 1847. Cambala lactarius Gervais (in part), Apteres, iv, 134, 1847. Reasia spinosa Sager, Proc. Acad. Nat. Sci. Phil., 109, 1856. Cambala lactaria Gervais, Exped. |’Amer. du Sud (Castelneau), Myriop., 17. “‘Reana chinosa Saeger,” Gervais, Exped. l’Amer. du Sud (Castelneau), Myriop., 14. Spirostrephon lactarius Wood, Myriop. N. Amer., Trans. Amer. Phil. Soc., Phil., Pl. ii, figs. 11, 11a, 192, 1865. 60 MEMOIRS OF THE NATIONAL ACADEMY OF SCIENCES. Spirostrephon lactarius Cope, Proc. Amer. Phil. Soc., Phil., xi, No. 82, 179. 1869. Trans. Amer. Ent. Soc., lil, 66, May, 1870. ISpirostrephon lactarius Ryder, Proc. U. 8. Nat. Mus., iii, 526, Feb. 16, 1881. Lysiopetalum lactarium Packard, Amer. Nat., xvii, 555, May, 1383. Not Cambala lactaria Gray, Griff., Cuvier An. King. Ins., Pl. 135, fig. 2, 1832. Not Cambala lactaria Newport, Ann. Mag. Nat. Hist., xiii, 266, April, 1844. Two males, two females. Body segments, exclusive of the head, 61, with 115 pairs of legs. Body and head horn color, usually mottled and banded with dark blackish horn color. The head usually with a broad, interantennal, black, conspicuous band inclosing and connecting the eyes. Hyes (compound) of 40 to 41 facets. Antenne dull, blackish brown; tip of the terminal joint pale, as also the other joints at their articulation. The body with a median dull yellowish dorsal stripe, and with a lateral row of concolorous diffuse spots, one on each longest lateral ridge (the spots vary much, sometimes covering four or five ridges and. extending low down on the sides of the seute. Each scute has, except those near the head and at the end of the body, about twenty-five prominent ridges, the dorsal twelve larger than those on the sides; these ridges are high, with concave valleys between them; the end of the ridges are acutely conical and project ones the ends of the scutes. Length of the entire body, 35™™; thickness, 2™™, : The above description was drawn up from the Louisiana specimens, which were highly colored, banded, and spotted. In the Massachusetts specimen the color is uniformly light brown, without the yellowish dorsal line and the lateral spots. The antennz are much darker, while the legs are paler than the body.- The head is much paler than the body; it is dusky on the vertex between the eyes, but there is no definite pean ene ane band asin the Louisiana examples. The Iowa specimens resemble in coloration those from Louisiana, but the yellowish dorsal band and lateral spots are not quite so distinct, though the interantennal blackish band is distinct. Massachusetts and McGregor, Iowa. Mus. Agricultural Department, Washington, D. C. (Prof. C. V. Riley); Palatka, Florida, and Milliken’s Bend, Louisiana (KE. Burgess); ‘‘ Eastern United States” (Wood); found under bark in the mountain regions of Tennessee and North Carolina (Cope),; Saint Louis (Theo. Pergande). Although this species is evidently the parent form of the cave-inhabiting Pseudotremia cavernarum, it has not yet been observed near the Indiana and Kentucky caves, though undoubtedly yet to be found in their vicinity, as it is a wide-spread species. It probably ranges through Central into South America. As Dr. Wood remarks: ‘‘I have seen a single specimen, a female, labeled as coming from New Grenada, which apparently belongs to this species.” This specimen I have seen in the museum of the Philadelphia Academy of Natural Sciences, but did not compare it closely with our species; it is much larger than individuals from the United States. PSEUDOTREMIA Cope. Pseudotremia Cope, Proc. Amer, Phil. Soc:, xi, No. 82, 179, 1869. Trans. Amer. Ent. Soc., iii, 67, May, 1870. Spirostrephon Cope, Amer. Naturalist, vi, 414, July, 1872. Pseudotremia Harger, Amer. Journ. Sci. and Arts, iv, August, 1872. Pseudotremia Ryder, Proc. U.S. Nat. Mus., iii., 524, Feb. 16, 1881. Body consisting of thirty segments; rather long and slender, with as many as fifty pairs of legs. Head with the muscular area (gena) behind the eye very full and swollen, globose, swelling out far beyond the side of the suc- ceeding scutum; front a little longer than wide. Eyes present, black ; the outline of the eye-patch narrow triangular, composed of about twelve to fifteen facets, arranged in four or five transverse oblique series. Antennze longer and slenderer than in any of the other genera of the family ; joint 3 is twice as long but not as thick as joint 2, bnt equals fifth in length, the latter, however, being very slender and clavate; the terminal seventh joint is unusually long, pear-shaped, and elongated towards the tip. The body constricts in a neck-like fashion behind the head; segments (scuta) 5 to 20 especially have a lateral shoulder or raised portion characteristic of the genus Lysiopetalum; this swollen portion has on each side about six longitudinal ridges, with deep valleys between; above, especiaily on the posterior half of the body, the dorsal portion of the laterally-swollen scuta is coarsely tuberculated instead of ridged, and the rounded tubercles are rather flat ~ and unequal in size. There are no set or lateral setiferous tubercles. The end of the body is as usual in the family, * the last segment with three pairs of small sete arranged one above the other. Above the middle of the side of the posterior scuta, especially the last 6, is a tubercle like those in Scoterpes and Zygonopus, but much smaller, from which a minute hair arises, and above, on the upper part of the shoulder, there are two rudimentary very smnall tubercles. The legs are long and slender, about one-third longer than the diameter of the body. Inthe male the eighth paic of legs are much less modified than in the succeeding genera; it consists of five joints, while in Trichopetalum, Scoterpes, and Zygonopus it is very rudimentary, consisting of but two joints. The basal joint is large and con- stricted near the middie, with a large setiferous tubercle on the inside; the constriction may represent an obsolete articulation, and thus the basal joint really represents the two basal joints of the other legs. The smaller multiarticu- late extremity of the:leg is composed of four well-marked joints, the basal as long as the three terminal ones without the claw, which is long and slender aud nearly as well developed as in the other legs. The male genital armature is well developed, nearly as much so as in the Julide. There is a median very long curved forked chitinous rod, a pair of median boot-shaped pieces, and a pair of lateral double blades or pseudorhab- o~ a! eee. ar MEMOIRS OF THE NATIONAL ACADEMY OF SCIENCES. 61 dites, composed of the usual lamina externa and lamina interna, which are variously spined and denticulated at their extremities, one supplementary spine being minutely and densely spinulated. The genus was characterized by Cope thus: ‘‘Annuli with two pores on each side the median line.” As already remarked, the so-called pores appear to be simply the lateral tubercles, giving rise posteriorly to minute sete, which are difficult to detect with a half-inch objective. The genus differs from Lysiopetalum in the slenderer, longer antennw, the rudimentary eyes, the more swollen and prominent lateral bosses or shoulders of the segments, while the body has about half as many segments as in Lysiopetalum, and is much shorter and more fusiform. The generic characters are very marked, though the species is clearly enough derived from the common out-of-door Lysiopetalum lactarium. PSEUDOTREMIA CAVERNARUM Cope. Plate VI, figs. 1, la-lu; 2, 2a—2e. Pseudotremia cavernarum Cope, Proc. Amer. Phil. Soc., xi, No. 82, 179, 1869. Pseudotremia cavernarum Cope, Trans. Amer. Ent. Soc., iii, 67, May, 1870. Pseudotremia cavernarum Packard, Amer. Naturalist, v, 749, December, 1871. Spirostrephon cavernarum Cope, Amer. Naturalist, vi, 414, July, 1872. Spirostrephon (Pseudotremia) cavernarum Harger, Amer. Journ. Sci. and Arts, iv, 118, 119, August, 1872. Pseudotremia cavernarum Ryder, Proc. U.S. Nat. Mus., iii, 526, February 16, 1881. Eyes black, conspicuous, forming a somewhat irregular, narrow triangular patch, with from twelve to fifteen facets. Antennz unusually long and slender, the joints pilose; joints 3 and 5 of the same length, or 3 a little longer; joints 2 and 6 of equal length; joint 7 elongate, pear-shaped, pilose, the extremity truncated, with two or three sense-setz uot so long as the end of the joint is thick. The first sentum next to the head is scutellate in shape, rounded on the front edge, somewhat produced anteriorly in the middle; the margin behind slightly sinuous; it is about two-thirds as long as broad. The second scutum is a little wider than the first, the third somewhat wider, while the fourth is much wider; dorsal face of first scatum smooth; the posterior part of the second scutum a little swollen; that of the third more so; that of’fourth scutum swollen and ridged much as in fifth and succeeding scuta. Scuta 5 to 20 are swollen high up on the sides into a shoulder, giving a quadrilateral instead of a circular outline to the segment, bulging out more subdorsally than below; the swelling has.six longitudinal ridges, while the posterior swollen end of the scuta above, especially on the posterior half of the body, is coarsely tuberculated, the tubercles being rounded rather than flat and unequal in size. No well-marked setiferous tubercles on the side from the middle of the body to the head; but on the last six segments there are on each shoulder or scutal swelling two minute rudimentary swellings or tubercles; but in my specimens I can see no set except on the two terminal segments of the body in male and female, where, on the end of the last seuta, there is a seta arising from a basal movable joint; there are three pairs on the lateral anal plates (thirtieth segment). Length, 18™™; thickness of the body, 1.5™™. The young when about half grown are white, the back of the antenne and anterior segments having a very slight dusky tinge. In numerous mature specimens from the Senate Chamber, Wyandotte Cave, 3 miles in, the body is white, with a slight flesh-colored tint. In numerous (150) specimens from this locality the head and dorsal side of the anterior segments are slightly dusky; the antenne are also usually slightly dusky, except the two terminal joints, which are white. There is thus seen to be a slight amount of variation in color in specimens collected at the same date in the same chamber in Wyandotte Cave. Among the 150 specimens taken at one time and place from Wyandotte Cave (Senate Chamber) and individually exaiined I could see none without black eyes, the pigment being well developed. There was a fair proportion of males. Four specimens which I collected in Little Wyandotte Cave were of exactly the same size as those from Great Wyandotte Cave; they were white-tinged, dusky on the head and forepart of the body, the eyes are black, and the eye-patch of the same size and shape, while the antennx are the same. > Six specimens from Bradford Cave, Indiana (which is a small grotto formed by a vertical fissure in the rock, and only 300 to 400 yards deep), showed more variation than those from the two Wyandotte caves. They are of the same size and form, bet slightly longer and a little slenderer, especially joints 3 and 5; joint 7 is decidedly longer than in any others; whiter, more bleached. The antennz are much whiter thanin those from the Wyandotte caves, and the head and body are paler, more bleached out, than in most of the Wyandotte specimens. The eyes vary more than in the Wyandotte examples, one having but twelve facets, another fourteen, and another fifteen, with a few minute rudimentary facets between the others. It thus appears that the body is most bleached and the eyes the most rudi- mentary in the Bradford Cave, the smallest and most accessible, and in which, consequently, there is the most varia- tion in surroundings, temperature, access of light, and changed condition of the air. Under such circumstances as these we should naturally expect the most variation. Professor Cope’s types were first found by him in Erhart’s Cave, Montgomery county, and Spencer Run and Big Stony Creek caves, in Giles county, Virginia; also in Lost Creek Cave, on the Holston River,in Grainger connty, Tennessee, and in other limestone caves of the valley of the Tennessee. Professor Cope afterwards (Amer. Natural- ist, vi, 14) discovered this species in Wyandotte Cave, remarking: ‘‘ The species is quite distinct from that of the Mammoth Cave, and is the one I described some years ago from caves in Virginia and Tennessee,” 62 MEMOIRS OF THE NATIONAL ACADEMY OF SCIENCES. Var. carterensis. A decided approach to S. laciarius is seen in certain brown specimens, only partly bleached, found in the Carter caves, Kentucky, viz: Bat Cave, X Cave, and Zwingle’s Cave, besides a cave across the road from the hotel, which is used as an ice-house. In the specimens from Bat Cave the antenne are slightly shorter, and a little slenderer, particularly joints 3 to 5; but joint 7 is much shorter and blunter than in the Bradford Cave individuals; the antenne, however, are of the same length, though slenderer than those living in Great Wyandotte Cave. The eyes form a nearly equilaterally triangular area, with from twenty-three to twenty-five facets. The segments behind the head are 30. They differ from the Wyandotte examples in the posterior or swollen portion being rather more prominent than in the former, forming more marked lateral swellings, with about eight ridges on the side of each boss, and the body is larger and thicker, but the legs are of the same length. The head is dark in front, mottled above and below with paler horn color. The antennz are concolorous with the head and body, but the terminal joints are paler, as are the legs, which are also paler at the articulations. The entire body is dark horp-brown, mottled and irregularly lineated. The smoother anterior portion of the scuta shows a tendency to be paler than the tuberculated portion, and of a bluish-gray tint. The tubercles are no more prominent than in the Wyandotte individuals. The segments in both the Wyandotte species and var. carterensis rapidly decrease in size, the penultimate segment being pointed, and each segment is provided with regular, high-raised, parallel, prominent ridges on the shoulder or lateral boss; about forty to forty-five on a scutum on the sixth segment from the end of the body. Length, 23™™; thickness, 2.5™™; the body being considerably larger and thicker than in the Wyandotte specimens. (See figs. 2, 2a—2c.) Two specimens from X Cave are exactly in size and color like those from Bat Cave. * Three specimens from the ice-house cave only differ from those in Bat Cave in being somewhat paler, but the eyes and antenne are the same. A large and a partly grown one from Zwingle’s Cave was collected by Mr. Sanborn August 23; these were also paler than those from Bat Cave. With them were associated a Ceuthophilus, with eyes well developed, and a Polydesmus. This form or variety would be, perhaps, mistaken for Lysiopetalum lactarium, but it is true in all the generic details to Pseudotremia; at the same time it is what may be called a twilight species, living in small caves in — situations partially lighted. It is probably derived from L. lactarium, or a closely-allied species. We doubt if it will ever be found living in the same situations as L. lactarium. It is evident that the var. carterensis is the ancestral form of the now fixed species P. caver- narum, and that the former has been derived either from L. lactariwm or from an allied species; hence the genus Pseudotremia has been probably derived from the genus Lysiopetalum, and was at first represented by carterensis, the latter being a twilight species, which gave origin to the cavernarum. We regard this series as the best and clearest proof of the derivative theory which we have observed, proof so clear as to amount almost to a demonstration of species- and genus- making by a change in the environment. In Mr. Hubbard’s collection from Wyandotte Cave (Aug., 1885) I found a single carterensis considerably larger ‘than any from the Carter Caves, but the body and tips of the antenne are perhaps a shade paler. The eyes are normal, black, while the two last joints of the antenne are a little slenderer than in my types of var. carterensis, but the last joint is decidedly shorter than in the form cavernarum; the specimen is much larger than any P. cavernarum from Wyandotte Cave. Genus SCOTERPES Cope. Spirostrephon (Pseudotremia) Pack., Amer. Naturalist, v, 748, December, 1871. Scoterpes Cope, Amer. Naturalist, vi, p. 409, 414, July, 1872. Scoterpes Pack., Proc. Amer. Phil. Soc. Phila., xxi, 192, 1883. Body very long and slender, not fusiform; consisting of thirty segments besides the head, and with about fifty- two pairs of legs, with the penultimate joint very long. Head rather large and unusually broad; no eyes present ; the genz unusually large, extending high up on the vertex, but not so globose as in Trichopetalum; the front is also carried farther up on the vertex than usual, and is much broader than long; the clypeus flat, slightly bilobed on the front edge. The antennez are moderately loug and hairy, with the sixth segment scarcely longer than in Trichope- talum, but more uniform in thickness, scarcely longer than thick; the terminal joint as long as the sixth, the end conical, more produced than in Trichopetalum or Zygonopus; at the tip are four rather long sense-setw. Body seg- ments becoming, as usual, smaller next to the head; the anterior of each division of the arthromere much swollen high up on the sides; each shoulder with three tubercles, which are arranged in a scalene triangle and bear much longer set than in the other genera, though not quite so long as the body is thick. The legs are long anda slender, much more so than in Trichopetalum, and somewhat more so than in Zygonopus. In the male the eighth pair of legs are rudimentary, being two-jointed, the second joint only one-fourth longer than the basal, and ending in a well- developed stout claw. The genital armature minute and very rudimentary, pale, scarcely chitinous; the outer lamina short and thick, with a stout external recurved spine, and two terminal obtuse points; the inner lamina shorter, MEMOIRS OF THE NATIONAL ACADEMY OF SCIENCES. 63 forming a truncated angular spine, and not much more than half as long as the outer lamina; between the inner and outer lamina, its base next to the inner lamina, is a middle spine ending in an irregular tuft of fine spinules. The genus is distinguished from Trichopetalum by its want of eyes, its broader head, its long slender body, with long set, by the eighth pair of female rudimentary legs ending inaclaw. From Zygonopus it differs in the shorter sixth antennal joint, its broader head, its slenderer legs, the sixth pairin the female not being unlike the others, and by the more prominent shoulders and longer set. The species of the two genera are of the same general form and size. The genus Scoterpes was proposed by Professor Cope for the present species in the American Naturalist for July, 1872, page 414. The very brief and incomplete characters given are the “lack of eyes and of lateral pores;” the absence of the latter having been ‘‘asserted by Dr. Packard.” Ignorant of the difference between the Mammoth Cave blind Myriopod and Lysiopetalum, the latter being the ouly genus of the family then known, we referred it to the genus Spirostrephon. SCOTERPES COPEL Cope. Plate VILI, figs. 1, la—1m. Spirostrephon (Pseudotremia) copei Packard, Amer. Naturalist, v, 748. December, 1871. Scolerpes copei Cope, Amer. Naturalist, vi, 414, July, 1872. Spirestrephon copei Harger, Amer. Journ. Sci., iv, August, 1872. Spirostrephon copei Packard, Zoology, edit. 1-3, 1879-81. Scoterpes copei Pack., Proc. Amer. Phil. Soc., xxi, 193, 1883. About twenty males and females examined. Body white, with no dusky discolorations ; thirty segments besides the head in specimens 11™™ in length; and fifty-two pairs of legs; in one female individual 8™™ long there were forty-nine pairs of legs, including the eighth orrndimentary pair; in other individuals 6™™ long there were twenty-four segments behind the head. The head is provided with short, fine, erect hairs of different lengths, especially on the sides of the gene. In the absence of a second species we can not distinguish all the specific from the generic char- acters. For minor specific characters the reader is referred to the figures. The males and females are alike in size and form. The specimens were most abundant in the Labyrinth in Mammoth Cave, but also occurred in other localities in the cave. It is also common in Diamond Cave, where I collected it, and was discovered by Mr. Sanborn in Poynter’s Cave, 300 yards from daylight. In one of the specimens from the last-mentioned cave the antenn# were rather more slender than usual. The genus Scoterpes and its single species cope? appears to be limited to Mammoth Cave and the others near, in apparently the same system of caves. It was erroneously reported by me to occur in Weyer’s and the Luray caves, as the specimens collected belong to Zygonopus whitei. Without doubt the genus is a modified Trichopetalum, which has become longer and slenderer in body, with longer legs and antenne, as well as sete; whether it is a descendant of Trichopetalum lunatum or not is uncertain; it may have descended from a different species, but there seems to be no reasonable doubt but that it is a modified form of a small hairy Lysiopetaloid form, with antenne exactly like those of Trichopetalum. (For details of which genus see Plate VII, figs. 2, 2a-2f.) ZYGONOPUS Ryder. Zygonopus Ryder, Proc. U. 8. Nat. Mus., iii, 527, February 16, 1881. Zygonopus Pack., Proc. Amer. Phil. Soc., xxi, 194, 1883. . Body rather slenderer than in Scoterpes. The head differs from that of Scoterpes in being much narrower and higher, the swollen sides or gene being much less swollen; the vertex is swollen; the front as broad as long, with the upper edge a little: hollowed, but quite distinct from the vertex itself. The eyes entirely wanting, as in Scoterpes. The antenn are rather thick, and in this respect approach Scoterpes, but the sixth and seventh joints are much longer and rather more setose; the sixth joint is about two-thirds as thick as long, and the last (seventh) joint nearly twice as long as thick. The sides of the segments are swollen subdorsally as in Scoterpes, and the setiferous tubercles are arranged as in that genus, but the set are shorter; the lower posterior edges of the arthromeres below the shoulder or hump is chased obliquely with fine impressed lines. The feet are Jess in number than in Scoterpes. The diagnostic characters of the genus lie in the remarkably swollen sixth pair of feet of the male, in which the second joint is rather thick, while the third joint is long, and with the fourth joint remarkably swollen, with a series of about nine oblique retractor muscles diverging from the proximal end of the terminal joint, which is long and slender and straight, witha well-developed claw. The seventh pair of the male are of the normal form. The rudimentary or eighth pair are like those of Trichopetalum, the second (terminal) joint not ending in a claw, thus differing from those of Scoterpes. The male genital armature is entirely unlike that of Scoterpes, though it is rudimentary and minute; the outer lamina consists of a basal subtriangular portion, ending in a long, slender, curved spine, beneath which is a stouter spine, shorter and less curved; a minute median setose lamina is present, while the inner lamina is a weak, slender, setose filamentary outgrowth. Mr. Ryder’s grneric characters are stated very briefly, as folluws: ‘Sixth pair of legs very robust, and with the third joint greatly swollen.” The generic characters are not contrasted by him with those of Scoterpes. This genus differs from Scoterpes in the remarkably swollen, clasping, sixth pair of legs and in the male genital armature, while either sex differs trom Scoterpes in the much narrower head and longer sixth and seventh antennal joints, 64 MEMOIRS OF THE NATIONAL ACADEMY OF SCIENCES. ZYGONOPUS WHITEI Ryder. Plate VII, figs. 1, la—lo. Zygonopus whiter Ryder, Proc. U. S. Nat. Mus., ili, p. 527, February 16, 1881. Zygonopus whitei Pack., Proc. Amer. Phil. Soc., xxi, 194, 1883. Spirostrephon copet Pack., Amer. Naturalist, xv, 231, March, 1881. Hight males, ten females. Body white, long, and slender; number of segments, 32. Head with scattered, fine sete ; autennze with the second joint not quite one-half as long as the third, which about equals the fifth in length, both being rather long ; the sixth is thick, barrel-shaped, not quite one-half as long as the fifth, but scarcely thicker ; the seventh joint is unusually long, a little more than three-fourths as long as the sixth joint; the end thick and well rounded, with the usual tactile large flattened setw; the third to seventh joints with long, dense setz, a few on the end of joint 5 longer than any on joints 6 and 7. The set on the body arise from tubercles arranged, as usual, in a scalene triangle, and the sete themselves are half aslong as the body is thick; they are considerably shorter and finer than in Scoterpes. The number of pairs of legs in the male is forty-seven in a specimep 8™™ in length; in the female there are forty- eight pairs. The sixth pair of legs of the male are somewhat longer and much swollen, the suture between joints 3 and 4 is very slight, the two joints together forming an ovate section of the leg a little thicker than the length of the second joint; terminal joint long and slender, considerably longer than joints 3 and 4 together. The two-jointed eighth rudimentary pair of legs are longer and larger than in Scoterpes copei, the basal joint nearly twice as long, while the second (terminal) joint is larger and swollen, and, besides being larger, ends in three or four fine minute setz instead of a short claw, asin Scoterpes. Length, 8™™. The male genital armature is very minute and rudimentary, and has already been described in a general way. With but one species as yet known it would be unsafe to assign their specific characters. The two inner lamine are quite unequal in jength and development, and the armature in general shows signs of degeneration, as though the species had originated from some form in which the male armature was more completely developed. Nine specimens were found by usin New Market and Luray caves, and about twenty in Weyer’s Cave, Virginia; ne Cave, Vir- ginia (Dr. C. A. White ; Ryder). This species in size and general appearance would be easily mistaken for Scoterpes copet, which we at first, from a too-hasty examination, supposed it to be. Mr, Ryder’s excellent description characterizes the species, but his fig- ures are indifferent, the third joint of the male is much more swollen in our specimens, and the normal leg (his Fig. 3) is drawn too slender, while the front of the head is not correctly rendered. In our specimens drops of a yellowish secretion were attached in alcoholic specimens to the base of many of the setz, indicating the presence of repugnato- rial glands, though no pores could be found. On breaking the body in two, nearly ripe eggs occurred in June; they were rounded-oval; length, about #™™. CAMBALA Gray. Julus Say, Journ. Acad. Nat. Sci., Phila., ii, 103, 1821. Cambala J. E. Gray, Griffith’s Cuvier’s Au. Kingd., xiv, Insecta, i, Pl. 135, fig, 2, 2a, 2b, 2c, no deser., 1832. Reasia R. Jones, Todd’s Cyc. Anat. Phys., Art. Myriopoda, 546. Cambala Gervais, Newport, Ann. and Mag. Nat. Hist., xiii, 266, toa Cambala Gervais, Apteres, iv, 137, 1847. Spirobolus (in part) Wood, Myr. N. Amer., 215, 1865. Cambala Cope, Proc. Amer. Phil. Soc., xi, No. 82, 181, 1869. Cambala Pack., Proc. Amer. Phil. Soc., xxi, 195, 1883. The essential, diagnostic characters of this genus are the linear eyes, the long slender body, with keeled scutes, while the antennz are short and thick, much as in Spirobolus. The body consists of fifty-nine segments; the scutes with high keel-like ridges. The eyes are arranged ina linear row of ocelli, forming a straight line, situated far behind the insertion of the antennz, next to the front edge of the first segment. The front of the head is somewhat longer than broad; the surface full and conyex asin Julus. Antenne are short and unusually thick, more so than in Julus or Spirobolus; seven-jointed ; joint 2 littie longer and thicker than 3; fourth shorter and more clavate than third; fifth rather thicker at end than fourth, but of about the same length; sixth thicker than any of the others, about as long as fifth; seventh very short, round, no longer than broad. The feet are slender, not quite so long as the body is thick. On the fourth lower large ridge is a whitish microscopic spot, which under a half-inch objective is seen to be a short acute tubercle; these are Say’s “stigmata,” but they occur on each segment, and are doubtless homologous with the setiferous tubercles in Trichopetalum, ete. The only species known has been mistaken for Lysiopetalum lactarium by Newport, Gray, and Gervais, hence the synonony of the two genera is somewhat confused. Newport, adopting Mr. J. E. Gray’s manuscript name Cambala, was the first to characterize the genus, remarking: ‘‘I have derived the characters of this genus from the specimens originally sent by Say to Dr. Leach.” It is probable that Say by mistake sent an example of his Julus annulatus instead of a L. lactarium, as the two species would be easily confounded, although his Julus annulatus must have been of course familiar to him, The mistake was a natural one, MEMOIRS OF THE NATIONAL ACADEMY OF SCIENCKHS. 65 CAMBALA ANNULATA (Say) Cope. Plate IX, figs. 1, la. Julus annulatus Say, Journ. Acad. Nat. Sci., Phil., ii, 103, 1821. Cambala lactarius J. E. Gray, Griffith’s Cuvier’s An. Kingd., Pl. 135, fig. 2, 2a, 2b, 2c; Insecta i, vol. xiv; vol. ii, 784, 1832. Cambala lactaria Newport, Ann. and Mag. Nat. Hist., xiii, 266, 1844. Cambala lactarius Gervais, Ann. Soc. Ent., France, 1844. Cambala lactarius Gervais, Apteres, iv, 137, 1847. Spirobolus annulatus Wood, Myr. N. Amer., 212, 1865. Cambala annulata Cope, Proc. Amer. Phil. Soc., xi, No. 82, 181, 1869. Cambala annulata Cope, Trans. Amer. Ent. Soc., iii, 66, May, 1870. Cambala annulata Pack., Proc. Amer. Phil. Soc., xxi, 196, 1883. Body very long but blunt at the end, consisting of fifty-nine segments besides the head; eyes consisting each of six ocelli arranged in a straight line. The first segment behind the head is smooth, about half as long’as wide, evenly convex, considerably broader than the head; the three sneceeding segments are of about the same length, and each are about half as long as the fifth and succeeding segments. On the first segment are about ten bead-like tubercles seen from above; on the third about eight longer tubercles can be seen from above; on the fifth and succeeding segments there are about nine dorsal and subdorsal high, prominent, thick, parallel ridges, becoming sharp behind. On the middle segments of the body about six sharp ridges with broad hollow valleys between can be seen from above. These are mounted on each side lower down by about twelve less distinct ridges, becoming towards the lower edge of the scuta less and less convex and distinct, until they are indicated by simple impressed lines. There are thus about thirty ridges in all on each scute. The segments (arthromeres) are short, and the smooth spaces between the rigid portions are very short above. The color of the body is horn brown, the head, feet, and antennz pale flesh- colored, and there is a dark median spot on the vertex between the eyes. The ridges are darker than the rest of the body. Length, 30™™, Little Wyandotte Cave, Indiana, and Cave of Fountains, next to Weyer’s Cave, Virginia (Packard); Zwingle’s Cave, Carter’s Cave, Kentucky (F. G. Sanborn); Spruce Run Cave, on the Kanawha River, Giles county, Virginia (Cope), Wyandotte Cave (Hubbard). One of the most abundant of the Myriopoda in the mountain region of -Tennes- see and North Carolina (Cope). This species is not unfrequently found in caverns, where Z. lactariwum more rarely occurs. This well-marked species may readily be distinguished from Lysiopetalum lactarium by the very short thick antenne, linear eyes, and by the slenderer body, which, however, ends much more obtusely. We know of but one other species of Julide with the eyes arranged in a linear series; this is the Prachyjulus ceylonicus Peters of Ceylon, figured by Humbert. The cave specimens which we have found are partially bleached, the result of probably a limited number of generations living in the darkness. Besides these true cave-myriopods, Mr. Hubbard found a single specimen of Polydesmus gran- ulatus Say, in Indian Cave, Barren County, Kentucky, not far from the mouth of the cave. It was bieached entirely white, although of nearly full size. Mr. Hubbard also found a bleached specimen of the same species in Lyon Cave, Kentucky. INSECTA. THYSANURA. LIPURA? LUCIFUGUS n. sp. Plate XVI, fig. 1. A small species of uncertain genus. with short, three-jointed antenne, and with distiuct eyes, occurred in Wyandotte Cave. TOMOCERUS PLUMBEUS Templeton, var. PALLIDUS. One specimen from Zwingle’s Cave was but slightly changed, being almost wholly plum- beous; it occurred one-quarter of a mile from daylight (Sanborn). In a number of other specimens from Zwingle’s Cave and others of the Carter caves the body is white, as well as the spring and the legs, but the tarsi retain a slight plumbeous tinge. The antenne are partly pale, the two basal joints being bathed with leaden gray. Ten examples col- lected by us had distinct black eyes, but minute and angular in outline, having suffered a considerable reduction in size. Specimens collected by us from the ice-house cave were white, with dusky antennz and black eyes, and were like those just described. Specimens trom X Cave were all bleached, like those from the other Carter caves, but in some examples the eyes were connected by a narrow black band, Mis. 30, pt. 2——5 66 MEMOIRS OF THE NATIONAL ACADEMY OF SCIHNCKS. Specimens from Weyer’s Cave and the adjoining Cave of Fountains were just like those in the Carter caves, being white, with small black eyes, and dull purple leaden antenne and tarsi. Those in the New Market Cave were white, with black eyes and dark lead colored antenne. In One Hundred Dome Cave, specimens said to have been collected one-quarter of a mile from the entrance were all dark, of the usual out-of-door plumbeous color. nee Remarks.—It is evident that the var. pallidus has been produced by the influence of its cave life. Var. pallidus occurs in a cave near Salt Lake, Utah, and the specimens do not differ from the bleached ones in the Kentucky and Virginia caves. The trunk becomes bleached, while the - extremities of the antenne and legs retain somewhat of the colors of the out-of-door form. None have been found without eyes. The shallowest caves, such as the ice-house cave, in Carter county, Kentucky, as well as the deeper ones, possess this variety. We also find the normal plumbeus in similar caves, though probably near daylight, but the inference that the pale bleached variety has been produced by want of light is a natural and the only possible one. It is proved by finding in Zwingle’s Cave a slightly-changed plumbeus associated with numerous pallidus. LEPIDOCYRTUS ATROPURPUREUS n.sp. Plate XVI, figs. 3; 3a, 3d. Body dark leaden-purple; eyes large and distinct; antenne smaller and shorter than usual, but little longer than the head; the fourth joint equal in length to the second and third together. The last pair of legs with rather long scales, rather slenderer than in ZL. marmoratus Pack.; the larger claw uot very broad, in this respect differing from DL. metallicus Pack., and being more as in LZ. marmoratus; the single tenent hair about as long as the inner smaller claw. The terminal joint of the spring (elater) is very short; the middle tooth much shorter than the terminal one, which is aiso short compared with that of LD. metallicus. Length of body, without the antenne and spring, 1™™. Diamond Cave. The species is described from alcoholic specimens in which the hairs have been rubbed off. In color it seems to be near ZL. metallicus; like that species it is stout-bodied, and has short antennze, but in the present species the antenne are still shorter, the entire antennz being scarcely longer than the head; in this respect and the shape of the end of the elater the present species differs from any of the other described forms. DEGEERIA CAVERNARUM n.sp. Plate XVI, figs. 2, 2a—2g. Whitish; with a slight yellowish tint; usually blind; no traces of eyes. Body of the usual form; antenne of great length, two-thirds as long as the body and more than twice as long as the head; basal joint longer than usual; fourth joint very long and slender. Legs: last joint with fine slender scales; the claws much as in D. grisea Pack., but the spines on the larger claw are less distinct and the tenent hair shorter; the spring long and slender; the second joint serrulate along the under side nearly to the base; third (terminal) joint long and slender, ending in three teeth; the terminal tooth claw-like, as usual. The collophore (Fig. 2c) is large and well developed. Length of body, without the spring, 3™™. : Fig. 2 represents an average specimen from New Wyandotte Cave, where we found this | species in great abundance. It also occurred in Bradford Cave, Indiana; in general form and color it was like the New Wyandotte form, but as seen in Fig. 2c the antennez are much shorter, especially the fourth joint; also the end of the spring is slenderer. It has no eyes; the body is of the same size, and the feet nearly the same. While the Lepidocyrtus may be only an occasional visitor to the cave in which it was found, the present species being eyeless and bleached, besides being very numerous, is, without doubt, a true troglodyte. Specimens found in the Carter caves, especially Zwingle’s, about one-quarter of a mile from daylight, had antenne (Fig: 2a)-with the fourth joint slightly longer than in the Bradford Cave examples, but much shorter than in those from New Wyandotte; and the spring (Fig. 2g) is less hairy. It also occurred in Bat Cave, the specimens being certainly eyeless. Specimens from Diamond Cave (Fig. 2d) were identical in form, color, and length of antennz with those living in the Carter caves. The color did not noticeably vary in specimens from any of the caves. MEMOIRS OF THE NATIONAL ACADEMY OF SCIENCES. 67 Unfortunately no specimens were found in the large Wyandotte Cave, if it occurs there. It does not appear to be an inhabitant of Mammoth Cave, though occurring in Diamond Cave. The antennz in this species are much longer than in any American form known to us, and are much longer than in any Huropean species figured by Lubbock, though in this respect it approaches nearest to D. annulata (Fabr.), which is a pale greenish-yellow species, with remark- ably long antenne. Remarks.—The result of the examination of specimens from different caves shows that this » species varies most perceptibly in the length of the antenne, especiaily the fourth or terminal joint, which is shortest in the specimens inhabiting the smallest, shallowest cave (Bradford), and longest in the deep, perfectly-dark New Wyandotte Cave. In caves like those of Carter county and Diamond Cave the individuals are nearly alike, the antenne being a little longer than in the Bradford Cave specimens. It should be noticed that, compared with our American out-of-door species, the New Wyandotte examples have remarkably long antenne. SMYNTHURUS FERRUGINEUS n. sp. Plate XVI, figs. 4, 4a, 4b, 4c. Ferruginous, mottled with bright orange-red; eyes distinct reddish, or deep brick-red; abdo- ‘men globular; antenne moderately long and slender, ten to twelve jointed, with verticils of hairs; the spring rather large; the terminal joint acute, serrulate beneath, except at the end, which is obliquely truncated; the tarsus is perhaps rather slender; the claws only moderately stout. Length,1.1™™. Common in New Market and Weyer’s caves, Virginia. The young when 0.8== Jong are white all over and with no traces of eyes. The antenne are slenderer than in S. rosea Pack., and the claws are slenderer than in an undescribed Smynthurus from Massachusetts. It is doubtful whether this should be regarded as a true cave species any more than Tomocerus _plumbeus, or Lepidocyrtus atropurpureus. Another white species occurs in Mammoth Cave. CAMPODEA COOKE! Pack. Plate XVII, figs. 1, la to 1%. Campodea cookei Pack., Amer. Naturalist, v, 747, December, 1871; Fifth Ann. Rep. Peab. Acad. Sci., Salem, 46, 1873. Closely allied to O. staphylinus Westw. (common to Europe and America), but it is much larger; the antennz are twenty-four instead of twenty-jointed, as in the other species, and are longer in proportion, reaching to the basal abdominal segment, while in C. staphylinus they reach only to the second thoracic; the terminal joints are much longer than in that species, the penulti- mate joint being one-third longer. The terminal joint (Pl. XVII, fig. 1e), as may be Seen by com- _ parison with that of the out-of-door species (Fig. 2), is very much slenderer, while the olfactory (?) area (ol.) at the extreme end is somewhat larger. Terminal abdominal segment longer and slen- derer; hind femora longer than in C. staphylinus. Caudal stylets (cercopoda) about twelve-jointed, very long and slender (Fig.1/). Entirely white, body somewhat pilose, some of the sete spinulose (Fig. 17). Length, .25 inch without the antenne and cercopods. Frequent in Mammoth, White’s, Salt, Diamond, Wyandotte (Packard), Indian Cave (Hubbard). MACHILIS CAVERNICOLA (Telikampf). Triura cavernicola Tellkampf, Archiv fiir Naturgeschichte, x, 321. Machilis cavernicola Pack., Amer. Naturalist, v, 747, 1871; Fifth Rep. Peab. Acad. Sci., Salem, 51, 1873. This species is described as being white. It is rare, and my single specimen seemed immature and was mislaid. Mammoth Cave.