Bes pee a Spsttis “Sesh hat pice HO ea tpie ¥ Is kat? fe q vist Reis eye earth ed ” Digitized by the Internet Archive in 2009 with funding from Ontario Council of University Libraries http://www.archive.org/details/proceedingsofaca33acaduoft 5 ae Tabs AIP ve \ a PROCEEDINGS OF THE ACADEMY OF NATURAL SCIENCES OF PHILAPELPHITA. LSS it. PUBLICATION COMMITTEE: JosEPH LeErpy, M.D,, Gro. H. Horn, M.D. Wiuiiam S. Vaux, THomas MEEHAN, Joun H. REDFIELD. Epitor: EDWARD J. NOLAN, M.D. acqt PHILADELPHIA: ACADEMY OF NATURAL SCIENCES, S.W. Corner Nineteenth and Race Streets. 1882. : ACADEMY OF NATURAL SCIENCES OF PHILADELPHIA, — February, 1882. I hereby certify that printed copies of the Proceedings for 1881 have been presented at the meetings of the Academy, as follows:— Pages 9to 24 .. : ; . May 10, 1881. Ob 25 to 56; —. : ; .» May 31, 1881. es 57 t0.88 . : : Lette 14, 1881. es 89 to 112 . a : - June 21, 1881. 118 to 144. : : . August 2, 1881. 9° 4 45) to 160 : : Bo ERM OFE Ui wcdltotiallfoso}ll “ 161t0o176°. . . . September 13, 1881. Sein abOse0Geie * : . September 27, 1881. ‘6-209 to 256 : : . October 18, 1881. s¢ 257 to 304 : ‘ . November: 8, 1881. s¢ 805 to 384 . : : . November 22, 1881. se 385 to 414 . : ; . December 6, 1881. {o £415 t0:446 -.- : : . December 13, 1881. «= 447 to 462 _— : : . February 7, 1882. ss 463 to 478. 6 E . February 21, 1882. EDWARD J. NOLAN, CD) ff Recording Secretary. br / kL } ‘ ro) PHILADELPHIA W. P. KILDARE, PRINTER. LIST OF CONTRIBUTORS. With reference to the several articles contributed by each. For Verbal Communications, see General Index. Arango, Rafael. Description of new species of Terrestrial Mollusca of COW once caren 125 Chapman, Henry C., M.D. Observations upon the Hippopotamus......... 126 On a Foetal Kangaroo and its Membranes........s.ceseeeeeeeeneeenseeeeererees 468 Heilprin, Angelo. Notes on the Tertiary Geology of the Southern United States........ dup deccves Sep O PO COCEE HOR OR EECA STOGT Bone? TOC EECHIOCG OPIECOEC 157 A Revision of the Cis-Mississippi Tertiary Species of the United SUMUCH oe seats asea rots det nen eWesesescdecssrcslcescecscsrers ss sccectactinscenclesdses eos 416 Remarks on the Molluscan Genera Hippagus, Verticordia and IRC CCWOliaeccertasccaccu mcs ccccevecsess .codaccravarescrrennccrdnocsav-ssiasadeascon=- 423 Note on the Approximate Position of the Eocene Deposits of Mary- NAIA eaten cctnk iodcetilas And atan cicoc owl conan sc odes se csescinvddecis sesredsaasccaeecsassrcaee 444 A revision of the Tertiary Species of Arca of the Eastern and South- ELM) OMILEM! SURLCS ccs cccccecseccecosccrccsrocceveveceerarncilscdanrseascirscnsacecse 448 ‘Lockington, W. N.. List of the Fishes collected by Mr. W. J. Fisher, upon the coast of Lower California, 1876-77, with descriptions of MOWAES PC ClOSaacanecesseeseacnsscnsesscs cee Peeecantedtoceasaactadesecscstiocsersre® 113 McCook, Rev. H.C. The Honey Ants of the Garden of the Gods............ 17 The snare of the Ray Spider (Epeira radiosa), a new form of Orb- WG] De ceecGons ecoacehoctlenccocac SoC ARC COOL CosbP bocboobcnastonencecuseuore eo coSscucenEdc 163 Ryder, John A. The Structure, Affinities and Species of Scolopendrella, 79 Stearns, R. E.C. Observations on Planorbis.............. Hagboce doasanceDOcKeCBOde 92 Wachsmuth, Chas. and Frank Springer. Revision of the Palzocrinoidea. Part IT, Family Spheeroidocrinida............cceccceecescncereeccereesssneeers 177 PROCEHHDINGS OF THE ACADEMY OF NATURAL SCIENCES OF PHILADELPHIA. 1881. JANUARY 4, 1881. Dr. Jos. Lerpy in the chair. Twenty persons present. Rhizopods as Food for Young Fishes.—Prof. Letpy remarked that last September he had received a letter from Mr. S. A. Forbes, of the Illinois State Laboratory of Natural History, Normal, Illinois, stating that the young of some of the suckers (Catasio- midz), Hypentelium, Myxostoma, etc., “ have the intestines packed with tests of Difflugia and Arcella.” Later, Mr. Forbes sent two slides, with some of the intestinal contents, for examination. The slide with food from the intestine of the large-scaled Mullet, Mysxostoma macrolepidotum, from Macinaw Creek, contained the following species: DIFFLUGIA GLoBULOsA. Shell of rather coarse sand, with larger grains around the mouth ; mostly in the shape of the segment of an oval, with the oral pole truncated. Most numerous form. Measurements of a number were as follows: 1. Shell 0.18 mm. long; 0.162 broad ; oral end, 0.102 broad. = * @ie.. * ey afi: ->:6* ee 0.102 ‘* eo © Saag 2. Gets ge ie 0.072 « x § O14 <- iae- -€ = Ea 1. Rie < « (1. « “ 0.096 «§ ey *, Qi « ee). et fs ee IOs >< DiFFLUGIA AcUMINATA. Shell mostly slightly unsymmetrical; some with a slight neck, straight or slightly everted at the mouth ; a few with 2 10 PROCEEDINGS OF THE ACADEMY OF [1881. two points to the summit; usually of minute sand and comparatively smooth. One oblique form noticed (No. 6), approaching D. constricta. Shell 0.18 mm. long ; 0.108 broad ; oral end 0.06 broad. . Shell of same size, but with a short neck, slightly erected and undulant at the border. wr 3. Shell 0.18 mm. long; 0.114 broad ; oral end 0.048 broad, Ae OMOS ee Seo = OOD ce “od 0:06 ss 5. 66 0.18 66 ce 0. 114 66 “ce 0.06 6é 6 SS Onl6e. ss $6 10,09)" £420.06 es Nos. 1-3 of fine sand, and smooth ; Nos. 4-6 of coarser sand. The slide with food of Hremyzon succetta. The material apparently consisted of the superficial sediment of the water, and contained entomos- tracans, rotifers, dipterous larve, desmids, diatoms, etc., together with the following : DIFFLUGIA GLOBULOSA. Shell 0.15 mm. long, 0.138 broad; oral end 0.078 broad. DIFFLUGIA LOBOSTOMA. Shell with trilobed mouth, 0.09 mm. long, 0.078 broad; mouth 0.03 wide. Several measured of the same size; others slightly smaller. The most common species present. DIFFLUGIA PYRIFORMIS. Shell 0.42 mm. long, 0.21 broad, at mouth 0.09 broad. ; ARCELLA VULGARIS. Variety with pitted shell. ARCELLA DISCOIDES. Shell 0.18mm. broad, mouth 0.026 wide. Another specimen 0.15 broad, with mouth 0.054 wide. Another rhizopod shell observed, was different from any pre- viously noticed. The shell had the form of that of Arcella dis- coides, with no trace of the structure characteristic of that of Arcella, but composed of a nearly colorless or pale yellowish chitinoid substance, incorporated with minute spherical granules of uniform size, darkly outlined, scattered irregularly, isolated, or in little groups or chains, straight or irregular, and in pairs, or up to five in number. The specimens measured about 0.105 mm. broad, with the mouth 0.03 wide. A chain of five granules of the shell measured 0.009 mm. long. It is certainly an interesting observation of Mr. Forbes, to dis- cover that the young suckers should use the rhizopod shells to obtain as nutriment their little stores of delicate protoplasm. + JANUARY 11. Dr. Joun L. Le Conte in the chair. Highteen persons present. A paper entitled “ Descriptions of new species of Terrestrial Mollusca of Cuba,” by Rafael Arango, was presented for Bab lication. 1881. ] NATURAL SCIENCES OF PHILADELPHIA. 11 JANUARY 18. The President, Dr. RuscHENBERGER, in the chair. Twenty persons present. J ANUARY 25. _, The President, Dr. RUscHENBERGER, in the chair. Twenty-one persons present. Jos. J. Knox and Geo. H. Rex, M. D., were elected members. Chas. Velain, of Paris, was elected a correspondent. FEesruary l. Mr. Gro. W. Tryon, JR. in the chair. Ten persons present. FEBRUARY 8. The President, Dr. RUSCHENBERGER, in the chair. _» Twenty-seven persons present. Note on Treeless Prairies.—Mr. THOMAS MEEHAN remarked that the absence of timber or arborescent growth on the grassy prairies of America, still continued to be a matter of controversy, but he believed that in the light of accumulating evidence, we might now come to a positive decision in regard to the question. The most prevalent belief had been that trees would not grow on these prairies—and we have had theories relating to soil or climate, to show why they could not grow. Then there were others who believed that trees did grow there in ancient times, but had been burnt off, and kept burnt off by annual fires. Mr. Meehan considered in detail, the authors who had pro- pounded various theories, and the distinguished men who had ad- vocated them, and said that it was evident climate could have nothing to do with the question, because in these prairie regions. there were often large belts of timber lands, projected like huge arms into the grassy regions, with precisely the same climatal conditions over both. That the soil was not unfavorable, was proved now by the artificial plantations everywhere successful, and that the soil was unfavorable to the germination of tree seed, as suggested by Prof. Whitney, was on the face of it untenable. 12 PROCEEDINGS OF THE ACADEMY OF (1881. from the fact that it required but the same conditions for the seeds of trees as for those of herbaceous plants, the number of species of which on the prairies was well known to be very large. Another great gain to our present knowledge, was that since the annual firing of the grassy prairies had been discontinued by the advance of civilization, the timber was everywhere encroaching on them. Among the facts which he offered in proof of this, was a reference to p. 505 of the Tth Report of the Geological Survey of Indiana, where Dr. Schneck shows how land which was once grassy prairie, is now covered with a luxuriant growth of forest trees; to the evidence of Major Hotchkiss, Geologist of Staunton, Vir- ginia, that the Shenandoah Valley, now heavily timbered, was clear of trees in the early history of Virginia; to the discovery of buffalo bones, in caves near Stroudsburg, Pa., by Dr. Joseph Leidy,—now a timbered region, the buffalo. only existing in open, grassy countries ;! and to various traditions of settlers: in:some valleys now timbered, that the land was originally clear of trees, He pointed out that in all known parts of the United States at the present time, except the arid regions, where only drought- loving plants could exist, the natural result of freedom was) the succession of forest growth. Seeds were scattered by winds or animals over acres of cleared land; if such land became neglected, these, again seeding in time, extended the forest area continu- ally. The tallest growing vegetation, like trees, crowded out the weaker, and the forest naturally crowded out the lower growing and weaker herbaceous plants. He illustrated this: by reference to the neglected cotton-fields of the Southern States. From all this, the speaker said that it was evident that there was nothing in Nature either now or in the past, to prevent the gradual encroachment of the forest over the grassy plains, till long before the white man came here, the whole would have been -com- pletely covered by arborescent growth. Were there any artificial causes equal to the exclusion of trees, and yet permitting an her- baceous growth? If we were to sow a piece of land’ in the autumn with some tree seed and some seeds of annuals, the latter would be up, flower, mature and scatter their seed to the ground before the next autumn, and many of these seeds would be washed into the earth, or drawn into the earth by insects or small animals: But tree seed would make young trees, which would not again produce seed for ten or more years. If now, at the end of this first season, a fire swept over the tract, the seeds of the annuals which had found “a slight earthy protection, would «come up again the next summer, again seeding and extending the area: The trees would be burned down, and though perhaps many would sprout, successive burnings would keep them -confined to ‘one place. In short, under annual burnings, herbaceous plants could 1 Since the reading of the paper, it has been brought to the attention of the author, that the bones may have belonged to the Wood Buffalo. 1881. ] NATURAL SCIENCES OF PHILADELPHIA. 13 still: increase their area annually, but trees could never get far beyond the line they had reached when the annual fire first com- menced. There could be no doubt that an annual burning in a tract destitute of forest growth, would certainly prevent the spread of timber, or of any plant that required more than a year to mature seed from the time of sowing. Now, if we look at the actual facts, we find that the Indians did annually fire the prairies. Father Hennepin, the earliest writer on Indian habits, noted that it was the practice in his time. There is little doubt but this practice of annual burning has been one extending long into the past. ‘What object had they in these annual burnings? They must have known that the buffalo and other animals on which they were largely dependent for a living, throve only on huge grassy plains, and that it was to their interest to preserve these plains by every means in their power. Low as their power of reasoning may be, they could not but have perceived that while grassy herbage throve in spite of fires, perhaps improved under the fiery ordeal, trees could not follow on burned land. What could be more natural than that they would burn the prairies with the object of retaining food for their wild animals? If we have no difficulty in reaching a positive conclusion so far, we may now take a glance at the early geological times. Mr. Meehan then re- ferred to the researches of Worthen, Whittlesley and others in Ohio, Illinois and other prairie regions. On the retreat of the great glacier, the higher lands and drift formation were probably high and dry long before the immense lakes formed from the melting and turbid waters ceased to be. It' was tolerably well understood that many species of trees and other plants which required a temperate atmosphere, retreated southwardly with the advance of the glacier, and advanced to higher latitudes on the glacier’s retreat. Thus these higher ridges would become timbered long before the lower lands became dry. Evidence accumulates that man existed on this continent, in the far west, not long after the glacier retreated, though “not long,” in a geological sense, may mean many hundreds of years. The lakes of glacial water would gradually become shallower from the deposit of the highly comminuted material brought down from higher land, from the wearing away of rocky breastworks as in South Pass, Illinois, as well as from the openings which would continually occur from nature’s ever varying plan of streams under ground. In all events, the drying of these lakes would be from their outward edges first. Aquatics would give way to marsh grasses, and these to vegetation such as we now find gener- ally spread over the prairie region. If now we can conceive of human beings such as we know the Indian races to be, al- ready in more southern latitudes—having learned the fact that firing would keep down trees and aid in the preservation of the chase—following the retreat of the glacier to the higher lands, and still as they advanced northwardly, firing the plains up to the 14 PROCEEDINGS OF THE ACADEMY OF [1881. water’s edge, it would certainly account for the absence of ar- boreal vegetation from these immense lacustrine lands from the very beginning of their formation. Of course with this view we should have to look for some evidences of man’s existence, both on the lands which were once under water, as well as those which were timber lands at his first appearance there. He did not know how many such evidences have been or may be found. Man’s traces in the past are at best but rare, and they would naturally be much more scarce in the lacustrine regions than in lands dry at the same epoch. At any rate, this part of his remarks he said, must be taken as mere speculation ; but, as we could see on the basis of sound scientific investigation why there could be no trees on these grassy prairies within the range of indubitable history, it was a fair inference that some such cause had continued from the beginning; namely, that annual fires had ever been the reason why arborescent vegetation had never had an existence there. The resignation of Mr. Edw. S. Whelen, as a member of the Council and Finance Committee, was read and accepted. FEBRUARY 15. The President, Dr. RuscHENBERGER, in the chair. Nineteen persons present. A paper entitled “The Honey Ants of the Garden of the Gods,” by Rev. Henry C. McCook, D. D., was presented for publication. FEBRUARY 22. The President, Dr. RuscHENBERGER, in the chair. Thirty-six persons present. The death of Harry C. Hart, M. D., a member, was announced. Mr. Isaac C. Martindale was elected a member of the Finance Committee, to fill a vacancy caused by the resignation of Mr. Edw. 8. Whelen. Mr. Charles P. Perot was elected a member of the Couneil for the unexpired term of Mr. Whelen. Robert P. Field was elected a member. John Brazier, of Sydney, N. S.W.; Rafael Arango, of Habana, and Chas. Mohr, of Mobile, Ala., were elected correspondents. The following were ordered to be printed :— 1881. ] NATURAL SCIENCES OF PHILADELPHIA. 15 DESCRIPTIONS OF NEW SPECIES OF TERRESTRIAL MOLLUSCA OF CUBA. BY RAFAEL ARANGO. Choanopoma acervatum Arango. Testa obtecte-umbilicata, ovato-oblonga, plerumque decollata, tenuis, plicis transyersis irregulariter et costulis spiralibus dis- tantibus undulatim et acervatim interruptis lamellata, albida, anfractus 6 (superstites 34-4) convexi, lente acrescentes ad suturam canaliculatam (ob plicarum extremitatem) denticulati, ultimus anfractui contigul eallo adnatus; apertura circularis, peritrema dupli- catum, internum rectum, externum patens, concentrice striatum et ob costulas spirales testee undulatum, ad anfractum contiguum callum formans, tum umbilicum lamina lata fornicata tegens. Long. 83-94; aa 43_5 mill. Differt ab Choan. Tri yont Arango primo visu (Gaal: spiralibus, ab sordido imprimis plicis teste undulata interrupta. Habitat.—Las Lagunitas prope Pinar del Rio partis occidentalis in plantatione dicta ‘‘ Vega de D. Manuel de Jesus Hernandez.” Cylindrella paradoxa Arango. Testa sine umbilico, fusiformi-turrita, solidula, albida, costis subrectis confertis munita, integra, apice obtusiuscula, anfractus 12-13 convexi, penultimus longitudinaliter in medio canaliculatus, ejusdem pars superior reliquis an- fractibus ignalis quoad sculpturam, inferior angustior, reducta, costulis: magis aproximatis, dein dilatata et umbilicum tegens. Apertura piriformis, obliqua, superne angustior ob carinam salientem partis superioris anfrac- tus. Peritrema continuum, undique equaliter expansum. Col- umna interna simplex. Long. 124; diam. 2} mill. Habitat.—Guane in loco “ Puerta de la Muralla ” dicto. Cylindrella incerta Arango. Testa quoad sculpturam et columnam internam simillima Cylindrelle czrulans Poey, sed forma ven- troso-cylindracea et anfr. superst. 8 (in testa integra 14). Long. 18 ; diam. 44, testa integra. Long, 144; diam. 44, testa fracta. Habitat.—Guane in loco “ Puerta de la Muralla” dicto. ” i 16 PROCEEDINGS OF THE ACADEMY OF [1881. Ctenopoma nodiferum Arango n. sp. Testa clauso-perforata, cylindraceo-turrita, decollata, tenuis, plicis zqualiter distantibus et liris debilibus decussata, cineracenti- albida, anfr. 34-4. superstites convexi, lente accrescentes, ad suturam canaliculatam subdenticulati; ultimus subdisjunctis pone aperturam callo lato cum penultimo anfractu junctus ; apertura verticalis, obliqua, ovalis ; peritrema duplicatum, internum rectum, externum subzequaliter patens, sed superne dilatatum, anfractum contiguum attingens et umbilicum claudens. Operculum typicum generis. Long. teste truncate 8; diam. 3; apert. 2 mill. Simile Ctenopomati nodulato ; sed differt testa plicis sequaliter distantibus (nec acervatim approximatis et liras decussata). Hatbitat.—Sub lapidibus circa oppitum Santo Cristo de la Salud prope Bejueal. Ctenopoma Wrightianum Gund. n. sp. Testa subperforata, cylindraceo-turrita, decollata, costis sequali- ter distantibus obtusis et liris subtilissimis decussata, cineracenti- albida, anfr. de superstites, convexi, lente accrescentes, ad suturam canaliculatam denticulati; ultimus subdisjunctus pone aperturam callo lato cum penultimo anfractu junctus; apertura verticalis, oblique ovali-rotundato ; peritrema duplicatum internum rectum, externum subzequaliter patens, sed superne non dilatatum itaque anfractum penultimum non attingens. Operculum typicum generis. Long. test truncate 11; diam. 5; apert. 24 mill. Proximum Clenopomati ruguloso sed distinctum costis teste obtusis (nec acutis) et liris decussata. Habitat.—Sub lapidibus loci Punta de la Jaula dicti in Provincia Pinar del Rio, partis occidentalis. 1881. | NATURAL SCIENCES OF PHILADELPHIA. 17 THE HONEY ANTS OF THE GARDEN OF THE GODS. By Rev. Henry C. McCoor, D. D. I.—GEOGRAPHICAL DISTRIBUTION. The peculiarity in the Honey Ants (Myrmecocystus melliger) which has attracted the especial attention of naturalists is that one of the castes or worker forms has the abdomen distended to the size and form of a currant or small grape, and entirely filled with grape-sugar or “ honey.” Very little of their habits has heretofore been known, and only the forms of the honey-bearer and worker-major. In order, if possible, to remove this reproach from Entomology, I started in the early part of July, A. D. 1879, for New Mexico, as the honey- ants have been found in the neighborhood of Santa Fe, and even as far north as Abiquiu, on the Big Chama River.! During a brief visit at the cottage of Gen. Charles Adams,? of Manitou, Colorado, which is located in the mouth of the Garden of the gods, in the course of some observations made upon the ants of the vicinity, a nest was discovered whose external archi- tecture was new to me. The sentinels were called out by the application of a straw, and their general appearance raised the suspicion that they might be Honey Ants, which, as I had never seen specimens, were known to me only by description. The nest was opened, and the delightful fact revealed that the objects of my. search were before me. I thereupon made an exploration of the vicinity, and found that the nests were present in sufficient numbers for purposes of study ; whereupon I abandoned my New Mexico outfit, encamped in the Garden of the gods, and began the observations of which the following paper is the record. Up to the time of my discovery, it had not been known that the Honey Ants were distributed as far north as Colorado. I found no formicaries at any other point in the State, although the opportunity to search for them was limited. There is little doubt, 1 At the latter point Prof. Edward D. Cope informed me that he had seenthem. Dr. Loew and Mr. Krummeck saw them near Santa Fe. 2 Gen. Adams has recently been widely known by his intrepid venture among the hostile White River Ute Indians, and rescue of their unhappy prisoners, Mrs. Meeker and others, at the risk of his own life. As a recog- nition of this service he has been appointed Minister Plenipotentiary for the United States to the Republic of Bolivia. 18 PROCEEDINGS OF THE ACADEMY OF [1881- however, that they may be found in favorable locations in the entire southern portion of the state, and perhaps also north of the latitude of Pike’s Peak.1 Mexico, New Mexico and southern Colorado, may certainly be designated-as the natural habitat of the Honey Ants. It is probable, however, that they may be found throughout the entire south-western portions of North America, especially the uplands. They will doubtless be found west of the Rocky Mountains, as I have recently found one female of this species among a collection of Hymenoptera sent to Mr. Cresson from southern California. The following facts can be presented concerning the vertical distribution :— Z LocaLity, ELEVATION. OBSERVER. City of Mexico, >. — . -7482 feet, . lave: 2 Matamoras, Mex., s Brownsville; U.8.,f°.° 50 Langstroth. Santa Me, Sse ae Loew, Kummeck. Abia, 38 4% <25920=" Cope. Garden of the gods, . . 6181 “ | McCook. It will thus be seen that. the points at which these insects have heretofore been found, lie for the most part upon uplands, ranging from 6000 to 7500 feet in height above sea level.. Mr. Langstroth’s find is recorded as ‘fin the vicinity of Matamoras,”? If this means the near vicinity, the fact prevents the generalization which one might otherwise have been tempted to form, limiting the ants to the upland, for Matamoras has but a slight elevation. II.—Nest SITES AND EXTERIOR ARCHITECTURE. The Honey Ants are domiciled in large numbers throughout the section of country known as the Garden of the gods. The conformation of the surface here appears to be an im- portant element in determining the habitat of the insects, and deserves a brief notice. The Garden of the gods embraces a_ 1 The matter of their distribution is a point to which the attention of entomologists and other naturalists is called, and any information bearing thereupon will be of value. 2 T could not lay hands upon the elevations of Matamoras, which cannot vary much from that of Brownsville, Texas, on the opposite side of the river. 3° ** Proceed. Acad. Nat. Sci. Phila.,”” vol. vi, 1852, p. 71. < 1881.] NATURAL SCIENCES OF PHILADELPHIA. 19 space of about two miles in length by one in width, the surface of which is broken into ridges crossing each other at various angles, and ‘crowned or bordered at the top by the red sandstone and conglomerate rocks, whose peculiar shapes and likenesses to heathen deities have probably suggested the name given to this bitof landscape. A- rude idea of the topography may be had by drawing a horse-shoe, the toe toward the north ; within the mouth of this let'a second horse-shoe be described, occupying about one- half the space in width and one-third in length. Unite the toes of Se Stee Vay Al ot (ier Ds 3 ih i Fig. 1.—Sketch-map of the Garden of the gods. the two shoes by a zig-zag line, and draw lines east and west, on either side from the interior figure. The western line of the out- side shoe will represent the Manitou ridge, which starts at the base of Pike’s Peak. The eastern line will indicate the cretaceous wall of the table-land known as the Mesa, and the two walls of the Red Canyon. The inner shoe has for its western line the Von Hagen ridge, for its eastern tlle Adams ridge; the east and west lines will represent the general course of the ridges which drop down from these two, from the broken central ridge, Prospect ridge, represented by the zig-zag line, and from the eastern face of Manitou ridge. These ridges are composed of red sandstone, which crops out freely, forming vast ledges and cliffs. The top soil, where the rock is not.exposed, is a heavy gravel, upon, which grow tufts of gramma grass, straggling bunches of grease wood, 20 PROCEEDINGS OF THE ACADEMY OF [ 1881. Spanish, bayonet, low cedars and pine, and in the little vales or nooks wild. sunflowers, wild roses, and numerous small thickets and clusters of a scrub oak ( Quercus undulata). _These localities are indicated diagramatically in the sketch-map at Fig, 1. All along the tops of these ridges, and on the eastern and south-eastern slopes, the nests of the Honey Ants are located (Pl. I, fig. 2). About ninety per cent..of those found were on the tops of the ridges, and every one on or near the summit or central line of the top. The choosing of such a site may, there- fore, be inferred to be a fixed habit of the ant. The advantage of this location. is apparent, at least in the points of dryness and warmth. I made several observations of the effects of the heavy July and August rain storms upon the exterior architecture, which is a low, gravel-covered moundlet, penetrated at the centre bya tubular gallery or gate three-fourths of an inch in diameter (PI.II, figs. 3,4). The large gravel-covered mounds of the Occidental Ant (Pogonomyrmex occidentalis, Cres- son), numbers of which were built in the valley of the Boiling Fountain Creek, and in the nooks between the ridges, were more or less damaged. by the wash of the water... Some were seriously injured, one wholly swept away. The only damage wrought upon the Honey Ant nests was a little beating down of the pellets of gravel within the gate. There was no injury from the wash of the water, and apparently no likelihood of any beyond that which the momentum of the rain-drops could inflict as they dashed upon the nest and within the gate. Throughout one storm, during the entire progress of which a nest was watched, several ants were stationed like sentinels within the gate around the upper margin GER Vi mic. Zo), hey rere evidently on the look out for any damages to their home. The disarrangement of a few pellets moved two of these sentinels to bring up bits of gravel and attempt repairs. But there was little occasion for this, although the force of the rain was great enough to cause a good deal of discomfort to the observer. In half an hour the rain ceased, the sun came out over Pike’s Peak, and a rainbow girdled the Mésa. One worker-major crawled upon the crest of the nest, stretched herself, reared her head as though to snuff the fresh air, then 1 This isa sketch of my camp, from the point at which the Adams and Von Hagen ridges meet. One of the ant-nests is seen in the foreground ; others are indicated by the white circles on the crests of the ridges. 1881. | NATURAL SCIENCES OF PHILADELPHIA. 21 hurried down the gravel side and started at aswinging pace along the trail to'a neighboring oak copse. An hour afterward she had not returned, and not another ant had left the nest. Several, however, came out, but apparently were disturbed by a gale which followed the rain, and returned. On another occasion, the slight disarrangement of the nest made by the rain was repaired immediately after the storm. It amounted to a closing up of the greater part of the entrance by some of the displaced gravel-stones along the crater. The exterior architecture has been referred to as a small moundlet of gravel.! The largest seen was one on one of the ridges quite within the Garden; it measured around the base thirty-two inches, in height three and one-half inches, length of northern slope four and one-half inches (Pl. 11, fig.3). The average dimension of the nests is something less than this. The base diameter varies from ten to three and one-third inches, the greatest number of nests measuring six and seven inches.” The ordinary height’ is from two to three inches. The shape of the nests is a truncated cone. The section across the top is about two inches in'diameter. In the centre is a tubular opening or gate, from three-fourths to seven-eighths inch in diameter.? Hil.—Posttion OF Honty-BEARERS IN THE NEST. Leaving the details of the architecture to a later period, that habit which attaches the greatest interest to this insect, viz., the storing of honey, may be considered. The first nest that was opened, and called the ‘‘ Bessie ” nest,‘ for convenience of notation, is on the terminal slope of Adams’ ridge, looking due south, and quite near to the valley of the creek Fontaine qui Bouille. The gravel had 1 Dr. Oscar Loew, ‘‘ American Naturalist,’ 1874, says of Melligera col- lected near Santa Fe, that “they make no hills, like other ants.” “A structure like'a crater indicates where they live underground.’”’ Every formicary seen by me had a decided elevation, 2 I succeeded in. bringing one of these mounds home nearly entire, having fixed the grayel contents by liquid cement. 3 Dr. Loew says of the nests near Santa Fe, that the openings were the size ofa quill. It seems strange that such a difference should exist within localities so near each other. 454 little girl, Bessie Root, a guest in Gen. Adams’ cottage, whom I had enlisted in the search for ant-hills, first reported to me the nest in which I found the Melligers. 22 PROCEEDINGS OF THE ACADEMY OF (1881. not been penetrated to a depth ot more than six inches before a honey-chamber was uncovered, and the presence of the honey- bearers indicated that-a home of the true Honey Ant had been found (Pl. III, fig. 5). Within a dome-roofed yault, about three inches in width and three-quarters to one inch in ‘height, hung the ~ honey-bearers, clinging by their feet to the roof. Their yellow bodies stretched along the ceiling, but the rotund abdomens hung down, almost perfect globules of transparent tissue, through which the amber-colored honey showed. They looked like a cluster of small Delaware grapes or large currants. Most of the abdomens were quite round, but they were in various stages of fullness. Upon some the external membrane of the abdomen was gathered in folds. A few of the abdomens, and especially those but little distended, were of a white instead of amber color. I have observed that the honey-bearers in my artificial nests show the honey, which has been gathered from white sugar, quite white and translucent. It is probable that the color becomes amber, and even a wine color, with age.- When the abdomen is full it fairly shines, reflecting the light that falls upon it from the lamp. With most of the honey-bearers the abdomens hang downward without touching the ceiling, except at the rotundity near the base, and often not even at that point. With some, however, the whole lower. part of the abdomen rests against the roof (Pl. IV, fig. 13). This appears to depend chiefly upon the contour of the perch, and not upon the relative degree of comfort to the ant in the two positions. The roof of the honey-chamber is different in structure from the floor, the latter being comparatively smooth, while the former is rough, being the natural granulated surface left after the picking away of the sandy soil. This character, of course, enables the honey-bearer to cling more easily and securely to her perch. This position is not held by the mandibles clasping the rugose dome with their sharp teeth, but almost exclusively by the feet, whose claws, hairs and pulvuli all doubtless contribute to the effect. Judging from observations upon artificial nests and from the utter unwieldiness and helplessness of the fully charged bearers, they are not much disposed to change their roost after once taking it, at least after they have reached a considerable degree of rotundity. But the statements generally made by writers, that they are wholly unable to move, and never change position, are 1881.] NATURAL SCIENCES OF PHILADELPHIA. 23 inferences without the facts. They are not unable to move, and in point of fact, do occasionally move their positions. Those whose abdomens are but half or even two-thirds the full globose, I have frequently seen coming out of their chambers, ascending the galleries and moying freely about the surface, Those with full globes can move about with no little agility when placed upon a table, or when exposed in their nests to some unusual danger or alarm. In the nests they slide along from point to point, moving their feet sidewise, and so make changes of position. FALLEN HoneEy-BEARERS Hetpiess.—If once they loosen their hold, however, and fall to the floor, they seem ordinarily helpless to recover. - Numbers of my full honey-bearers dropping from various causes, or shaken down by thoughtless visitors, laid upon the fioor helpless, resting upon the rotund abdomen, bodies up, antennz and feet in motion, and seeming exceedingly uncomfort- able. -Those who so fell as to have some object upon which to lay their feet,as a clod or the surface of the jar, fared better. In very favorable positions a few recovered their roost. But as a rule they were helpless, remained stationary, and so passed their lives, which were evidently shortened by their position; although some of them lived thus several months (see Pl. VI, fig. 32). TV.—Sovurce or Honey-Supptry. The rotunds do not elaborate the honey, as has frequently been asserted. I was not for a moment misled by this fancy, being satisfied that, in the nature of things they were sedentary, and that their immense abdomens were charged by regurgitation from the workers who were the honey gatherers. But whence do they obtain their supplies? Not from Aphides, at this season of the year at least. I searched every bush and shrub in the vicinity, including large numbers of wild rose bushes, but failed to find any of these familiar and useful Emmet “herds.” Certainly, at least, the honey ants were not there drawing supplies from them. It was not possible to trace the ants to these or other sources of food supply during the day, for I found very soon that they were nocturnal insects. Their nests were as silent, and to all appearance empty, as an abandoned habitation, during the daytime. I accordingly stationed myself beside a nest to await the nightfall. This nest was located upon the summit of a ridge which from a peculiar formation of a 24 PROCEEDINGS OF THE ACADEMY OF [1881. rock upon it I named Eagle-head ridge, and the nest Eagle-head nest. At 7.30 P. M., the sun was set, and darkness had begun to gather. A few ants appeared within the gate. They advanced to the top, followed by others ; they pushed out upon the gravelled sides of the mound, over which a goodly swarm of yellow insects was soon gathered. There were no rotunds or semi-rotunds among these mustering squadrons; all were workers, with normal abdo- mens. Presently an ant left the mound and started over the ridge northward. Another—several—a score followed, until within a brief time a vast column was seen trailed along the ridge, all moving in the same direction. The evening had now become so far advanced that it was difficult to trace the column, but by stooping down close to the earth and using care not to alarm the ants, I was able to do so. The trail was somewhat winding, but on the whole seemed to be chosen with some regard to avoiding the inequalities of the ridge. I was not impressed, however, with the engineering skill of the insects in this matter. At the distance of about fifty feet from the nest, the column turned down the slope and entered a copse of scrub oak (Quercus undulata, var.). I traced a number of ants to a bush several feet within the thicket, but failed to unravel the secret that night. The next night a similar experience awaited me. After a long, careful, but vain search, I retired to my tent baffled. The third night (July 29), the ants of Eagle head next came out at 7.23 P.M. Those on Toad-stone ridge, to which I had assigned my assistant, Johnson, came out at 7.25 P. M., but did not begin to move until 7.44. Johnson followed them, but failed to find their feeding ground. They moved north and eastward, as did those of the Eagle head. These latter began to move almost as soon as they came out. They followed the same trail as on the previous evening, the track having beem marked by me. The movement was somewhat slower than before, perhaps because the trail had been washed by a heavy rain during the afternoon. There was no leader. A dwarf worker kept in advance over the greater part of the track, then a worker minor took the head of the column. The two were separated from each other, and the van of the column about eight to ten inches. There was, however, not the slightest evidence of any leadership at any time, in any part of the moving line, although I carefully looked for such. 1881.] NATURAL SCIENCES OF PHILADELPHIA. 25 The ants directed their movements to the same tree as on for- mer forays, reaching it in seventeen minutes, at 7.40 P,M. They distributed themselves along the tree, hunting trunk, branches, leaves. I could trace their forms, but when it is remembered that I was wedged in among the thick, low branches of this dwarf oak, holding up a lantern with one hand, and using the other to clear space for it; that the necessity to avoid alarming the timid insects compelled me to retain very inconvenient positions fora long time, it will not seem strange that I could find nothing satis- factory until between nine.and ten o’clock. At last, in course of the slow investigations, I reached the extreme end of a branch on the south side of the tree, and found a number of ants engaged upon clusters of brownish-red galls. The ants were moving from gall to gall, not tarrying for any length of time upon any. They applied their mouth organs to the galls frequently. The dimness of the light, and the distance which I was compelled to keep, pre- vented me from seeing anything more than this. But it was plain that they were obtaining honey stores, for in the lantern light it could be seen that their abdomens were already much distended by the sweets which they had lapped. The branch was carefully cut off without disturbing the ants, taken to my tent, and the. movements of the insects observed during the remainder of the night, the branch having been so placed as to prevent the escape of the ants, who were yet easily under view. They, however, were so preoccupied with their honey gathering, that they made little effort to escape. Directing attention to the galls, it was seen that some of them were gradually exuding minute globules of a white transparent liquid, which the ants greedily licked. I tasted the liquor, and found that it was very sweet and pleasant. The object of the nocturnal expedition of the ants, and the source of their honey supply, were thus revealed, These galls are of various sizes, from that of a currant downward. Most of them were of a Turk’s-head shape, some flattened spheres. They are placed in groups of two and more along the stems of the branches ; they are commonly of reddish-brown color, marked with black patches, but some of them are of a brighter tint, almost rose-color, Some of a livid yellow marked with black, some almost green. By cutting off a few of the clusters and removing them from the ants, I saw that the sugary sap issued from several points upon the gall, which in some 3 26 PROCEEDINGS OF THE ACADEMY OF [1881. cases became beaded with six or more globules, several times larger than a pin-head. By removing these beads successively, I found that during the night one gall gaye out at least three series. The continual flitting of the ants from branch to branch and gall to gall, was thus explained: the successive exudations invited their frequent return to the galls from which they had formerly fed. When the branch had first been brought to the tent, some of the gall-bearing twigs had been clipped off and placed within the artificial nests, but received no attention from the ants. Some of the bleeding galls were now introduced, which were instantly covered by the ants, and soon cleaned of their beaded sweets. An examination of the first galls explained the reason for their neglect—they were sapless. NECTAR-PRODUCING GALLS.—A number of galls of various sorts and sizes, was collected for dissection. They were readily divided into two classes (1), the livid and greenish galls which were soft and entire; (2) the darker colored ones which were hard, unyield- ing to the touch, and pierced at one side by a small, smooth, reg- ular, cylindrical cavity. It soon appeared that the bleeding of honey-sap was confined to the first class. Upon cutting away the soft pulpy fruit (@f it may be so termed), a hard whitish-green ovoid cell, not unlike a cherry seed, was found at the centre. It was about one-eighth inch in diameter. Lying outside of and against this, in a little cavity, I found in one gall a minute, living erub (Pl. IIT, fig. 12), The body was white, of eleven segments, the head tipped with a brownish hue. The inner cell when opened, showed a spherical cavity in which was a very minute gelatinous pyriform object, which adhered to the side of the cavity. I had no microscope with me, and in lieu of facts, can only conjecture that this may have been an embryonic form of an insect, which matures later in the season. The hard galls were next dissected. They are all pierced on one side, invariably near the base (figs. 10, 11), by a circular open- ing made by the matured gall-insect in its escape. Fig. 11 repre- sents one of these, a turban-shaped gall, magnified about three times the natural size. A section view of the gall (fig. 10) shows that the exit hole Teh) penetrates the interior cell-case, which must therefore serve as the cocoon in which the pupa transforms. Inside of some of these cells I found traces of a flossy texture, The cells are commonly spherical, but (as in fig. 10) sometimes 1881. | NATURAL SCIENCES OF PHILADELPHIA. a evo-shaped. They are separate from the rest of the gall, from which they quite differ in appearance, and are of a firmer sub- stance. In fig. 10, the gall is three-sixteenths of an inch in length, of which the cell occupies two-thirds, that is, one-eighth of aninch. The largest gall observed had an outside measure- ment of three-eights inch long and the same across the top. In one of the galls opened, I found an imperfect insect (imago), which is identified by Mr. E. T. Cresson as of the genus Cynips, a true gall-fly. The specimen would not permit further identifi- tification. j At the meeting of the American Association for the Advance- ment of Science, held in 1880, at Boston, I had the pleasure of presenting the substance of this paper to the Entomological Section. My account of the extravasation of the galls, as above, caused much comment, the result of which was to confirm the accuracy of the observation which had been challenged. Prof. C. V. Riley, well known as an entomologist, declared upon his own observations, that many galls exude saccharine matter, citing among others, those of certain Phylloxere on Hickory, one of which he had named carye-gummosa on account of the abundance and stickiness of the exudation. Mr. H. F. Bassett, who has made extensive and careful studies of galls, said that he had found many species of galls visited by ants ! - Specimens of the oak-gall visited by Melliger were sent to Mr. Riley, concerning which he says: The gall is one that is found quite commonly in the Rocky Mountain region on Quercus undu- 1 American Entomologist, Dec., 1880. The following additional remarks will be interesting in this connection: Mr. E. P. Austin remarked that the chemical composition of sugar and woody fibre are the same, and that sugar could be produced by conyersion from woody fibre in the plant. Dr. J. L. Le Conte said that he understood tannin to be a conjugation of gallic acid and sugar. Mr. B. P. Mann suggested that some light might be thrown upon this food-supply of the ants, by the nature of niuch of the moisture which appears occasionally at night in great abundance on the leaves and other portions of plants, and which is usually mistaken for dew. This moisture, it is said, differs from dew in being produced under circumstances which would not account for the formation of dew, and in containing a perceptible quantity of sugar. It is the ordinary watery ex- cretion from the surface of the plant, which, under favorable conditions of the atmosphere, collects in beads or in drops, instead of evaporating as rapidly as it is formed. 28 PROCEEDINGS OF THE ACADEMY OF [1881]. - lata, as determined by Dr. Engelmann, who sent me the same gall in 1874, though I had previously collected it myself. It is, un- doubtedly, an undescribed gall, and a yery similar one occurs on the Quercus macrocarpa in the Mississippi valley. It has the ordinary woody texture that belongs to so many Cynipidous oak- stem galls, and the architect develops in a paler cell that occupies a large part of the interior of the gall. When fresh, the gall is quite bright-colored, inclining to crimson or scarlet. It seldom attains a larger size than an ordinary pea, and differs from similar galls in my cabinet by having frequently a rather broad, flattened crown, though this character is by no means constant.! NocrurnaL Hasirs.—It has already been said that the ants collect the oak-gall nectar by night. Observations daily re- peated upon a number of nests, determined that they leave their nests for the oak thickets at or near 7.30 o’clock P. M., and between that hour and 8 o’clock, which is about the time of sun- set in July and August. Previous to the departure, the crater, gate and exterior of the mound become gradually covered with swarms of insects whose yellow bodies quite hide the red gravel surface of the nest. The marching of the honey-gatherers.has already been described, but always there remained a yery numer- ous force at home, who were seen at all hours, of the night. on guard within and around the gate. (PI. V, fig. 25,) .The return home began about or a little before midnight, and continued until between four and five, which was near daylight at that season. One or two extracts from my field notes will indicate the facts on this point. “11.30 P.M. Some ants returning home; the moye- ment very slow and deliberate. . ..., 12.30. Quite a number are now returning. Some are also still going outward... Numbers of workers patrol the mound and vicinity challenging nearly all in- comers, who have to stand the test and give the required satisfac- tion. What is the antennal password? None of the returning repletes are tolled by the home sentries. . ..... This morning at 4.10 A. M. the ants were seen coming in from the oak bushes, most of them well laden, but others not so full. There are evidently degrees of success in honey-gathering among them. Some of the dwarfs had very full abdomens. . ... 4.30 A. M. 1 He suggests for the gall the name Cynips quercus-mellaria, | Am. Ento, Dec. 1880. 1881. ] NATURAL SCIENCES OF PHILADELPHIA. 29 The ants are returning in numbers and rapidly moving from the brush to the nest. It is about daylight.” In these night observations the light of the lantern seemed to cause the ants in column no little disturbance. They would go toward the lantern as it sat on the ground near the trail, appear to examine it, then move away. It really seemed to confuse their ideas of locality, and shake their confidence as to the site of the trail, although no one was finally thrown off the track thereby. _ The sentries at home were always more or less excited by the light, and delicate manipulation was everywhere required in order to preserve the natural conditions and get the natural behavior. At no time were the ants seen during the day except when it rained, and then only a few sentinels appeared at the gate. Ordi- narily the entrance, as far as the eye could see, was entirely abandoned. It is doubtful if Melliger can endure a great amount of sunlight and heat. While excavating a nest, a number of specimens were collected in a large empty glass bottle, which was set aside for further use. Not more than three minutes afterward when I took up the vessel to insert more specimens, those already collected were dead. The sun had killed them. I was surprised at this quick fatal issue, and tried to revive the insects; but no, they were quite dead. The sun was of the usual August tempera- ture, but the bottle was large, and such a result in so brief a time argues extreme sensitiveness to the heat. I have observed that the agricultural ants! always avoided the noonday heats of Texas, which are certainly intense; and indeed all ants appear to me to shun, more or less, the midday fervor of the sun. But Melliger doubtless is more susceptible to solar influences than most of her fellows. It cannot therefore be wondered at that she seeks her food under the shelter of night. V. QUALITY OF THE ANT HONEY. A number of the honey-bearers were unavoidably injured and their abdomens broken during the excavations of the nests, and I observed from these the quality of the honey. It is very pleasant, with a peculiar aromatic flavor, suggestive of bee-honey, and quite agreeable tome. Dr. Loew describes it as having “an agreeable taste, slightly acid in summer from a trace of formic acid, but perfectly neutral in autumn and winter.” It contains, according 1 Op. cit., p. 18. 30 PROCEEDINGS OF THE ACADEMY OF [188]. to this writer, a little more water than the honey of bees, and has therefore somewhat greater limpidity. Fortunately, the composition of this ant-honey has been sub- jected to a thorough chemical analysis by a competent authority, Dr. Chas. M. Wetherill.! The experiments were made at the request of Dr. Leidy, from specimens of IM. melliger-mexicanus collected by Mr. Langstroth at Matamoras, Mexico.”» These ants showed the variations observed by me in the distension of the abdomen, and the amount and color of the honey. Six of the average-sized honey-bearers were weighed, and showed the average weight of the honey-bearer’s body alone (without honey) to be 0-048 grammes, and the average of honey in a single ant 0°3942 grammes. The amount of honey was therefore 8-2 times greater in weight than the body without the honey. The density calculated for the ants filled with honey was 1:28, and for the bodies alone 1:05. Dr. Wetherill’s calculations expressed in English Troy weight would allow about six grains for the weight of each honey-bearer. © It would thus require about one thousand (960) honey-bearers to yield one pound of honey (Troy weight), or about twelve hundred (1166) to yield a market or avoirdupois pound. The syrup extracted from the ants had an agreeable sweet taste, and an odor like that of the syrup of squills. When set aside as removed it showed no trace of crystallization to the naked eye or under the microscope. Under high powers fragments of organic tissue were seen. When evaporated by the heat of steam, it dried to a gummy mass, which did not exhibit traces of crystallization after standing for a couple of weeks. This mass was very hydroscopic, becoming quickly soft from the absorption of water from the atmosphere. It dissolved with- out residue in ordinary alcohol, leaving a residue in nearly absolute alcohol. These solutions did not crystallize when set aside. They had exactly the smell of perfumed bay rum. After various tests, which are described, Dr. Wetherill analyzed by combustion with oxide of copper and chlorate of potassa a portion of the gum- ke substance which resulted after the ant-honey had been left in vacuo for two weeks. As this was not perfectly hard, but ofa sticky nature, it was necessary to introduce it into the combustion 1 Proc. Acad. Nat. Sci. Philad., Vol. VI, pp. 111, 112, 1852. 2 Thave some of these still in good condition after twenty-nine years’ preservation in alcohol. 1881. ] NATURAL SCIENCES OF PHILADELPHIA. 31 tube upon a piece of glass. 0497 of honey gave 0.306 of water, and 0.684 of carbonic acid, corresponding to a percentage of C — 37:535, H — 6.841, O by loss —55'634. This corresponds, as nearly as could be expected under the circumstances of the analysis, with the formula of crystallized grape sugar, C,, H,, Ou. Dr. Wetherill, who in this analysis was especially seeking light upon the origin of the ant-honey, thus announces his conclusion : “It results, I think, from these experiments, that the honey con- tained in the Mexican ant is a nearly pure solution of the sugar, so called, of fruits, which is in a state of hydration, isomeric with grape-sugar, Cy Hy Ou, and differing from grape-sugar in not erystallizing.” It is certainly an interesting confirmation of the value of this reasoning from analysis, that the ants have been proved by field observations to have collected their honey-dew as Dr. Wetherill concluded, from the nectar of plants. Thus the methods of cabinet and laboratory, and the objective studies of the field, confirm and complete each other. With regard to the acidity of the ant-honey, which has been referred to, Dr. Wetherill found that it reacted slightly acid to blue litmus paper, but want of material prevented satisfactory experiments. He was in doubt as to whether it was formic acid, or acetic from the oxidation of the alcohol in which the ants were preserved. A portion of the alcohol (reacting acid like the honey) neutralized by caustic potassa, when distilled with sulphuric acid, gave an aqueous acid liquid, which, on addition of nitrate of silver, gave a whitish precipitate, becoming black on boiling, ren- dering the supposition of formic acid probable. The uses to which the Mexicans and Indians put this ant-honey are various. That they eat it freely, and regard it as a delicate morsel is beyond doubt. Prof. Cope, when in New Mexico, had the ants offered to him upon a dish as a dainty relish. The Mexi- cans (Loew) press the insects, and use the gathered honey at their meals. ,They also are said to prepare from it by fermenta- tion an alcoholic liquor. Again, they are said (Edwards) to apply the honey to bruised and swollen limbs, ascribing to it great healing properties. Dr. Loew’s suggestion to bee-keepers to test the commercial value of these ants as honey producers is wholly impracticable. The difficulties of farming the colonies, gathering the supply, and the limited quantity of the product, would prevent a profitable industry.. The greatest number of honey-bearers in a Oo 2 PROCEEDINGS OF THE ACADEMY OF [1881. large colony, taking my observations as a standard, will not exceed six hundred, which, counting six grains of honey to the ant, would be little more than one-half pound avoirdupois. Besides, the senti- ment against the use of honey thus taken from living insects, which is worthy of all respect, would not be overcome. ‘The Mexicans and Indians will therefore probably not be disturbed in their monopoly of the honey-product of the nests of Melliger. VI. IntTERIOR ARCHITECTURE. GATE ARCHITECTURE.—In order to determine the gate archi- tecture—a term by which I characterize the structure of the nest nearest-to the entrance—several formicaries were car efully opened and studied. Four of these are here given as fair types of all. It will be seen from these that a general similarity of plan preyails. The gate itself is a single tubular opening in the centre of the mound, from three-fourths to seven-eighths of an inch in diameter. It issmooth within, and penetrates the mound and the earth perpen- dicularly to a depth varying from three and one-half to six inches. This gate is funnel-shaped at the top, and the funnel (P1.1V, fig.14, F) is gravel-lined, differing therein from the lower part or nozzle of the gate (fig. 18, N).. The nozzle descends perpendicularly, or with a slight slope, for three inches, more or less, and then deflects at an-angle more or less abrupt, cee an arm (A.) usually shorter than the nozzle. This leads into a series of radiating galleries and rooms, and the point of deflection may be called the vestibule, V. These galleries and rooms appear to extend quite habitually beneath and. chiefly in one direction from the gate. There are indeed galleries immediately surrounding the gate on every, side ; but these appear to be limited except in the one direction, within a radius of about eight to ten inches, and to the same distance in depth. These general statements may be illustrated and expanded ‘by the following details of particular nests. 1. Nest No. 7, fig. 19, was,a small nest three and one-third inches in diameter. The gate had a perpendicular depth from the surface of three inches... Thence at nearly a right angle it bent south-east for two and one-half inches, forming the arm, A, and meeting at V a series of branching galleries, a, b, c,d. Gallery a, bore westward, terminating under the gate; 6, bore southwest, appearing to run upward toward the surface; c, extended down- 1881. ] NATURAL SCIENCES OF PHILADELPHIA. 33 ward and southward at a sharp inclination, entering a long room. E, was a small circular chamber, at one end of which was a beau- tiful gallery, f, running deep downward and inclining slightly west. It was entered near by and above by another gallery, d, running toward the surface. 2. Nest No. 6, fig. 18. The depth of the gate, G, was three inches; the length of the arm, A, two inches. The gallery into which A opened toward b, divided at one end with two branches separated at their mouths by a little column of two stones resting one upon the other. The gallery, c, could be traced at least six inches downward, and a gallery opened directly downward at a. 3. Nest on Eaglehead Ridge, Pl. V, fig. 20. This nest, from which many of my night studies were made, was finally opened, and the section view, fig. 20, taken. The vestibule, as in the above examples, also opened into a main gallery, b, which led to the northeast, and joined a circular gallery which passed around the vestibule and terminated in an ovalroom, A. At the other end it entered a circle, which widened upon one side into a bay-room, and sent off a couple of branches, one of which, ¢, was a chamber. Two galleries, g g, opened downward. Beyond this, southward, was a long waved gallery, D D, which ended at e e, and branched at h. Galleries, g g, in this series, also led downward. No. 4. Fig. 23, Nest No. 4, on Adams Ridge. The diameter of this mound was three and one-half inches at the top and seven inches at the bottom. ‘The vestibule sloped eastward from the summit, downward three inches to the main gallery, which had three branches, x, y and z; x was followed six inches northeast and upward; y, extended southwest and downward; z, southeast and downward. A gallery, 1, ran upward from z, and connected with «. Another, 2, opened on the southwest into a room, A, six inches long and three inches wide, at the west end of which were galleries dividing north and south. A third gallery separated from z at 3, and bent northward, apparently uniting with a room, A, five inches long. This room was entered again by a widened mouth, Be, about one-half inch above z. At the vestibule and upper part of z were a number of cocoons. The room, A, was five inches below the surface of the ground at G. GALLERIES AND Honry-Rooms.—The last figure gives an idea of the relation of some of the honey-rooms to the gate and the upper series of galleries. These rooms lie at least as near to the 34 PROCEEDINGS OF THE ACADEMY OF [1881. surface as six and eight inches. They vary in size, but for the most part, are about five or six inches in length and three or four in width. They are irregular in their outlines, but havea general tendency toward the oval... One of the most irregular is figured at Pl. V, fig. 21, HR, a large chamber which lay nearly underneath the gate. The gallery, g g, into which the vestibule opened, debouched into this room, and a portion of the gallery roof unbroken is shown at ug. At B, appeared a bay-room, or enlargement of a gallery, which penetrated the earth horizontally at one end and at the other seemed to wind into the vestibule. The height of the rooms at the walls or sides is from one-half to three-fourths of an inch. The roof is vaulted, thus causing the height to increase gradually until at the centre it is one. and one- half inches, which is the greatest distance that, 1 measured, FLoors AND Roor.—The floors and walls are well nigh invari- ably smooth, quite smooth some of them. The roof, on the con- trary, is rough, presenting the natural condition after the sandy pellets of earth and the little pebbles had been picked out by the workers. This can hardly be otherwise than by purpose, precisely as with the smoothness of the floors. The roughness of the roof evidently greatly favors the use to which the honey-bearers put it as a perch. So the smoothness of the floor and walls much. better adapts them for the use of gangways. The amount of travel to and fro must be enormous, it is true, ina large formicary ;. but I cannot think that the resulting friction will account for the smooth- ness, independently of the purposed masonry of the ants... Inthe galleries the entire surface, above and below, is smooth, a condi- tion which might be anticipated on the ground of adaptation, GALLERIES AND Rooms.—The galleries are tubular openings, varying somewhat in size, from one-half to three-fourths of an inch, and even more, in diameter. A vertical section, however, uniformly shows a quite perfect circle. The underground formi- cary may be described in general terms as a system of galleries and rooms, arranged in seyeral horizontal series, one above another, approximating the order of “stories” in a house, and intercom- municating at many points by vertical galleries. The character of the interior architecture can, perhaps, be best shown further by giving somewhat in detail my studies of one nest. The nest selected for exhaustive exploration was situated upon the summit of Adams Ridge, just above the nook within which my 1881. ] NATURAL SCIENCES OF PHILADELPHIA. 35 camp was located. Three entire days, besides other portions of time, were spent in this work by myself and assistant. The nest interior sloped eastward, and toward the base of the hill, and occupied a space (in round numbers) eight feet long, three feet high and one and one-half feet wide, the whole tunneled through the soft red sandstone rock of which the ridge consists. This rock is much of it quite friable, crumbling readily under the pres- sure of the hand, but packs tightly under the stroke of mallet and chisel, thus making difficult mining for men if not for ants. _ Most of our work was done with the chisel, and the galleries and rooms had to be worked out with knives.! These thirty-six cubic feet of rock were fairly honeycombed by the series of galleries and cham- bers above referred to. The dimensions of the exterior nest are as follows (see Pl. IV, fig. 15): Height, north side, 24 inches; west side, 13 inches ; east side, 1Z inches; south side, 12 inches; distance across the top, @ ¢ —10 inches; distance around the base, @ 7 e ¢ = 29 inches; distance around the crater, mon r = 8 inches; eastern ridge,of the crater, vn — 1} inches; western ridge of crater, ms = 1 inch; distance across the gate at x z—1 inch, at sv— 4% inch; depth of the gate before bending, 4 inches. The mouth, as appears from méasurement, was ovate (Pl. IV, fig. 14), but the entrance beyond was a circular tube. The mound was removed and the soil carefully scraped away. Close to the surface, at the distance of one-half to three-fourths of an inch, openings were found of various sizes, from one-fourth to one inch in diameter. These openings occurred at various dis- tances from the gate, on all sides, four and one-half, five, five and one-half, eight. eight and one-half inches and upwards to ten inches on the northwest side, eighteen inches on the south side, and eighty-two inches on the southeast, in which direction the formi- cary extended. Toward the termination of the nest, however, they did not appear so near to the surface. Section viev's were next had by cutting across the nest. On the north side I found no galleries at a greater depth than eight inches. On the south side, the first cutting was made east and 1 While engaged upon this part of my work, I was pleasantly surprised by a brief visit of Prof. A. S. Packard. I am glad to be able thus to refer to his valuable testimony in confirmation of some of the statements of this paper. 36 PROCEEDINGS OF THE ACADEMY OF [1881. west, and thereafter the rock cleared away outwardly, until the end of the nest, when the cutting was made inwardly from the starting point toward the gate. The character of the architecture is the same throughout the entire nest, so that the following views. will suffice to typify all the interior. The figures Pl. VI, fig. 35, and Pl. V, figs. 16,17, give views of vertical and horizontal see- tions made from the gate (southeast), the bottom of the section being twenty-one inches below the surface and the distance of the furthest point from the gate twenty-three and one-half inches. Fig. 35, Pl. VI, is a front view of galleries looking south, and exhibits a surface about seventeen inches in length by seven in height.! The main series of galleries: within this area are accu- rately shown, but. the connecting vertical galleries were broken awayin the excavation, and are not figured. Fig. 16 is a vertical section showing the southwest and south- east sides of the excavation at the same point as the preceding figure, part of which is included in this view.” There are here shown the general tendency of the galleries (g, 9, g) toward stories, arranged one above another; the relative position of the honey- rooms (R, R), and the relation of the series to the large honey- rooms, C, D, E, shown fully at fig. 17. The broken lines, ¢ 1 d, and eh k, show a series of rooms. some of which were occupied by larve and some by honey-bearers, The large rooms, C D E, Fig. 17, belong to the lowest series, and are figured and described as fairly typical of all the honey-rooms and other chambers. They were carefully uncovered by chisel and knife, and after being sketched, a plaster cast was taken of them, which is preserved in my collection.*? These rooms were of 1 Detailed measurements.—a to G = 5% inches; G tod = 11 inches; e to f = 10 inches; 4toz—11 inches; & to/—11 inches; m to n=38 inches ; b to ¢ — 25 inches; 0 to f—3% inches; p tog —1% inches; g to k — 2% inches; 7 to 1 = 2% inches; 0 to r = 3% inches: 1 tos = 64 inches. 2 Fig. 16 measurements.—a to b = 4% inches ; ¢ to d ="10 inches; eto f = A4inches; 7 to¢ — 4% inches; 7 to k = 45% inches; c to 7 — 10 inches. The gallery, j, appeared to connect upward with the lowest series of rooms, ef hk. 3 T succeeded by vast painstaking and labor in securing a number of fine specimens of the architecture, which were carefully packed in boxes and committed to the Express Company at Colorado Springs. The company received a heavy bill for transportation, and delivered my beautiful and 1881. ] NATURAL SCIENCES OF PHILADELPHIA. 37 an irregular oval shape; in length five, three and one-half, and six inches successively (C, D, E),; and were of an average width of about four inches. They were not built upon a level, the origin of ©, at b, being three and one-half inches above the middle point of D,and six inches above the termination of E. A side gallery, g g, skirted two of the rooms, and appeared to open-upon a fourth chamber at F, which, however, was too much broken in the digging to beidentified. Of course, only the floor and part of the side walls of the rooms are shown, but. the roofs were vaulted and rough, as already described, and rose to the height of three-fourths to one and one-fourth inches... Within them, clinging to the roofs, were packed the rotunds... The number in each room averaged about thirty ; and as there were at least ten chambers thus occupied, the number of rotunds in the nest was certainly not less than three hundred. Of far the greater proportion of these the abdomens were distended to a perfect sphere. THE QuEEN Room.—TI had the good fortune to capture the fertile queen of this colony.. She was found quite near the extreme end of the formicary, in'a nearly circular room four inches in diameter. The series of galleries and honey-rooms which composed the formicary terminated in ‘a single gallery (fig. 22, g g g), about. eighteen inches long, three-fourths inch wide and one-fourth inch deep. The gallery sloped sharply with the slope of the hill-side on which the nest was made. Near the middle partthereof was the queen-room (C), being seventy-two inches from the central gate and twenty-eight and one-half inches below the surface of the hill. Besides the queen the room con- tained a large number of naked grubs, callows, honey-bearers and workers. It is not improbable that the queen habitually dwelt in or near this room; but it may be that during. the successive attacks upon the nest, the workers bore their-queen still further and further from the point of danger until the limit was reached. Ten inches below the queen-room, the gallery, g gg, was con- tinued until it finally terminated in a small circular chamber (EK) or bay-on the one side, and on the opposite side a narrow gallery (tg), which curved upward.- This was the end of the formicary. costly specimens at the Academy broken in pieces! It was an act of gross carelessness, Which merits this notice, as some specimens brought home in my trunk survived even the ‘‘ baggage smashers.”’ 38 PROCEEDINGS OF THE ACADEMY OF [ 1881. It was eighty-two inches from the central gate, forty and one-half inches below the level of the main nest gate, and twenty-nine and one-half inches below the level of the hill-side. The entire length of the formicary from northwest to southeast was thus seven feet eight inches.! VIT. QuEEN LIFE. The captured queen of the large excavated nest was transported to Philadelphia, placed in one of my artificial nests, a large glass globe, and afforded several interesting observations upon. her habits. Her Bopy-Guarp.—After the usual custom of ants, she was continually surrounded by a guard of workers. (Pl. VI, fig. 29) varying in number, but usually as many as twelve or. twenty. These attendants quite enclosed her, and restricted her moye- ments, apparently watching and guarding her with great careful- ness. On one occasion when she escaped to the upper surface of the nest, she was followed and seized by a worker-major, who interlocked her mandibles with the queen’s (Pl. VI, fig. 26) and dragged her down the gateway into the interior. The royal lady gave only a passive resistance, holding back somewhat heayily. Depositine Eaas.—I quote from my notes the deseription of this process, the various stages of which I was also able to sketch. “The queen has been laying a small heap of eggs. She is now on a little elevation of earth, surrounded by a number of workers of all castes,some of whom lick her abdomen, especially beneath and at the apex. One, meanwhile, gives her food in the usual way, by regurgitation. J see the tongues of the two insects overlap in the 1 While preparing clay models of some of the above examples of ant architecture for my cabinet of Insect Architecture in the Aeademy of Natural Sciences, it was suggested that moulds be made, from which plaster casts could be taken, for the benefit of such other scientific collec- tions and public museums as they might be wished for. This I had done, and the moulds are now in the hands of the Curator in charge of the Academy, by whom they will be furnished, upon proper order, at the cost of reproduction and packing, as nearly as may be. Five specimens are cast, viz., those figured at PI. Il, fig. 4, and Pl. V, figs. 16, 17, 22 and 23. They are cast natural size, except fig. 16, which is half size. The cost, painted natural color, will be $10 for the set, unpainted $6. Orders should be sent to Charles ¥. Parker, Curator in charge Academy of Natural Sciences, Philadelphia, Pa., U. 5, A. 1881.] NATURAL SCIENCES OF PHILADELPHIA. 39 act. The queen’s abdomen is raised high, her head is stooped, she lifts the abdomen up and down. -The workers have clustered under her body, giving her somewhat the appearance of a success- ful candidate undergoing ‘a chairing.’ She has changed her position; the workers follow, quite surrounding her. ‘Two are beneath the abdomen, which is depressed now, the head being elevated. The attendants sit down patiently to watch. They keep their antenne moving continually, while they amuse them- selves by cleansing their persons. The queen moves; a dwarf seizes a fore-foot and attempts to control her course. This and ‘‘nipping ” with the mandibles, is the common mode by which the guard directs the queen’s motions. The eggs laid are in an irregular mass about one-eighth of an inch thick. There are twenty to thirty minute yellowish, ovoid objects, which adhere to each other. The workers surround the mass, some appeared to lick it. The queen straggles over the eggs, places a foot upon the mass. A dwarf seizes the foot hastily and draws it back, while another worker catches up the egg-mass and draws it aside.” The observation was made at 11.20 P.M.; at 1 A. M., when I retired, no change had occurred. This is as much of this interest- ing behavior as I was able to observe in this female. I have, however, seen the actual deposition of the eggs by a queen of Camponotus pennsylvanicus.! VIII. Acrs of BENEFICENCE. In the natural sites the workers showed great interest in the preservation and removal of the rotunds, dealing with them very much as with the larve. As the honey-rooms were opened and the rotunds disturbed from their roosts, the workers of all castes rushed eagerly to them, and dragged them into the unbroken interiors. Sometimes several ants would join in removing one rotund, pushing and pulling her along. One sketch (PI. VI, fig. 27) made in my notes, represents a major pulling a rotund, whom she has seized with her mandibles by the outer abdominal wall, while a dwarf-worker is mounted upon the globe, standing upon her hind legs “ a-tip-toe,” as it were, pushing lustily. Another sketch (Pl. VI, fig. 36), caught on the spot, represents a worker- major dragging a rotund honey-bearer up the perpendicular face 1 See a note in ‘‘ Proceed. Acad. Nat. Sci. of Phila.,’’ 1879, p. 140. 40 PROCEEDINGS OF TIE ACADEMY OF. [1881. of a cutting made in the excavation of the nest. The mandibles of the two insects were interlocked, and the worker backed up the steep, successfully drawing her protegé, This interest is maintained in the daily life of the formicary. The workers were continually seen hovering about the rotunds as they hung from the roof of my nests, or as they lay upon the floor cleansing their bodies. It is evident that these creatures are regarded as dependents, and, as with the queen, virgin females, males and larve, are fed and tended by the active members of the community. In all these cases the same communal instinct would of course control action, giving at least the semblance of beneficence. ; Lack or Iypivipuat BENEFICENCE.—But a great number of ex- amples fell under notice which go to throw doubt upon the pos- session of any personal or individual sentiment as towards special eases of need, outside of the above limit. Some of these may deserve permanent record. 1. In making up my artificial nests, I placed in the natural soil, which was closely packed down, and then introduced the ants, knowing that they would work out their own habitations. The honey-bearers were thus mingled upon the surface with the workers, upon whom fell the entire task of digging galleries. In this work, and in the distribution of the excavated pellets, there was much room for the exhibition of individual carefulness and tenderness toward the honey-bearers. Not a single such instance was noted, although I watched closely and with some anxiety to discover such excellencies in my little friends. On the contrary, the exhibitions of an apparent cruel neglect and positive cruelty were many. The grains of sand and soil were heaped around the rotunds (Pl. VI, fig. 31) until the poor creatures were literally buried alive. It would have been easy for the busy masons to draw their fellows aside and thus carry on their work. But it either never occurred to them to do so, or the disposition was wanting. 2. Again, as the openings were made into the earth, most of the rotunds, not prevented as above, managed to roll down the galleries and secure a place in the honey-rooms. They were not observed to be aided in this by the workers, and I believe that they attained their perches unaided: Some of them, on the route, became fastened in the gang-way in most uncomfortable positions, 1881. | NATURAL SCIENCES OF PHILADELPHIA. 4] heads downwards, bodies awry, etc. The workers passed by and over them continually, for many days, without the slightest apparent concern, and certainly without a single observed effort to relieve their comrades, who could readily have been extricated and drawn into the chambers. 3. It frequently happened that the rotunds dropped or were shaken down from their perch against the roof to the floor. These creatures remained in the positions in which they fell, except when they chanced to so fall as to be able to clasp with their claws some clod of earth, or bit of gravel, or the rough surface of the project- ing walls or roof, In such case, they either recovered their perch, or placed themselves in comparatively comfortable postures. The greater number, however, fell upon the round abdomen in such wise that the body stood up quite erect (Pl. VI, fig. 32), leaving the legs thrust out unsupported. These unfortunates were faith- fully attended, often cleansed and caressed, but in no single instance did the workers attempt to right them and restore them to the roof. Yet they were abundantly able to do so, with little effort, and the fallen rotunds were in sore need of help. Some of these lived for two months and longer in this awkward position, but it was very evident that they were extremely uncomfortable. When it was practicable to extend my help to those near the surface it was eagerly accepted, the offered stick or quill clasped by the mandibles, sometimes assisted by the feet, so firmly as to enable me to transfer the heavy creatures to any point, even to lift them out of the nest. Here again the idea or at least the act of helpfulness was lacking. If we are to suppose the power of communicating their distress and desires to have been possessed by the bearers, we must think the workers even yet more lacking in feeling and intelligence. 4, One honey-bearer was partly buried under her perch, that portion of the roof having fallen. Her abdomen was quite covered by the fine sandy particles at the margin of the little landslide. The task of rescuing her would have been easy to the workers, but it was never undertaken. A sketch (PI. VI, fig. 28) was made, shortly after the occurrence, which shows one worker- minor standing before the rotund with head and body erect, antennz atent, with every mark of curious interest in her deport- ment. She watched the struggles and mute appeals (as it seemed to me) of her unhappy comrade, who by great exertion had suc- 4 42 PROCEEDINGS OF THE ACADEMY OF [188]. ceeded in heaving up the clod, and then “ passed by on the other side.”? Meanwhile a second worker was perched atop of the clod, coolly and cosily combing her back-hair and antenne! This tableau is simply characteristic of the ordinary behavior of the workers. An apparent exception was noted in the case of a semi-rotund who was overtaken in a gangway by water with which I was sup- plying the community, and stuck fast ina bed of mud. Fora long time the workers, who were incited to masonry, as usual, by the water supply, dug and traveled around and over the imbedded ant without notice of any sort. Finally one stopped and licked the antennz and head of the prisoner, who began to struggle, and so dropped down a little into the gangway. Meanwhile the first- comer had left. A second ant stopped, applied the tongue a moment, gave a little tug at the unfortunate, and was off. Still the stream of workers passed on. Finally, an additional pull from below was given by a concealed worker, but when I closed the observation the ant was still imbedded in the mud within the gangway. It was impossible to decide in this case whether the helpers noted were moved by personal kindness, or rather (as is most likely), by the same impulse which directs them in ordinary mason operations and toward supposed dead comrades. Sir John Lubbock, who has made interesting experiments and observations with a view to testing the presence of benevolent feeling in ants, does not have a very high opinion of emmet charity, but concludes that there are “ individual differences,” and that among ants, as with men, there are Priests and Levites, as well as Good Samaritans. I am much inclined to the view that anything like individual benevolence, as distinguished from tribal or communal benevolence, does not exist. The apparent special cases of beneficence, outside the instinctive actions which lie within the line of formicary routine, are so rare and so doubtful as to their cause, that (however loth), | must decide against anything like a personal benevolent character on the part of my honey-ants. Such an example, indeed, as one of those cited by Lubbock, viz., the neglect on the part of co-formicarians to remove the decapitated heads of enemies from the limbs to which they are firmly clasped, does not seem to me as remarkable as it does to 1 Journal of the Linnzan Society, Zoology, Vol. XII, p. 497. 2 Op. cit. p. 492. 1881.] NATURAL SCIENCES OF PHILADELPHIA. . Sir John. I have often observed the same fact among various species, and, knowing by experience, the difficulty of unloosing those formidable jaws, clasped by their immense muscles in the rigor of death, would charge it to inability rather than indisposi- tion, that these adhering death’s-heads are not removed by kindly offices of comrades. But such examples as are here recorded, together with kindred ones given by Lubbock, may fairly be quoted against the existence of a personal benevolent character in ants. | However, the question can by no means be regarded as settled. CLEANSING AND FEEDING LARv#.—One or two miscellaneous observations may, perhaps, be allowed a place in this connection. The solicitude of the workers for the helpless larvz was a matter for continual admiration. The oflices of nurse do not seem to be confined to-any one caste, but the burden of duty appeared to be assumed by the dwarfs, and next to them the minors. When the grub is to be cleansed it is taken in the mouth, turned by the fore pair of legs, the antennz meanwhile touching and apparently aiding, while the mandibles are applied over the grub, their teeth apparently working chiefly within the annular divisions of the several joints. Doubtless this motion is accompanied by a free use of the tongue, but this I did not observe. When the grubs are to be fed, the workers pass from one to another, striding over them, and standing among them (PI. VI, fig. 34) as they lie in little groups. The wee white things perk up their brownish yellow heads, which they stretch out and move around, evidently soliciting food. ‘Their nurses move from one to another, apply the mouth for a moment, and pass on. At the slightest alarm the grubs are seized and hurried into the recesses of the nest. Their position is frequently changed, from higher to lower, from outer to inner rooms, and the reverse, with- out any purpose which I could discover or imagine. When this sort of transfer was not going on, the nurses would often be engaged in shifting the position of their charges, flitting restlessly among them, picking them up, turning them around, putting them down again, with an aimless uneasiness that bore an amusing like- ness to the dandling which human infants undergo at the hands of certain young mothers. Torter Hasirs.—It has been said that the honey-bearers are cleansed by the workers. This is the rule; but the rotunds are 44 PROCEEDINGS OF THE ACADEMY OF [1831]. not wholly dependent for this upon their fellows.. In one of my formicaries, the rotunds when placed within the light, began to cleanse themselves, without leaving their perch.. They held con to the roof by the two hind legs and one of the middle pair, and used the other middle and the two fore legs in the usual manner of ants.|. They were quite able thus to draw a leg through the spur-comb of one of the fore-feet ; to brush the head, ete. In one case I even saw a honey-bearer performing the offices of the toilet upon a worker. The latter held her mandibles apart, while the rotund licked the mouth parts; and from thence pro- ceeded to the vertex of the head. Both insects were in a semi: rampant posture the meanwhile. FRATERNAL RELATIONS WITH SISTER CoLONIES,— Revue et Magazin de Zoologie, Tome XII, 1860, p. 271. . 1881. | NATURAL SCIENCES OF PHILADELPHIA. 65 (Mexricanus) has of course yielded to that of Llave, moditied, however, from JJelligera to Melliger. The Colorado insects, upon which the studies of this paper are based, I have ventured to regard as a new variety, and have named Myrmecocystus hortus- deorum, and thus have retained Wesmael’s name as a variety name. Tt will be well to state briefly the facts in the economy of these insects indicated in the foregoing and other papers, in order to mark precisely the new facts which have now been communicated here. Llave’s information was all at second hand, he having made no personal observations of the habits of Melliger. From a person living at Dolores, a village in the vicinity of the city of Mexico, he learned : 1. That the ants were popularly known under the name of Busileras; 2. That they do not erect heaps of earth at the entrance to their nests ; 3. That on opening the nest, a species of gallery is reached, to the roof of which certain ants are suspended, packed one against the other ; 4, That these ants cover the roof as well as the wall of the gallery. '5. The women and children of the valley know these nests per- fectly well, and frequently open them for the sake of the honey- béarers, or rotunds. The honey is sucked from the abdomen of the rotunds, with great relish, at the nests; or, if it is wished to pre- serve them, they are lifted by the head and thorax and placed upon plates, in which they grace the village feasts, and are eaten as delicacies. 6. The rotunds when thus placed together, stir around, lay hold of and tear one another, and finally end life by bursting. 7. The skin of the abdomen, which binds the segments together, is so thin, and the upper coat so distended, on account of the quantity of honey which it encloses, that the least pressure suf- fices to cause the ants to disgorge. 8. When they do not so disgorge, that is, by elevating the head and thorax, the honey diminishes, and the ants eat it. 9. Dr. Llave observed, moreover, from specimens of the ants sent to him, that there were different castes of workers and degrees of distension in the abdomens, and v a PROCEEDINGS OF THE ACADEMY OF , (rset 10. That the honey in the rotunds varied in color from a crystal whiteness to a wine-color. Several of the above statements, as has been seen, are without foundation, but the majority of them are confirmed in whole or part by my observations. Wesmael,! who made his study from specimens sent him from Mexico by the Belgian Envoy, Baron Normann, records his credence of the theory announced by that gentleman, viz., that the honey-bearer elaborates the honey and deposits it in certain reservoirs, analogous to the cells of bees, for the nurture of the formicary. Baron Normann was unable to obtain examples of these reservoirs to send to Europe, or rather failed to do so under the conviction that they would be destroyed during shipment. In point of fact, such reservoirs exist only in imagination. One of the most perplexing accounts of the honey-ant is that of Mr. Henry Edwards.” The statements recorded are made at second hand from the verbal narrative of a Capt. W. B. Fleeson, whose observations were made at or near Santa Fe. They are so extraordinary and contradictory of my own experiences, that I am compelled to withhold credence, until some experienced observer shall have corroborated them, a result of which I have little expectation. According to this account, no exterior mound- let surmounts the formicary, but simply two openings into the earth. Within the nest, at a depth of about three feet, “a small excavation is reached, across which is spread, in the form of a spider’s web, a network of squares spread by the insects, the squares being about one-quarter inch across, and the ends of the web? fastened firmly to the earth at the sides of the hollowed space which forms the bottom of the excavation. In each one of the squares, supported by the web, sits one of the honey-making workers, apparently in the condition of a prisoner, as it does not appear that these creatures ever quit the nest.” But the marvels of this strange story are not exhausted. ‘‘’ The 1 Bulletin de l’ Acad. Roy. des Sci. et Belles lettre de Bruxelles, Tome Vp. 770. Pl. XIX, figs. 1-4. 2 Proceed. California Acad. of Sciences, Vol. V. 1873, p. 72; ‘* Notes on the Honey-making Ants of Texas and New Mexico.”’ % Of course, this is pure fiction, as no ant makes a web, or anything that could well suggest one. The cutting ant does make out of fragments of leaves a ‘‘comb”’ of more or less regular cells, resembling the nests of the paper-making wasps. ‘ 1881. ] NATURAL SCIENCES OF PHILADELPHIA. 67 inmates of the formicary are composed of two distinct species, apparently even of different genera, of ants. There are the ordi- nary yellow workers and .honey-bearers of Melliger, and besides, black workers, who act as guards and purveyors. One column of the blacks surrounds the openings on three sides, attacking, driving off or destroying all intruding insects. Another column bears, through the unguarded side of the hollow square, fragments of flowers, aromatic leaves and pollen, which (adds our author), by a process analogous to that of the bee, the honey-makers con- vert into honey.” One can hardly refrain from the thought that Capt. Fleeson was testing the credulity of the writer by one of those jokes of which naturalists are occasionally the victims. But, if the narra- tive is to be taken in good faith, I can only explain the facts by supposing, first, that the observer happened upon a nest of cutting- ‘ants (Atia fervens), within whose boundaries a nest of Melliger had chanced to be established, and had confounded the habits of the two as those of one formicary; or, second, that the cutting- ant, or some other species of similar economy, has really acquired the habit of kidnapping and domesticating the honey-ant for the sake of its treasured sweets, precisely as many ants domesticate aphides; or, as the slave-making ants, Formica sanguinea and “Polyerqus lucidus, domesticate Formica fusca and F. Schauffussi.* One of the latest accounts of the honey-ant, and so far as it goes, one of the best, is that of Mr. Saunders, the editor of the Canadian Entomologist,? who communicates to his journal some observations made by Mr. Kummeck, at Santa Fe.* According to this observer, considerable numbers of these insects are found in the mountains of that vicinity. Hesat by a nest six or seven 1 One may not be over rash in refusing belief even to facts that go counter to all past experiences, for the marvels of Nature are ever widening within our view. While, therefore, I am inclined to reject the whole story, I await the observation of some trained naturalist, giving the account the benefit of the above possible explanations. 2 Can. Entom., 1875, Vol. VII, pp. 12-13. 3 I may be permitted to explain why I did not go to New Mexico, to attempt on the spot a solution of some of the questions raised by these accounts. I had made every arrangement to do so, after my studies in the Garden-of-the-gods were completed, but on the morning that I was to break camp, was taken with a sudden and violent illness which compelled me to abandon my journey. 68 PROCEEDINGS OF THE ACADEMY OF (1881. hours and noticed the workers carry home leaves of different plants to feed, as he supposed, “the others that produce the honey.” This would seem to confirm the leaf-bearing habit quoted by Mr. Edwards from Capt. Fleeson. The inference as to the use of these leaves is, however, quite unwarranted, as the portage of leaves, etc., into nests is not an uncommon habit among ants of divers species. . Without stopping to discuss the question whether such material may contribute to the food supply of the formicary, it may be remarked that its most probable and ordinary use is for purposes of architecture or nest-building. Mr. Kummeck also makes the remark, which I had not seen at the time my own conclusions had been reached, that “ in early life none of these insects present any unusual distension of the body, but when arrived at a certain period of maturity some individuals begin to show a distended abdomen.” The ant honey has no commercial value among the New Mexi- cans. It hasa place, however, as a remedy in the domestic thera- peutics of the native Indians, who compound a drink by mixing three to four drachms of the honey with six ounces of water. The drink is used in cases of fever. The honey is also applied as an unguent in eye diseases, especially cataract. . To the above may properly be added two accounts of my own studies published in the London Journal of Science.! These are reports made by Mr. Morris, of the verbal communications in which my observations were originally announced to the Academy of Natural Sciences of Philadelphia. They were made and printed without any oversight or responsibility on my part, but are admirably, and in the main, accurately done. They haye been reproduced with various degrees of fulness in other journals. e. Such other notices of this ant as I have been able to find, and have had occasion to use, will be found properly referred to in the text of this paper, where those who are interested in the literature can readily find them. 1 Jour. Sci., February, 1880, “Living Honey Comb ; a novel phase of Ant Life.’ By Mr. C. Morris. Ibid. July, 1880, “ Habits and Anatomy of the Honey-bearing Ant.’? By Charles Morris. 1881. ] NATURAL SCYENCES OF PHILADELPHIA. 69 XIV. DEscrRIPTION OF SPECIES. FORMICARLA. Family FORMICIDZ. Subfamily CAMPONOTID& (Forel). Genus MYRMECOCYSTUS, Wesmael. Cataglyphis, Foerster, Verh. d. Nat. Ver. d. Rheinl., 1850; Mayr, Europ. Formic., 1861; Norton, Wheeler’s Report, Vol. V, Zool., p. 734. Monocumbus, Mayr, Verh. d. Zool.-bot. Ver. in Wien, 1855. Myrmecocystus, Forel, Etudes Myrmecologiques, Bull. Soc. Vaud. de Sci. Nat. M. melliger, Liave. 1. Var. mevicanus, Wesm. 2. Var. hortus-deorum, McCook. Workers.—Three castes, major, minor and minim or dwarf. Color, a uniform light yellow; the body is covered quite thickly, the legs more thickly, with short yellow hairs. The maxillary palps are very long, six-jointed, third joint longest; they. are covered, especially beneath, with long hairs, curved backward, Labial palps four joints ; mandibles with nine teeth. The head is quadrate, in the worker-major more rounded at the sides than with the minor and dwarf; wider than the thorax. Clypeus smooth, rounded, slightly flattened in front of the frontal area. Frontal area smooth, shining, triangular, somewhat truncated posteriorly. Ocelli sufficiently prominent ; a tuft of hairs on the face beneath, directed forward. The body is of good length, narrow and compressed beneath at the mesothorax; metanotum as high as, or slightly higher than the pronotum. The node cordate, cleft at the tip, thickened at the base, set perpendicularly upon the petiole. Anus strongly ciliated. Length, worker-major, 8mm.; worker-minor, 7 mm.; worker-minim, 55 mm. Honey-bearers —A sedentary class or caste distinguished by abdomens distended into spherical form by expansion of the crop filled with grape-sugar.. The length (including abdomen) is 13 mm. (one-half inch); the proportions and description of the head and body are those of the worker-major, of which it may be a developed form. Female.—Virgin queen, total length, 13 mm., as follows: Man- dibles, 1 m.; head, 2 mm.; body,5 mm.; abdomen,5mm. Width 70 PROCEEDINGS OF THE ACADEMY OF [1ssi. of abdomen, 3 mm.; of prothorax, 2 mm. Color, livid yellow. Fore-wing, 14 mm. long; venation as in Pl. X, fig. 77. Male.—Length, 5 mm.; length of fore-wing, 54 mm. Color, livid yellow; the head, upper part of thorax and dorsum of abdo- men blackish. The mandible has one feeble tooth at the tip, and two others shorter and feebler. Habitat.—Southern Colorado, occupying subterranean formica- ries with small gravel-covered exterior moundlet, pierced by one central gallery. ALPHABETICAL KEY TO REFERENCES IN THE PLATES. The references are uniform in application throughout all the figures. References which occur only once, and are explained in the ‘‘ Explanation of Plates,” are not placed in the Key. ab., abdomen. ab. pl, abdominal plate. ab. pl. d,abdominal plate dorsal. ab, pl. v, abdominal plate ven- tral. an, anus. an. sp, scape of antenna. ant, antenne. be. s, buccal or mouth sac. b. gz, bowl of gizzard. bn. gz, button of gizzard. 6. ms, branched muscles. CL, Clypeus. _ C. ms, crop muscles. Col, colon, Cy. gz, cylinder of gizzard. D, dorsal. E, epithelium. £.%. s, epithelium imbricated, serrate edge.] £. ab. pl, exterior abdominal plate. Epe, epicranium. f. ar, frontal area. Jem, femora. Jig, flagellum of antenna. fm, foramen. gz, gizzard. hy, hypopygium. 7. ab. pl,interior abdominal plate. al, ileum. al. v, ileo-secal valve. im, intestine. 1b, labium. lb. p, labial palps. lbm, labrum. mb, mandibles. mo, mouth. 1881. ] NATURAL SCIENCES OF PHILADELPHIA 71 mpg, malpighian tubes. ms, muscles. m. th, mesothorax. met. th, metathorax. m. tr, metatarsus. mz, maxilla. max. p, maxillary palpi. nd, node of petiole. pr. th, prothorax. py, pygidium. px, pharynx re, rectum. re. gl, rectal glands. s. gz, sepals of gizzard. stm, stomach. str, strie. nk, neck. su, sucker. 0c Pp, Seciput. tub, tibia. oc, ocelli. to, tongue. (i, cesophagus. ir, tarsus. p. ms, pharyngeal muscle. V, ventral. EXPLANATION OF PLATES.! PuatTE I. Fig. 2. View of my camp in the Garden of the gods, showing the site of some of the nests of the honey ants studied. The view is taken from the rocks at the junction of Adams and Von Hagen ridges (see Fig. 1, p. 19), and looks towards the south, and the eastern face of Pike’s Peak. One of the nests is shown in the foreground, and the sites of others are indicated by the white circles on the tops of the ridges. My tent and booth are seen near the centre of the sketch, and just opposite, on the right, is the oak copse in which the ants were discovered feeding on the exudations of galls. Page 19. Prate II. Fig. 3. Elevated gravel cone of a honey-ant nest; the gravel is of red sandstone, and the rocks around are bits of quartz of several colors, giving a pretty effect. This nest is the largest seen, and measures three and one-half inches high and thirty-two inches around the base. Page 21 1 Mr. JoszpH JEANES, a member of the Academy of Natural Sciences of Philadelphia, contributed the money required for the illustration of this paper, and thus has greatly added to whatever value it may possess, 72 PROCEEDINGS OF THE ACADEMY OF | 1881. Fig. 4. A nest built partly around a tuft of gramma grass, and less conical in shape than the above. Pirate ITI. Fig. 5. View of honey-bearers as seen in natural site, clinging to the roof of a honey-room. About natural size. Page 22. Fig. 6. View of honey-bearers in same position, drawn from one of my artificial nests. Mingled with them are seen ordinary workers, and semi-rotunds, or workers apparently in process of transformation into honey-bearers. About natural size. Pate IV. Fig. 7. Sprig of dwarf oak, Quercus undulata, with galls of Cynips quercus-mellaria, showing the beads of sweet sap. Page 25. Fig. 8. The same galls enlarged. Fig. 9. Another cluster of the same galls. Fig, 10. Section of gall showing the inside cell, c, and the exit hole of the gall-fly, eh. Page 26. Fig. 11. Turk’s-head gall, showing exit-hole, eh. Fig. 12. View of inside of a gall, showing a globular cell, and a small grub domiciled against it. Page 27. Fig. 13. A honey-bearer clinging by her feet to the wall of a honey-room. Page 22. . Fig. 14. The crater of a gate to an ant’s nest, showing the grav- eled funnel, F, and the smooth nozzle, N. Page 32. Fig. 15. Outline of the elevation of a formicary. Page 35. PLATE V. Fig. 16. Double section view of the interior of a nest, drawn from a point in the excavation twenty-one inches below the sur- face. Nest made insoft,red sandstone. g, 9,9, galleries arranged in stories. R, R, R, vertical sections through honey-rooms and chambers for nursery purposes. C, D, E, the floors of a suite of honey-rooms, showing their connection with the general system. Page 36. Fig. 17. The three honey-rooms C, D, E, above referred to, and the indication ofa fourth, F. Length of C from a to b = 5 inches ; D, from c to d = 34 inches; E, from e toh — 4 inches, Eleva- tion of b above « — 34 inches; of b above e — 6 inches. A little stairway united D with C and F; g,g,a gallery. Page 36. 1881. ] NATURAL SCIENCES OF PHILADELPHIA. 73 Fig. 18. Section through middle of nest, showing the gate architecture. G, gate; N,nozzle; A,arm of the gate gallery terminating in the vestibule V. a, b, c, branching galleries, Page 32. Fig. 19. A similar section of another nest. Letters as above; E, a small room, with gallery /, leading downward. Fig. 20. Similar section of another nest. The main gallery branches to the right, and passes behind the gate, b, b, b, into room A. E,C, small bays or rooms; D,D, ee, curved and branched gallery on the same plane, with openings downward 9g, g, g. Page 33. Fig. 21. A honey-room, HR; g; gallery leading into the gate gallery, G; ug, unbroken part of same; B, small bay-room. Page 34. _ Fig. 22. Termination of excavated nest, 6 feet 10 inches from gate, 2 feet 5 inches below surface. gg, gallery entrance; C, Queen-room, 4 inches diameter. E, Small bay-room, apparently beginning of a chamber; ¢ g, terminal gallery, running upwards as though the ants were in process of excavating a room resem- bling C. Page 36. ; Fig. 23, Sloping section through middle of nest, showing rela- tion of gate to the upper series of galleries and rooms. A, B, honey-rooms; x, y, z, main galleries; 1, 2, 3, side openings. Page 33. Fig. 24. A honey-bearer regurgitating honey from her crop at the solicitation of hungry workers. Page 46. . Fig. 25. Sentinels on guard at the gate. Page 20. PLATE VI. Fig. 26. A queen dragged home by a worker. Page 38. Fig. 27. A honey-bearer dragged and pushed by a worker-major and dwarf from a broken room into a gallery. Page 39. Fig. 28. A honey-bearer under a “ landslide,” one worker look- ing on, curious but inactive, another on the clod at her toilet. Page 41. Fig. 29. Queen surrounded by her ‘court ” or body-guard of attendant workers. Page 38. Fig. 30. Workers carrying a pebble up the mound. Fig. 31. Honey-bearer partly buried alive under pellets brought up by mining workers. Page 40. 6 74 PROCEEDINGS OF THE ACADEMY OF [188]. Fig. 32. Honey-bearer fallen from her perch, being cleansed by a worker, who reaches down from the wall. Page 40, Fig. 33. Honey-bearer with (apparently) morbid abdomen. Page 58. Fig. 34. Worker nurses feeding and cleansing larve. Page 43. Fig. 35. View of vertical section of a nest, showing galleries arranged in stories. See Pl. V,fig. 16. G,location of gate ; a—t, e—i, k—l, galleries; R, R, sections of honey-rooms. Page 36,and foot-note. Fig. 86. A worker dragging a‘honey-bearer up a perpendicular surface into a gallery. Page 39. Piatt VII. Fig. 37. View of the under side of the head of Myrmecocystus hortus-deorum, showing the mouth organs. X 20, Page 20. The letter-references in this and subsequent anatomical figures are uniform throughout. The Key to References, therefore (p. 70), will apply to all figures. Fig. 38. Face sculpture of same. XX 20, Page 20. Fig. 39, Side view of head of worker to show parasitic mites clinging thereto. The mites are about natural size. Page 63. Fig. 40. Dorsal view of mites greatly enlarged. Fig. 41. Ventral view of same. Fig. 42. One of the suckers, sw, contracted. Figs. 43 and 44, the same further drawn out, Fig. 45. Muscles of the honey crop, showing their netted and branched character. X 30, Page 54. Fig. 46. The same, from margin of the crop. C. ms, crop muscles ; 6. ms, branched muscles. Fig. 47. Third leg of M. hortus-deorum, worker-minor. X 10. Fig. 48. Section of segmental plate of abdomen of honey ant, showing hexagonal cells of epithelium, and a bristle-like hair, or seta, arising therefrom. Fig. 49. Profile view of segmental plates of Camponotus inflatus, showing the overlapping of the same, and the imbricated epithelial cells, forming a rachet-like structure which suggests the possibility of a sound-producing organism. e. ab. pl, exterior abdominal plate ; i. ab. pl, interior ditto. Page 61. Fig. 50. Profile view of abdominal plate of M. hortus-deorum, to show the same. 1881.] NATURAL SCIENCES OF PHILADELPHIA. TO Fig. 51. After Lubbock. Section through the head of Zasius niger, to show site of buccal sac, bc. s, the pharynx, pz, ie’ its muscles, p. ms. X 36, Page 52. Fig. 52. View of ile cesophagus of a worker of MM. hortus- deorum. One side of the thorax and petiole are cut away in order to show the cesophagus in site. >< 18, Page 53. Fig. 53. Abdomem of honey ant, showing the segmental plates both dorsal (D) and ventral (V) in normal condition of the crop. X 16, Page 53. Big 54. Same, when separated by partly pues crop. Page 53. Pirate VIII. Fig. 55. Entire crop with gizzard and stomach. Dissected from a honey-bearer with morbid abdomen. % 14, Page 53. Fig. 56. Crop, gizzard, stomach, malpighian tubes and intestine. From honey-bearer. X 14, Page 54. Fig. 57. Enlarged view of gizzard. X 50, Page 55. Fig. 58. After Forel. Topographic, somewhat diagrammatic representation of the organs opening into the cloaca of Bothrio- myrmex meridionalis 2, enlarged 18 times. 4, 5 and 6, optical section of the tergal chitinous pieces of what are really the 4th, 5th and 6th abdominal segments (nodes of the petiolus reckoned as one segment). Opposite and beneath there are shown the sterna of the corresponding segments. do, dorsal vessel; an. v, right anal vesicle; an. gl, right anal gland; Can, intestinal canal (intestine and rectum); po. v, poison vesicle with gland; ac. gl, accessory gland of the poison apparatus ; Ov, rudi- mentary ovaries with vagina; ab. g, the last three abdominal ganglia of the ventral cord with their commissures. Between 6 and the corresponding sternal plate (6/), lies a cleft (shown wide open in the figure) which leads into the cavity of the cloaca. In this cayity one finds, reckoning downwards from 6 to 6’: 1. O,the common opening of the anal vesicles. 2. an, anus (opening of rectum). 3. 7. st, rudimentary sting, into which the poison vesicle opens, and then lower down, the accessory gland of the poison apparatus. 4. 0. sa, opening of the rudimentary female sexual apparatus. Fig. 59. Crop in normal condition, from a virgin queen. The junction, jn, of the abdomen with the petiole is bent-over, showing ‘ 16 PROCEEDINGS OF THE ACADEMY OF [1881. a part of the esophagus as drawn from the petiole. The contin- uation of the same, @.c, within the abdomen is shown; also the relation of gizzard to both crop and stomach. X 14, Page 53. Fig. 60. View of the intestine from the posterior pole of the abdomen to the anus. X 35, Page 56. Puate IX. Fig. 61. Synthetic figure exhibiting the entire course of the alimentary canal, from mouth to anus. Page 56. Figs. 62-70 compose a series illustrating the progressive disten- sion of the crop. from the normal condition to that of the honey- bearer. Page 56. The series begins with Figs. 63 and 66, where the crop is normal ; in fig. 66 the crop has shrunken after distension. Fig. 64. Worker crop, half filling abdomen. Figs. 62, 65. Workers-major, or semi-rotunds, with distension of crop still further advanced. Fig. 67. Abdomen of a worker-minor, showing same process ot distension. Fig. 68. Abdomen of a honey-bearer, opened at the slit, s, to puncture the crop and exhibit by its shrinking away the fact that the crop fills the cavity of the abdomen. Page 57, Fig. 69. Full crop of honey-bearer, with the lower part of the alimentary canal shown through the abdominal wall against which it is pressed, and evidently in healthy condition. Page 57. Fig. 70. Abdomen of honey-bearer, the full crop pressing the gizzard, stomach, etc., into the cloacal cavity. Page 57-8. Fig. 71. Abdomen of the Australian carpenter-ant, Camponotus inflatus, exhibiting the characteristic distension of J. hortus- deorum. Drawn from an alcoholic specimen. The figure is somewhat flattened by pressure ; other abdomens in my possession are quite spherical. The gizzard, stomach (ruptured and stretched) and intestine are shown in the same relative position as in the honey ant. Page 58. PLATE X. Fig. 72. Side view of honey-bearer, JZ. hortus-deorum. X 3. Fig. 73. Dorsal view of same. X 3, Page 69. Fig. 74. Honey-bearer of Camponotus inflatus, dorsal view X 3, Page 59. Figs. 75, 76. Male of IZ. hortus-deorum. XX 5. “I tI 1881. ] NATURAL SCIENCES OF PHILADELPHIA. 9 Figs. 77, 78, Winged female, or virgin queen of the same. > 3. Fig. 79. Worker-minor of I. hortus-deorum. * 5. The workers- major and minor or dwarf are exactly similar in form, only longer in the proportions given in the description. Page 69. Figs. 80, 81. Node or scale of the petiole queen of honey ant, side and front views. > 10. Fig. 82. In part, after Swinton; to show the striz, str, supposed stidulating organs, upon the junction of the abdomen and second node, 2. nd, and also on junction of second node with the first (i. nd,) of Myrmica ruginodis. -T io 8) PROCEEDINGS OF THE ACADEMY OF [1881. Marca 1. . The President, Dr. RuscHENBERGER, in the chair. Eighteen persons present. MARcH 8. The President, Dr. RuscHENBERGER, in the chair, Fifty-eight persons present. A paper entitled “On the Structure, Affinities and Species of Scolopendrella,” by J. A. Ryder, was presented for publication. Prof. Angelo Heilprin delivered the introductory to his course of lectures on Invertebrate Paleontology. Marcu 15. The President, Dr. RuscHENBERGER, in the chair. Prof. Henry Carvill Lewis delivered the introductory to his course of lectures on Mineralogy. The following was ordered to be printed :— 1881. | NATURAL SCIENCES OF PHILADELPHIA. 19 THE STRUCTURE, AFFINITIES AND SPECIES OF SCOLOPENDRELLA. BY JOHN A. RYDER. Inasmuch as a notice,! published by me in the American Natu- ralist for May, 1880, has awakened renewed interest in relation to these singular types, and because the ordinal division proposed by me for their reception has apparently been accepted by Dr. Latzel? in his revision of the Austro-Hungarian species of Myrta- poda, I venture to offer the results of my studies for the use of those interested. Since the publication of my notice above referred to, I have met with a paper unknown to me at-the time mine was written, which in many respects anticipates the observations made by the writer,and relied upon as characters of ordinal value. The publication here alluded to is entitled “ Myriapodender Umgegend von Danzig”? by Menge, in which the author discusses at length the anatomy of the Scolopendrella immaculata Newp. From his plates and text I find that, while he confirms my observations in many respects, in others his interpretations conflict with mine. Not only is this the fact with regard to my observations, but also with those of others who have studied the genus. Taken as a whole, the monograph of Menge is, however, by far the most valuable which has yet appeared. The following synopsis of Menge’s observations will, I think, be found useful. I will preserve as nearly as possible the phrase- ology of the German text, which has reference to S. immaculata. “ Head compressed, ovoid, little longer than wide; : antenne 40-42, articulate, inserted immediately behind the la- brum, the joints compressed fusiform, urn-shaped, the basal ones thicker than long, becoming gradually longer towards the tip, so that the apical joints are more than twice as long as thick; ter- minal joint acorn-shaped. Each joint is encircled at its middle by a circlet of outwardly divergent hairs. . . . . Behind the insertion of the antenne, at the sides, are two little prominences on the epicranium and alongside and above them on each side is placed a round black eye, visible only under the microscope. 1 “Scolopendrella as the type of a new order of Articulates (Symphyla).”’ 2 R. Latzel, ‘‘ Die Myriapoden der oesterreichisch-ungarisch Monarchie. Erste Halfte, Die Chilopoden,”’ pp. 228, Pls. 10, 8vo. Vienna, 1880. 3 Neuste Schriften der naturforschenden Gesellschaft in Danzig, LV, 4tes Heft, 4to. 1851. 80 PROCEEDINGS OF THE ACADEMY OF [188]. “Mouth parts arranged for biting, Labrum forming the ante- rior portion of the epicranium, and divided into two rounded extremities anteriorly by a mesial emargination, both halves ter- minated by three pairs of teeth directed towards each other. Mandibles beneath labrum, somewhat exserted, one jointed, elon- gate, curved inwards, the outer side finely pubescent, the apex armed with four large and five small, hard, brown teeth. Maxillte two-jointed, basal joint cylindrical, curving slightly inwards, sur- mounted by two apical pieces, the outer, longest and most slender piece may be regarded as representing a palpus, while the inner one, which is shorter, terminating in a series of bristles, may be regarded as the second joint of the maxille. The labium is an oblong plate divided in the middle by a suture. The anterior margin of each half bears three conical teeth. “The body consists of twenty-three somites; twelve principal ones, to which the legs are attached, and ten smaller intermediate and a caudal somite. Each principal or leg-bearing somite has a quadrate sternum which is divided in the middle into two halves by a median furrow; the narrower intermediate somites have an elongate, undivided sternum, while the sides are covered by a triangular chitinous piece... . .-. Attached’ to every. leo bearing somite except the first, behind and alongside of the inser- tion of the legs are a pair of simple, hairy appendages. The anal opening is on the ventral side of the body covered by a rhomboidal almost semicircular plate, the lateral extremities of which extend somewhat beyond the edges of the last dorsal seute. The dorsal surface is covered by fifteen scutes which are slightly imbricated. The hind margins of the scutes are but slightly emarginated. '.°. “ Legs in twelve pairs, or three less than the number of dorsal scutes; no pairs of legs corresponding to the fifth and eighth dorsal and the caudal scutes. The first pair of legs are the smallest, and including the tarsal joint, are four-jointed. The fourth joint is the longest, the tarsal the shortest. ‘The latter is simply a little conical prominence on the outer face of which there are attached two hooked claws. . 2°. 4 “The conical caudal appendages are simply an efferent appa- ratus connected with two long caecal pouches which are filled with a viscous spinning material. The efferent duct ends between two terminal bristles. ‘“* Besides the dorsal and sternal scutes there are pairs of liga- 1881.] NATURAL SCIENCES OF PHILADELPHIA. 8] mentous bands which join the terga and sterna; in the hinder somites these bands are joined together dorsally and form an arch. The anterior ones are joined medially below by lateral processes. They also exist in the head. “The digestive apparatus consists of a straight canal which extends from the mouth tothe anus. The pharynx passes between the ligamentous bands of the upper cephalic plates. The zsopha- gus is provided with annular folds and longitudinal and annular muscles The stomach is decidedly widened and consists of an elongated cylindrical sac and is covered with brownish hepatic cells ; these cells have finely granular contents. There are no cilia on the inner surface of the stomach or alimentary canal. The stomach is usually of a yellowish brown color because of the contained food, which consists of particles of brown mould or humus, which could not be taken in by a sucking apparatus. The cavity of the small intestine is very much more contracted than that of the stomach, and at its commencement four vermi- eularly coiled malpighian tubules open into it. ‘‘ Immediately above the anus lies the opening of the oviduct and ovary, the latter consisting of a simple cylindrical canal with thin transparent walls. Eggs in different stages of development -Inay usually be found therein. ‘“* | did not see the male sexual organs, as all of the specimens dissected were females. “I was not able to study the nervous system satisfactorily. From the head a simple cord passes backwards, which has scarcely noticeable ganglionic swellings at every somite, from which simple pairs of nerves pass to the legs. “+The vascular system consists of a simple straight canal just below the dorsal scutes beginning just behind the head and passing backwards to the caudal appendages where it divides, each branch ending blindly in the latter. The presence of valves in the dorsal vessel, trachea or tracheal openings was not revealed by an amplifi- eation of 450 diameters with a Nobert microscope of fine quality. “The function of the ventral and caudal appendages is not cer- tainly known. I have on several occasions found specimens of Machilis which had eggs attached to the caudal styles, and I have supposed that these appendages in Scolopendrella have the same use. If a needle is brought into contact with the tips of either of these appendages at the opening of the spinning organ a long 82 PROCEEDINGS OF THE ACADEMY OF [1881. thread may be drawn out. It is believed that this spinning appa- ratus is used in fixing or attaching the eggs of the animal. “The pairs of legs and the number of joints in the antennz are variable. J found but eleven pairs of legs and seventeen joints ina young animal. The first pair of legs was wanting. In other specimens with twelve pairs of feet I found twenty-five, in others thirty-two, and in others still, forty-two joints in the antenne. the last seems to be the number in full-grown specimens.”’ 3 In conclusion our author observes that, ‘‘ It will have been con- cluded from what has been said, that Scolopendrella is distin- guished from Lithobius as well as from Geophilus by the very different manducatory apparatus, the double tarsal claws, the ventral and the caudal appendages with the spinning apparatus, and that it does not naturally fall into the same family with either of those genera. On the contrary the animal agrees in its princi- pal characters (excepting the spinning organs) and especially in its habits with Campodea ; is distinguished from it, however, by the greater number of pairs of legs and the dorsal seutes. I believe, accordingly, that Scolopendrella may be regarded as the type of a genus or family intermediate between the six-footed Lepismide and the Scolopendridex.”’ The foregoing paragraph shows how very nearly Menge had. concluded thirty years ago that these singular animals should be separated from the Myriapods proper. The parallel between his conclusions and my own are very striking, as will be seen from the following words from my notice already alluded to. “ This form, as interpreted above, becomes of the highest interest to the zoologist, and if the writer is not mistaken, the biunguiculate legs and their nearly complete correspondence in number with rudi-. mentary abdominal and functional thoracic limbs of the Thysa- nura, especially Machilis and Lepisma, which also have basal appendages to the legs, indivate as much affinity with insects as with myriapods, and may indeed be looked upon, perhaps, as representing the last survival of the form from which insects may be supposed to have descended. Iname the new group Symphyla, in reference to the singular combination of myriapodous, insectean and thysanurous characters which it presents.” Our conclusions as to its zoological position being nearly the same, upon the details of the anatomy we disagree. I stated in my note my interpretation of the ventral openings on the third or 1881. | NATURAL SCIENCES OF PHILADELPHIA. 83 fourth body segment. (The first condition occurs in immature specimens with less than twelve pairs of legs, the last in adults). * Genital orifice on the ventral side of the body opening on the third or fourth body-segment in both sexes. In one sex the open- ing is a simple pore, in the other a longitudinal cleft, closed by means of an oblong chitinous piece on either side, the two together occupying a sub-quadrate space. Heart dorsal; tracheal system represented by a series of simple tubular arches, without a spiral filament, which arise from openings on the ventral surface of the animal, inside the bases of the legs, widening and passing upwards to and apparently in close relation with the dorsal vessel. Intes- tine straight, with two very long, tortuous Malpighian tubules opening into it at the posterior third (8. notacantha).” The main points of disagreement are in regard to the position of the genital organs and the supposed tracheal arches. Menge states that the oviduct opens posteriorly and above the anus, and claims to haye seen the eggs in the latter and the ovary. As to this point, I did not confirm his observations, although I do not deny that he may have seen real ova. Nor do I now affirm posi- tively that the ventral opening seen by me is genital; the only evidence being the circumstance that I found two kinds in different individuals. Its function may be that of the ventral sucker of Collembola. Menge also says he saw no males, which is a curious fact. His statement that the caudal stylets will adhere to a sharp point brought into contact with their tips, I can confirm, and I have alse seen a thread drawn from them in S. notacantha. He is confident that what I took for tracheal arches are simply chitinous rods or ligaments which serve to join the sterna and the scutes. . He is mistaken, however, when he affirms that the posterior ones form a continuous arch, since in all the specimens examined by me the arch was broken at the dorsal vessel, the widened ends of the opposite halves of the arches seeming to lie against its sides. The walls of these arched tubes showed double contours under the microscope, which proves them to be hollow. He also finds four Malpighian tubules in S. 7mmaculata, whereas I find but two in S. notacantha. He finds as few as seventeen joints in the antenne to as many as forty-two. I find from four- teen to twenty-eight in two species. Newport,! speaking of the species studied by Menge, finds the joints of the antennz to vary 1 Monograph of the Class Myriapoda, Order Chilopoda. Trans. Linn. Soc. XIX, pp. 349-489, 1 Pl. 1849. 84 PROCEEDINGS OF THE ACADEMY OF [188]. from twelve to twenty-eight, and finds specimens of different ages with nine, ten, eleven and twelve pairs of legs. This variability in the number of pairs of legs I have noticed in both the American forms studied by myself. Newport also at first thought the creature was nearly related to Geophilus, but afterwards placed it between Lithobius and Scolopendra, but he at last considered it the type of a family, a conclusion which Gervais! did not accept. Wood? says he never saw any specimens of the family, and gives the characters assigned by Newport. The first species described was by Gervais in 1839, from speci- mens found in the environs of Paris. This species was made the type of the genus. In his description he disagrees with Menge in the distribution of the legs. This may however be on account of the difference of the species. Order SYMPHYLA. Amer. Nat. XIV, p. 375-6. Head, antenne and mouth parts thysanuriform. Trachea as tubular arches without spiral filament. Spiracles within the bases of thelegs. An orifice on the ventral side of the body opening on the third (young) or fourth (adult) body-segment ; present in some individuals as a pore, in others as a longitudinal cleft, closed by means of an oblong chitinous piece on either side, the two together occupying a subquadrate space. Two Malpighian tubules (four Menge). Legs five-jointed, terminated by a pair of claws. Ven- tral appendages at the bases of each pair of legs except the first. Caudal stylets containing spinning glands which open at their tips. Ovary lying dorsad of the rectum (Menge). Famity SCOLOPENDRELLIDA., Newp. Transac. Linn. Society, XIX, p. 374. SCOLOPENDRELLA Gerv. Comptes Rendus, 1839. S. notacantha Gerv. Aptéres, IV, 301, Pl. 39, fig. 7; Ann. Sci. Nat-, Zool. IT, 1844, p- 70, Pl. 5, figs. 15-17; Ryder, Am. Nat., p. 375, 1880. Hab. France and ? Pa. and Md. S. immaculata Newport. Trans. Linn. Soc. XIX, pp. 373-374, Pl. XU, figs. 4, 4a, b,c; Menge, Neuste Schr. d. naturf. Gesell. Danzig, IV, 1851, Pls. 2, Hab. England and Germany. ‘ 1 Apteres. Suite 4 Buffon, Walckenaer et Gervais, t. IV, p. 301-803, Paris, 1847. . 2 Monogr. North American Myriap., Trans. Am. Philos. Soc., XIII. New Series, 1869. 1881. ] NATURAL SCIENCES OF PHILADELPHIA. 85 §, americana Packard. Proc. Bost. Soc. Nat. Hist., XVI, p. 111, 1873. Name only. Hab. Salem, Mass. 8. gratie Ryder. Am. Nat.. XIV, p. 375, 1880. Name only. This species (Figure 1.), may be distin- guished from S. immaculata by the presence if ofa pair of stout hairs which diverge outwards from the sides of the body at each segment. Head wider than body, not cordate but sub- a Sasa pentagonal from above. A single pair of eyes SS au ~ on the sides of the head behind the antenne, § . ea BN I dedicate this handsome species to my y ae Gk sister. 7a oon eal 9 ef gee §. microcolpa Muhr. Zoolog. An- ' zeiger, LV, 1881, pp- 59-61, ~ Fig. 1.—S. gratia. figs. 1, 2 and 4. Enlarged 25 times. E Is near S. notacantha, but is said to have no ventral appendages at the bases of the legs. I would remark, how- ever, that in the specimens thought to be nota- cantha, I find these appendages present, but they are extremely small and may easily be overlooked. Muhr’s paper is a valuable con- tribution however to the anatomy of the mouth parts of a form near the species first described. Habitat, Prague, Bohemia. Figure 2, representing an American speci- men of the same, or nearly the same, as S. notacantha, has. a very suggestive resemblance to Japyx in the shape of the body; whether this is more than a resemblance I forbear to suggest. No doubt now remains in my mind that dissimilar as Lepisma, Machilis, Lepismina, Nicoletia, Campodea and Japyx at first appear upon comparison with each other, their principal characters suggest in Fig. 2.—S. notacantha. Enlarged 25 times. 86 PROCEEDINGS OF THE ACADEMY OF [1881. the most forcible manner, an affiliation with Scolopendrella. This is most strongly indicated in the mouth parts, legs, varia- bility in the number of antennal joints and habits of life in all of which Scolopendrella exhibits the strongest resemblances to the Thysanura, with very marked affinities to the Myriapods as well. The position of the ovary is that in Geophilus, but spinning organs are also characteristic of the male Geophilus and Polydesmus; a female specimen of the latter, while being kept in confinement, spun a web about its eggs in a jar in which I had confined it. I never noticed that any American female Geophilus spun webs about their nests, though I have frequently encountered masses of their beautiful amethystine-colored eggs, over which they kept faithful watch. Whether the proposed order Symphyla is sufiiciently well characterized may be a matter of doubt; this can only be decided by a more elaborate investigation of its anatomy, which the writer hopes to be able to carry out at no distant day. 1881. ] NATURAL SCIENCES OF PHILADELPHIA. 87 Marcu 22. The President, Dr. RuscHENBERGER, in the chair. Twenty-six persons present. The death of Jos. A. Clay, a member, was announced. Marcu 29. The President, Dr. RuscHENBERGER, in the chair. Twenty persons present. APRIL 5. The President, Dr. RuscHENBERGER, in the chair. Thirteen persons present. The deaths of John Gould, of London, a correspondent, and of Thos. W. Starr, a member, were announced. Aprin 12, The President, Dr. RuscHENBERGER, in the chair. Twenty-four persons present. The death of Col. T. M. Bryan, of Vincenttown, N. J., a cor- respondent, was announced. APRIL 19. The President, Dr. RuscHENBERGER, in the chair. Twenty persons present. A paper entitled “Observations on Planorbis,” by R. E. C. Stearns, was presented for publication. On the Variations of Acmexa pelta, Escholtz.—Mr. Tryon read a portion of a letter from Mr. Henry Hemphill, of Oakland, Cali- fornia, referring to certain specimens of Acmza, collected by him, and presented to the Academy this evening. “T will now call your attention to Nos. 457, 458.459 and 460. I have made two trips to Monterey, Cal., this winter. During my 88 PROCEEDINGS OF THE ACADEMY OF (1881. first visit I collected a few specimens of Acmza pelta and its vars., and when IJ returned home and began to clean the specimens I was very much puzzled over some specimens of No. 458. Several years ago I had collected shells of No. 459 at Monterey, which at that time I called Nacella instabilis, but these half and half varie- ties did not appear at that time. After a little reflection on the matter, I began to suspect the true condition of the subject and became so much interested in it, that I concluded to go to Mon- terey again and try to work it up,and I think I have done so. It is simply a question of station. “ When the young of A. pelfa stations itself on the kelp (Phyl- lospora Menziesti, Ag.), it assumes the aspect of Nacella, and as long as it remains on the kelp it does not change its color in the least, and only varies its form to suit the shape of the stems of the kelp to which it attaches itself. But when from any cause it leaves the kelp and takes to the rocks, it seems to begin imme- diately to paint up and ornament itself after the fashion of the specimens I have sent you. “When it remains on the kelp a long time and completes its growth, we then have Nacella instabilis, and if living in an exposed position its apex becomes worn, the sculpture faint, etc. When the young station themselves on the rocks they do not assume the Nacella aspect at all, but commence immediately to adorn them- selves in gay and beautiful colors as you will see by the fine series of No. 457. Now for the facts and reasons why I came to this conclusion. I collected about 200 living specimens on the kelp in all stages of growth, and out of the 200 I found but two speci- mens that varied their color at all; one was a very young and small specimen, with a few light dots on or near the apex, and the other was a large specimen with a tesselated border on the inside. T also collected about 200 on the rocks, and ev ery one was more or less variegated with either the square dots or alternate rays of white and blac k, while those that had evidently been on the kelp had their tesselated borders well advanced. We must also take into consideration the fact that Monterey is the most southern point at which Nacella instabilis has been found, and it is quoted by Dall as rare there. The water of the bay where these half and half No. 458 are found, and also others, is comparatively smooth to what it is on the outside where the typical Nacella is found, which will account for the preservation of the apex and sculpture and may have something to do with the form, and undoubtedly is the cause why it is rare at Monterey. “To show the effect of station, with probably other causes, I send you a full series of a very pretty var. of A. pelta from Olym- pia, Puget Sound, W. T., 461 to 466 inclusive, that I collected last summer. Station, on Mytilus edulis. “This very pretty var has almost the exact form and looks like a huge Nacella peltoides, Dall, and is a very interesting addition to our limpets.” 1881. ] NATURAL SCIENCES OF PHILADELPHIA. 89 APRIL 26. The President, Dr. RUSCHENBERGER, in the chair. Twenty-six persons present. A paper entitled ‘List of Fishes collected by Mr. W. J. Fisher, upon the coasts of Lower California, 1876-77, with descriptions of new species,’ by W. N. Lockington, was presented for publication. The death of Dr. J. Dickinson Logan, a member, was announced. Motility in Plants—Mr. THoMAS MEEHAN remarked that com- paratively little knowledge of motion had been gained since the time of Linnzeus. The recent work of Mr. Darwin on the motions of plants, was a valuable contribution to the subject, though con- fined to motion in roots and leaves. He thought it would serve the cause of science to note that the presence or absence of light in itself could not, as so often assumed, account for all the phe- nomena of motion. He had made numerous and careful observa- tions, this season, on motility in Draba verna, which plant, so far as he knew, had not been observed to have any peculiarities. The petals are usually closed during the early season, though the pedi- cels are erect in the daytime, drooping so as to form almost a perfect circle at night. These pedicels become erect about three hours after sunrise when there is about twelve hours of sun in the day, commencing to droop at about two o’clock in the afternoon, This diurnal motion in the pedicels continues some days after the petals have fallen, and apparently as long as the silicle continues to grow. Later in the season, on clear days, the petals com- mence to open early in the morning, contemporaneously with the rising of the pedicel; by the time this was erect, the petals would be nearly expanded. The expansion, when the sun rose at half- past five or six, would be complete by nine A.M. Strange to say, no matter how clearly the sun might continue to shine, the petals commence to close about noon, and by about two P. M., are com- pletely closed. During the course of his observations, there was a period of four days cloudy, and no attempt at expansion was made. The fourth day, however, was so slightly cloudy, that the eye could scarcely look at the sun through the thin cloudy veil, The amount of absolute light could be little less than on some days earlier in the season, when the sun was wholly unclouded, but still there was no attempt at expansion of the petals. Continued observa- tions seemed to show that not mere light, but clear sunlight, was necessary to the opening of the flower. One evening there was a heavy thunder shower; the next day 7 90 PROCEEDINGS OF THE ACADEMY OF [1881. was densely cloudy, warm and moist, but the flowers of the Draba expanded just as well as under the bright sun of previous days! These facts show that we cannot refer the opening of the flowers either to light or sunlight alone. Mr. Meehan believed that plants not only behaved differently at different times, but in different countries; and as no one, not even Mr. Darwin, seems to have noted the expansion of the petals of the Draba in England, it is possible that under those cloudy skies, they do not expand at all. So far as he had noted here, the self-fertilized flowers of the closed Drabas produced seed just as well as the expanded ones, which might possibly be occasionally cross-fertilized by the small sand wasps, which visited the open flowers freely for pollen. How habits change at times, Mr. Meehan illustrated by speci- mens of Lamium amplexicaule, a common introduced weed in gardens. Dr. Bromfield, in his Flora of the Isle of Wight, notices that the flowers vary in size during the season, but that the earliest ones are the largest. Here it is reversed. The speci- mens exhibited had already flowered from six verticels, and had mature seeds in many, but the flowers had never expanded in any case. Indeed, very rarely had the closed corollas been produced - beyond the calyx. They were essentially cleistogene. As showing how unceriain were the laws influencing this condition, when usu- ally about the end of April, the perfect flowers appeared, some plants would have them a week or more before others alongside produced any. ‘To all appearances, external influences were the same. As somewhat bearing on the laws of motion, the angle of diver- gence in branches was referred to. Mr. M. exhibited branches of Salix caprea. Normally the branches separated from each other at a very acute angle, but the fertile ament on these branches was vendulous. Under no external influence, so far as we could tell, an individual appears with pendulous branches. This has been increased by grafting, and is known in nurseries as the Kilmarnock weeping willow. But the aments have retained their normal con- dition as regards the branch. The catkins are erect on the pen- dulous branches, while pendulous on the erect ones. Morpho- logically a catkin is but a modified—an arrested—branch, but we see by this that whatever cause induced the change from the normal condition of divergence, it was purely local, and ceased to exist before it reached the arrested branch or ament. These facts were offered to show that in studying motility in the various parts of plants, it would be well to remember that ex- ternal causes had but a limited influence, and that in these cases a combination of circumstances often controlled the influences at- tributed to one. As, therefore, the facts would vary with various observations, —those of one observer sometimes seeming rather to conflict with than to confirm another,—it was too soon to form any just conclusion as to the motive cause. What was desired 1881.] NATURAL SCIENCES OF PHILADELPHIA. 91 was not so much these speculations, but an increase in the number of observers, and a correct record of well authenticated facts. The resignation of Dr. Henry C. Chapman, as a member of the Council was read and accepted. Jesse S. Walton and Harry Skinner were elected members, The following was ordered to be published : 92 PROCEEDINGS OF THE ACADEMY OF (1881. OBSERVATIONS ON PLANORBIS. BY ROBERT E. C. STEARNS. I. Are the Shells of Planorbis Dextral or Sinistral ? Incidental to an investigation into the relations of certain fresh- water snails, upon looking through the books, I find that authori- ties differ on the point, whether the shells GE Bienen are dextral or sinistral. While Say,! Swainson,? G. B. Sowerby, Jr.* and Reeve?* regard them as sinistral, or reversed, and properly figure the shells, in their works, in a sinistral position, and not “ upside down,” as in many of the books, Macgillivray® says “the shell is dextral, as several observers haye proved; not sinistral, as many have alleged ;”’ and Woodward,® H. and A. Adams,’ W. G. Binney § and others also describe it or refer to it as being dextral. Dall remarks in a foot-note to his paper “On the Genus Pom- pholyx and its Allies,”’ “if we consider the shells of this group as dextral, they offer the peculiarity of having the genitalia as in most sinistral shells; Pompholyx presents the same conditions and is certainly dextral.” Dr. Philip P. Carpenter, referring to Planorbis, says, “it lives in a reversed position.’’? Tt will Je aictase ed from the above that eminent wiiters, are divided, and that we have substantial authority on both sides of the question. My own observations thus far prove the shells to be ae sinistral," but as I have examined but comparatively few of the whole number of species, it may be that the shells in some species are dextral, and in other species sinistral. Say ex Binney, L. and F. W. Shells of N. A., Part II, p. 103. Treatise on Malacology, p. 337. Conchological Manual, p, 245. Conchologia Systematica, Pl. CXC. Molluscous Animals of Scotland, p. 114. Manual of Mollusca, second ed., p. 302. Genera of Recent Mollusca, Vol. II, p. 260. Smithsonian Miss. Pub. No. 148, p. 103. Annals of Lyceum of Nat. History of N. Y., Vol. [X, March, 1870. Lectures on Mollusca, 8. I. Report, 1860. 1 The figures of Say’s larger species in Gould’s Invertebrata of Mass., first ed., are most excellent. o on - oof @ pet on 7 ) 1881.] NATURAL SCIENCES OF PHILADELPHIA. 93 G. B. Sowerby, Jr., in comparing Planorbis with certain Am- pullarie (Marisa), says: “It is further to be remarked that the discoidal Ampullariz are dextral shells, and the Planorbes are sinistral or reversed ; and although the latter are sometimes so flat and orbicular that it is difficult to know which is the spiral side, it may nevertheless always be ascertained by a careful examina- tion.””! While the anatomy of Planorbis in its principal characters, is presumed to be sinistral, and indicates, with the sum of other characters (including habitat), a most intimate .relationship to Physa, whick has, as is well known, a sinistral shell, yet some of the authors who affirm the sinistral character of the soft parts or body, say also, that the species of this genus (Planorbis), have a dextral shell, an inclusive and broad statement which applies to all of the species, and apparently repeat this tradition, or else assume that it is so, because the great majority of gasteropodous mollusks which have shells at all, have dextral shells, the excep- tions being comparatively few. If we consider what are regarded as apical characters in forms about which there is no question, and it is permitted to reason, from analogy in this connection, it will be seen that some species of Planorbis have sinistral shells, and I submit as examples sustaining this position, the larger West American forms known as P. ammon, (fig. 1) Gould (+ P. Traskii, Lea,), P. trivolvis, (fig. 2) Say (+ P. var. occidentalis, Cp. MSS.), P. tumens, Cpr., P. subcrenatus, (figs. 3-3a) Cpr.*, P. corpulentus, (figs.4-4a) Say, and © P. tumidus, Pfr., from Nicaragua, also P. corneus. L., Britain, as shown in authentic specimens received from an experienced and _ ? Conchological Manual, p. 245. 2 Pacific Coast specimens. 3 This species more nearly resembles P. cornevs, than does any other American form. Many of the smaller so-called species (American) are exceedingly close to the smaller forms of Europe, and it is not unlikely a careful investigation would place some of them under the names previously made by the earlier authors. Mr. W. G. Binney writes of Physa hypnorum, ‘it is one of the species common to the three continents ;”’ and of Limnea, he says, “ It seems certain that the boreal regions are inhabited by several species common to similar latitudes in Asia and Europe, such as Z. stag- nalis and L. palustris.”?> This remark will ultimately be found to apply with equal truth to species of Planorbdis. 94 PROCEEDINGS OF THE ACADEMY OF [1881. trustworthy correspondent near London. This latter is the only large European species, I am familiar with. While it is neither proven nor asserted herein, that all species of Planorbis have sinistral shells, neither is it known that the anatomy of all the species is sinistral. The relations of Planorbis to Limneza are not so remote as to make it altogether unwarrantable to look for a divergence in that direction. The extremesvariableness of Planorbis has undoubtedly led to the making of too many species; specimens which are conceded Fig. 2. Fig. 3a. Ml \) Ty UTES P. trivolvis, Say. P. ammon, Gld. P. subcrenatus, Cpr. P. corpulentus, Say. to be of the same species, from different though adjacent ponds, ete., vary more or less, and this is particularly the case-with West American forms which are in various degrees affected by the character of the water, temperature, etc. While it is quite certain that the specific names herein given would be reduced by a careful and philosophical comparison, at this time I can only refer to them as they now stand in the books; I may mention P. tumens from near Petaluma as a dwarfed variety of P. corpulentus; varie- ties of the latter are frequently confused with P. ammon. 1881. ] NATURAL SCIENCES OF PHILADELPHIA. 95 It may here be remarked that Macgillivray, an enthusiastic observer, who has described some of the British species with ereat fidelity, lays much stress on the shape of the mouth, com- paring it with those of Helix and Zonites, as conclusive of the dextral character of the shells in Planorbis.' If analogies in form of mouth are werthy of consideration, though this point may not have much weight without other and corroborative evidence, we have in this character a stronger argu- ment the other way, by, more properly a comparison between nearer related forms like Physa (that is the more globose species), and most of the forms of Planorbis I have given, holding the latter ina normal sinistral position, when the tendency to the physoid mouth, the ovate shape and sag of the aperture will be readily noticed. Frequently, immature, half-grown, and less than half-grown shells of Planorbis have been brought to me by collectors who were quite confident they had made new discoveries, and it is not improbable that young shells as above have been described and published as new species of Physa. Tee I would further Fie. 5, paae suggest a compari- R son of the apertures aN (y | ( ) of our larger Cali- we) Vy fornian (adult) Ameria scalaris, Jay. Physa globosa, Physahumerosa, Gld. shells of Planorbis, Eiatde held in a sinistral position with Ame- ria scalaris (fig. 5)? (= Paludina sca- laris, Jay),a curious Florida form; Phy- sella globosa (fig. 6), Hald., a Tennessee Fie. 8, Fig. 9. Physa ancillaria, Say. Physa ampullacea, Gld. 1 It is presumable that the shells of Planorbis, by which Macgillivray was impressed and which were the most familiar to him, were the small species of his own country, which are flat, symmetrically coiled, regular in form, and gradual in growth, being in striking contrast with the sturdier, ventricose West American forms I have cited—which also more conspic- uously exhibit sinistral characters. ? Dall says: ‘‘A careful examination of a number of specimens of this singular form, shows that it is distinct, and not a young Planorbis, as has 96 PROCEEDINGS OF THE ACADEMY OF [1881. species; also with other species of Physa, like P. humerosa (fig. 7), P. ancillaria (fig. 8), P. ampullacea (fig. 9), etc., et sic de similibus. We shall, however, find more satisfactory testimony on the sinistral point by analyzing the apical characters. If, as in other shells, we consider that to be the upper end or spire in which we can follow the volutions through the entire shell from tip or nucleus to the ultimate or basal whorl and mouth— then it is impossible to ayoid the conclusion that some species of Planorbis have sinistral shells. In some of the smaller forms, like P. vortex, P. contortus, P. glaber, P. carinatus, and P. spir- orbis, all of which are British species, and in which the whorls are (in comparison with larger American forms) quite evenly coiled and in which also, the increase in size of whorl is quite gradual, the difference between the two sides of the shell, apical and umbilical, is not as readily perceived. The largest British species, P. corneus, confirms my view, being sinistral. The rapid enlargement of the whorls in some of the West American species is in marked contrast with even the shells of P. corneus of the same diameter; the height of the latter as compared to P. corpu- lentus being as ‘31 to ‘54, while the length of the aperture is in still greater contrast, being as -42 to ‘76. These measurements were made from average-sized perfect specimens of both species placed apex up, with the mouth to the left, It is easy to perceive that in those forms where the ratio of increase is great.as between the last whorl and the preceding volu- tion and so on, whorl compared with whorl, through the whole, as for instance in P. ammon, that the depression of both spire and umbilicus is most marked; but nevertheless the umbilicus is the more profound as may easily be proved by counting the volutions first on one side, and then on the other. Again, if additional proof is wanted, take any one of the larger forms! of the species herein named, and carefully, by degrees, burn oft the under side (which can easily be done, by pouring a little been suspected.—Annals N. Y. Lyceum Nat. Hist., Vol. UX, p. 356; foot- note. 1 As distorted individuals, with the volutions exceedingly irregular, are of frequent occurrence, in making the test suggested, such monstrosities should be rejected, as they would as often waduly favor one side of the question as the other. 1881. ] NATURAL SCIENCES OF PH"LADELPHIA. 97 acid ina saucer or watch-glass), until the shell is eaten off to a line which obliterates the umbilicus, when the nucleus of the spire, the tip, and half a whorl to a whorl and a-half will be found re- maining. Those who insist on the dextral character of the shells in Planorbis, unless they except the species I have named, are thus compelled to demonstrate how in the sequence of growth the umbilicus can precede the nucleus. Though specimens of the forms under discussion, in various embryonic ‘stages, have frequently excited my attention, yet the material, so far as adolescence is connected with the present line of inquiry, was at the moment, unfortunately, inaccessible. I have therefore been compelled, in order to present such structural features of the shells as are related to the direction (right or left) of the volutions, the form of the aperture, etc., to use adult speci- mens, and by breaking back, piece by piece, and whorl after whorl, towards the nucleus, until the larger whorls are sufficiently removed, so that the apex or spire ceases to be either concave or depressed, and is simply flat. It would be almost, if not quite impossible to do this with the smaller species, owing to their diminutive size and exceeding fragility, and difficult to obtain the necessary sections for illustration herein, by the use of acid. The figures (10) are drawn from specimens of Planorbis corpu- lentus collected in Oregon, also in Clear Lake, Pres 8: California, by that indefatigable collector, Mr. \ } C.D. Voy. Before manipulation they measured Ly as follows: PAE ue a Largest diameter, : . °94 inch. nuclear whorls. Height, : j LS - 88) meh: Number of whorls, four and a-half, whieh were broken back to one and a-half whorls, with a diameter of ‘10 inch.; height -15 inch. The umbilicus in one instance was still discernible—in the others, destroyed. This species is widely distributed and occupies an extensive geographical area, on the western side of the continent, from the Columbia River in the north; easterly to Lake Winnipeg; and southerly to Cape St. Lucas.’ Binney says, “ P. corpulentus is catalogued from Guate- mala by Mr. Tristram.” 1 Prof. Geo. Davidson collected specimens at this place. 98 PROCEEDINGS OF THE ACADEMY OF [188]. P. Traskii, Lea, which Mr. Binney makes a synonym of P. ammon, belongs to the western corpulentus form. While the foregoing figures (10) show the shells of one species reduced to one and a-half whorls, the following figures exhibit the form of Planorbis tumidus, Pfr., which measured in Largest diameter, 5 : : - 68 mich, with five whorls, reduced by breaking down to two and a-half whorls and a diameter of -16 inch. This last is a more southern Ati. species ; numerous specimens were [ 2 collected by the late Thos. Bridges, who found them abundant in ele- vated pools, small lakes, etc., amid the forest slopes of Mombacho, in Nicaragua. It closely resembles (ee a) more northern forms and should AT hardly be called a species. ci aa cea Eo, ae It will be observed in the figures (10) that we have in the embryonic shells of Planorbis corpulentus a near approach to Physa, a close resemblance to a Physa with a flattened spire; suggesting such physoid forms as P. humerosa, while the adolescent stage of Planorbis tumidus (at half its adult size) also suggests an umbilicated Physa with a flattened spire, somewhat like (N. W. American) Physa Lordi (fig. 12), with the spire cut off, and an umbilicus punched in, back of the mouth. The first figures (10) explain | De Kay’s “ Physa planorbula” (fig. 13), and also suggestively point towards Ameria scalaris. These figures also exhibit the physoid mouth, and show that there is neither necessity nor propriety in leaving closely related forms for more distant analogies. In this connection it should also be kept in mind that certain species of Physa, included in Ehrenberg’s subgenus Isidora, are more or less umbilicated. Of the smaller species of Planorbis which have passed under my examination, I have seldom found it difficult to determine the sinistral characters by a com- parison of the two sides of the shell. If we could unroll a specimen of, say, Planorbis spirerbis, and Fie, 11. Fie. 12. Fig. 13. Physa planorbula, Physa Lordi. 1881. ] NATURAL SCIENCES OF PHILADELPHIA. 99 then straighten it out, it would resemble, in minature, an acutely elongated conical tube, in a general way like the following figure : Fie. 14. x ML nt A of which N represents the nucleus, A the aperture or mouth, and ML a median line. Now it will readily be seen that such a tube, if simply wound up, or made into a flat coil, and during the pro- cess of winding, kept horizontally and laterally in = P1@® plane with the central or median line which divides the tube into equal parts, would, in an exceedingly small shell, make it somewhat difficult to determine which was the apical or the umbilical (that is the upper or under) side of the shell, as the nucleus and nuclear whorls in such a case would be equally as perceptible on one side of the shell as op the | | other, and the concavity or depression of both | “os sides would be the same, being equal to one-half | of the diameter of the tube as seen at X. | The Californian species to which I have re- | ferred, instead of being represented by an attenu- ated tube like the preceding figure, which very slowly increases in circumference from nucleus to aperture, would if unwound, give us a more robust form, a more rapidly enlarging, conical tube, like this (fig. 15): | N being the nucleus, A the aperture, and ML the median line. It will be seen that if this tube, commencing at N the nucleus, was evenly coiled upon the median line, the nucleus as in the first instance, though very much more depressed, owing to the greater diameter of the tube as seen at X, could be equally well seen on the two sides, the umbilical and apical i depressions being the same. When the line of coil is other than median, and the greater portion of the tube or shell is below the | line of coil, as is the case with the species I find to Zz be sinistral, then of course the umbilicus is the more and the 100 PROCEEDINGS OF THE ACADEMY OF [1881. apex the less depressed, and therefore the latter is more easily discerned. Another point too important to be overlooked, is the form of the tube as exhibited in a transverse or cross section; whether circular, semilunar, or horizontally or perpendicularly ovate. Some of the larger species have tubes, which, in cross section, are of the latter shape, hence the physoid aspect of the aper- ture both in adult and embryonic specimens, Since the foregoing was written, the shells which appear in the list appended hereto as from Lake Simcoe, came to hand. The lot embraced three species of Planorbis. Several specimens (of rather small size) of P. trivolvis, are of the western P. corpu- lentus character, and sinistral. Twenty-five specimens of P. campanulatus, Say, are also sinistral as described by that author, and three specimens of P. bicarinatus are dextral, though des- cribed as sinistral by Say. Of the smaller American species glanced at by me in the course of investigation, I find P. vermie- ularis, from Utah Lake, U. T., is sometimes dextral. From what is presented above it will, I think, be admitted that some species of Planorbes have shells whose structure is in har- mony with the sinistral characters of the anatomy, as might reasonably be presumed, and it is not unlikely that such will prove to be the rule and not the exception, when an extended and critical examination of the whole group has been made. I do not propose, at this time, to inquire into the origin of the related forms referred to in this paper; but the suggestions, which have incidentally occurred in, or grown out of the consideration of the simpler points discussed, impress me as inviting investigation. The following species of Planorbis from the localities stated, have been especially examined in connection herewith. * P. trivolvis, Erie Canal, N. Y. a Tinker’s Creek, Lake Co., Willoughby, O * &¢ Foot’s Pond, Woodburn, near Cincinnati Ohio. % os Lake Winnebago, Wisconsin. me “ Wabash River, Posey Co., Indiana, * ‘ Covington, Kentucky. + és Washoe Lake, Nevada, Note.—I am indebted to the. courtesy of the Smithsonian Institution for all of the figures herein, except 10, 11, 14 and 15, which are original. 1881. | NATURAL SCIENCES OF PHILADELPHIA. 101 t P. trivolis, Near Salt Lake, Utah T. § sf Utah Lake, U. T. * P, between lentus and glabratus, } * P. approaching glubratus. } * P. approaching Carthage, Ohio. Cumberland Co., Tennessee. Cumberland Co., Tennessee. lentus. t P. near tumens, Tuolumne Meadows, Tuolumne Co., Cal. * P. near corpulentus, Indian River, Texas. x2 ee a Bexar County, Texas. * < “6 Miami, Florida. +t 3 << - Lake Simcoe, Canada. t+ P. corpulentus, Near Portland, Oregon. tt ‘ Lake Simcoe, Canada. + ff var. Traskii, Clear Lake, Lake Co. Cal. ah ee Oregon, Mus. Stearns. ¢ P. var. occidentalis, Cp. Russian River, near Ukiah, Cal. tt “ “ King’s River, Cal. t+ eS a6 Mountain Lake, near San Francisco, Cal. t a x Santa Cruz, Cal. ¢ P. tumens, Los Angeles, Cal. tt} P. tumidus, Nicaragua. * P. glabratus, East Tennessee. * P, bicarinata, Erie Canal, N. Y. * ue _ McHenry County, Ills. t ss Portland, Oregon. + eS Lake Simcoe, Canada. * P. lentus, Clear Lakes of Indiar River, Florida. * P. campanulatus, Orono, Maine. - ff Henry County, Ills. |. PB. ammon, Salinas River, Cal. “|. P..corneus, Great Britain, many localities. The note marks above refer to the following parties from whom the material examined was received: * Prof. A. G. Wetherby ; + CO. D. Voy; { Henry Hemphill; ++ A. W. Crawford; § Dr. Edward Palmer; {{ the late Thomas Bridges; || O. Button; § the late W. W. Walpole, Esq., from all of whom I have received most generous assistance. As to the validity of the species, or determinations as above, it is not necessary to discuss the matter in this paper, as it is not pertinent to the objective point, but as may naturally be supposed by any one who has had occasion to examine into the literature relating to the group, and to make a critical comparison of mate- 102 PROCEEDINGS OF THE ACADEMY OF [1881. rial, I have found much that is unsatisfactory, and it is not asserting too much to say, that too many species have been made, and that a careful revision is required ; and in this connection, which shows the well-known variability of the group, I may men- tion the examination of a parcel of specimens from a single ‘‘ pond back of Covington, Kentucky,” kindly furnished by Professor Wetherby, which admits of a separation resulting in three species and a remainder which readily connects all three ; for this reason I do not claim that the determinations as given are always con- sistent, though carefully considered; as before remarked, how- ever, the point I have endeavored to present is not affected thereby. II. On certain Aspects of Variation in American Planorbes. In the course of the preceding inquiry various aspects of varia- tion, as exhibited in the material under examination were con- stantly recurring. Without presuming to explain such phenomena, which would quite likely .be a difficult matter, even if all the peculiarities of environment in each case, or of each lot of shells examined, were known, and without such data, quite hypothetical, yet a few notes and comments suggested by the forms referred to, may be worth a passing notice. The larger so-called species of Planorbis may for convenience in discussion be grouped as follows : First. Those in which the whorls are rounded; that is to say Rrassieand 17. if the tube or cone, as represented in the pre- ceding paper, was cut transversely, the section would show a rounded (not round) outline. Examples—The typical P. corneus, L.,' of Europe; P. Guadaloupensis, Sby.;? P. sub- crenatus (figs. 16, 17), Cpr.,? and P. tumidus, Pfr.,f of Nicaragua, a quite persistent form, not, however, quite as rounded as the others. Second. Those in which the whorls are Ty either planulate, angulated, carinated or sub- P. subcrenatus, Cpr. carinated, which includes most of the larger 1 Woodward’s Manual, Pl. XII, fig. 34; Sby.’s Manual, Pl. XIV, fig. 311; Reeve, Conch. System., Pl. CXC, fig. 1. 2 Tid, fig. 2. 3 Binney, L. and F. W. Shells, N. A., Part Il, figs. 176, 178. 1881. ] NATURAL SCIENCES OF PHILADELPHIA. 103 North American species. Examples—P. corpulentus (figs. 18,19), Say,! P. Traskit, Lea., P. occiden- talis, Cp. and P. bicarinatus (fig. 20), Say. In these the tube, if cut transversely, would present an outline more or less angulated. Forms like P. trivolvis (fig. 21), é Say, connect the two groups; for P. bicarinatus, While in some instances this Ba: species exhibits the rounded whorls of the first, it imperceptibly differenti- ates from the above to obtuse angulation, and thence to the subcarinate forms of the second Fias. 18 and 19. Fia. 20. P. corpulentus, Say. group. P. ammon (fig 22), Gould, must be mentioned here, as it illus- trates another aspect of Fia. 21. See variation, that of a. more WG . YZ rapid enlargement of the f>\z ; (©) = whorls, the result of a more obtuse cone than in frv- volvis; this, when flattened above or angulated, gives us the form P, Traskii, the most striking of all the American Planorbes ; it is the extreme or culmination of the flattened or planu- lated aspect in the second group of species, of which P. corpulentus is a well known form and more widely distributed than the other ; Dr. Cooper’s P. occidentalis being an inter- mediate link between typical P. trivolvis and ordinary average specimens of P. corpulentus. Southern specimens of P. trivolvis seem to be nearer the southern form of P. lentus than do average specimens of these alleged species from northern Stations ; and both of the above from southerly stations approach more closely to the European corneus than do northern specimens of the same ; the same may be said of the Nicaraguan P. tumidus. _———_ Z P. trivolvis, Say. P. ammon, Gld. ‘ Binney’s figures, ibid. 104 PROCEEDINGS OF THE ACADEMY OF [1881. P. bicarinatus, the cone of which is less robust than that of P. ammon or P. Traskit, being in that respect nearer to the typical trivolvis, exhibits the culmination of the carinated or keeled character of the second group, and appears to mark the limit in this direction, having reached what may be termed a permanent point. This species is usually quite persistent as to plane of coil; though in Binney it is reported from a single station as far south as “Northern Georgia”—it seems to prefer northerly regions. It is apparently of rare occurrence west of ite Rocky Moun- tains. Mr. Hemphill informed me that he detected two or three individuals at Antioch, California, a station peculiar in its enviro- mental characters, pene at a point where the Sacramento and San Joaquin incre meet and unite the drainage waters of the two great valleys of the same names; mingling in combined volume their fresh water with the salt tidal-water from San Fran- cisco Bay. Other forms are sometimes found at this point; they seem unable to obtain a foothold or to establish a permanent colony or settlement. The region is one of marshes, which sus- tain a rank growth of coarse vegetation, especially what is known as tules, which sift, as it were, the waters, and hold for a time forms which, during the great floods of excessively wet winters, are swept from their native haunts through the submersion or overflow of the ponds, lakes and streams of a vast interior region. Thus Mr. Carlton! found a few juvenile specimens of Carinifex here in May, 1869, which, like Mr. Hemphill’s specimens of P. bicarinatus, had never before nor, so far as I can learn, have never since been reported from this place nor any point in the ad- joining region. In fact, the only habitat west of the Rocky Moun- tains, I believe, from which this Planorbis is reported on good authority, is Oregon; I have specimens from Portland, collected by Mr. Hemphill. A frequent aspect of variation in the forms falling within the first group, is that of occasional bulgings or swellings, as seen in P. glabratus, Say, and P. tumens, Cpr., suggesting periodicity in growth, or rather periods of hibernation or rest, and periods of activity, at the termination of which a mouth or expansion of the 1 Proc. Cal. Acad., Vol. IV, p. 50. 1881. | NATURAL SCIENCES OF PHILADELPHIA. 105 aperture, analagous to a varix, is formed—and this repeated as the animal advances towards maturity, imparts to the shell its special feature. It will be readily seen by this, that any of these forms, scattered or distributed over a wide region in northerly or extremely elevated stations, where the season of cold reaches a maximum, against which protection must be sought by hiberna- tion, might in some of their colonies be subjected to such condi- tions, and hibernation be the only protection, as in the land snails of arid regions, against seasons of excessive drought, and in other regions against the cold of winter; and bulged or varicose varia- tion be produced in a perfectly simple way, that is, in harmony with or through the operation of a general law ; and this variation be perpetuated for some time in colonies migrating from such stations to a more genial habitat; until after awhile, some of the descendants of these varicose ancestors reach places where hiber- nation is unnecessary by reason of a permanency or mean of con- ditions—temperature, supply and quality of water being in equilibrium with the usual requirements of these animals—and the ordinary smooth, evenly-grown shells again prevail through re- version to the original form. To return to the groups, as above, the Covington Pond speci- mens referred to in part first, connect said groups, being what may be called “ ¢trivolvis, with variations ”’—that species or general form being, through its plasticity, the connecting link. Still another aspect of variation is shown in Ingersoll’s! P. plexata, from St. Mary’s Lake, Antelope Park, Colorado. Here we have a variation not unusual in the various Planorbes, and not confined to any of the larger species, that of irregularity in winding, as if through extreme torsion the coil cockled; the whorls twisting off the line or plane of volution. P. plexata is an eccentrically coiled trivolvis, the deviation from plane of volu- tion having somewhat of regularity of occurrence, and not im- probably owing to the same cause as that to which I attribute the bulging in the glabratus form, namely—to recurring seasons of hibernation and activity, when the new growth hardly makes a “good joint” as a mechanic might say, with the edges of the previous mouth; the heavy water plants at the bottom of the lake described by Mr. Ingersoll, quite likely perform a part, in causing or contributing to the irregular winding of the shell at ' See Hayden’s Reports Territorial Surveys, 1874, p. 402. fae 106 PROCEEDINGS OF THE ACADEMY OF [1881. the time when a new growth commences—when the shell which is to be is hardly more than plastic membrane, not backed up with a stiffening of lime, as it is after the fabric is perfected. In Mountain Lake, near San Francisco, a few miles west of said city, curiously distorted pond snails of the genus Physa occur, which at one time, some years ago, excited attention. The season of their growth is the summer, and its generative warmth is accompanied with the trade winds, which blow across the lake with considerable violence ; the plastic shells of the Physz are forced against plants, chips and various fragments, odds and ends afloat in or around the lake; and the outer lip thus gets dented and bent, giving a curious twist to many of the individuals. A figure (128) illustrating a distorted specimen from the foregoing locality is given by Mr. Binney in his L. and F. W. Shells of North America. However, I have no reason to believe that this deformity is transmitted, as only a small proportion of the multi- tude are affected. The specimens on which Mr. Ingersoll’s species is based, were found by him, as stated, in a snow-fed pond of small size, between or among high cliffs. As before implied, the vacillations in plane of coil may be owing to interruption of growth by recurring periods of hibernation, the characters in the environment, men- tioned by Mr. Ingersoll, affording a reasonable solution of the phe- nomena. Such ponds are subject to marked climatic contingen- cies; and sometimes, or rather in some years, their basins are nearly or quite dry—and again, fluctuations of temperature, according to the volume of water, which is an important factor, are far more critical in small ponds than in lakes or large bodies of water, where the extremes of temperature, as well as other conditions, as quality of water, are less variable or extreme. These two aspects of variation, bulging and irregularity in coiling are exhibited with more or less frequency in all of the larger American species, and in a greater or less degree, through- out the entire area inhabited by Planorbis; occurring oftener, perhaps, among colonies which inhabit elevated stations, than with those living at altitudes nearer the level of the sea. I am of the belief, too, that these aspects of variation are less frequent among colonies inhabiting southerly and semi-tropical regions. All of the variations referred to are, when present, more con- spicuous in the larger forms west of the Rocky Mountains, for 1881. | NATURAL SCIENCES OF PHILADELPHIA. 107 the reason that some of them reach a size very much in excess of the largest individuals of the same species, from points east of said range. The carinated and planulate forms seem to be freer from the bulging or variceal peculiarity than the others. As to the relations of the various species to each other, or their interrelations, it is quite evident that many of them have an immediate common ancestry. P. trivolvis (+ P. trivolvis var. fallax + P. lentus) of the American species appears to be the dominant stock-form, and may be regarded as Americanized corneus, if a semi-political term may be used in a _ physico- geographical sense; its presence in the company of such forms as Limnexa stagnalis, L. palustris and others, of circumpolar dis- tribution, indicates a geographical identity with the European species. While the Planorbes attain their maximum of size in that part of North America west of the Rocky Mountains and north of latitude 30° N., the number of supposed species, or of forms which present characters more or less distinct, are more numerous east of said range. There is apparently no relation between altitude of habitat and size of shell. The quantity of West-coast material accessible at this moment is too small to enable me to give a satisfactory exhibit of measurements. The following will, however, convey a fair idea of the robust proportions of the more conspicuous West- American forms, the first and second being P. trivolvis and the third P. ammon. 1. Utah Lake, U. T., elevation 4498°5 feet. Greater diameter 1°41; lesser, 1°04 inches. Long. of aperture ‘71; longitudinal diameter of whorl at juncture of parietal callus 45 inch. 2. Washoe Lake, Nevada, elevation 5006 feet. Greater diameter 1:30 ; lesser, 1-01 inches. - Long. of aperture ‘60; long. of whorl at junction of parietal callus -47 inch. 3. Salinas Valley, Cal., elevation 100— feet. Greater diameter 1:24; lesser, 98 inch. Long. of aperture -90; long. diameter of whorl at junction of parietal callus 55 inch. 4, Clear Lake, Cal., elevation 1323 feet. Greater diameter 1°05 ; lesser,"74 inch. Long. of aperture ‘77; long. diameter at junction of parietal callus -76 inch. This last (4) is a typical Traskii. A comparison of the meas- 108 PROCEEDINGS OF THE ACADEMY OF [1881. urements of the aperture and of the whorl at the junction of the parietal callus in this and 3 (P. ammon), with similar measure- ments in 1 and 2 (P. trivolvis), will give a good idea of the relative obtuseness of their cones or tubes, as well as of the inflation or patulous aspect of the aperture, and of its effect on the physiog- nomy of the shell. A typical P. corneus (British specimens) of 1:10 inches greater diameter measures ‘87 inch lesser diameter, while the longitude of aperture is -42, and longitudinal diameter at junction of parietal callus is -30 inch. The following original figures from nature,show the Washoe Lake Fie. 23. form as above (fig. 23); while fig. 24, from the same locality, also illustrates the distortion resulting by deviation from or eccentricity in plane of coil. The sinistral or dextral inquiry led to eee See the discovery that P. brea rinanias is gue ee times right and sometimes left; this is an {fm : interesting fact, because said species ex- ; = hibits certain analogies with other peculiar and characteristic forms. The relations of the dextral Carinifex, whose planorboid character led Dr. Lea? P. trivolvis. Original (distorted). to describe it as a Planorbis, are appar- ently closer to P. bicarinatus than to any other species. The exceeding variability of Carinifex is seen by the figures here given, LEE Fie. 26. which, however, do not fully represent a | the range of variation. (Figs. 25, 26, 27). In the light of our present knowledge it should perhaps be regarded only as a ee coincidence that the very territory from ete Lea’s types. which P. bicarinatus is with the two J Klamath and f ; ae F if) Canoe Creek exceptions of Hemphill’s Antioch, and specimens, Portland (Oregon) localities, entirely absent, is the territory inhabited by Carinifex, and in which it has been found, either recent or fossil. When the From | immense area of this territory is con- Wasindth eed {eon sidered, the number of localities in Ceeumene Pit River, Cal. which it has been detected, are few; still Carinifex Newberryi,and varieties. these few are so related the one to the 1 Binney’s L. and F. W. Shells of N. A., Part II, p. 74. Fig. 27. 1881.] NATURAL SCIENCES OF PHILADELPHIA. 109 other as to indicate a wide and general distribution within its boundaries. These localities are as follows, commencing at the easternmost station : 1. Utah Territory; near Utah Lake, in Wahsatch Mountains, collected by Dr. Edward Palmer. Museum Stearns. (Semi- fossil.) 2. Nevada (Tertiaries) ; as Vorticifex Tryoni, Meek, in King’s Survey. 3. California; Owen’s Valley, collected living by Hemphill ; “The most southern locality. The animal undistinguishable externally from that of Planorbis ammon.” Cooper. 4, California; Klamath Lake and Canoe Creek, living; Dr. J. S. Newberry. S. I. Collection. 5. California; Pitt River, Dr. Cooper; living. S. I. Collection. 6. California; Clear Lake, living; Dr. Veatch. S. I. Collec- tion. Cooper makes a var.’ ‘‘minor” of specimens from this place. 7. California; Antioch; living. Carlton. ‘A few very young ones, perhaps a dwarfed southern variety like those from Clear Lake.” 8. California; Livermore Valley, Alameda County; “in the hills north of Martin’s, near Tassajara,” Cooper, fossil; label marked “ Planorbis, etc., Tertiary?” in State Geol. Survey Coll., Univ. of Cal. The Utah specimens, though small, are mature, and include the form rounded above like Meek’s Nevada species, as well as the more flattened and grooved features of the Tassajara' specimens, which approach in size and general characters more closely to Lea’s type, fig. 25. Some of Palmer’s Utah specimens are elevated, and vary in the direction of fig. 27, though not terraced or keeled as much; the lot of only a dozen specimens, exhibits a remarka- ble range of variation. There are striking analogies between the shells of Carinifex in its varieties, and the Australasian brackish water Amphibole. A new species has been made on one of the varieties by Mr. Smith, of the British Museum, which he named C. Ponsonbit ;* 1 Tassajara is the name of a stream which is frequently dry in the latter part of summer. 2 Proc. Zool. Soc., 1875, p. 586. Also Quar. Jour. Conch., Vol. I, p. 150. 110 PROCEEDINGS OF THE ACADEMY OF [1881]. the specimens were collected by Lord Walsingham, in California. There are several varieties still undescribed, which challenge the attention of those who are ambitious in this direction. Carinifex exhibits many of the variations in form of tube or cone, in cross-section, which are seen in Planorbis, without the bulgings of the varicose forms, and plus the elevation of spire which is seen especially in extreme individuals like the figure ; the outline of the mouth is very much like that of P. bicarinatus, and in some of its varieties suggests a P. bicarinatus, with the umbilicus deepened by pushing up the spire from below. With the discovery of new localities, and ample material both recent and fossil, without doubt the sequence of variation will be traced, and its relation to meteorological, geological and chemical changes, within the area of its distribution partially indicated. In this connection I would direct attention to Prof. Hyatt’s interesting letter to Mr. Ingersoll, referring to Steinheim fossils, and to the Valvate of Lawlor’s Lake, Nova Scotia, in Prof. F. V. Hayden’s Report, 1874. 1881. ] NATURAL SCIENCES OF PHILADELPHIA. Ett May 3. The President, Dr. RUSCHENBERGER, in the chair. Thirty-four persons present. Dr. Geo. A. Koenig was elected a member of the Council to fill the vacancy caused by the resignation of Dr. Chapman. May 10. The President, Dr. RUscHENBERGER, in the chair. Twenty-eight persons present. The deaths of Jos. 8S. Lovering, a member, and of Dr. James Lewis, of Mohawk, N. Y., a correspondent, were announced. May 17. Mr. Tuos. MEEHAN, Vice-President, in the chair. Twenty-nine persons present. The following papers were presented for publication : “ Quercus rubra L. var, Texana,” by S. B. Buckley. “Quercus Durandii,” by 8. B. Buckley. ‘* Rhus cotinoides,” by S. B. Buckley. "May 24. The President, Dr. RuscHENRERGER, in the chair. Twenty-nine persons present. A paper entitled *‘ Revision of the Paleocrinoidea, Part II, Family Spheroidocrinide, including the subfamilies Platycrin- ide, Rhodocrinide and Actinocrinide,” by Charles Wachsmuth and Frank Springer, was presented for publication. The death of Thos. A. Scott, a member, was announced. Sexual Characters in Fritillaria atropurpurea, Nuttall—Mr. THomMAs MEEHAN noticed the occurrence of separate male and hermaphrodite flowers in Fritillaria atropurpurea, cases of bi- 112. PROCEEDINGS OF THE ACADEMY OF NATURAL SCIENCES. [188]. sexuality being rare in truly liliaceous plants. His specimens were from western Nevada, and the characters now noted were exhibited in 1880; but as the plant had only been received a short time before, the absence of female organs might be attributed to weak- ness, but observation this season shows it to be a normal condition. The plant produced four spikes, flowering on the 15th of May. The spikes were 15, 14, 13, and 11 inches high, respectively, and had from twelve to ‘fourteen narrow elaucous: leaves each, and all of apparently equal strength. The tallest spike had but one per- fect flower; another had the lowermost perfect, but with four others above, all purely staminate. The other two had three flowers each, all staminate. These staminate flowers had antlers as large and as polleniferous as the hermaphrodite ones, and the perianth seemed in every respect as perfect, the only difference being in the total absence of all traces of a gynecium. It was evidently the normal condition of the species, “which, by the way, seems to run closely into Fritillaria parviflora. Mr. Meehan said the facts were interesting, as drawing still closer the well-known relationship of Melanthiacex and Liliacee. A tendency to diclinism had hitherto been supposed to be the special characteristic of the former order, although occasionally, as in Asparagus, there were indications of the same characteristics in Liliacez also. May 31. Ther President, Dr. RuscHENBERGER, in the chair. Twenty-five members present. A paper entitled ‘“ Observations on the Hippopotamus,” by Henry C. Chapman, M. D., was presented for publication. Alexander Biddle, M. D., W. Norton Whitney, M. D., and John G. Lea, M. D., were elected members. Thomas T. Bouvé, of Boston, was elected a correspondent. The following were ordered to be published : 1881. | NATURAL SCIENCES OF PHILADELPHIA. 113 LIST OF THE FISHES COLIECTIED BY MI. W. J. FISHER, UPON THE COASTS OF LOWER CALIFORNIA, 1876-77, WITH DESCRIPTIONS OF NEW SPECIES. BY W. N. LOCKINGTON. In April, 1876, the schooner Harvest Queen, W. J. Fisher in command, left San Francisco on a collecting expedition to Lower California. The intention was to work down the western coast, and then along the eastern to the head of the Gulf of California ; and the object was to collect seals, birds, fishes, mollusks, crus- tacea, radiates, etc., for sale or exchange. Financially the expedition proved a signal failure; few skins or shells of value were procured, and most of the former were spoiled before reaching San Francisco; but the scientific results were far from insignificant. Chiefly in consequence of the free use of the dredge in depths under fifteen fathoms, large numbers of small crustacea, many of them new or little known, were procured, also a few rare and three or four new mollusks, and some interesting fishes from the Gulf, the latter mingled with the crustacea and other objects dredged at moderate depths. Besides these fishes, which will be described in the following pages, several better known species were taken in considerable quantity at Magdalena Bay, upon the Pacific coast of the peninsula. As these fishes have already formed the subject of two short papers in the Proceedings of the California Academy of Sciences, and one or two were by error described as new, they will be in- eluded in the list herein given, with references to my previous papers. The reptiles obtained were catalogued in the American Natur- alist, April, 1880, p. 295. Nothing new was found among birds, and the only mammals taken were a fine female Elephant seal (Morunga proboscidea), killed at Ascencion Island, on the western coast of the peninsula; several sea-lions (Humetopias stellert) ; and the skull of a species of Orca. Diodon maculatus Lac. Gulf of Cal. Tetrodon politus Gill. Magdalena Bay. Tetrodon punctatissimus ? Gunther. Gulf of Cal. Antennarius leopardinus Guthr. Gulf of Cal. Hypleurochilus gentilis (Grd.) Gill. La Paz, Las Animas Bay, Gulf of Cal. Pholidichthys anguilliformis Locktn. Gulf of Cal. Cremnobates altivelis Locktn. Gulf of Cal. 9 114 PROCEEDINGS OF THE ACADEMY OF [1881. Clinus phillippi Steind. Gulf of Cal. Apodichthys univittatus Locktn. Gulf of Cal. Microdesmus dipus Guthr. La Paz. Gobiesox rhessodon Rosa Smith. Gulf of Cal. Gillichthys mirabilis Cpr. Magdalena Bay. Pimelometopon puleher (Ayres) Gill. Magdalena Bay. Semicossyphus pulecher Locktn, Proc. Cal. Ac. Sci., 1876. p. 87. Cynoscion parvipinnis (Ayres) J. & G. Magdalena Bay. Menticirrus undulatus (Grd.) Gill. Magdalena Bay. Girella nigricans (Ayres) Gill. Magdalena Bay. Sparus brachysomus Locktn. Proc. U. 8S. National Museum, 1880, 283. Magdalena Bay. Pristipoma melanopterum Cuv. Gulf of Cal. Serranus nebulifer (Grd.) Steind. Magdalena Bay. Paralabrax nebulifer Locktn. Proce. Cal. Ac. Sci., 1876, 86. Serranus clathratus (Grd.) Steind. Magdalena Bay. Serranus maculofasciatus Steind, Angeles Bay, Gulf of Cal. Serranus gigas (Ayres) J. & G. Magdalena Bay. Centropomus undecimalis Bloch. Ascencion Island, I. C. Centropomus viridis Locktn. Proc. Cal. Ac. Sci., 1876, 109. Selene vomer (L.) Liitken. Magdalena Bay. Argyreiosus pacificus Locktn. Proc. Cal. Ac. Sci., 1876, 84. Trachynotus carolinus (L.) Gthr., Cuy. and Val. Magdalena Bay. Trachynotus ovatus L. Lae. Magdalena Bay. Trachynotus ovatus Locktn. Proc. Cal. Ac. Sci., 1876, 86. Trachurus saurus Raf. Magdalena Bay. Sphyrena argentea Grd. Magdalena Bay. Sphyrena argentea Locktn. Proc. Cal. Ac. Sei., 1876, 88. Atherinopsis californiensis (Grd.) Gill. Magdalena Bay. Myxus harengus Ginther. Las Animas Bay, Gulf of Cal, Albula vulpes (L.) Goode. Albula conorhynchus Locktn., loc. cit. 83. Heterodontus francisi (Grd.) J. & G. Magdalena Bay. Cestraciun francisi Locktn., loc. cit. 85. Mustelus hinnulus (Blainville) J. & G. Magdalena Bay. Mustelus californicus Locktn., 1 ¢. 87. Triacis semifasciatus Grd. Magdalena Bay. Triacis semifasciata Locktn., 1. ¢. 87. ‘° Sphyrna zygena (L.) Raf. South of Cape St. Lucas. Branchiostoma lanceolatum (Pallas) Gray. Angeles Bay, Gulf of Cal. Tetrodon punctatissimus ? Giinther. Cat. Fish. Brit. Mus., VIII, p. 302, 1870. D. O,5As 839P. 30750. 27 19 Body compressed, short ; the dorsal profile from tip of snout to eye slightly coneave, thence to origin of dorsal gently convex, 1881]. NATURAL SCIENCES OF PHILADELPHIA. 115 thence descending moderately, somewhat concave from posterior margin of dorsal to caudal. Abdominal outline slightly concave in front of eye, thence boldly convex to origin of anal, thence cor- responding to dorsal outline. Head 3} times, greatest depth 2,% in total length, the greatest depth above the centre of the pectoral base; greatest breadth (at gill-openings) less than the length of the head, and 3,3, in total length. Snout 14, eye 4 times in length of head, width of inter- orbital area about 13 times the diameter of the eye. Depth of caudal peduncle nearly 22 times in greatest depth. The back behind the head forms a keel. Snout truncate at the tip, concave above and below, higher than wide ; nostrils inconspicuous, a single opening on each side on a minute papilla; mouth small; eyes round, lateral, entirely in the posterior half of the head, the upper orbital margins slightly raised, so that the interorbital area is somewhat concave. , Teeth smooth, sharp-edged, line of junction distinct. Gill-opening small, almost perpendicular, distant from the eye about an eye-diameter. , Dorsal fin arising at a distance from the gill-opening equal to the length of the head, with nine articulated rays, the first simple, the others once or twice divided. Base of dorsal less than one- third of the length of the snout, height rather more than one- third of length of head. Anal slightly posterior to dorsal, with eight rays, the first ver- tical with the posterior margin of the dorsal; height and length of base equal to dorsal. Caudal very slightly convex on posterior margin when opened out, all the principal rays divided, some of the central ones four times ; length of fin about 54 times in total length. Pectoral base oblique, its upper axil near the upper end of the gill-sht, thence backwards and downwards at an angle of about 45°. Pectoral fin fan-shaped, rather short (the tips of the rays in the specimen are slightly: injured), the uppermost ray short, simple; the next (longest) simple; the others once or twice divided, slightly diminishing in length downwards. Top and sides of head, breast, and abdomen to anal fin with prickly papille, most numerous upon the breast and under side of the snout. Rest of body naked. Color purple, thickly sown with subcircular or subelliptic 116 PROCEEDINGS OF THE ACADEMY OF [1881]. lighter spots; these become larger upon the flanks, until the ground color fades out, leaving the abdomen, breast, and under side of snout dirty yellow. Spots of sides and upper, portion light purplish. Behind the anal fin the spots can be traced all round the body. No color bands or spots.are now evident upon the fins. Specimen in alcohol since 1876. Angeles Bay; Gulf of California. From 7. furthii, Steind. (Sitz. Akad. Wiss. Wien, 1876. Icthyol. Beitrage, V, p. 22) this species may readily be distinguished by the great length of the snout, two-thirds of that of the head, while in the 7. furthii it is only 24, of the head. Other differences are the absence of a skin-fold on the tail, the inconspicuous. nasal papilla, and the smaller development of the spinules upon_the back. The coloration is also different. J. furthii has indistinct transverse bands, and a dark band on the pectoral base The short description given by Dr. Gunther of 7’. punctatisst. mus agrees, so far as it goes, with the present species. His specimens were from Panama. Should the present species prove distinct, I propose for it the name of oxyrhynchus. Length of. specimen, 3°06 inches. Antennarius leopardinus, Gnthr.? Trans. Zool..Soc., 1864, 15 7. Fishes Cent. Amer. in Trans. Zool. Soc., Vol. VI, 484, Pl. VII. This species was originally described from Panama. Two examples were obtained by Mr. Fisher by dredging Pia: a depth of 22 fathoms, among beds of pearl oyster ( Margaritophora) off San José Island, Amortiguado Bay, Gulf of California, The spots have in most cases faded into light yellow, leaving the bright vermilion of all the rest of the body and fins unchanged,, A black spot persists upon the ninth ray of the dorsal, and there are traces of black upon the sides and head. The under side of the abdomen, in advance of the anal, is light yellowish with dark spots. Total length 1°95 inches; width from tip to tip of pec- torals. 1°70 in. Cremnobates altivelis, nov. sp. D. 4.25, A. 21, P. 13, C. 13, V. 2, L. lat. 37. Body compressed, greatest depth a little behind pectoral axil, greatest. thickness at gill-covers, dorsal and abdominal profiles of 1881. | NATURAL SCIENCES OF PHILADELPHIA. 117 similar curvature, decreasing regularly to the caudal fin. Profile of occiput and superorbital regions convex; snout somewhat produced, its upper outline slightly concave. Head one-fourth of total length, greatest depth a little less than length of head, caudal peduncle about one-fourth of the greatest depth. Eye round, lateral, with a slight direction upwards, its diameter less than the length of the snout. Interorbital area nearly equal in width to the diameter of the eye, concave transversely, upper orbital borders slightly raised. A short nasal tentacle slightly anterior to the front margin of the eye. A large fimbriated tentacle on each side of the first dorsal ray. Cleft of mouth oblique, the lower jaw the longer; the posterior convex extremity of the club-shaped maxillary about vertical with the centre of the pupil. Teeth of the outer row regular, sharp, incurved, the largest in front, gradually decreasing along the lateral portions of the jaws, and not extending much past the middle of their length. A narrow band of small teeth in the rear of the outer row. Vomer- ine teeth. Branchiostegals six. Gill-openings continuous, membranes not attached to the isthmus, Distance from the first ray of the dorsal to the posterior margin of the eye equal to the length of the snout. First two rays of the dorsal much developed, the first slightly the longer, and nearly equal in height to the distance of its base from the tip of the upper jaw; third ray about half the length of the first; fourth very short; succeeding rays to the twenty-sixth longer than the third, the three last somewhat decreasing. _ Anal commencing under the eleventh dorsal ray; coterminous with, and equal in height to the dorsal. Caudal with thirteen simple jointed rays, the longest in the centre, posterior margin convex. Pectorals narrow, lanceolate, the fifth and sixth rays longest, and four-fifths the length of the head. Ventrals inserted in advance of the pectorals. Lateral line with thirty-seven simple pores, parallel with dorsal outline to opposite the origin of the anal, where it is deflected 118 PROCEEDINGS OF THE ACADEMY OF [1881. almost perpendicularly downward to the middle of the side of the body, along which it continues to its termination. Scales rather large, about ten in a transverse row in the central part of the body, their posterior margin membranous. No scales on fins. A line of pores around the margin of the orbit, another along the posterior margin of the pre-operculum, connected to each other and to the lateral line by a line from the centre of the hinder border of the eye. Color (in alcohol) bright pink above, becoming dusky below, under side of head light olivaceous, lower lip blackish. Dorsal pink, dusky on its margin, a black spot on the fourth ray, and another on its hinder part upon the 24—25th rays, the latter spot extending on to the body. Membrane of anal black: Occipital tentacles black. Total length 1-9inch. A single specimen from La Paz, dredged at a depth of 22 fathoms. This species may be distinguished from C. monophthalmus, Gnthr., by the much greater development of the first two dorsal rays; by the longer and concave snout,and by the coloration; and from C. marmoratus, Steind. (Sitz. Akad. Wiss., Wien, 1876, 174), by its more elongate form, shorter cleft of mouth, and longer first dorsal ray. In C. marmoratus the second dorsal ray is longest. Pholidichthys anguilliformis nov. sp. Body exceedingly elongate, much compressed, naked, upper profile of head forming a continuous convex curve to the tip of the snout, which is about equal in length to the eye, Head six and two-fifths, greatest depth sixteen times in total length. Eye lateral, round; interorbital space about two-thirds of the diameter of the eye, convex transversely. Posterior extremity of maxillary vertical with the hinder margin of the eye. Tip of snout a little below the level from the centre of the eye; mouth moderately oblique, lower jaw slightly the longer. Teeth of lower jaw in a close-set row, the largest in front, diminishing along the sides. Teeth of upper jaw similar but smaller. Palate smooth. Vertical fins continuous but distinct, dorsal entirely spinous, anal commencing a little behind the middle of the entire length of 1881. ] NATURAL SCIENCES OF PHILADELPHIA. 119 the fish.. Ventrals two-rayed, very slightly in advance of the pectorals, which are about equal in length to the distance of their base from the eye. Color (in spirits) dark blackish brown, mingled with white upon top, sides and lower parts of head. Interorbital area and top of snout white. A single specimen dredged off San José Island, Amortiguado Bay, Gulf of California. Total length 1-60 in. "Head 0:25 in. The example is broken across, the branchiostegals are defec- tive, the caudal fin broken, and some fin-rays missing, so that the fin formula cannot be accurately given. The dorsal fin has above sixty rays. The body is much more slender than that of P. leucotenia Bleeker, and there is no trace of the longitudinal bluish white band of that species. Apodichthys univittatus, nov. sp. D. circa 95. A. ca. 1-40. Body elongate, much compressed, band-like, preserving almost the same depth to about the posterior fifth of the body, thence tapering more rapidly to the caudal fin. Head seven, depth nearly ten times in the total length; depth of caudal peduncle about one-half of that of body. Snout obtuse, about two-thirds as long as the diameter of the eye, the upper profile of the head a continuous curve from snout to occi- put. Interorbital area highly convex transversely, about equal in width to half the diameter of the eye. Eye entirely lateral, round, contained entirely in the anterior half of the head; iris golden. Mouth small, the posterior extremity of the maxillary reaching to the anterior margin of the eye. Teeth small. Branchiostegals five. Dorsal fin continuous with but distinct from the anal, arising vertically above the tip of the operculum, and composed of spines only. Anal preceded by a long, sharp, slender spine of V-shaped transverse section, the hollow side anterior, the length of the spine equal to about half the depth of the fish. Distance from anal spine to tip of operculum a little more than to tip of caudal. Caudal with numerous accessory rays, so that its sides are almost straight, posterior margin broken in specimen, all the rays simple. General color (in spirits) light reddish, the vertical fins rather bright, and the top of the head reddish brown. Tip of snout brown. A silvery band (possibly bluish in life) from the tip of 120 PROCEEDINGS OF THE ACADEMY OF [1881. the snout, across the lower part of the eye, cheeks and opercles, terminating at about the middle of the length of the operculum ; this band bordered above by a narrower brown band. A single specimen. Lower California, probably from the Gulf. Length 1°88 in. Length of head .0:27 in... Greatest depth 0°19 in. Snout to anal spine 1:10 in. The peculiar vitta upon each side of the head at once dis- tingnishes this species from the two other described forms. Microdesmus dipus, Giinther, Trans. Zool. Soc., 1864, p. 26. A single specimen of this rare species was obtained at La Paz, near low-tide level. Hemiramphus unifasciatus, Ranzani. D,.14.- -A..16. oe With some hesitation I refer two specimens taken in Las, Animas Bay, Gulf of California, to this species rather than to H. rose J. & G. The proportion of the jaw to the length of. the body, as well as the number of dorsal rays, agree. with unifasciatus. On each side of the median line of the back a row of elongated spots of dark greenish tint, one on each scale, forms a narrow band, and between these bands is a median series of dark spots formed by thickly aggregated black dots. Between the narrow dark-green band and the lateral band similar close aggregations of - dark spots form a series of spots, elongated transversely. The silvery lateral band is bordered above by a narrower greenish stripe. 1881.] NATURAL SCIENCES OF PHILADELPHIA. 121 QUERCUS DURANDII, Buckley. BY 8S. B. BUCKLEY, PH. D. Thad spent several years in studying the trees of the United States in their native places, when, in the month of September, 1859, as I was walking from Camden to Allenton, in Wilcox County, Alabama, I saw an oak, different from any I had ever seen. It was ina dark, rich, limestone soil, on the right hand side of the road, about three miles from Allenton. The bark of its trunk and limbs was scaly; leaves lanceolate, entire or slightly lobed or repand; acorns small, ovate, obtuse; cup very shallow, about one-eighth as long as the acorn. The tree was 1 foot in diameter and about 25 feet high, and the only one there. I showed specimens of it to the gentleman with whom I staid all night. He told me it was a rare oak in that country, growing near the banks of streams and swamps, that its wood was close grained and very tough, making excellent screws for cotton-gins, firm and durable wagon-hubs, etc., also splints for making baskets for the cotton-field, when the cotton is being gathered; that it was seldom more than 3 feet in diameter and 60 feet high. This oak was the Quercus Durandii, a new species which I de- scribed in the Proceedings of the Academy of Natural Sciences in 1861. It is named after the late Elias Durand, who was for many years chairman of the Committee on Botany of the Academy. In October, 1859, I saw it again at Shreveport, in Louisiana, on the bluffs of the Red River, in the upper part of the town. There its acorns were larger and longer than in Alabama. Here there were several trees of it, one of which was 4 feet 2 inches in dia- meter at 3 feet from the ground. These were low, spreading trees, growing in an open space. In Southern Texas, on the Colorado River, in Fort Bend and other counties, it is a large tree, often 2 to 3 feet in diameter and 60 to 70 feet high. In 1874, in Milam County, Texas, I saw many large, tall trees of this oak in the bottom lands of Little River, a tributary of the Brazos. There it is sometimes called the “ bastard white oak”’ because the bark of its trunk resembles that of the scaly forms of white oak. Durand’s oak grows on the banks of Shoal Creek, one-half to 122 PROCEEDINGS OF THE ACADEMY OF [1881. two miles above Austin, also near Mount Bond, about three miles from the city. - In 1872, Mr. Elihu Hall, of Athens, Illinois, was at Austin col- lecting plants. I showed him Durand’s oak, which is referred to as a good species in his “ Plantes Texanz,” published in 1873, Mr. Charles Mohr, Botanist for the Tenth Census Department of Forestry in the Gulf States, was here, last November, and obtained sections of the wood, and specimens of the leaves and acorns of Q. Durandii for the Department of Forestry under the direction of Prof. Chas. S. Sargent, for the United States Government. Lastly, Prof. Sargent came here three weeks ago, and together we visited the banks of Shoal Creek, where there were several small trees of this oak, the largest being nearly two feet in diameter. He was satisfied it was a good species. I have been thus particular in giving the history of this oak to the present time, because all who have seen the living trees regard it as a good species, which is not the case with some who have not. In Dr. Engelmann’s ‘ Oaks of the United States,” published at p- 392, “ Transactions of St. Louis Academy,” speaking of the forms of Quercus undulata, he says: ‘“ The var. Gambelii runs into variety Gunnisoni on one side, and on the other into variety breviloba (Q. obtusiloba var. brevifolia, Torrey, Bot. Bound. and probably Q. Durandii and San sabea, Buckley).” Dr. E. thus regards it as probably identical with a marked variety of Quercus undulata, an opinion which a better acquaintance with living trees would surely lead him to revise. Without extensive field expe- rience it is almost impossible to decide on the limits of species in oaks, as well as in grapes and other difficult genera. 1881.] NATURAL SCIENCES OF PHILADELPHIA. 123 QUERCUS RUBRA, L., var. TEXANA. BY S. B. BUCKLEY, PH. D. Quercus rubra is distributed over a larger area than any other oak in North America. According to Dr. Richardson, it is the most northern of oaks; he found it on the Saskatchewan and the rocks of Lake Namakeen, in British America. It is in Nova Scotia, and southward through the United States to El Paso County, in the northwest part of Texas. The writer saw it in the coves of the mountains near Fort Davis, in the summer of 1875, at elevations of from 5000 to 6000 feet above the sea. The differences of soil and climate in which it grows cause it to vary so much in size, wood, leaves and acorns that the two extremes of difference considered apart from intermediate forms, would make two very good species. The Texas form, growing on limestone hills and coves and little valleys in the vicinity of Austin and westward, has been called Q. palustris by Torrey and Gray, in the Report of the Botany of the Pacific Railroad, Report of Capt. John Pope, p. 173; also in other reports of theirs on the Botany .of Texas. It has also been called Q. palustris by Dr. Engelmann, when he named the plants collected by Elihu Hall in Texas in 1872. (See Hall’s Plante Texanz, p. 21, Nos. 604 and 605.) Hall obtained specimens of it here in June, and I sent him acorns of it in the fall, and he informed me that Dr. Engel- mann regarded it as Q. palustris. I have never seen the true Q. palustris farther south than the vicinity of Washington. Prof. Sar- gent and Mr. Mohr both inform me that they do not know of its being in the Southern States, and so says Michaux, in his “ North American Sylva.” It is notin Chapman's “ Flora of the Southern States.” These things in part, joined with the characteristics of the oak as growing here, caused me to describe it as a new species. (See Q. Texana, in Young’s “ Flora of Texas,” p. 507.) If nota good species, it is a well-marked variety of Q. rubra. Let it then be called Q. rubra var. Texana. Last December, Mr. Charles Mohr and myself got sections of its wood, etc., which he sent North for the Department of Forestry of the Government Census Bureau. We then thought it to bea good species. A few days after, on the bottom lands of Walnut Creek, about six miles below Austin, we found the acorns and 124 PROCEEDINGS OF THE ACADEMY OF [1881]. leaves of Q. rubra and also those of Q. Texana on and beneath the same tree; and also many intermediate forms. It was not only one tree, but many, which showed these differences. Quercus rubra attains a large size in Eastern Texas; but west of the Trinity River it is seldom more than two feet in diameter. Its wood here is harder, firmer and better than that of the Northern red oak. About two miles from Raleigh, North Carolina, on the Fayette- ville road, in 1858, I measured a Quercus rubra which was 20 feet 8 inches in circumference at three feet from the ground. It was a low tree, with a remarkable spread of very large limbs, whose length on the south side was 72 feet and 71 feet. The longest on the north side was 66 feet. Near the base of its limbs at ten feet from the ground it was 27 feet in circumference. The cireum- ference of the largest limb, at two feet from the trunk, was 9 feet 7 inches. Another limb was 7 feet 7 inches in circumference. In September, 1859, I measured a Q. rubra in Wilcox County, Alabama, which was 24 feet 7 inches in circumference at three feet from the ground. It was a tall, well-developed, healthy tree. Another one not far distant was 18 feet 2 inches in circumference at three feet. In the town of Romulus, Seneca County, N. ¥., I measured another, in 1865, which was 17 feet 2 inches in in cir- cumfererence at three feet high. Most of the large red oaks of the Northern States have been cut down to make staves for flour barrels, ete. 1881.] NATURAL SCIENCES OF PHILADELPHIA. 125 RHUS COTINOIDES, Nutt. BY S. B. BUCKLEY, PH. D. This sumac was discovered by Nuttall, in the autumn of 1819, on the Grand,a large tributary of the Arkansas River, and within the limits of the present Indian Territury. It was on high broken rocky banks at a place then known as the “ Eagle’s Nest.” It grew there only as a shrub, and was not again found by Nuttall. He gives a description of it, with a plate, at p. 71, vol. ii, of his addition to Michaux’s ‘* North American Sylva.” He there calls it Cotunus Americana. Nuttall’s specimens were in fruit only. I found this same #hus on the 6th of April, 1842, in descending from the table lands of North Alabama to the Tennessee River, on the Huntsville road. Here were large shrubs of it growing in-clumps from 8 to 10 feet high. They were in fruit and strikingly beautiful. I collected many specimens of them Soon after, I came to the river and staid all night ata Mr. Chun’s.. Next morning, I crossed at “ Ditto’s Landing,” and went on the Madison turnpike to Huntsville, nine miles distant. After dinner, I went twelve miles from Huntsville to a Mr. Bailee’s, in Madison County, Alabama, on the road to Winchester, Tennessee, via Salem. From Mr. Bailee’s house I walked up to the top of a low mountain to get specimens. Near its base and on its sides, in wooded ravines, I saw several trees of 2. cotinoides in flower, and with larger leaves than those seen before south of the Tennessee River. These trees were about 38 feet high and from 8 to 12 inches in diameter, with trunks and larger limbs coated with light gray and deeply-furrowed bark resembling the bark of the larger trees of the common sassafras (S. officinale). I climbed a tree to obtain specimens. The branches were brittle, and when broken emitted a yellowish sap, the odor of which was highly aromatic ; to me very unpleasant. This sumac, when in fruit, resembles the common Venetian sumac (Rhus cotinus), as is indicated by its specific name. It must be very local in its habitats, and also very rare. I think it has not been found elsewhere than in the Indian Territory and North Alabama. In Chapman’s “ Flora of the Southern States,” p. 70, it is stated that I found it in the interior of Alabama. Dr. Chapman was led into this error from our correspondence and exchange of plants when I was living in Wilcox County, Alabama. 126. PROCEEDINGS OF THE ACADEMY OF [1881. JUNE 7. The President, Dr. RuscHENBERGER, in the chair. Highteen persons present. The following was ordered to be printed : OBSERVATIONS UPON THE HIPPOPOTAMUS. BY HENRY C. CHAPMAN, M. D. On several different occasions, before and during the reign of Augustus and of his successors, Antoninus, Commodus, Heleoga- balus, ete., the Hippopotamus was exhibited at Rome. Naturally one would suppose, therefore, that among the writers of those times a truthful description of this interesting animal would be found. Pliny’s! account, however, is only a restatement of the imperfect and erroneous descriptions of Herodotus and Aristotle, with some mistakes of his own added, while those of later Latin writers like Tatius, though better than Pliny’s, are still obscure and contain errors. Indeed, the Hippopotamus, as described by Herodotus? and Aristotle,? is so unlike the animal known at the present day, that either these usually most accurate and trust- worthy observers could never have seen the Hippopotamus or else they must have described some other animal under that name. About the middle of the sixteenth century it is said that Belon saw the living Hippopotamus at Constantinople, but even so late as the time of Cuvier‘ the living animal had not been seen in Western Europe. The London Zoological Garden, I believe, has the credit of having been the first in modern times (during 1850) to exhibit the living Hippopotamus. So far as I know, the first dissection of this animal was made in 1764, by Daubenton.® The specimen, however, being a female 1 De animalibus, Lib. viii, cap. xxxix ; Lib. ix, cap. xiv. 2 Historia, Lib. ii, cap. 1xxi. 3 Historia Animalium, Lib. ii, cap. iv. 4 Ossemens Fossiles, Tome deuxieme, p. 383. 5 Histoire naturelle, &c., avec la description du Cabinet du Roi. Tome douzieme, 1764, p. 50. Supplement to Buffon. - 7 1881.] NATURAL SCIENCES OF PHILADELPHIA. 127 foetus, it was questionable how far the description would apply to the adult animal. Nothing further was added to the above account until, in 1844, Vrolik! described the stomach of a half-grown indi- vidual from drawings sent from the Cape of Good Hope. A few years afterwards Peters,” in his Travels, gives a short but valuable account of the appearance that the viscera presented in the adult animal. It will be seen, therefore, that the knowledge of the soft parts of the Hippopotamus was very limited up to quite a recent period. During 1867 there appeared the elaborate monograph of Gratiolet*? on the anatomy of this animal, and the important observations of Crisp.‘ Gratiolet’s description was derived from his dissection of the two young animals, male and female, that were born and had died in the Jardin des Plantes. Science is indebted to Dr. Alix for the publication of this important work, Gratiolet. dying before its completion. This distinguished anatomist had, however, before his death, communicated to the Académie des Sciences,’ an abstract of his researches. A third young Hippopotamus having died in Paris during the preparation of the work just referred to, Dr. Alix had a further opportunity of supplementing and confirming Gratiolet’s views. Dr. Crisp’s specimen was a male and about fourteen months old; it was burnt to death in the fire that destroyed the Crystal Palace in London, and was the first Hippopotamus dissected in England. Dr. Crisp® refers to Gratiolet’s abstract in Annales des Sciences Naturelles for 1860, but does not mention that in Comptes Rendus for 1860. His observations were, therefore, uninfluenced by those of Graticlet. In 1872, Mr. J. W. Clark’ published the “ Notes on the Visceral Anatomy of the Hippopotamus ” that died in the London Garden. This animal was a female, and only a few days old. It is well known that in addition to the ordinary Hippopotamus, there is a rarer species from the Western Coast of Africa, first 1 Amsterdam Verhandelingen, x, 1844, p..240; Recherches sur la Baby- russa. * Reise nach Mosambique, 1852, i, p. 180. 3 Recherches sur anatomie de l’Hippopotame. Paris, 1867. * On some points connected with the anatomy of the Hippopotamus. Proc. of Lond. Zool. Soc., 1867, p. 601 and 689. > Comptes Rendus, 1860, pp. 524, 595. Op. cit.,-p. 601. 7 Proc. Zool. Soc., London, 1872, p. 185. 128 PROCEEDINGS OF THE ACADEMY OF [1881]. made known by Morton,! and called by him Hippopotamus Liberiensis. Its osteology was afterwards fully described by Prof. Leidy,? who showed that this species differed so much from the ordinary one that a distinct name, Cheropsis, was given to it, as indicating that the supposed new species was really a new genus. Prof. Leidy’s views have since been thoroughly corrobo- rated by other anatomists, particularly by Milne Edwards,? in his recent beautiful monograph on this animal. The only living example of the Cheropsis Liberiensis ever seen outside of Africa was the female specimen only three or four months old that died a few minutes after arriving at the Zoological Gardens in Dublin in 1874, and that formed the subject of a paper by Mr. Alex. Macalester.* Since then, within a year, the late lamented Dr. A. H. Garrod > communicated to the Zoological Society of London the results of his dissection of the adult male Hippopotamus that had lived twenty-eight years in their admirably conducted Garden. It will be observed from this resumé of the literature of the subject that, with one or two exceptions, the Hippopotami that have been dissected were young animals; some not more than a few days or weeks, others about a year old,and that with the exception of the Chewropsis examined at Dublin, they were of the ordinary kind, or the Hippopotamus amphibius. While the general results of these various observations are con- firmatory of each other, nevertheless, on account of the difference in the age and sex of the individuals dissected, it is still important that whenever the opportunity presents itself of examining a full or half-grown Hippopotamus the results of such disseetion should be compared with those already made for the sake of confirming, supplementing, or further illustrating them. It is with this object that I bring before the Academy the results of my examinations of the female Hippopotamus which recently died in the menagerie of Mr. Adam Forepaugh, to whom I am indebted for the oppor- tunity of dissecting it; and of the male specimen that died in New 1 Proc. Acad. Nat. Sciences, vol. ii, p. 14; Journal, vol. i, 1849, p. 231. 2 Journal Acad. Nat. Sci, vol. ii, 1852. 3 Recherches sur les mammiferes. 4 Proc. of Royal Irish Academy, 1874. The anatomy of Cheeropsis Libe- riensis. 5 Trans. of Zoo. Soc. of London, 1880. On the Brain and other parts of the Hippopotamus. 1881.] NATURAL SCIENCES OF PHILADELPHIA. 129 York on its way to the Zoological Garden of Philadelphia. I take the occasion also of thanking Mr. Arthur E, Brown, Superintend- ent of the Zoological Garden where the dissections were made, for materially assisting me in the investigation. Both the animals were examples of the ordinary species, the Hippopotamus amphibius, and measured about 5 feet 6 inches in length. The female was both the taller and heavier of the two. Her height at the shoulder being 28 inches and weight 550 pounds. She was probably older than the male. The condition of the skin in the female suggested the idea that it had not been sufficiently bathed during the past winter. It is well known that the health of the skin, and of the animal generally, depends upon the free use of water, either in the form of a bath, or where that is not practi- cable, by constant sponging, ete. With the exception of some slight inflammation of the fourth stomach and an apparent hyper- trophy of the left ventricle of the heart, the organs were healthy. The male animal died from an inflammation of the stomach and intestines, the epithelium and submucous tissue in parts of the stomach being stripped off, while portions of the intestine were gangrenous. The immediate cause of death was a large well- organized clot in the heart. As the myology of Hippopotamus and of Cheropsis have been described and figured by Gratiolet! and Macalester? respectively, I will not dwell upon this part of the subject, but pass to the consideration of the internal organs. Alimentary System, etc_—The tongue of the Hippopotamus (Pl. XI, fig. 1) is a long, flattened organ expanded and rounded off at the top rather than tapered. It measured 14 inches in length, in breadth 34 inches at the middle and 5 inches at the top. At the back of the tongue where one finds the circumvallate papille in man, in place of these are seen what might be called elongated, thorny papille. They do not correspond to either the human filiform or fungiform papillz. The latter were well developed. I did not notice anything peculiar about the submaxillary gland, the sublingual however was absent;! the parotids were present, but not very well developed, as Gratiolet states was the case in the animals examined by him. The small size of the parotids in the Hippopotamus may be due to the habit of passing so much time in the water; the necessity of the secretion not being felt, 1 Op. cit., Planches IV to VIII. 2 Op. cit., pages 496, 500. 3 Op. cit., p. 384. 10 130 PROCEEDINGS OF THE ACADEMY OF [1881. as is the case in fishes. According to Gratiolet,! it is doubtful if they were present in the very young animal. What at once struck me, on exposing the larynx, etc., was the space (Pl. XI, fig. 1) intervening between it and the tongue, and the large size of the back of the tongue as compared with the epiglottis. Through the flexibility of this space the larynx when elevated can be thrust up into the posterior nares; this space, together with the tongue, effectually cutting off the cavity of the mouth. This can be well seen in the living animal. It is possible that this disposition of the parts may be of advantage to the Hippo- potamus when sunk in the water. Under such circumstances, the nose only appearing, the air can pass into the external nares: and so back directly into the larynx. Further, as the external nares are extremely flexible and close very tightly, it may be that the animal before sinking under the water can take in a considerable quantity of air into the nose and retain it there until needed, when it is then drawn into the larynx. On looking over the literature of the anatomy of the Hippopotamus, I find that Gratiolet? and Clark’ are the only ones who dwell particularly upon this part of its economy. Clark gives figures of the spaces I have referred to, and points out what appears to be the probable function of the parts. The larynx and its muscles have been well described by Gratiolet,* Crisp’ and Clark.® It would be superfluous therefore for me to dwell upon them. I will, however, call attention in this connection to the fact of the epiglottis (Plate XI, figs. 1 and 2 e) being small as compared with the larynx (Plate XI, fig. 2), the former measuring 245 inches in length, and 2 inches wide, the latter being 64 inches long and 64 in circum- ference, and that the nares, epiglottis, ete., of the Hippopotamus reminded me rather of those of the Manatee than of those of the Cetacea. The vocal cords were situated obliquely, the anterior ends being lowermost; they measure 2 inches in length and 4 inch in depth. There was nothing peculiar about the lungs; they were not divided into lobes or subdivided into lobules recognizable by the naked eye, as described by Gratiolet.’ The stomach in the Hippopotamus is subdivided into four distinct 1 Op-,cit.,6p: 64. 2 Op. cit., p. 375. 3Op. cit., p. 188. * Op. eit., p. 305. > Op. cit., p. 608. ® Compare Gratiolet, op. cit., p. 368. * Op. cit., p. 374. 1881. | NATURAL SCIENCES OF PHILADELPHIA. 131 compartments, b, c,d and e (Plate XII); the first, b, however, not being so apparent externally as the other three (shown in Plate XI, fig. 3). The cesophagus, a (Plate XII), opens freely into the compartment 6, which is situated posteriorly, and which might be easily overlooked unless opened. I propose calling this compartment the first stomach, as the food can pass from the cesophagus into it without necessarily passing into either of the other two stomachs, ¢ and d, whereas the food must pass through a small part at least of b in order to get into ¢ ord. This is due to a peculiar disposition at the entrance of the stomachs ¢ and d (Plate XIII). At this point the lining membrane is raised up into two valvular folds, g and h (Pl. XIII), of which the former is the best developed. The fold g almost divides the second stomach into two parts. These folds are 10 and 4 inches in length respectively, and about the } of an inch in breadth, and contain muscular fibres. When these folds are approximated the cesophagus, a, and first stomach, b, are completely shut off from c and d. When, however, the valvular folds are separated, then the food can pass from the esophagus, a, or from stomach, 3, over the edges of the folds, g and h, into either the stomachs, c ord. As the compartment d passes into e, which is continuous with the intestine, /, it appears to me that the two compartments may be appropriately called the third and fourth stomachs, in which case ¢ would be the second one. From a simple inspection of the stomachs of the Hippopotamus, one would be disposed to conclude that the animal was a rumi- nant. As the act of rumination, however, has never been observed in the Hippopotamus, either in captivity or in the wild state, so far as is known, the inference must be that the food passes either directly from the esophagus into the second or third stomachs, as is probably the case with liquids, or into the first stomach, and then indirectly into the second or third, when more solid articles are introduced. The four stomachs differ considerably in size, the third, d, being by far the largest ; it measured from right to left 27 inches, as seen in situ in Pl, XII; it overlaps, when viewed from the anterior sur- face, the second and fourth stomachs, ¢ and e, and, to a great extent conceals the first stomach, b, especially when the latter is empty. The first stomach, 6, measured 15 inches from right to left, and is so closely united to the third one, d, that externally 132 PROCEEDINGS OF THE ACADEMY OF [188]. the.two look like one when empty, and their distinctness does not become evident until they are forcibly separated and opened. The first stomach is also connected laterally with the third and fourth ones. These are about the same length, 7 inches, measured from right to left. The third stomach communicated with the fourth by a narrow aperture, which measured 3 inches in diameter. In situ the second stomach was situated in the left hypochondriac region; the fourth stomach on the right; the third stomach lying between the third and fourth and in the same plane, and in front and partially concealing the first stomach. The difference in the four stomachs of the Hippopotamus viewed internally are even more marked than those observed externally. The smooth mucous membrane of the cesophagus contrasts strongly with that of the first stomach, in which the mucous membrane exhibits parallel folds or ridges. In the second stomach the ridges are seen, but here they consist of rows of villi, averaging the } of an inch in height; the villi are not so closely set on the rows but that they can be readily distinguished. The villi in the third stom- ach, however, are densely packed and smaller than those of the second stomach. In addition the mucous membrane is thrown into eight large (the seventh divided into two) folds that run at right angles to the long axis of the stomach. The mucous membrane of the fourth stomach differed from that of the others in being the only one containing the gastric glands; according to Dr. Hunt, these measure in length 4, of an inch, in breadth ;}, of an inch. The food did not seem to be digested to any extent in the first three stomachs, but lay as a sodden mass. In the fourth stomach, how- ever, the food was softened, and its general appearance differed from that of the other stomachs. As the animal died shortly after eating, digestion had not been going on any length of time, so that any great change in the food could not have taken place. In the case of the male, the fourth stomach had been affected by dis- ease and the food appeared almost unchanged. The small intestine in the female measured 70 feet, the large intestine 11 feet. There were no valvulz conniventes in the small intestine but the mucous membrane was villous and exhibited the Lieberkuhnian follicle and the Peyer’s patches in the lower two-thirds. There was no very sharp line of demarkation between the small and large intestines, the beginning of the latter being indicated by a slight enlarge- ment. A small transverse fold was the only indication of an ileo- 1881. ] NATURAL SCIENCES OF PHILADELPHIA. 133 excal valve. As might have been expected on account of the large and complex stomach, there was no well defined czecum. It is an interesting fact, however, that the peculiar glandular-like structure in the czecal end of the colon of the Giraffe first described by Cob- bold, should be present in the Hippopotamus. There was nothing very peculiar about the pancreas or the spleen. The duct of the former pierced the duodenum separately from the ductus choledo- chus. The latter was closely bound to the greater curvature of the stomach by a fold of peritoneum. The liver was a quadri- lateral mass not subdivided to any extent into lobes. The gall bladder was absent in the female; in the male, however, it was present and measured 6 inches long. On comparing my observations with those of the anatomists already referred to, I find, that while some of the descriptions accord yery well with mine, others differ considerably. Thus Daubenton’s description is very good, especially when it is remem- bered that it is based upon the examination of a feetus. His! figures give a very good idea of the relations of the four stomachs when they are separated from each other by division of their con- necting bands. Peters’? and Vrolik’s* accounts are very fair. Unfortunately, however, no figures are given. On the other hand, I cannot say that the figure given by Crisp* of the stomachs illustrate the specimens examined by me. His descriptions, how- ever, of the colic gland, spleen, liver and pancreas accord very well with my observations. Clark® gives four figures, illustrating the stomach described by him, that by Gratiolet, and of one pre- served in the Museum of the Royal College of Surgeons. The figure of the latter gives a much better idea of the stomach examined by me than either that of Clark’s or Gratiolet’s. Garrod ® states that ‘he could find no confirmation of the peculiar position of the different parts described by Mr. J. W. Clark in his speci- men.” Possibly these differences observed in the stomach may be due to age, sex, to the extent to which the different stomachs had been separated, or to the amount of food that they contained, ete. As all of these conditions will influence greatly the form of the organ, it need not occasion surprise that I find the accounts Op. cit., figs. and 2, Pl TV. 2 Op. cit., p. 180. 3 Op. cit., p. 240. * Op. cit., fig. 3, p. 604. 5 Op. cit., figs. 4, 5, 6, 7, p. 190. © Op. cit... p- 16. 134 PROCEEDINGS OF THE ACADEMY OF | 1881. somewhat discordant, without, however, intending to throw dis- credit upon any of them. A Peccary having died at the Zoological Garden the same day that the Hippopotamus arrived there, a favorable opportunity presented itself of comparing the stomachs of the two animals. While externally the stomach of the Peccary is not subdivided to any great extent, internally through the elevation of the mucous membrane into two ridges, three compartments, cardiac, middle and pyloric, may be distinguished. The cardiac portion further subdivided at its termination into two blind pouches, opens into the middle division of the stomach; the latter receives the cesophagus and communicates with the pyloric part. Conceive the ridges and the cardiac pouches in the stomach of the Peccary greatly enlarged and we would have the stomach of a small Hippopotamus. On the other hand, diminish the first two stomachs of the Hippo- potamus to mere blind pouches, at the same time increasing the constriction between the third and fourth ones and we have, with- out any stretching of the imagination, the stomach of the Manatee. Beginning with the Pig the transition from that form of the stomach through the Babyrussa’ to that of the Peccary is an easy one. The latter again, leads to the Hippopotamus, which in turn anticipates on the one hand the Manatee and on the other the Ruminant type. Vascular System.—The circulation of the blood in the Hippo- potamus was first studied hy Gratiolet. The result of his careful investigation was the subject of a special communication to the Academy of Sciences, which appeared in the Comptes Rendus,? several years before the publication of his more general work by Dr. Alix, A good account of the heart is also given by Crisp.? With the exception of the above accounts, little or no attention seems to have been given to the study of the circulation by those anatomists who have dissected the common variety of Hippopota- mus, Daubenton* devoting merely a few lines to the heart, while the later writers do not mention the circulation at all. Macalester ® mentions one or two peculiarities about the blood-vessels in the Cheropsis. Although I have nothing particularly to add to Gratiolet’s excellent description, inasmuch as the subject of his 1 Vrolik, op. cit., p. 240. 2 Tome li, p. 524, 1860, 1867. 5 Op. cit., p. 609. * Opi cit:, ps ole 5 Op. cit., p. 495. 1881. ] NATURAL SCIENCES OF PHILADELPHIA. 135 dissection was only a day old, it was important that the heart and blood-vessels in a more fully developed animal should be examined with reference to determining whether the circulation was in any way modified by age. On opening the thorax of the animal it appeared to me that in both sexes the heart was large in proportion to the size of the animals. This is ina great measure due to the thickness of the walls of the left ventricle. In the female Hippopotamus, which was the first examined, I suspected this might be due to hyper- trophy, but finding it to be the case in the male also, perhaps this is normal. The heart, in an empty condition, measured, from base to apex, 9 inches, and in circumference 14 inches. The wall of the left ventricle measured 1 inch in thickness, that of the right } of an inch. According to Gratiolet,’ the heart in the young Hippopotamus terminates in two points, the ventricles being separated by a little groove, reminding one of the form of the heart in the Manatee and the Dugong. There was no indica- tion of this groove in either of the Hippopotami examined by me. With the exception of the absence of the corpora arantii on the semilunar valves of the pulmonary artery and their very slight development in those of the aorta I did not notice anything peculiar about the interior of the heart. The aorta gave off the coronary arteries first, which were very large and then an innomi- nate and the left subclavian. The innominate divided into the right subclavian and a trunk which bifurcated into the two com- mon carotids. The external carotid as well as the ascending cervical and occipital arteries were all rather slender vessels in proportion to the size of the head and neck. The external carotid artery was very much larger than the internal. Ossemens Fossiles, I, p. 492. 6 Paleontologie Francaise, p. 143. 7 Geological Magazine, 1875, p. 423. 8 De Zigno, Sopra un nuovo suienio fossile. Reale Acad. dei Lincei, 1877-78. 148 PROCEEDINGS OF THE ACADEMY OF (1881. as those of the Hippopotamus. It is proper to mention, however, that the same distinguished observer considers the teeth of the Manatee and the Prorastomus, another extinct Sirenean, to be rather allied to those of the Tapir and Lophiodon, but this quali- fication does not really invalidate the supposed affinities between the Sirenea and the Hippopotamus. For the Artiodactyla and Perisodactyla are probably offshoots of a common stock, and hence we may expect to find in these two groups certain characters common to both, inherited from their Lophiodon and Coryphiodon- like ancestors. The affinities of the teeth of the Manatee with those of the Tapir—the first an embryonic Artiodactyle, the second a generalized Perisodactyle—would be examples of the above view. I do not mean to imply that the Manatee has neces- sarily descended directly from the Hippopotamus, though extinct intermediate forms may in the future show this to be so, for possibly they may be the descendants of a common ancestor. To many such speculations may appear mere waste of time, we being unable, from the nature of the case, to experimentally prove or disprove the truth of the hypothesis advanced. It seems to me, however, that the only explanation of the structure of the living forms and of the petrified remains of the animals referred to in these observations is the hypothesis of there being some generic connection between them. 1881. } NATURAL SCIENCES OF PHILADELPHIA. 149 JUNE 14. The President, Dr. RUscHENBERGER, in the chair. Twenty-one persons present. A paper entited “ Notes on the Tertiary Geology of the South- ern United States,” by Angelo Heilprin, was presented for pub- lication. The death of Baron Maximilien de Chaudoir, a correspondent, was announced. Some new Generaof Fresh Water Sponges.—Mr. E. Ports re- ferred to a recent paper by H. J. Carter, F. R. S., (Ann. Mag. Nat. Hist., Feb. 1881,) entitled, ‘‘ The History and Classification of the Known Species of Spongilla,” in which the writer has dis- tributed the species, heretofore grouped under one generic title, among five genera, founded upon the differences in form and ar- rangement of the spicule surrounding the statospheres. He spoke of the arrangement as a timely step well taken in ad- vance, in the history of this branch of the animal kingdom, He believed that the characteristics of the statospheres and their spicule were those which furnished the only reliable dis- tinctions among fresh water sponges ; but the recent discovery of novel forms in American waters had already required an increase in the number of genera and seemed to make it desirable to modify the terms of some of those already established. In illustration he referred to several forms observed in this neighborhood, resembling in many points the English Spongilla lacustris, (taken as a type of the genus Spongilla in the new ar- rangement), in which, however, the spicule were not acerate, but irregular in shape; were not placed “tangentially” upon the surface ; or were altogether wanting. Specific names were suggested for these, but were held under advisement, awaiting a decision as to whether it would be better to create new genera for them or to enlarge the scope of those already defined by Dr. Carter. The two new genera already decided upon were then described. Under the generic head Meyenia, Dr. Carter has grouped those species, in which the statosphere is surrounded by birotulate spicule, radiately arranged; one disk resting upon the surface. Throughout the genus as already constituted, the shafts of these spicules are of a nearly uniform length ; and the outer disks nearly or quite touching at their edges give the appearance of a second coat to the statosphere. In two species, however, observed by Mr. Potts last summer, this uniform series was broken by another, of about double their length, much fewer in number, somewhat regularly arranged, interspersed among them. He 150 PROCEEDINGS OF THE ACADEMY OF [1881. proposed to group these under the genus Heteromeyenia, as H. argyrosperma and H. repens ; suggesting that the latter may pos- sibly be the same as Bowerbank’s Spongilla, now Meyenia Baileyt. Another new genus had been formed and dedicated to Dr. Carter under the name Carterella, to include the singular form described by the speaker last yearin the Proceedings of the Acad- emy, and then called Spongilla tentasperma; changed later to S. tenosperma. The distinguishing peculiarity of this genus is that the tube surrounding the foramen of the statosphere is elongated and divides into 2-5 long, curling or twisted tendrils by means of which during the winter the statosphere remains attached to the stems or roots upon which the sponge had grown. This will now be Carterella tenosperma. A second species has been added to this genus, the discovery of Professor Kellicott and Mr. Henry Mills, of Buffalo, under the name of Carterella tubisperma. In this, the tube is much longer than in any sponge heretofore described, terminating in several straggling, rather weak tentacles much shorter than in the former species. The birotulate spicule in the two forms are quite differ- ent, and the species are unquestionably distinct. JUNE 21. The President, Dr. RuscHENBERGER, in the chair. Nineteen persons present. JUNE 28. The President, Dr. RuscHENBERGER, in the chair. Fifteen persons present. Jerome Gray was elected a member. M. 8S. Bebb, of Rockport, Ill., and Chas. 8. Sargent, of Brook- line, Mass., were elected correspondents. The following was ordered to be printed : 1881.] NATURAL SCIENCES OF PHILADELPHIA. 151 NOTES ON THE TERTIARY GEOLOGY OF THE SOUTHERN UNITED STATES. BY ANGELO HEILPRIN. In the following notes the author makes no pretense at unravyel- ling the many knotty points connected with the Tertiary geology of the southern United States; he has merely brought together such facts, old and new, and certain conclusions drawn from these facts, as may possibly serve to facilitate the inquiry into this as yet imperfectly known branch of American geological history. It is with this view of rendering the material treating of the subject more accessible to the working geologist that some of the pub- lished sections are here reproduced. A convenient starting-point in Eocene stratigraphy is afforded by the famous bluff exposed on the Alabama River near Claiborne, Ala., and which has yielded the fossils known to geologists and paleontologists as those characteristic of the ‘‘ Claiborne Group.” Section of Claiborne Bluff.—Probably the most trustworthy section of this bluff is that afforded by Tuomey (‘“ First Biennial Report of the Geology of Alabama,’ 1850, p. 152), as follows: Feet. g Red sand, loam, and pebbles. 30 of Mottled clay. 8 | é Limestone, with grains of green sand. 54 d Ferrugirous sand ; numerous fossils. | ec Whitish limestone. 62 b Bed of clay 15 feet thick, with, seam of limestone on top.) 15 Norte. —Tuomey does not give the thickness of bed “‘d,”’ but it appears from the concurrent statements of different observers to be about 17 feet. The total height of the bluff above the Alabama River would therefore appear to be in the neighborhood of 190 feet. 152 PROCEEDINGS OF THE ACADEMY OF [1881. The measurements and descriptions of Conrad (“ Fossil Shells of the Tertiary Formations,” 1833, p 32; Proceedings of the National Institution, 1841. p. 174), Hale (“Geology of South Alabama,” American Journal of Science, new ser., VI, p. 354), and Lyell (Journal of the Geological Society, IV, p. 10, et seq.) do not differ very essentially from the data given by Tuomey. The arenaceous bed “d,” about 80 feet above water level, has yielded the vast majority of the fossils for which the locality is famous, and is that which has been identified as the equivalent of the “ Calcaire Grossier’’? (Upper Eocene) of France (et conseq., Bruxellian of Belgium, and Bartonian of England). To what extent the deposits either below or above this bed can be cor- related with the remaining deposits of the Paris or London series has not yet been determined ; nor has it yet been conclusively shown what exact portion of the American Eocene formation is represented in this Claiborne exposure. Although formerly con- sidered to be near the base of the system, there are now very strong grounds for concluding that these beds are underlaid by older Eocene strata having a thickness of at least 200 feet, and possibly even much more. The age of the limestone bed “ e,” although perhaps the character of its contained fossils does not permit absolute determination, is in all probability Jacksonian— at least in part—and will doubtless be found to correspond witha portion of the bluff exposed at St. Stephen’s on the Tombigbee River, about thirty miles almost due west of Claiborne. At any rate, a portion of the white, or so-called “ rotten” limestone im- mediately west of Claiborne has been found to contain several of the characteristic fossils of the Jackson group, and these asso- ciated with the remains of Zeuglodon; there is, therefore, no doubt as to the age of at least this portion of the white limestone, nor can there be any reasonable doubt as to the continuity existing between these deposits and the similar ones exposed on Claiborne bluff. Section on Bashia Creek, Clarke Co., Ala.—Probably the section representing the oldest Eocene deposits of the State of Alabama is that exposed on Bashia Creek, Clarke Co., and detailed by Toumey in his report on the geology of the State (First Biennial report, p. 145): 1881. ] NATURAL SCIENCES OF PHILADELPHIA. 153 1 Hard Limestone. 4 feet. 2 Marl, highly fossiliferous. 25 fost: 3 | Blue sand. yA Variable. 4 Lignite and clay. | Gfeet. 5 Laminated clay, sand, and mud. Thickness undetermined. 6 | Lignite, do. dont Notre.—Beds 5 and 6 do not properly belong to the section, but ‘‘repre- sent beds seen on another part of the stream below the preceding ”’ (Tuo- mey, loc. cit,, p. 146). Beds corresponding to No. “2” of the above section are like- wise exposed on Cave and Knight’s branches, tributaries of Bashia Creek, and have been shown by Dr. Eugene A. Smith to underly the base of the ‘“‘ Buhrstone” proper by nearly (if not more than) 200 feet (Heilprin, “ Proc. Acad. Nat. Sciences of Philadelphia,” 1881, p. 369). The relations of these various beds will be best understood by a reference to the sections exposed on the Tombigbee River. Sections on the Tombigbee River.—At Wood’s Bluff, near the mouth of Bashia Creek, we have the following exposure : ! No. Feet. v4 Orange sand, or stratified drift. 10-20 | 6 Grayish or greenish laminated clays, colored brown by 10 iron. | 5 Ledge of bluish or greenish sand, fossiliferous—capped 2 | by a ledge of hard nodules. | 4 | Bluish laminated clay, with few fossits. 5 3 | Indurated greenish sand, full of the same shells as 3 | marl bed No. 2. | | 2 | Greensand marl, quite soft, and full of shells, | 3 | 1 | Indurated greensand with shells, and a stratum of 10-15 | oyster shells at water’s edge—said to extend 10 feet | | further down. 1 I am indebted to the kindness of Dr. Eugene A. Smith, State Geologist of Alabama, for the use of this heretofore unpublished section, 154 PROCEEDINGS OF THE ACADEMY OF [1881. Bed No. 4 of the preceding section is considered by Dr. Smith to be most closely related in the character of its fossil remains to the fossiliferous strata exposed on Cave and Knight’s branches, and it is therefore not unlikely that the series 1-5 corresponds in the main with No. 2 of Tuomey’s Bashiasection. The basal lignite would then probably be found to underly the lowest stratum exhibited at the Bluff (Heilprin, loc. cit., p. 367-8). Bed No. 6 (Wood’s Bluff section) can be traced down the river for a distance of two to three miles, when it dips beneath the water’s level. Somewhat below this point, and beyond the mouth of Witch Creek, the stratigraphical relation of the different beds is beauti- fully exhibited in a prominent cliff (‘“* White Bluff”), rising from 250 to 275 feet above the river. The upper portion of this bluff is constituted by the characteristic siliceous clay-stones and silici- fied shell deposits of the southern “ Buhrstone ”’ formation, which make up fully 100 feet of the vertical height. Laminated lignitic clays (bearing numerous leaf impressions), with occasional inter- calated beds of pure lignite, enter mainly into the composition of the intermediate portion, 7. e., from the water’s level to thé base of the buhrstone above mentioned. Allowing a uniform southerly dip of 10 feet to the mile, which appears to be consistent with obtained data, it is manifest that at this point the lower fossiliferous strata exposed at Wood’s Bluff (and consequently, the equivalent deposits on Bashia Creek and its tributaries, Cave and Knight’s branches) must lie from 175 to 200 feet below the base of the siliceous mass constituting the true buhrstone ; or, in other words, we have here a series of deposits aggregating about 300 feet in thickness, which can be shown to be of an age anterior to the deposi- tions of the Claiborne fossiliferous sands. At Baker’s Bluff, a few miles above St. Stephen’s (which is situated about twenty-eight miles south of Wood’s Bluff), the buhrstone, according to Tuomey, appears in a vertical escarpment rising only 50 feet above the water, a low height perfectly in accordance with the loss occa- sioned by the general dip extending over nearly twenty miles. At this point, moreover, and occupying a position above the buhrstone, Tuomey (Joc. cit., p. 148) identifies a bed of green sand (8 feet in thickness) as the equivalent of the Claiborne fossilifer- ous sands “d” of his section), and containing numerous fossils identical with those found at Claiborne. Still further south, and occupying a considerably lower level, the same bed is described 1881.] NATURAL SCIENCES OF PHILADELPHIA. 155 as having an extent of 12 feet, and immediately above St. Ste- phen’s, was seen to dip beneath the water’s edge. At this last locality we have a beautiful exhibit of what has generally been designated by the name of “ White Limestone.”! There can be not the least doubt, however, that this “* White Limestone,” which has most frequently been taken to represent strata of Vicksburg age, is in reality, as has been insisted upon by Winchell (Proceedings of the American Association, 1856, Part I, p. 85), a combination of strata belonging to two distinct (at least, as now recognized) groups of deposits. The lower moiety, dipping into the river, and resting upon the subjacent Claiborne sands (Tuomey, Joc. cit., p. 157; Lyell, Journal Geol. Soc., Lon- don, IV, p. 15; Hale, A. J. Science, new ser., VI, p. 359) is the true “ White Limestone,” an exponent of the Jacksonian group of deposits, as may be inferred from its position, and the charac- ter of its contained fossils.2 Moreover, were it otherwise the case, it would have been very difficult to explain the total disap- pearance over a distance of only thirty miles (and with but ex- ceedingly moderate dip) of the equivalent beds exposed on the Alabama River at Claiborne. The upper moiety, on the other hand, is a portion of the well known Orbitoide ( Vicksburg or Oligocene) rock, and is that which alone contains specimens of Orbitoides Mantelli (Winchell, loc. cit., p. 85). From the data herewith presented, a section of the Tertiary strata traced along the Tombigbee River from Wood’s Bluff to St. Stephen’s, may probably, with considerable approach to truth, be constructed as follows: 1 ] have been unable to discover the exact height of this bluff. Neither Lyell nor Toumey mentions it ; Conrad, in the appendix to Morton’s ‘ Sy- nopsis’’ (p. 23), states itis about 100 feet. * Spondylus dumosus and Ostrea panda, originally described as charac- teristic fossils of the Newer Cretaceous (upper Eocene) deposits of the southern United States, have been found abundantly near the base of the bluff. (1881. PROCEEDINGS OF THE ACADEMY OF 156 Ul es eee ee ee ee ee ee ee 288 f te rs anya sored KE WANT v0qS1qmoy, WHO], a “Spuny aYASnE pee oes suaydayg 4g haf: Late Bae -~ ~ aa of or Ag — "te BEqnogo uoroage fo yreap a $ N poy OTM aruba ie wrysVg, "3S O0/ £81O wey al mm Td SANS S 74 §, poo onusry : VLA ik M. (s/os) se10 ~ | sroysayng 1881. ] NATURAL SCIENCES OF PHILADELPHIA. 157 An examination of the foregoing section shows almost conclu- sively that the Eocene deposits of Alabama have a thickness of very nearly 400 feet; and, indeed, I am informed by Dr. Smith that there are good grounds for supposing that Tertiary beds exist in the northern part of the State, whose position would be still 150-180 feet below the Wood’s Bluff marl bed. It will fur- ther be seen that the Claibornian (or Claiborne proper, as charac- terized by the fossiliferous greensands) holds a position decidedly near the top of the series, a position almost precisely similar to that occupied by the ‘“ Calcaire Grossier’’ (Parisian) of France, and more properly Upper than Middle Eocene, which last it has very generally been considered. What relation beds *b” and ‘6c of the Claiborne Bluff holds to the sub-Claibornian (** Buhr- stone ” in part) deposits of the Tombigbee River has not yet been absolutely determined ; but there can probably be no legitimate doubts that they represent, in a modified form, the upper moiety of those deposits. Although the marked difference in the lithological character of the strata of the two localities as compared with each other (and indeed it must be confessed, this is much greater than could have been reasonably inferred from the general constancy of the deposits in this region) would seem to militate against such a view, there is, nevertheless, sufficient evidence, both stratigraphical and paleontological, to support this conclusion. In the first place, by determining the position of the buhrstone rock near St. Stephen’s as immediately underlying the highly fossiliferous greensand layer, TTuomey has proved that the two series of deposits (the Buhrstone on the Tombigbee, and bed “c¢” on the Alabama) hold relatively the same position to the true Claibornian, lying immediately below it. In the second place, the argillaceous strata at the base of Claiborne Bluff (bed “4” of Hale’s series) have been identified by Hale, both on litho- logical and paleontological evidence (A. J. Science, new ser., VI, p. 356), as occurring at Coffeeville Landing on the Tombigbee River, about 14 miles north of St. Stephen’s, what might very readily have been expected from an inspection of the general lay of the different formations.! No data are given relative to 1 A line uniting Claiborne and Coffeeville Landing would run almost precisely parallel with the line marking the junction of the Cretaceous and Tertiary deposits lying hence due north. The contour lines traced by Tuomey would indicate a true dip west of the southerly line, and that this 158 PROCEEDINGS OF-THE ACADEMY OF [1881. the position of the Buhrstone at this last locality, but hypotheti- cally considered (as deduced from its position at White Bluff, and its general dip), its summit must still oceupy a position fully 100 feet above the level of the river ; and this thickness (100 feet) coincides very closely with the thickness (80 to 90 feet) of the deposits below the true Claibornian (bed ‘‘d”) as exposed at the bluff on Alabama River. And finally, that at least a very con- siderable portion of the inferior beds at this last named locality represent strata of a different lithological character in other por- tions of the state —and consequently, that they are local deposits— is proved by the concurrent statements of Hale (Joc. cit., p. 356) and Winchell (loc. cit., p. 86), both of whom assert that the cal- careous deposit below the arenaceous bed (not the “ White Lime- stone”) is not known to occur at any other locality.! Admitting the conclusions reached in this paper, it will be seen that the Alabama Eocene deposits comprise :— 4. “ White Limestone ”’ (Jacksonian), best exhibited at Claiborne (upper portion of bluff) and St. Stephen’s (lower moiety of bluff), not very abundant in fossils—Pecten membranosus, P. Poulsoni, Ostrea panda, Spondylus dumosus, ‘ Scutella” Lyelli, etc.—50 —? feet. 3. The fossiliferous arenaceous deposit (Claibornian), best shown at Claiborne—subaqueous at St. Stephen’s—very rich in fossils, and of the age of the “ Calcaire Grossier ” of France —I7- feet. 2. “ Buhrstone”’ (Siliceous Claiborne of Hilgard), comprising siliceous clay-stones (buhrstone proper) densely charged with fossils or their impressions, laminated clays, sands and calcareous deposits—beds “b” and “tc” of the Claiborne section, the cliff at White Bluff, and the so-called “ Chalk is actually the case is proved by the difference (80-90 feet) between the actual heights at which the equivalent beds at St. Stephen’s and Claiborne are placed. This also accords with Hilgard’s observations in Mississippi, where the dip of the Jackson and Vicksburg strata was found to be about 10 to 12 feet per mile 8. by W. (A. J. Science, new ser., XLIII, p. 36). 1 It is greatly to be hoped that under the present able management of Dr. Smith, the survey will be able to yield much more satisfactory data connected with the geology of the State than have heretofore been rendered. 1881. ] NATURAL SCIENCES OF PHILADELPHIA. 159 Hills” of the southern part of the State. At Claiborne the representative beds consist of aluminous and calcareous deposits, poor in fossils, but containing occasional layers of Ostrea selleformis—about 250 feet? 1. The Wood’s Bluff and Bashia (with Cave and Knight’s Branches) deposits (Eo-lignitic), consisting of alternating dark clays, greenish and buff sands, and numerous seams of lignite, partly very rich in fossils, and as far as is yet posi- tively known, the oldest Tertiary deposits of the State.—50 — ? feet.! It is the intention of the writer to discuss in a future paper the relations of these various Alabama deposits to those of other sections of the United States, and to correlate them, as far as possible, with the Eocene deposits of the typical European basins. 1 Tt appears to the author that it would be convenient to designate these lower deposits, which hold a rather constant position at the base of the Eocene series in different parts of the eastern and southern United States, by a term which could be readily applied in adjective form, and which would at the same time in some manner express the relation of the beds referred to. He therefore proposes the term ‘‘ Eo-lignitic,’’ which, while it to some extent indicates the general character of the beds so designated, is not restricted in its definition to the character cf the deposits of any one single locality. The ‘‘ Buff Sand’’ of Winchell (loc. eit., p. 89), probably falls into this group, but its exact position, or its correspondent, does not appear to be as yet definitely determined. It is seen to underly the ‘‘Buhrstone,’”’ and is considered by Winchell to represent the absolute base of the Tertiary system of the State. At Black’s Bluff, Wilcox Co., it is stated to repose directly on the subjacent Cretaceous limestone, but in a foot-note (p. 90), we are informed that, according to Tuomey, the characteristic fossil of this limestone, an Ostrea, is probably Tertiary. 160 PROCEEDINGS OF THE ACADEMY OF { 1881. JuLY 5. The President, Dr. RuscHENBERGER, in the chair. Seventeen persons present. A paper entitled, ‘‘ The Snare of the Ray Spider, Epeira radi- osa, a new form of Orb-web,” by Rev. H. C. McCook, D. D., was presented for publication. Sarcodes sanguinea.—Mr. THoMAS MEEHAN referred to discus- sion among members at former meetings, as to the true character of parasitic plants. They were believed to be in the main of two classes,—one which might be represented by the common mistle- toe, with woody stems continuing from year to year,—the other like the Arceuthobium, or pine parasite, which died to the surface of the wood, but continued to grow up from the same spot every year—a sort of parasitic herbaceous plant. It was a question how far oot parasites partook of these several characters. There were some plants, as Castilleia and Comandra, which might be said to be in a transition state between an ordinary terrestrial plant and a parasite. Usually they were as other plants, but some of the roots would attach themselves to other roots, and form as perfect a union as genuine parasites, and, by the decline in vigor of the victim root beyond the point of union, evidently showed they were really parasitic, deriving nourishment from the attach- ment. Aphyllon uneflorum, germinated on the annual fibrous roots of Asters and Solidagoes, as had been clearly traced, and perhaps on other plants ; and after germination formed a mass of innumerable coral-like spongelets, drawing moisture and perhaps some other elements of nutrition from the surrounding medium. Epiphegus Virginiana behaved precisely inthe same way. Mono- tropa and others had also this mass of pseudo-roots, or spongelets, and had been supposed to germinate and live wholly on half de- cayed vegetation, but he believed from analogy they would be found, as in Hpiphegus and others, to germinate at first on living roots. Conopholis was the only root parasite he had found any reason for believing to be a perennial. This had been found at- tached to quite large roots, evidently coming up from the same spot from year to year as Arceuthobium does. Having correspondents in regions where grows the beautiful Snow-plant of the Sierras—-Sarcodes sanguinea—about which nothing but its aerial character has been so far known, he had set them to watching for him, their appearance and final end. The places where they grew were carefully marked, and with the fol- lowing results :—Mr. John M. Hutchings, of Yosemite, found the bottom of the old plants 10 to 14 inches below the surface, with not the slightest signs of attachment anywhere. To him it ap- 1881.] NATURAL SCIENCES OF PHILADELPHIA. 161 peared no more than an ‘ordinary annual plant of great beauty.” Of course an ordinary annual growing from seed, could not push through the ground at so great a depth. The vital power spent in overcoming so heavy an obstruction, would be exhausted long before the growing point pushed through a foot of soil to the sur- face, as observing seed growers of experience would testify. Only a parasite, or bud, having an unlimited supply of food to draw on could accomplish this feat. But the matter was settled by an- other observer, Mrs. Ross Lewers, of Washoe Valley, Nevada, who, together with her observations, had sent the dead spongy _ mass from the last year’s plants, which Mr. M. exhibited, and these _were found to have a slender pine root through the mass around which the spongy mass had grown, and as it was dead, there was little doubt that it had been the matrix from which the seed had made its original start, and which the plant killed in the end. The dead, spongy mass of pseudo-roots was larger than he had seen in any other species of root parasite. Altogether it might be said that Sarcodes sanguinea was an annual, germinating on small pine roots, and subsequently obtaining subsistence from the earth as Aphyllon, Epiphegus, &ce. Talinum teretifolium.—Mr. THoMas MEEHAN remarked that the point he made recently in regard to Draba verna, that mere light alone evidently failed to account for the special opening-time of flowers, was confirmed by recent observations on Talinum teretifolium. When preparing the chapter on this plant for his ‘‘ Native Flowers and Ferns of the United States,’ a few years ago, he had watched plants almost daily through the whole season and found that without any exception they opened at 1 Pp. M. and closed at 2. This season he had watched them again and found that though the time of opening was the same, 1 P. M., they never on any occasion noted, closed at 2, but continued open sometimes to half-past three or half-past four,and on one occasion were found closing at half-past five. Mr. Meehan said he had endeavored to associate these variations with some atmospheric changes, such as heat, light, or moisture, but in no case did these endeavors prove satisfactory. In the *‘ Proceedings of the Kansas Academy of Science,” Professor Smyth had contributed materials fora “floral clock” for Kansas, and Talinum teretifolium was set down in the list as opening at 11 a. M. in that State. Mr. Meehan believed that the laws influencing this peculiar class of motion in flowers, were completely hidden from us, and that the subject offered an inviting field to the biologist. Mr. Redfield suggested that perhaps the age of the plants made some difference in their habits. Mr. Meehan replied that the patch in his garden was much larger now than at first, from addition through self sown seeds ; but all the plants behaved precisely alike. He did not suppose 12 162 PROCEEDINGS OF THE ACADEMY OF [1881. that external circumstances had no influence ; but that the con- dition of the living material on which they acted, decided the final character, and that Mr. Redfield’s question was in the right line. ASiotine 1Ip2 The President, Dr. RuscHENBERGER, in the chair. Ten persons present. The death of John P. Brock and of Alfred D. Jessup, members, was announced. The following was ordered to be printed :— 1881.] NATURAL SCIENCES OF PHILADELPHIA. 163 THE SNARE OF THE RAY SPIDER (EPEIRA RADIOSA), A NEW FORM OF ORB-WEB. By Henry C. McCook, D. D. In the vicinity of Philadelphia, June 14, 1881, I found a number of spiders grouped not far from each other on Epeiroid webs, which proved to be of a type hitherto unknown, and which I designate as the Actinic or Ray-formed Orb-web, The spider appears also to be new to science, and is named EH peira radiosa.! I. CHARACTER OF THE WEB. The first example or two of the spiders collected seemed to be upon nests that had been broken by ordinary wear and tear in Fig. 1.—The Ray Spider seated in her snare, just before drawing the trap-line. capturing insects; bat the repetition of the form in a third snare, particularly of the peculiar open cen‘ral, caused more careful examination. The result was the discovery of the remarkable 1 Radiosa, full of rays. A closer study of the spider may compel the change of its generic position. 164 PROCEEDINGS OF THE ACADEMY OF [1881]. form of web here described. On account of the continually changing form of the snare, it will be necessary to present it from various points of view, and as seen in different stages of its diurnal changes. Fig. 1 presents a view of the snare in a partially relaxed con- dition. The spider is seen seated in the centre of a series of rays, i, il, ill, iv, v, which are grasped by the third and fourth pairs of legs. There is no hub, properly speaking, but the axes of the rays may be seen at times united upon a central point, as at H, fig. 2. The general tendency is to four or five main divisions or Fie, 2. rays, as may be seen by studying the figures presented. But there is more or less variation, and in the course of the day’s usage in capturing prey two sections will become interblended upon one axis, as appears to be the case in fig. 2, and also in fig. 4. The central space is a large irregular opening, constituting about one-third of the entire snare, whose diameter is usually from three to five inches (see fig. 3). The central circle, meshes, and notched spirals which so generally characterize the Orb-webs are thus wholly wanting here. 1881.] NATURAL SCIENCES OF PHILADELPHIA. 165 The orb may be said to be composed of a series of independent rays or sectors, each ray composed of several spirally crossed radii, and the whole series united into an orb by cross-lines or spirals like those which unite the radii. In the shifting of the section-lines above referred to, this separation of the orb into in- dependent rays, is always quite evident. The spirals are covered with viscid beads, as in most orb-webs The radii do not all pass to the Hub or Centre, as do those of orb-webs generally, but converge for the most part upon the axes of the rays as repre- 9 sented at figs. 1, 2, 3. These axes themselves converge upon a Fie. 5.—View from front. Web taut. Perspective not shown. Central opening exact. single strong thread, a trap-line, T ( figs. 1 and 2, and succeeding cuts), which is attached to some part of the surrounding surface, of rock, earth or plant. When the snare is flat or relaxed, as was the case with the one drawn at fig. 2, and as appears in fig. 1, the trap-line is often about perpendicular to the plane of the orb, as is the handle to the rays of an open Japanese umbrella. This, how- ever, depends somewhat upon the environment; a convenient point for the attachment of the trap-line will cause the animal to . divert the thread more or less from the perpendicular. 166 PROCEEDINGS OF THE ACADEMY OF [188]. We may now suppose the spider placed as in figs. 1 and 3, at the point where the rays converge, grasping the axes with the four hind feet. The two front feet seize the trap-line and draw it taut. Then, precisely as a sailor pulls upon a rope, “ hand over hand,” the little arachnid’s feet move along the trap-line, one over another. The axes of the rays, held firmly in the hind feet, follow her ; the centre of the snare bears inward, the other parts are stretched taut, and the web at last has taken the form of a cone or funnel (figs. 4, 5). In this position the snares continually suggested to me the figure of an umbrella with 1ibs reversed by the wind and the covering Fic. 4.—Side view of Ray Spider’s snare, when drawn taut or bowed. Seen within a cavity. stripped loose from the top of the handle, Fig. 4 gives a side view of the web when thus bowed or drawn taut; another snare is shown at fig. 5, as seen from behind. In this example (fig. 5), the spider has moved quite down the trap-line to the surface of the little twig (projecting into the cavity) to which it is attached. It will thus be seen that the snare is more or less a plane surface, or more or less conical, according to the position of the animal upon the trap-line and the degree of tension thereof. 1881. | NATURAL SCIENCES OF PHILADELPHIA. 167 II. MopE oF OPERATING THE SNARE. When an insect strikes the snare, the spider has two modes of operating. The first somewhat resembles that of the ordinary orb-weaver in that the insect is simply permitted to entangle itself and is then taken, swathed, returned to the centre and eaten. There is, however, this difference: before going to the insect, the axes of the snare are twisted or knotted, by a rotary action of the body and movement of the legs, so that the untouched parts of the orb remains taut. Fig. 2 represents a snare thus “ locked,” Fie. 5. or, perhaps I might more properly say, ‘‘ keyed.” The trap-line is now relaxed, although its elasticity is such that the change can scarcely be noticed. The spider then moves upon her victim, quite habitually cutting out the spirals with her mandibles as she goes. When the insect is ensnared well towards the circumference of the web, and indeed, for the most part, in other cases also, it results that the ray or sector upon which the entanglement had occurred, is quite cut away. The spider thereupon proceeds to operate the remaining parts of her snare, which, in time, is thus destroyed by sections, as will be fully illustrated hereafter. 168 PROCEEDINGS OF THE ACADEMY OF [188]. The second mode of operation resembles that of the Triangle spider, Hyptiotes cavata, Hentz, which has been so admirably described by Wilder, and which I have very frequently and fully observed in the suburbs of Philadelphia and throughout Pennsyl- vania. It is at this point that the habit of our Ray spider becomes particularly interesting. The Triangle spider makes a triangular web, which is in fact an orb sector, composed with unvarying regularity of four spirally crossed radii converging upon a single line T (fig. 6,a). Upon this line the spider hangs back downward, grasping it with all her feet, and having a portion of the line, SI (fig. 6, b), rolled up slack, between her two hind, or sometimes, Fia. 6.—Triangle spider hanging upon taut snare. a, Spiderin position. SI, Ball of slack-line. b, Enlarged figure of spider, showing the mode of grasping the line. apparently, her fore and hind feet. Thus the forward and back parts of the trap-line are taut, while the intermediate part is slack. The spiral parts of the snare are also taut. When the web is struck by an insect, the spider suddenly releases her hind feet, the slack line sharply uncoils, the spider shoots forward, the whole web relaxes, as at fig. 7,and the spiral lines are thrown around the insect. This is repeated several times before the prey is seized. 1881. } NATURAL SCIENCES OF PHILADELPHIA. 169 Precisely the same action characterizes the Ray spider. Her ordinary position, or at least the one in which I most frequently observed her, is a sitting posture, back upward, as shown at fig. 1. The axes of the rays are held in the third and fourth pairs of legs, the fourth commanding the upper, the third the lower series, quite habitually, as it appeared to me. A sort of “ basket,’ system of connecting lines, shown at figs. 1, 9, unites all the feet, seeming to converge toward the fore-feet (perhaps, upon the second pair), where they grasp the trap-line. It is upon this foot- basket that the spider sits when her net is bowed. This, however, is not the invariable posture ; in the reconstruc- tion of the rays and shifting of the axes, as the day’s work tells Fig. 7,—Triangle spider, Hyptiotes cavata, with slack-line uncoiled and snare relaxed. upon the snare, the spider will vary her posture to that of fig, The trap-line generally has a direction downward rather ane upward, so that the head and fore-feet tend to be depressed below the abdomen, and this depression may gradually result in the complete inversion of the animal, fig. 5,so that she assumes the natural position of orb-weavers. I have even seen individuals with the back turned downward, fig. 8,as is the habit with the Triangle spider and with all those species who make a dome or horizontal orb-web, as the Basilica spider, EL. basilica McCook, and the Orchard spider, #. hortorum Hentz. If now the feet‘of the spider be carefully examined with a good glass, a coil of slack-line will be seen, precisely as in the case of the Triangle-spider. This is illustrated at fig. 8, where a, b, c, are \ the axes of several rays, grasped in the third (3) and fourth (4) 170 PROCEEDINGS OF THE ACADEMY OF [1881. pairs of legs, and Sl is the coil or slack-line between these and the fore pairs (1 and 2), or simply between the pair of fore-legs, 1 and 2. As the spider does not exceed one-eighth of an inch in body length, and the position of the snare is within cavities and inter- stices of rocks, where the light does not bring out the delicate tracery of the fine webs, the observation of these and other points of like character, is a matter of some difficulty. But, although the exact relations of the coil to the feet were sometimes in doubt, and indeed seemed to vary somewhat, the existence of the coil and its general relations were determined beyond doubt. It is also certain that the slack-line sharply uncoils and straightens when the spider releases her grasp upon the trap-line, and that the web unbends and shoots quickly forward. It is instantly changed from the bowed or conical form of figs. 4 and 5 to the circular plane of Fig. 8.—Ray spider (greatly en- fios. 1 and 2. larged) in position on taut snare. = © To show the slack-line coil, Sl. The following points, however, long evaded my observation, before webs were found which presented the conditions for successful study. But at last I was well satisfied, although I hope for further and fuller verification during the present summer. The “springing ” of the snare is caused by the sudden releasing of the trap-line from the fore-feet, instead of the hind-feet, as with the Triangle spider. The polarity of the two arachnids relative to their webs is reversed, Hyptiotes having her fore-feet, but Radiosa her hind- feet towards the web. The slack-line is therefore coiled between the two fore-feet or between the fore and hind-feet of Radiosa, but between the two hind pairs (as a rule) of Hyptiotes. A glance at fig. 6, b, will suggest the manner in which Hyptiotes is affected when her two hind feet are released from the trap-line. The coil, Sl, straightens, and the whole body of the spider shoots forward. If now we turn to Radiosa, as represented at fig. 8, or again, as shown somewhat better at fig. 9, we observe that if the fore-feet , 1, 2, fig. 9, are released suddenly from the trap-line, T, the whole body shoots backward, although still toward the snare, as with Hyptiotes. This was the action which I observed. The determination was finally accomplished by first carefully 1881. | NATURAL SCIENCES OF PHILADELPHIA. 171 sketching the arrangement of the basket stretched between the feet (2, 3,3, 4, 4, fig. 9). With this chart in one hand, and in the other hand a magnifying glass focused upon the feet, I watched until favored with several successive and unsuccessful springings of the net. As the spider only leaves her seat when she thinks that an insect is well entangled, and again bows her net by pulling on the trap-line if no prey be ensnared, the above conditions en- abled me to compare my chart of the basket, with the basket itself as seen under the glass. I found that the outlines on the paper and the lines under the animal’s feet exactly corresponded. There had therefore been no change in the relative positions of the hind- feet, mandibles and palps, perhaps also of the second pair (2) of feet. There had been an ac- tual (not seeming) motion of the body with and in the di- rection of the web, and this had been caused by releasing the first pair of legs (1) from the trap-line. The importance of this determination seems greater from the fact that I had at first concluded that the Radiosa actually operated her snare by sections. That is, instead Fria. 9.—Ray spiderin position showing slack of springing the whole orb at Poe ea eae Teak Ba cE Ghy ay yi once, as above described, she simply sprung the ray struck by an insect, by unclasping the foot holding the axis of that ray. Thus, ray ii, fig. 9, would be sprung by releasing the axis of ii, from No. 3, the third foot. This is probably not done when the snare is in complete form (as at figs. 1, 8, 4), but I now believe that it 7s done when the web has been partially destroyed, and is reduced to two rays or sectors as at fig. 11. This I hope to determine accurately during the current summer. If it should be verified we shall have another resemblance between the habits of Hyptiotes and Radiosa. III. GRADUAL OBLITERATION OF THE WEB. The fragmentary condition of Radiosa’s web after contact with insects has already been referred to. The snare is gradually 172 PROCEEDINGS OF THE ACADEMY OF [1881. obliterated, a conclusion to which the spider herself very curiously contributes. When an insect strikes the snare, as at fig. 10, ray I (broken ray), Radiosa first “keys”? the snare by twisting together the foot-basket and the parts adjoining (C), including the end of the trap-line This maintains the compact condition of the snare after the spider has left the central point at which she has held all parts together in the manner heretofore described. Then the insect is sought. Creeping along the axis of the ray upon which is the entanglement, she cuts away the cross-lines as she goes, leaving the bare skeleton of radii, as shown, fig. 10, I, broken ray. The insect is then brought back to a point (D) near the centre, but (in this case at least,) above it, where it is eaten. While the feast goes on, not unmindful of future supplies, the spider (S) clasps the ad- joining axis and (C D) the connecting lines, which appear to bein condition for operating somewhat in the usual way. When the insect is eaten, the former position is resumed, the trap-line clasped, and the net bowed and tightened. After a morning’s trapping, if the game has been plenty, and generally towards the Fie, 10.—Ray spider. Action when an insect middle of the afternoon, Ra- SAT aa a ee aC diosa’s snare will be found reduced to one or two rays or fragments of rays. I have seen it reduced to a bare skeleton. In fig. 11, there are one ray (I), and two fragments of two others united into a new ray, and these are placed in opposite parts of the orb. Again, one-half of the orb may be eliminated (fig. 12), leaving two radii (i, ii) to operate with. Xadiosa was also observed to construct or adopt a new trap- line, thus changing, so to speak, her base of operations. This action is illustrated at fig. 12, where Ta is the original, and Tb the new trap-line. This is not a frequent occurrence, as the necessity for changing the original line does not appear to arise frequently. 1881. ] NATURAL SCIENCES OF PHILADELPHIA. 178 IV. THe AFFINITIES OF THE ACTINIC ORB WITH OTHER ORB WEBS. Not the least interesting and valuable feature of the Ray- spider’s industry, is that it discovers a connecting link between two forms of snare which stand at the very opposite poles of the spinning-work of the (Orbitelariz) orb-weavers, At the one ex- treme is the familiar circular snare or full orb of the ordinary garden spider, as, for example, that of EHpeira domictiliorum, Hentz, fig. 13. At the other is the orb-sector of the Triangle spider, figs. 6,7. A glance at these will show how far they are apart in structure. The same separation appears in the habits of the two araneads.! As opposed to the Hyptiotes, the spiders of which Z. domiciliorum is a type hang head downward in the Fic. 12.—Ray spider. Half of , orb eliminated and a new trap- Fic. 11.—Ray spider’s snare line, Tb, formed. after usage in taking prey. centre of the orb, with their feet grasping small groups of the radii; or sit in a silken den, or crevice, holding to a taut trap-line which is connected with the centre. There is no slack coil, and no springing of the net as with the Triangle spider. The industry of Radiosa, it is now seen, is united to that of the Full Orb makers (Z. domiciliorum, et al.), on the one extreme, by the completeness of the circle ; but with that of Hyptiotes, on the other extreme, by the independent character of the rays, the nature of the trap-line, and the entire mode of operating the snare. The facts necessary to trace their affinities I have already given. ' T hope that I shall be tolerated in the invention of this general word for members of the Order Aranew ; ‘‘ Arachnid,’ the class term, is toe general ; ‘‘aranead’’ is needed for the true spiders. 174 PROCEEDINGS OF THE ACADEMY OF [ 1881. Some of the striking differences I have also recorded, and they may thus be summarized. The web of Hyptiotes is a single sector ; that of Radiosa has four or more, united. Hyptiotes commands one line with her feet, the trap-line and its continuation; Radiosa commands several axes, which are connected with, but not con- tinuous of, the trap-line. Hyptiotes has her head, Radiosa her abdomen towards her snare. Hyptiotes habitually hangs to the trap-line, back downward; Radiosa generally sits upon a foot- basket of lines, back upward. Hyptiotes shoots forward when her net is sprung; Radiosa shoots backward—but both spiders move toward their webs. Hyptiotes holds her slack coil between the two hind-feet (apparently); Radiosa between the fore-feet. In these differences, the points wherein Radiosa varies from AAU Z2== F1a@, 13.—Full-Orb snare of Epeira domiciilorum. Hyptiotes show a quite apparent approach to the behavior of £. domiciliorum and the Full-Orb makers. Thus the distance which heretofore had separated between the far-away extremes of the spinning-work of the Orbitelariz, has been completely bridged over by the industry of our little indigenous aranead—the Ray spider. It is to be remarked that while structurally the Triangle 1881.] NATURAL SCIENCES OF PHILADELPHIA. 175 spider isas widely removed from the Domicile spider, as economi- cally, the Ray spider is more closely allied structurally to the latter than the former. V. NATuRAL HABITAT AND ENVIRONMENT. The first specimens of Radiosa taken were hung in large open- ings left between the breastwork stones of a very old mill-dam. The wall had crumbled and quite fallen away in places, leaving large cavities, within whose moist, cool shelter, among ferns and mosses, this, with several species of spiders, had domiciled. The brook poured over the middle part of the wall, making a pretty waterfall; briers, bushes, ferns and various wood plants grew out of the wall and stretched over a deep pool 12 or 15 feet in diame- ter, into which the fall dropped. On the lower bushes and branches above the stream, and continually agitated by the splashing of the water, was a colony of Stilt spiders, Tetragnatha grallator, stretching their long legs along their round webs, and dancing with the motion of the waves; the beautiful nests of Phillyra riparia, Hentz, nests of Tegenaria persica, Lyniphia communis, L. neophyta, Epeira hortorum, and one or two species of Theridiords, were in close neighborhood. The whole pretty scene was embowered in a grove of young trees. A more charm- ing habitat could not well have been found. Another colony, not far away, was established within the cavi- ties formed underneath the roots of a large fallen tree, and beneath the ledges of some rocks over which the roots turned. In several similar positions were found the same nests, and also among the rocks in a wild ravine through which ran the stream Lownes’ Run.! Further explorations of the surrounding country showed that the spider was largely distributed, and in similar conditions. I found numbers in ravines, on the broad leaves of the skunk cab- bage, Symplocarpus (or Ichtodes) fotidus, the snares stretched over the brooklet, and beneath the shelving banks. They were also found among the rocks of Crum Creek over the beautiful drive to Howard Lewis’ mill. The habitat of the Ray spider may therefore be described as moist, cool, shaded cavities and recesses among rocks, roots, beneath banks and foliage, over or near run- ning water. 1 Since writing the above I have found Radiosa in similar environment at Mineral-spring Glen, New Lisbon, Ohio. 176 PROCEEDINGS OF THE ACADEMY OF [188l. JuLy 19. The President, Dr. RuscHENBERGER, in the chair. Twelve persons present. A paper entitled “ A Revision of the Cis-Mississippi Tertiary Pectens of the United States,’ by Angelo Heilprin, was pre- sented for publication. JULY 26. The President, Dr. RuscHENBERGER, in the chair, Twelve persons present. A New Form of Fresh-water Sponge.—A note was read from Mr. Epw. Ports, reporting the discovery in Chester Creek of another curious form of fresh-water sponge, a third species of Carterella, resembling C. tubisperma in the character of its birotulates and the length of its foraminal tubes, but much more robust than that species. The tendrils are nearly as long as those of C. tenosperma, but broad, flat and ribband-like. Thus far it is the most conspicuous and peculiar of our Ameri- ean forms. He proposed for it the name Carterella latitenta. John G. Graham was elected a member. The following was ordered to be printed : 1881.] NATURAL SCIENCES OF PHILADELPHIA. 177 REVISION OF THE PALEOCRINOIDEA. BY CHARLES WACHSMUTH AND FRANK SPRINGER. Part II. Famity SPH AZROIDOCRINIDA, INCLUDING THE SUB-FAMILIES PLATYCRINIDA, RHODOCRINIDA, anv ACTINO- CRINID&. The first part of this work was published contemporaneously with the “3d Lieferung” of Professor Zittel’s ‘‘ Handbuch der Palzeontologie,” which embraces the Crinoidea. In his classification, this distinguished author follows Johannes Muller, and divides the Crinoidea into three orders: Hucrinoidea (Brachiata, Muller), Cystoidea, and Blastoidea ; subdividing the first into the Tesselata, Articulata, and Costata. The“ Tesselata ”’ agree in general features with our Palzocrinoidea, and the Articu- lata with the mesozoic and recent Crinoids, for which we have proposed the name Stomatocrinoidea ; but while we treat these groups as of the same rank with the Blastoidea and Cystoidea, they are, according to Muller and Zittel, mere subdivisions of the “ Brachiata.” } Zittel divides the Tesselata into twenty-six families, among 1 While this was in press, we received from Dr. Etheridge, Jr., and P. Herb. Carpenter, an interesting paper upon the genus Allageerinus, a new form from the Carboniferous of Scotland, which they consider to be *‘tesselate’’ in the younger, ‘‘ articulate”’ in the adult state. In a dis- cussion upon Miiller’s terms, Tesselata and Articulata, they arrive at the conclusion, that at the present state of our knowledge of these Crinoids those names are inappropriate and should be abandoned. . They adopt our name Paleocrinoidea, but object to Stomatocrinoidea, as they think it possible, that also Crinoids of the other group might have possessed an external mouth. They consider the irregular arrangement of the plates in the calyx, against the almost perfect symmetry which is found throughout the other group, and the vault structure, to be better and more persistent characters for distinction than the condition of the mouth. We can only notice here this important paper, but shall take pleasure to refer to it at some future time. 13 178 PROCEEDINGS OF THE ACADEMY OF [1881. which his Ichthyocrinide and Taxocrinide substantially agree with our Ichthyocrinide, except that he included among the Taxo- crinide the genera Lecythocrinus and Gissocrinus, which we refer to the Cyathocrinide. Our Cyathocrinide include his Poteriocrinide, Heterocrinidz, Cyathocrinide and Hybocrinide, except that he places among the first of these families, Agassizo- crinus and Belemnocrinus, which we think belong to other families. Zittel’s classification, in its general results and conclusions, does not differ materially from our own; but instead of subdividing the Palocrinoidea at once into a number of small groups which he calls families, we separate them at first into comparatively few well-marked groups, which we subdivide when necessary. Our families are not based upon mere differences in the arrangement of the plates, but are expressions of important modifications in the structure of the animal, which must have affected the whole organism, and consequently form the basis of well defined natural divisions. The groups which we recognize as Ichthyocrinide, Cyatho- crinide and Spheroidocrinide existed at the beginning of the geological record, and flourished side by side until they became extinct. They are so well defined by nature, that once under- stood there is no difficulty in identifying them. The smaller groups into which we have divided the Spheroidocrinidz, are likewise of early origin, but they follow more or less the same general plan in the arrangement of their plates, as well as in their mode of development, individually and paleontologically. There have been several interesting publications on Crinoids during the past year, containing, among others, descriptions of new species of both Ichthyocrinide and Cyathocrinide. These species will be noticed and systematically arranged in an appendix at the end of this work. We are under special obligations to Prof. Spencer F. Baird for access to a number of rare and valuable books in the Smithsonian Library; to Dr. C. A. White, of the National Museum, for numerous favors received during the preparation of this work ; to Prof. A. H. Worthen, of Springfield, [l., and to a number of other gentlemen for the liberal loan of books and specimens, and for other valuable information. 1881.] NATURAL SCIENCES OF PHILADELPHIA. 179 I11.—SPH HROIDOCRINIDA. The above name is proposed as a family designation, to include such forms of the Palezocrinoidea, in which both calyx and vault are constructed of a large number of immovable plates and these forming inflexible walls ; with several orders of radials, and one or more of interradials on both the oral and aboral sides. The Spheeroidocrinide differ thus conspicuously from the Ichthyo- crinide with their flexible walls and squamous vault, and from the Cyathocrinide with their uniform elements of three rings of plates in the calyx, without interradials, and with simple oral plates in the vault. The family, as thus defined, will include genera with underbasals, and genera without them. In this we differ from most authors, who make the presence or absence of these plates a marked family distinction, and who place together within the same family Cya- thocrinide, Ichthyocrinide and Rhodocrinide. Pictét, Traité de Paléontologie, vol. iv, included in his “ Cya- thocriniens ” our Cyathocrinide, Rhodocrinide and. partly our Ichthyocrinide ; from the former, however, he excluded Graphio- erinus, which had been described by De Koninck and Lehon with a single circle of plates beneath the radials, and from the latter Forbesiocrinus and Taxocrinus, in which underbasals had not been discovered ; while he admitted the allied genera Ichthyocrinus, Lecanocrinus and Mespilocrinus in which they had been observed, Similar opinions were held by d’Orbigny, Hall, Miller, Austin and others, not including, however, Roemer and Schultze, who made the Rhodocrinide a distinct family. In the first part of this work we have discussed somewhat fully the relations of the underbasals, which we took to be the product of growth in geological times, introduced gradually by interpola- tion between the basals. It is very remarkable that, although the introduction of underbasals dates back to the Lower Silurian, as a rule, the genera in which those plates are found differ at no time materially from those in which they are wanting. Even as late as the Subcarboniferous, we find such species as Actinocrinus Whitei, and Rhodocrinus Wachsmuthi, both from the Burling- ton limestone, so strikingly similar in every respect, both in the structure of the body and arms, that the species cannot be 180 PROCEEDINGS OF THE ACADEMY OF [1881]. separated except by means of the basal portions. There are sim- ilar examples in the Silurian and Devonian, which will be noticed later. If, therefore, it be true that the underbasals had no important bearing or influence upon the general structure of the Crinoids, there is no good reason for making their presence or absence a family character, and basing thereon a division to rank with the Cyathocrinide and Ichthyocrinide, whose fundamental structural plans offer broad and unmistakable distinctions. Nevertheless, they are not without importance in classification, and certainly characterize a group of more than generic value. We have accord- ingly brought together the genera of this family in which these plates exist under the name “ Rhodocrinide.”’ The great family Spheroidocrinide includes a vast variety of forms, and a mere separation of these into genera does not meet the requirements of a systematic classification. We find that the genera fall naturally into groups which are well defined. These subordinate groups, which are three in number, we consider to be sub-families, and have arranged them in the following manner : 1. PLatycrinip£.—Underbasals wanting; basals and first radials forming the greater part of the calyx ; succeeding primary radials very small or rudimentary; all higher orders of radials embraced within free rays; mterradial system but little developed. 2. ACTINOCRINIDZ. — Underbasal wanting; calyx -composed of basals; two or more orders of radials: well developed interradial, and often interaxillary series. 3. Ruopocrinip&.— Underbasals present; calyx composed of basals and several orders of radials; interradial system well developed. Before proceeding to the more detailed consideration of these groups, we will consider the different parts of which the body in the Sphzeroidocrinide is constructed, and this will throw additional light upon the relations of the subdivisions. 1881. ] NATURAL SCIENCES OF PHILADELPHIA. 181 1.. UNDERBASALS AND BASALS. The genus Glyptocrinus Hall, from the Lower Silurian is one of the earliest, most beautiful, and most instructive types of the Paleocrinoidea, and fortunately is often found in excellent preser- vation. Looking at the great number of plates which compose its body, at its elaborate ornamentation, one is naturally inclined to consider this as one of the most maturely developed forms in the whole family, but in other respects, it possesses in a marked degree the characters of the young crinoid of fater geologic times. Glyptocrinus was originally described with five basals and no underbasals. Hall afterwards discovered in Gl. decadactylus small pieces concealed within the basal cavity, so rudimentary, however, that both he and Meek hesitated to call them basals, although both authors apply that term to the proximal plates in all other cases. Meek distinguished them as ‘‘ subbasals.”” We have examined the plates in question very carefully in the species named, and find them, although very rudimentary, placed within the basal ring, hence they are, according to our terminology, true underbasals, and not as Hall describes them a “ quinquepartite ” upper joint of the column.! In some other species of this genus the underbasals seem to be altogether wanting, at least are not developed externally. In Glyptocrinus Dyeri no trace of them can be discovered, though we have examined with reference to this point, some most perfect specimens. If the underbasals were elements of family import- ance, Gl. decadactylus and Gl. Dyeri would be representatives of distinct families. Glyptocrinus is exclusively a Lower Silurian genus. The two species from the Upper Silurian, referred to it by Hall, have been transferred by us to other genera. One of these, Mariacrinus Carleyt, is another interesting case illustrating our view that the underbasals have no important effect upon the general structure of the body. WM. Carleyi would be an excellent Glyptocrinus were it not that the calyx below the radials is composed of a single 1 The underbasals cannot be developed from a columnar joint, or their sutures would correspond with the sutures of the column; whenever this is divided, the division occurs alternately with that of the underbasals, and as a rule alternately with the proximal ring of plates. 182 PROCEEDINGS OF THE ACADEMY OF (1881, ring, and this of only four pieces. Periechocrinus, like Mari- acrinus an Upper Silurian genus and like it without underbasals, has instead of four or five basals, only three. In all other respects these genera agree so perfectly with Glyptocrinus that they cannot be distinguished, proving again how closely Rhodo- crinide and Actinocrinidee are linked together, and that they are in fact variations of one great group. Where underbasals are unrepresented, families have frequently been created upon the number of the basals, and Angelin based his entire classification upon the number of proximal plates, whether basals or underbasals. Convenient as this scheme of classification may seem, it is altogether artificial, and combines forms which are widely different, while it separates others which are clearly allied. We have in the introduction to this work, page 17, dwelt at some length upon the basals or first ring of plates below the radials ; and believe we have shown that the basal disk, whether composed of one, two, three, four or five pieces, can almost invariably be reduced to five elementary pieces, and that all deviations from this number have been produced by anchylosis of two or more of the original segments. This, of itself, is a strong argument against a classification based upon the number of these plates. Among the Actinocrinide, only a few genera with the original five basal plates are known, and these are confined to the Silurian ; indeed we have good reason to believe that only the very earliest representatives of this group possess a base divided into five pieces. Genera with four basals commence in the Silurian and terminate in the Devonian; while genera with three basals are found from the Upper Silurian to the close of the Warsaw lime- stone where the family becomes extinct. The genera with four basals have been referred by us partly to the Actinocrinide, and partly to the Calyptocrinide. The latter family has four basals throughout, but even here this number cannot be considered a family character, since Melocrinus and Mariacrinus, which have four basals, belong undoubtedly to the Actinocrinide. Species with three basals are found among both Actinocrinide and Platycrinide, and the latter are by no means restricted to this number, as Dichocrinus, which has been by most systematists placed in the same group with Platycrinus and Hexacrinus, has but two basals, 1881.] NATURAL SCIENCES OF PHILADELPHIA. 183 It is apparent from these facts that neither the existence of underbasals, nor the modifications which took place in the basal disk, had any such corresponding effect upon the general structure of the crinoids as to entitle them to be considered characters of family importance, though in distinguishing subordinate groups they may possess some value. The radial and interradial plates are elements of far greater value. 2. RADIAL PLATEs. In our nomenclature we have proposed different terms for special parts of the ray, discriminating between radials, brachials and arm plates. We designate as “ radials” the whole succes- sion of plates above the basals radially situated and enclosed within the body walls. The “arm plates”? form the movable portion of the ray; the “ brachials,” while radials in position, are arm plates in construction, being free and distinctly articu- lated. The term “ brachials” is therefore purely a conventional one, employed for greater convenience in description. We desig- nate as “primary radials” those below the first bifurcation, while the “ secondary radials” compose the first branches of each ray, the so-called “ Distichalia’”’ of Muller and other writers; and those of succeeding bifurcations are distinguished by referring them to the order in which they stand in succession. Miiller, in applying the term “ distichalia,” was evidently not aware, that there are many fossil Crinoids in which these plates give off branches which likewise form a part of the test, otherwise he would not have regarded the arms as commencing at the “ axil- lary distichals.” Schultze, in his Monograph Echinod. Hifl. Kalk, p. 5, improved upon Muller’s views, and asserted that ‘‘the commencement of the arms begins invariably with the first distinct articulation of the ray.” Like Muller, he uses the term “distichalia”’ for the secondary radials, and proposes no name to designate the plates of the higher branches within the body. Schultze differs from Miller in designating the free radials as arm plates, and in this he agrees with us; except that we distinguish those arm plates which are in direct vertical line with the radials as “ brachials.” De Koninck and Lehon regarded the arms as beginning from the first bifurcation in the ray, no matter whether they became 184 PROCEEDINGS OF THE ACADEMY OF [1881. free at this point, or remained included within the body walls for some distance. They, however, characterized those plates which are immovably united with the calyx—the ‘“‘distichalia” of Muller—as “pieces brachiales,” and the movable joints as “articles brachiaux.” P. Herbert Carpenter, in his late work on the Actinometra, p. 22, states that the views of De Koninck upon the relation of these plates were unquestionably correct in the case of the “Articulate Crinoids,” but that their application to the ‘fossil Tesselata ” was beset with some difficulties. We do not exactly coincide with any of these views. It seems to us that either the entire radial series of plates within the calyx, eventually up to the sixth division, or even higher, must be called radials, or this term must be restricted to the first radial plate. The first primary radials are the only plates, besides the basals, which form a part of the calyx in all Crinoids, and which can be homologized with the apical plates of other Echinoderms. All succeeding plates in the series are, in our own opinion, originally arm plates, which by growth during the life of the individual— chiefly, no doubt, in the embryo—and by development in geologi- cal time, were enclosed within the walls, and became thus modified into radials, the change being produced by growth and the devel- opment of additional interradials and interaxillary pieces. That this was the case in the higher orders of radials can be clearly demonstrated, and we feel confident, from analogy, that the same rule extends to the plates throughout the ray, which in turn suggests the idea that the arms fundamentally commence with the second radials, and not with the axillary plate as intimated by Carpenter, nor with the distichalia of Muller. In practice, how- ever, and in this;we agree with Carpenter, it is more convenient to regard the arms as commencing with the first free plate beyond the calyx. The radials, as we designate them, consist throughout the Sphe- roidocrinide, of five rows of plates, of two to three each, longi- tudinally arranged. The upper one bifurcates and supports upon its upper sloping sides two rows of one or more smaller plates— the secondary radials—which in turn, either support the arms directly, or divide again, and are followed by radials of the third order. In species of the latter kind, the upper plate in one or both divisions is axillary and supports on each sloping side another 1881. | NATURAL SCIENCES OF PHILADELPHIA. 185 row of radials. The formation of higher orders of radials takes place in a similar manner, only one branch generally bifurcating, and one remaining single. In species with but two orders of radials the number of primary arms is limited to two in each ray, which, however, sometimes branch after they become free. Species with tertiary radials may have either three or four arms to the ray, depending upon whether one or both divisions bifurcate. Upon the same principle species with four orders of radials may have five to eight arms, and those with a fifth order may have from eight to twelve and even sixteen. The first number can only occur when all the branches remain simple; the latter when they all divide. Hence the number of primary arms is dependent upon the number of orders of radials represented in the species, and whether part of them bifureate again or remain simple, It is important to note that simple arms are always given off from opposite sides alternately. The number of arms is most frequently ealy of specific import- ance, but in cases where certain rays throughout a number of species are distinguished by a smaller or greater number, the arm formula may become almost a generic character, as for instance in Batocrinus, where the posterior rays are generally more developed. In Hretmocrinus the anterior and two posterior rays are less developed than the antero-lateral rays. In some genera the arms appear to be limited to a certain number, in others they vary. In the typical Actinocrinus there are species with four, five, six, seven and eight arms to the ray, but the normal number is uniform in the different rays. In all these cases, how- ever, as well as not unfrequently within the limits of species, there are exceptions. 3. INTERRADIALS AND ANALS. The interradial and anal plates occupy the intermediate spaces between the five radial divisions or rays in the body, and their number and shape depends altogether upon the number, position and proportions of the radial plates. They vary from a single plate to twenty or more, but are represented even in the young crinoid by at least one plate. In species in which the secondary radials of adjacent rays abut laterally, the number of interradials is naturally small, but when higher orders of radials are present, and especially when the rays are widely separated, the number is 186 PROCEEDINGS OF THE ACADEMY OF [1881]. comparatively much larger. When the second and third radials are short, the number of interradials decreases, and it increases when they are long and narrow. This increase in number takes place by interpolation, contrary to the radials in which the increase is from the distal ends of the rays. The interradials are designed to fill up vacant spaces in the test, and this accounts to some extent for the great diversity which is found in their number within the limits of a genus, and within species at different stages of maturity. The first interradial is always larger than any of the rest, and is situated between the upper sloping margins of the adjoining first radials, except in some species of the Rhodocrinide in which it rests directly upon the basals, separating the ring completely. There are generally two plates in the second series, and two or three in each succeeding one, but in forms where the secondary or tertiary radials form an arch over the interradial spaces, as in Batocrinus, there is often only a single plate in the second or third series. The plates decrease in size upwards, those of the uppermost row being sometimes barely visible to the eye. The posterior or anal area is readily recognized in most of the Spheroidocrinidee by its greater width, and by having a larger number, and a somewhat different arrangement of plates. In most of the genera the first anal plate is in line with the first radials, resting upon the basals. In our remarks upon the Cyathocrinidz we noted the fact that in that family the anal plates, with a few exceptions, are directed toward the right side. In the Sphxroidocrinide, on the contrary, the symmetry is always bi- lateral, in some cases almost perfectly pentahedral, and a vertical section through the median line of the anal area, the anal aper- ture, the central dome plate, and along the anterior ray, divides the body invariably into two equal parts, and this symmetry, which extends to the arrangement of the arms, is one of the most char- acteristic features of the family. The term “anal plates,” as now used, designates the entire series of plates which compose the posterior interradius. Properly speaking, however, this is not quite correct. Careful examina- tion shows clearly that the majority of these plates are in the true sense of the word “ interradials,” while only a few of them are actually ‘‘anals,” by which we understand plates supporting the anus, or that can be accounted for as being in any way, directly 1881. } NATURAL SCIENCES OF PHILADELPHIA. 187 or indirectly, connected with that organ. In the earlier genera, Glyptocrinus, Reteocrinus, Glyptaster, Archxocrinus, Eucrinus, and in all genera up to the Subcarboniferous in which the anal opening is lateral, the posterior area is distinctly divided into two equal parts by a single median row of plates. This row, which often extends to the anal opening, is composed of the true anal plates, but the plates on either side of it are interradials. By considering the latter, without regard to the median row, it will be found, that the two sections taken together correspond exactly in number and general arrangement with the interradials of the other areas, or at least differ not more than the other four differ among themselves. But it must be observed that in genera in which the first anal plate rests directly upon the basals (Glyptaster, Eucrinus, Dorycrinus, etc.), the first true interradial in the pos- terior area is divided, and is represented by two smaller plates, separated by a special anal plate (Pl. XIX, fig. 2). In these genera the anals proper extend from the basals to the anal aperture. Glyptocrinus and its congeners, in which there is no anal piece represented between the first radials and in which the first interra- dial range, consists of a single plate in all five spaces, the special anal plates begin with the second range in which there are three plates. In later geological times, when the anal opening became more central, the special anal plates decreased to two or three, and in the typical Actinocrinide in which we include the genera Actinocrinus, Amphoracrinus, Strotocrinus, Physetocrinus and Steganocrinus, they are reduced to a single plate. Some of the Platycrinide have no special anal piece, end the posterior side differs merely by having a somewhat larger inter- radial, others however, as Dichocrinus and Hewxacrinus, have a very large special plate. In some of the Rhodocrinide, like the Calyptocrinide, the symmetry of the calyx is almost regularly pentahedral, and none: of the plates of the posterior side are actually anals. 4. INTERAXILLARY PLATES. The space within the axil of the secondary radials is frequently filled by plates, for which we proposed in the first part of this work the name “ axillary plates.” This designation is undoutedly appropriate, but finding that it had been previously used by several authors for the bifurcating plates, we have thought best, in order 188 PROCEEDINGS OF THE ACADEMY OF | 1881.. to avoid confusion, to use the term “interaxillary ” for these plates. In the Platycrinidee, which have rarely more than a single order of radials within the calyx, interaxillary plates are not represented. In the Actinocrinidz and Rhodocrinide they may be present or absent in the same species, and sometimes in different rays of the same specimen, their number, like that of the interradials, increas- ing with age. Roemer and Joh. Miiller considered the presence of interaxil- laries as of generic value, and the latter proposed a division of the genus Actinocrinus, placing all species having those plates under Pyxidocrinus; but it is evident that such a division is alto- gether artificial and not warranted by the facts, and if carried out would produce confusion. 5. VAULT. One of the writers, in a paper upon ‘the internal and external structure of Paleozoic Crinoids ” (Am. Jour. Sci., Sept. 1877), discussed the importance of the vault with reference to classifica- tion. It was noted that in a large number of genera, among them Actinocrinus, Rhodocrinus and Platycrinus, and their allies, the ventral covering is composed of strong plates closely cemented together, and that these form a free arch which braces the entire oral side of the body without the aid of oral plates. This is the general character of the vault in the family under consideration. The vault in the Sphxroidocrinide is usually well preserved, owing to its solid structure, and is capable of accurate definition. Its plates vary from a few to many hundred; but, notwith- standing this diversity in number, their arrangement is governed by definite rules. Certain of these plates, which we have termed the “ apical dome plates,’ are represented in every species of this group. They consist of a central piece, occupying a position directly above the oral centre, which in this family is quite uniformly the centre of the disk. It is surrounded by six prox- imal plates, interradial in position, of which four are large, and equal, and two smaller. The four large plates are placed above the four regular interradial spaces respectively ; the two smaller plates which are equivalent to and take the place of one large plate, are directed posteriorly, being separated from each other by anal plates or the proboscis. These seven plates are easily recognized in species with comparatively few summit pieces and a lateral 1881. ] NATURAL SCIENCES OF PHILADELPHIA. 189 anal aperture (Pl. XVIII, figs. 7 and 9), but their identification is often difficult in forms in which a large subcentral anal tube is interposed between the two small plates, pushing them toward the anterior side, while the central piece rests against the side of the anal tube. (PI. XVIII, fig. 8). There are other vault pieces occupying a radial position which are 2ither in contact with those just described, or, as is more frequently the case, separated from them by a belt of small pieces. Their number varies considerably among species, and depends upon the number of primary arms, without reference to the number of bifur- cations after they become free. They increase in proportion to the number of primary arms, in the same manner and on the same principle as the plates of the calyx, each order of radials has its corresponding plates in the vault.’ Therefore, in adult specimens, with some practice the number of arms can be ascertained as well from the dome as from the calyx. In species with two arms to the ray, there are two ranges of corresponding radial plates in the dome ; the first or upper being a large bifurcating plate, equiva- lent to the primary radials of the calyx. This is followed by two other plates, which take the place of the secondary radials, one over each arm base, with athird plate—an interbrachial—between them. When there are three arms to the ray, there are three ranges of radial dome plates, two plates in direct succession from the large bifurcating plate toward the single arm, and a second bifur- cation, with one plate in each branch, toward the division with two arms. In species with four arms to the ray, there are two secondary bifurcations, producing radial dome plates of a third order, leading to each arm base,and so on. There are also interradial plates represented in the summit, occupying intermediate spaces between the radials, but their arrangement is very irregular and their number variable. In some genera the number of vault pieces is enormous, notably in Strotocrinus, which has a large number of arms. Looking at such a specimen with its vast number of apparently irregular vault pieces, one would scarcely expect to find this multitude of plates arranged upon a definite plan; and this the same that prevails in the calyx. We have called the principal plates in the vault apical dome plates, because they correspond to the apical plates of the aboral side. The six proximal plates surrounding the central piece repre- 190 PROCEEDINGS OF THE ACADEMY OF [1881. sent the basals or genitals, and the radial dome plates, the radials or oculars. The centre piece may perhaps be compared with the underbasals, or the subanal plate of the Echini. The apical dome plates, as the apical plates of the calyx, are fundamental elements, and are represented in the vault of all Spheeroidocrinide both in the young and the adult, from the Lower Silurian to the Subcarboniferous. They are generally larger than the other dome plates, and more prominent, frequently nodose or spiniferous, though in some species they cannot, at least in mature specimens, be readily distinguished from the other dome-plates which have attained equal size. In some genera, for instance, Culicocrinus, Muller, they occupy almost the entire ventral disk; in Glyptocrinus and Rhodocrinus, on the contrary, they fill only the median part. In some species of Vorycrinus the central piece is spiniferous and the radials nodose ; in others, all these plates are spiniferous. In Amphoracrinus only the four larger proximal dome plates are nodose or spiniferous ; in Agari- cocrinus all apical plates are tuberculose; in Batocrinus the entire dome is composed of nodose plates. The proximal plates in some species are attached to the centre plate, in others separated from it by a ring of small accessory pieces, and in still others the centre piece is entirely isolated by the wide belt of minute pieces. The latter is frequently the case in large specimens, and in genera with but few primary arms, like Megistocrinus. In this genus it is inter- esting to find in very young specimens and in the smaller species the central and proximal plates in contact, while in the larger and adult specimens all are isolated, even the proximal plates being separated from each other. The radial dome plates are sometimes attached to the other apical plates, frequently so in young speci- mens, and generally in the Platycrinide. In Platycrinus the radial series of the dome is s composed of two rows of pieces alternately arranged, which decrease in size toward the arm bases, and of which the first and larger plate fits in the angle of two adjacent proximal plates. The vault of the Platycrinidz differs in several particulars from that of the other Sphzroidocrinide, and in these same char- acters it approaches the Cyathocrinide. We elsewhere suggest that the Platycrinoid is the simplest form of the Spheroido- crinide, and that it represents the younger stage of the family. This is indicated by the construction of the calyx, but not less by 1881. ] NATURAL SCIENCES OF PHILADELPHIA. 191 the structure of the vault. In the genus Coccocrinus Muller, one of its earliest forms, the vault is composed of five large oral plates, resting upon the upper truncate side of a single interradial, and, as found in the fossil, it has a central oral opening and lateral grooves for the ambulacral furrows. Zittel has already noted the close resemblance of the above structure with the recent genus Hyocrinus Wyville Thomson, and the larval state of Comatula, calling it very appropriately “ einembryonales Stadium von Comatula in persistenter Form.” The similiarity to Hyo- erinus is probably merely superficial, as the lateral grooves in Coccocrinus were evidently closed by additional plates as in other Platycrinide, while they are open in Hyocrinus. The oral plates of Coccocrinus have been, by several authors, confounded with an apparently similar superstructure in Symbathoerinus, Triacrinus and other forms, but there is really no analogy between the two structures. The parts which enclose the opening at the oral centre in the latter forms are radial in position, and therefore not oral plates, but merely extended articulating facets of the radials. In those genera, the central space, like the opening at the centre of the oral plates, is also closed in perfect specimens by apical dome plates, which rest directly upon the extended pro- cesses. This group will be introduced hereafter as a separate family under the name Symbathocrinidz. The ventral disk in Coccocrinus bears a close resemblance to that of Cyathocrinus, but while the former has an additional inter- radial interposed between its radial and oral plates, in Cyatho- crinus the intermediate plate is absent, and the oral plate rests against the incurved upper margins of the radials. In Platycrinus and similar genera, the two series of alternate plates which, as mentioned before, cover the radial regions of the dome, are inter- posed between three and sometimes five interradial plates, which in Coccocrinus as oral and interradial plates occupy the same position. This suggests the question whether these plates in the Platycrinidz, and the interradial dome plates in the Spheroido- crinide generally, are not the homologues of the oral plates, which are here broken up, and represented by several plates instead of one. This interpretation seems to us the more probable because Coccocrinus is one of the earliest known forms of the Platycrinide, and may be considered an embryonic type of the family. ~The homology in Platyerinus, however, extends only to the 192 PROCEEDINGS OF THE ACADEMY OF [1881]. second or upper row of interradials, the first interradial, which exceptionally in this group is placed almost within the dome regions, is identical with the outer interradial plate of Coccocrinus, and as such forms part of the apical and not of the oral system. This view differs somewhat from that expressed by us in Part I, p. 138, where we stated that the oral plate of the Cyathocrinide had “no representative in the vault of the Actinocrinide,atleast not externally ;’’ we were not at that time acquainted, with the genus Coccocrinus, which has given us new light upon the subject. We have thought heretofore that perhaps the triangular porous structures arranged around the inner test of many Actinocri- nid, might be the homologues of the oral plates. The vault throughout the Spheroidocrinide is perforated with a single opening, which in all of them is more or less excentric ; in some lateral and placed toward the periphery of the disk; in others sub-central leaning toward the posterior side of the body. The construction of the parts at the inner surface of the dome proves that the opening communicated with the posterior side of the visceral cavity, not with the digestive organs. It is separated from the ambulacral and oral systems by a strong partition attached to the inner surface of the vault, and hence the opening represents the anus, and is not the oral aperture, as has been supposed by the earlier writers. The anus is either in the form of a simple opening through the vault, or is prolonged into a tube, which in Batocrinus sometimes attains a length of three times the height of the body including the arms. The tube is in all cases composed of heavy, generally nodose, wedge-form pieces, which are firmly put together, giving but little flexibility to the structure. It has no openings or pores through its plates or at the sutures, but has in the centre a comparatively small passage, with a minute outlet at the extremity. In cases where the anus is not extended into a tube, the aperture is generally situated within the centre of a wart-like inflation com- posed of very small pieces. It is possible that in such cases the small inner plates formed a little pliable tube, which could be drawn in by the animal like the anus in recents Crinoids,! but a 1 Tn the genus Codonites of the Blastoidea, we find in connection with the anal opening a similar little tube, which we found in one specimen extended outward, while in another, traces of its little plates are left within the opening. ' 1881. } NATURAL SCIENCES OF PHILADELPHIA. 193 contraction of the long solid tube of Actinocrinus, as has been suggested by Austin, is wholly impossible. There has been considerable difference of opinion as to whether species with a solid anal tube should be separated from those with a simple opeaing. Considering the slight distinctions upon which many of the genera have been founded, it would seem that the tubular structure ought to be of suflicient importance to justify a generic separation; but when we consider that various generic groups, after being carefully restricted with reference to all other characters, include both forms, its value as a full generic character must be somewhat doubtful. We once supposed that it might be a sexual difference, but the specific relations of the forms thus distinguished do not sustain that supposition. Both forms are not found in all the groups, though they exist in many, and throughout all divisions of this family. In some eases generic separations have been made upon this character, as for instance Physetocrinus has been divided from Actinocrinus, Alloprosatlo- erinus from Agaricocrinus, etc., while in other cases as Platy- erinus, Glyptocrinus and Strotocrinus both forms have been retained in the same genus. It must also be observed, in this connection, that in some cases, especially species with a very slender proboscis like Batocrinus retundus, we find oecasionally specimens in which the tube seems to have been accidentally broken away during the life of the animal, and in which the fractured edges of its base had become absorbed and rounded, giving it the appearance of a naturally simple opening. That the simple opening could have been produeed in a like manner in Strotocrinus and other genera, no one would for a moment suppose after examining good specimens. The fact that the crinoid lived on without the tube, at least proves that this structure had no important influence upon the general organization of these animals. A tube is more frequently found in genera in which the arms are arranged in a continuous series around the body, while in species with a simple lateral opening the arms are arranged more or less in clusters, leaving wide spaces between the rays. In the former case, the long tube could discharge the excrements free from the arms, and in the latter it was not needed as the refuse matter could be easily discharged between the bases of the arm clusters. ‘In view of these facts, we think a subgeneric division 14 194 PROCEEDINGS OF THE ACADEMY OF [1881. sufficient to mark the two structures, but this should be done uniformly and we shall accordingly propose subgenera where necessary for this purpose. The vault does not completely cover the calyx, but leaves along the line of junction a row of oval or circular passages which have been called arms or “ ambulacral openings.” The belt in which they occur is known as the “arm regions,” and their distribution in the different rays is expressed by the “ arm formula.” Through the arm openings which are very conspicuous in the Spheroido- crinidex, food entered the body, and they served as passages for the ambulacral vessels. In a mature specimen the number of primary arms can be ascertained by counting the arm openings ; but not always in young specimens, or in species in which the radial portions are extended into free rays. By “ free rays” we mean lateral extensions of the body, com- posed of a succession of radials, unconnected by interradials, and covered with similar plates, as solidly and in the same manner as the radial portions of the dome proper. These free rays, whether composed of only a few plates as in Platycrinus, or extended almost to the full length of the arms as in Hucladocrinus and Steganocrinus, are actually portions of the body, and the arms are given off from them in the same manner as from the body in other cases. | In the Platycrinide generally, there are within the calyx primary radials only, all the higher orders of radials being included within the free vays. In the body of a Platycrinus in its ordinary preser- vation, we find but five arm openings, but whenever the bifurcating plate is well preserved at its distal end, ten openings are visible, and these form passages for the free rays which here divide, each branch giving off arms laterally and from opposite sides. The free rays are rarely preserved in the fossil unless the arms are attached, when they really appear like arms and have been described as such. That they are not arms is proved by the fact that their ventral side is not provided with a furrow, but is covered in the same way as the vault proper. In these forms, as might be expected, the number of arms cannot be determined from 1 This proves that P. Herbert Carpenter is correct in saying that the division of the arms actually begins at the middle of the bifurcating plates, (on Actinometra, p. 22). i 1881. ] NATURAL SCIENCES OF PHILADELPHIA. 195 the arm openings, unless. the full length of the ray is preserved ; any fracture of the ray, whether cutting off one arm or a dozen, shows in the specimen only a single opening. The Platycrinoid with its simple form, is similar in structure to the young Actinocrinoid, in which some portions of the ray are yet in the condition of free appendages. The young Actinocrinoid, at this stage, has the same number of arms as the adult; but in species with numerous arms, the upper divisions of the ray, which in the mature animal are incorporated in the body walls by means of interradial and interaxillary plates, form free appendages in the younger stage, and consequently the number of arms is comparatively less than in full grown individuals. In Stroto- crinus, which has the greatest number of arms, we find in very young specimens only four arm openings to the ray ; approaching maturity there are eight, afterwards twelve, and in fully matured individuals there are perhaps sixteen or more. Specimens in different stages of growth, have frequently been described as distinct species on account of variation in the number of arm openings ; and this has even been extended to genera. As growth progressed, the upper branches gradually lost their free character by being absorbed into the body walls through the interpolation of interradials and interaxillaries, both in the calyx and in the vault. The plates which covered the ventral side of the free appendages were thus drawn into the vault, and became. at length a part of the main body; but the Platycrinide, which have no interaxillaries nor increase of interradials, retain the free append- ages during life. As a general rule it may be asserted, that in the Sphzeroido- crinidz the plates of the vault increase in number outward ina similar manner to the plates of the calyx. The various plates of which the body is composed might be separated into two classes : 1. Plates which either do not increase in number, or do so only at the distal ends, and not by interpolation, including the basals, radials and arm plates. 2. Plates which increase by interpolation only, including the underbasals, interradials, anals and interaxillaries, to which we might add the joints of the column. 196 PROCEEDINGS OF THE ACADEMY OF (1881. 6. ARMS AND PINNULES. The arms in the Spheroidocrinide are either simple or branch- ing, and are constructed either of a single or double series of joints. Single arm joints are restricted to the Silurian, all Devonian and Subcarboniferous genera have two rows of alter- nately arranged joints in the arms. In the Cyathocrinide, almost to the close of the Subcarboniferous, the arms are composed of single joints, but in the Kaskaskia Limestone and Coal Measures a few species occur with double-jointed arms, intermingled, how- ever, with species apparently of the same genus, in which the arms are constructed of single wedge-form joints alternately arranged. These two structures run so closely into each other through transition forms, that we have been compelled to arrange them in some cases under the same genus, although we have in other cases considered the arm structure to be of generic import- ance. For tiiis we have been criticised by Prof. Wetherby, of Cincinnati, who thinks it ‘‘a singular statement, that a character in forms of the same geological age may be generic in one case, and only specific in another.’’ He evidently overlooks the fact that all Crinoids in their young stage have single-jointed arms, and that the double-jointed feature is a product of growth which primarily was only an individual variation, but which, by becom- ing fixed and constant in certain forms, attained generic value, especially when taken in connection with other characters. The best specific and generic and even family characters, originated in individual variation, and at some period in the paleontological history of the organism were without value as a means of classi- fication. Near the close of the existence of the family Cyatho- crinide, the double-jointed arm structure began to be developed, in some forms irregularly, in others to such a degree as to be constant, and to form a distinguishing characteristic of many species, which thus fell naturally into a generic group. This process is illustrated in Hupachycrinus, Hydreionocrinus and Frisocrinus, and there are similar examples among the Spheroido- crinide. The young Platycrinus ‘and Actinocrinus have a single series of cuneiform arm joints, which are constructed exactly like those of the adult Poteriocrinus, and the pinnules are given off in a like manner. At a more advanced stage the joints begin to enlarge laterally in such a manner that the sharp inner angles 1881. ] NATURAL SCIENCES OF PHILADELPHIA. 197 interlock. This process commences at the extremities of the arms, and gradually involves the lower portions down to the bases. During this stage we often find the lower arm joints quadrangular, with parallel sutures, followed by wedge-form and cuneiform pieces, and finally the tips*constructed of a double series of plates. The same development, which thus took place during the life of the individual, is observed to go on in geological times, but not contemporaneously in different families. In the Actinocrinide and Platycrinide it became complete in the Silurian, and is found invariably in all succeeding forms. In the Cyathocrinide, that structure appeared only at the close of the Subcarboniferous, shortly before the family became extinct. In this group, the arm pieces attained that marked wedge-form which everywhere pre- ceded the double joints in the Burlington limestone, and here in some species of Poteriocrinus and Celiocrinus the plates began to interlock already at the tips of the arms. This became more frequent and more conspicuous in the Kaskaskia group, where in some few cases, it extended to the entire arm. The different stages of individual growth, as they became gradu- ally introduced paleontologically and fixed, undoubtedly form excellent generic characters, but we must not forget that there was a time in the life of the crinoid at which the arms were neither single- nor double-jointed, but at which the joints began to interlock, and when probably a very few days brought about important changes in the arms of the growing animal. This stage is represented paleontologically among the Cyathocrinide by Eupachycrinus, Erisocrinus and Hydreionocrinus, and in this view of the case it is not difficult to understand how this arm structure may be of generic importance as a rule, but scarcely of specific value in exceptional cases. It has been stated that the double-joint structure was introduced in the Spheeroidocrinide in the Silurian, and this occurred under exactly the same conditions as it did later on in the Cyathocri- nid. By far the greater number of species in the Lower Silurian have single arm joints, and these, with a few exceptions, consist of quadrangular pieces with parallel sutures. In the Upper Silurian we find a few forms with single joints, and along with them arms with cuneiform joints—either interlocking or not—associated in the same strata with species having double series of arm plates, and we find all intermediate gradations between the two extremes. 198 PROCEEDINGS OF THE ACADEMY OF [1881. In some of these species the two structures are found combined, in others so closely associated that it is next to impossible to separate them; and in still others the extremes are separately represented. Generic divisions based upon these variations, unless accompanied by other distinctive characters, seem to be unneces- sary and inexpedient. It would require the creation of a large number of new genera, which would probably have to be sub- divided to accommodate other transitional forms, and so on with- out end. The double series of joints resulted from the increasing width and outward growth of the arm ; hence arm joints which originally were simple and cuneate did not in the mature animal extend through the full width of the arm, but gradually interlocked by their sharp angles, so that joints on each side, which at first were separated by another joint, came by degrees to be partly in con- tact and to rest upon each other. Therefore,in double-jointed arms every joint at each side bears a pinnule, while in those with single joints the pinnules are found only on alternate sides. The arms in the Actinocrinide and Platycrinide divide rarely after they become free, the branching as a rule taking place in the body or in the free rays, Megistocrinus, Amphoracrinus, and Periechocrinus form the only exceptions. In the Rhodocrinide, on the contrary, the arms branch as a rule beyond the calyx. The ventral grooves in the arms of this family are less deep, but comparatively wider than those of the Cyathocrinide. They are bordered on each side by a row of long pinnules, which cover them perfectly. Whether the grooves were closed by marginal plates seems to us doubtful, although Prof. Wetherby states that he has observed such plates in Glyptocrinus, they probably occur below the bifurcation where the arms should be regarded as free rays, or perhaps they are restricted to oral arms, such as P. Herbert Carpenter describes in Actinometra. The pinnules throughout this family are long, closely crowded together laterally, and the two rows with which each arm is pro- vided fit together so neatly, and cover the arm furrow so perfectly, that additional plates were scarcely needed. Each pinnule is com- posed of anumber of joints, which differ in form in different genera. In some they are of equal width and height, outwardly convex; in others higher than wide, with the outer surface flat; in some they are entirely smooth, and in others provided with a peculiar 1881.] NATURAL SCIENCES OF PHILADELPHIA. 199 hook; but in all cases they lie so close together side by side, that it appears as ifthe pinnules were laterally attached. In Actino- erinus and Strotocrinus each pinnule is furnished near the middle of its outer surface with a tooth-like spine which curves abruptly upwards ; these spines are short and obtuse near the arms, but gradually increase in length toward the tips of the pinnules. As a rule the pinnules are deeply grooved on their inner surface, and in perfect specimens the grooves are covered by a double series of very minute pieces, though owing to defective preservation this covering is rarely observed. In young specimens, while the arms are composed of a single series of joints, the pinnules are not in contact, and are only given off from the alternate joints, but when the alternate arm joints meet by lateral growth, and the pinnules attain their full size they become gradually connected. D’Orbigny’s genus Hdwardsocrinus was founded upon a young Platycrinus, whose arms and pinnules were in their transition state. 7. INTERNAL CAVITY. The construction of the interior of the body of all Paleozoic Crinoids is best known in the Spheroidocrinide, among which, specimens preserving some parts of the delicate organs have occa- sionally been found. The inner surface of the vault is often deeply grooved toward the brachial zone, producing corresponding elevations outwardly on the test. There are generally five large grooves, each branching into two smaller ones, the former corresponding to the five rays, the latter to their main divisions. This kind of vault is found most frequently among the Rhodocrinide. Among the Actino- | crinidz external ridges are rarely observed, but in their place the vault within is strengthened with bars or braces radiating from near, but not joining at the centre. The braces widen toward the arm bases, where they fold over to form regular tubes, corresponding with the natural grooves in the vault just described, and they branch as those do. In genera in which the rays are extended into free appendages, and in which but five ambulacral canals pass out from the vault proper (Platycrinus, Steganocrinus, etc.), the grooves are deep and in some cases were evidently closed and formed into tunnels, leaving, however, in either case beneath the median portion of the dome and in front of the anus, a space which is occupied by narrow grooves, meeting 200 PROCEEDINGS OF THE ACADEMY OF [1881. in the centre, and following the median line of the radial depres- sions and galleries below the vault to the arm openings. In some specimens of Actinocrinide, almost the entire test is lined with a delicate calcareous plexus or network. This lining is not in contact with the test directly, but connected with it by small partitions, producing innumerable little chambers, which communicate with each other and with the visceral cavity. There is, at least, one such partition or support from each plate of the test, generally arising from the median portion of the plate. (Pl. XIX, fig. 16). The plexus is very delicate in some specimens, in others—mostly in large specimens—-rather dense and rigid, but in all of them perforated with conspicuous pores or passages, whose arrangement corresponds with the direction of the sutures between the plates of the test. There is one pore at least to each angle of the plate, but sometimes additional ones in large individuals. The structure extends but little below the regions of the second radials, leaves passages at the arm openings, and toward the vault reaches to a place near the median portion of the ray, leaving at the centre an open space in the test which is occupied by the central vault piece. From this space five wide avenues, corresponding with the grooves on the inner surface of the vault, pass out toward the arm bases. The avenues produce five subtriangular interradial or interpalmar fields, which are raised conspicuously above the floor of the vault. They are of rather dense texture at the borders, the inner side somewhat thickened, while the surface of the median portions is rough and uneven, perforated with large and small passages which communicate with the avenues. Four of the interpalmar fields are equal, the posterior one fre- quently larger and penetrated by the anal aperture. In species with a lateral opening all five fields have about the same form, and the central space between them is of pentangular outline. The ease is different in species with a subcentral anus, when frequently the posterior field is larger and encroaches deeply upon the middle space, giving to it a lunate instead of asubcircular or pentangular outline. Inspecies of this kind, the anus is placed near the edge of the interpalmar field, but separated from the central space by a partition which forms the border of the field. In species with secondary radials, the avenues divide, sending a branch to 1881. | NATURAL SCIENCES OF PHILADELPHIA. 201 each arm, and forming thereby, between the interpalmar fields, a smaller intrapalmar! one similarly constructed. A calcareous lining, such as described, has been observed by us with slight variations in Batocrinus, Dorycrinus, Teleiocrinus, (Meek and Worthen’s Strotocrinus B), Agaricocrinus and Hretmo- crinus, and probably existed in many other genera. In the Actinocrinites, or typical Actinocrinide, the inner framework was either less developed, or was of a more perishable nature. In the genus Actinocrinus it is indicated on the inner floor of the test by little roughened places, which we take to be traces of the pillars which supported it. In Physetocrinus even these have not been observed, but it is a characteristic feature of that genus, that the plates of the calyx have along the sutures at each angle very distinct indentations resembling pores, which give to the outer surface of the test almost the appearance of the inner plexus in species of Batocrinus or Dorycrinus. Whether there was any communication with the visceral cavity through these indenta- tions cannot be ascertained from the fossil. The test at these points is exceedingly thin and transparent, but we have never detected an actual passage. It must also be noted that the vault in that genus has similar indentations, but these, contrary to the others, open out from the inner floor of the test, being arranged along the radial grooves, not interradially as those along the calyx (Pl. XIX, figs. 5 and 16). In Strotocrinus, an internal framework has been observed in connection with the calyx, but none with the vault, and its typical form had apparently, like Physetocrinus, slight indentations along the inner floor of the vault. The general internal structure indicates a concentration of organs toward a point beneath the centre of the vault, in front of the anus, but not tothe anusitself. The latter is situated distinctly outside the radiation, 7.¢., interradially. The grooves which we have noticed in the vault were figured by De Koninck and Lehon in their Recherches Crin. Carb. Belg., but they seem to have regarded them as muscular impressions. Billings, in the third and fourth Decades of the Canada Geological Report, was the first to treat ! The term ‘‘zzterpalmar Felder’’ was used by Joh. Miiller for the ‘‘in- terradialen Felder zwischen den Tentakelrinnen im Perisom des Penta- crinus, ‘ Jntrapalmar Felder’ for the interdistichal Felder’’ (Monatsber. Berl. Acad. 1841, p. 218). 202 PROCEEDINGS OF THE ACADEMY OF [1881. of them in connection with the ambulacral system. He showed how impossible it was that the ambulacral canals in some Paleo- zoic Crinoids could be continued along the outer surface of the vault, and he reached the conclusion that they passed into the body at the arm bases. It is remarkable that Billings, after mak- ing this important discovery, in connection with which analogy suggested that the food entered the body in the same manner, clung as late as 1870 to the old theory that the subcentral passage in these crinoids—interradiately situated as we have shown— served both as mouth and vent. This view was advocated by him in a series of interesting articles, published 1869-70 in the Am. Jour. Science and Arts, Nos. 142, 145, 149, as “ Notes on the Structure of Crinoidea, Cystidea and Blastoidea.” Since that time, it has been most generally conceded that the interradial opening was the anus only, and that the oral centre or mouth in the earlier crinoids was hidden from view by external structures.! Billings’ views with regard to the ambulacral passages were con- firmed by Wachsmuth’s discovery of radiating tubes beneath the vault,2 which, as he ascertained, connect with the ambulacral furrows in the arms. We have since examined these tubes in several other specimens, both of Actinocrinus and Strotocrinus, and are enabled to give additional information regarding them. The radiating tubes are attached to the vault, running parallel to its inner surface. They consist of five main trunks, which follow the direction of the five main avenues which separate the interpalmar fields. They bifurcate in the same way and until a branch connects with every arm: They are composed of four rows of plates, two below and two above. The two latter touch with their edges the inner surface of the vault, are alternately arranged, and grooved along their median line, leaving a tun- neled passage between the walls of the tube and the vault. The trunks of the two lateral sets of tubes on the same side not un- 1 The following writers have expressed this opinion: Schultze, 1866, Monog. Echin. Eifl. Kalk, p. 7; Meek and Worthen, 1869, Proc. Acad. Nat. Sci., Phila., p. 323; Lovén, on Hypomene Sarsi, reprinted Ann. and Mag. Nat. Hist., Sept., 1869 ; Wachsmuth, “*On the Internal and External Structure of Paleozoic Crinoids,’’ Am. Jour. Sci. and Arts, Aug. 1877, p. 115; Zittel, 1879, Handb. d. Paleontologie. 2 Described by Meek and Worthen, Geol. Rep. TIl., vol. v, p. 329, and Wachsmuth, Am. Jour. Sci., Aug., 1877, p. 119. 1881. ] NATURAL SCIENCES OF PHILADELPHIA. 203 frequently meet before reaching the peristome, but the anterior ray is always distinctly separated. Whenever tubular canals have been observed, they are pre- served only to the border of the central space, but none of them have been found to join in the centre. In a specimen of Actinocr. glans, however, the tubes before terminating bend down- ward toward the visceral cavity, give forth lateral processes, as if disposed to branch, and expose two openings at the extremities directed to opposite sides. The openings indicate that the tubes may have been connected with each other by lateral passages, and formed a ring around the centre. This is evidently the structure of Actinocr. Verneuilianus, in which a circular vessel is observed beneath the centre at a short distance from the vault; no ambu- lacral tubes are attached to it, but there are small radial openings + with which they might have been connected. The lower portion of the ring is composed of minute interlocking pieces, with five additional openings interradially situated. This ring is com- paratively large, enclosing within its circumference the contracted upper part of the convoluted digestive organ, which is well preserved in the specimen from which these facts were obtained. A tubular skeleton, as above described, has thus far been observed only in the Actinocrinide, but a tubular passage beneath the vault, in connection with the arm grooves and oral centre, has been traced in most groups of the Paleocrinoids, and no doubt existed also in the Blastoids. In Cyathocrinus, and probably in the Cyathocrinide generally, the tube is constructed of two series of pieces overlying the oral plates, and these again are covered by two similar series of plates,which forma part of the vault. In Granatocrinus the tubes follow the pseudambulacra, being covered by three series of small plates which must be considered extensions of the vault(Pl. XIX, fig. 3). It is now generally conceded that the tubular canals beneath the vault contain the same organs which in modern crinoids are exposed on the ventral disk, and like them embrace the food passages, and certain other vessels in connection with the ambu- lacral system. In this view of the case, it is reasonable to further consider that the annular vessel, above described, served as a water-vascular ring. The relations between the vault and the ventral covering of recent Crinoids are not so close as has been sometimes supposed, 204 PROCEEDINGS OF THE ACADEMY OF [1881]. they are indeed different things, although there are certain analo- gies between some of their parts. Among these are the oral plates, which are represented in some of the later Crinoids, but absent again in the fully grown Pentacrinus and in the Comatulide. They are also absent in the Sphzeroidocrinide, unless we consider the interradial vault pieces to be their representatives. We have already suggested that allintemadial plates in the dome—exclu- sive of the proximal pieces—may perhaps have been modified oral plates which, either by division or interpolation, gradually increased in number. The dome of Coccocrinus has a single oral plate to each interradial field, while the corresponding spaces in most species of Platycrinus are occupied by three and occa- sionally five pieces each. It is as easy to consider the single plate of the former to be represented by three in Platycrinus, as that the three are sometimes replaced by five within the limits of the same genus, the plates occupy the same position in both cases, but in some groups the true orals meet laterally whichis not the case with the interradial dome pieces of Platycrinus or Aclt- nocrinus, nor with the undivided plates of Coccocrinus. In Cyathocrinus where the orals are very conspicuous, they join beneath the radial groove, and form the floor upon which the am- bulacral tube rests. The bottom of the tube is composed of two series of pieces, which are covered directly by vault pieces in two alternate rows, whose lateral margins rest upon the upper edges of the two orals; while in Platycrinus the corresponding yault pieces abut laterally against the sides of the interradial—oral— plates in an unbroken succession. In Platycrinus the interradial plates thus take exactly the same position as the exposed part of the oral plates in Cyathocrinus, while the covered parts are unrepresented. In Coccocrinus, a covering of the ambulacral groove has not yet been observed, but judging from the fissure between the oral plates, it probably rested just upon their edges. and formed an intermediate link between the vault structure of the Cyathocrinid and Platycrinide. In the Actinocrinidze and Rhodocrinide, the alternate dome plates are not so readily distinguished, as in the Platycrinide and forms with free rays, in which they are well marked in the extended parts. In the recent Crinoids the alternate plates are represented by the “ Saumplattchen,” which, however, instead of — 1881.] NATURAL SCIENCES OF PHILADELPHIA, 205 forming a part of a solid vault, are movable, and line the lateral margins of the tentacle furrows. The proximal and central dome plates are altogether unrepre- sented in recent Crinoids. This is best perceived by comparing Coccocrinus as usually preserved, with Hyocrinus or other recent genera in which the oral plates are developed. In both cases, there is at the oral centre an opening at which the grooves con- verge, surrounded by the oral plates ; but, while in Hyocrinus and all recent genera this opening is unobstructed by solid parts, in Coccocrinus, Cyathocrinus, and the Palzocrinoids generally, it is covered by the apical dome plates. The central piece generally occupies the median portion of the vault, and always indicates the centre of the oral system. We have already noted narrow grooves upon the inner surface of the vault, which meet on the central piece, and follow the median line of the radial depressions and galleries to the arm openings. Only three main grooves meet at the centre, those of the two lateral rays are uniting before reaching that point (Pl. XVIII, fig. 1). The grooves are best observed in natural casts of the interior, in which they appear on the surface in the form of narrow bands or ridges (Pl. XIX, figs. 5 and 9). The position of the grooves indicates that they may have contained axial cords in connection with a nervous system located beneath the central plate. The location of the nervous system within the regions of the ambulacral centre is in analogy to the structure of other Echi- noderms, except the Comatulidz, in which, according to P. Herbert Carpenter, the principal nervous systems are located at the apical side, and in connection with the quinquelocular organ which occupies the cavity of the centrodorsal plate.' The interpalmar fields are composed of a soft skin, but although this is more or less incrusted with limestone particles, which sometimes almost look like vault pieces, they have no affinities with the plates of the vault. The plates of all recent Crinoids are perforated with numerous pores for the introduction of water into the body, a function which could not well be performed by the interradial pieces, but much less by the solid undivided oral plates of Cyathocrinus and Coccocrinus. In the Cyathocrinide, these functions may have been performed by the ventral sac ' On some points in the anatomy of Pentacrinus and Rhizocrinus. Jour. Anat. and Phys., vol. xii p. 35. 206 PROCEEDINGS OF THE ACADEMY OF [1881. which is profusely punctured, but evidently not by the simpler ventral tube of the Sphzeroidocrinide which is destitute of such openings. Nor can we imagine that there was any such communi- cation through the dome proper, its plates are perfectly con- nected at their sutures, and the interradial series especially are strengthened by strong braces within. There are evidently closer relations between Cyathocrinus and Hyocrinus or genera with oral plates, than between the Penta- crinide and Spheroidocrinide, in which those plates are either unrepresented or greatly modified. The latter two types form the extremes, and are probably more distant in their relations with each other, than most Blastoids and Cystideans from the Paleocrinoids. The affinities of the Paleocrinoids with the Blastoids, become more apparent by our recent discovery of hydrospires in a speci- men of Zeleiocrinus. Their exact construction has not yet been fully ascertained, but that such organs existed in some of the Actinocrinide is now demonstrated beyond a doubt. The speci- men is fragmentary, it was obtained from a narrow cherty band of the Upper Burlington Limestone, and is itself silicious. The interior is solid, with the exception of a natural concavity beneath the vault, at which point it was broken in quarrying, exposing a part of the upper face cf the tubular skeleton. Portions of two tubes only are visible, and these are broken transversely after their second branching, the fracture giving a cross-section of the tubes and surrounding parts. In Yeletocrinus as in Strotocrinus proper, the lateral rim contains radiating tunnels formed by par- titions between the divisions of the rays. The tunnels, as observed by us in several specimens, are divided transversely into two com- partments, of which the upper one is occupied by the ambulacral tubes (Pl. XIX, figs. 7band 8). In the specimen under considera- tion the lower or dorsal compartment has a semicircular outline, and within this, below one of the branches of the ambulacral tubes, there are visible two distinct folds, closely resembling the folds in the hydrospires of Granatocrinus (P|. XIX, fig. 3). Beneath the adjoining branch, the folds cannot be so well distinguished, but the outlines of the hydrospires are also there faintly indicated. Considering that the arms in the Blastoids are inverted and recumbent, and that their calcareous portions represent not only he solid parts of the arms, but also a part of the test, it will be 1881. ] NATURAL SCIENCES OF PHILADELPHIA. 207 seen that the hydrospires above noted, and those of the Blastoids, have not only a similar form, but also a very similar position." The hydrospires of the Crinoids, like those of the Blastoids, are placed in close proximity to the arms with which they were probably in communication, close to the test and within the general cavity of the body. The above is, to our knowledge, the only case in which hydro- spires have been observed among the Spheroidocrinidx, but they were probably present in other genera, and perhaps in the Palo- crinoids generally; while these organs are unknown in all later and recent Crinoids, and in other groups of Echinoderms. It is a fact worthy of note that all Cystideans and Blastoids, and so far as known, all Paleocrinoids which possess hydro- ' The vault in the Blastoids, as we understand it, consists not merely of the plates which cover the oral opening, but extends all along the median portions of the pseudambulacra (Pl. XIX, fig. 3), forming underneath a good-sized tunnel, which we take to be the homologue of the ambulacral tube of the Crinoids. If this interpretation is correct the structure bears the closest similarity to that found in those Crinoids in which the vault is extended into free rays, and in which these extensions combine to some degree the characters of the arms and body. The recumbent arms of the Blastoids are, according to this, lateral extensions of the body which take the place of true arms; but while in the Crinoids the radial exten- sions give off regular arms, in which the ambulacral tubes are converted into grooves, the corresponding parts in Blastoids remain attached to the body, and the pinaules form the only free appendages. It is possible, how- ever, that in the Blastoids the lateral furrows which traverse the ambulacral fields were not covered by plates, and that these correspond to the open arm grooves—respectively arms—in Crinoids. Dr. Hambach (Contributions on the Anatomy of the genus Pentremites, p. 7) is probably correct in supposing that the pinnules of the Blastoids were not connected with tke pores, as hitherto believed. We think it probable that they rested in the funnel-shaped pits which alternate with the pores, and which communicate with the lateral grooves of the pseud- ambulacra, while the pores probably communicated with the hydrospires. This view coincides with what we have heretofore suggested, that the upper face of the pseudambulacra corresponds to the grooves within the arms of the Crinoids, and indicates that there are close affinities between the ambulacral field itself and the solid portion of the arms, The passage directly beneath the field is probably the dorsal or axial canal, which by the inverted position of the arms became the inner instead of the outer passage. The hydrospires in the Blastoids are placed beneath the canal, and extend along the perivisceral cavity of the body, like in the case of Teleiocrinus. 208 PROCEEDINGS OF THE ACADEMY OF [1881. spires had a subtegminal mouth, and a solid test built up of plates so closely fitted together that expansion or contraction was impossible. Expansion in some parts, however, was neces- sary to produce circulatory currents for the introduction of food. In most Echinoderms, including all recent Crinoids, this is accom- plished by means of the pliant test and soft appendages which surrounded it. The Echini alone, like the Paleozoic Crinoids, have a rigid test, but they possess an external mouth, and in addition to their numerous soft appendages a movable actinal membrane, capable of considerable expansion, even in some cases beyond the line of the actinostome. It seems to us not unlikely that the hydrospires served the purpose of gills, producing by their con- tractions and dilatations the requisite circulation to introduce food and expel the refuse matter. This would account for their absence in the recent Crinoids and other Echinoderms, and would suggest that they were probably connected with numerous soft appendages along the arms, arranged perhaps in like manner as the pores along the ambulacra of the Blastoids, but not as in the Cystideans, in which the pores which connect with the hydrospires are distributed over different parts of the body. A better knowl- edge of these organs, as they exist among the three great divisions of the Paleozoic Crinoids, would doubtless afford far more satis- factory characters for separation than we now possess. In the abdominal cavity of the Paleocrinoidea, the only structure which has been observed consists of a peculiar skeleton located beneath the tubular canals, which from its position, in analogy to other Echinoderms, has been referred to the digestive apparatus.! In its usual preservation, it is a large convoluted body resem- bling the shell of a Bulla, open at both ends. The upper end is placed beneath the centre of the vault, and the lower directed toward the base. It is dilated above; contracted below; its surface about parallel with the walls of the visceral cavity. In some species it is subcylindrical, with the vertical axis the longer ; in others globular or even depressed globose ; but it is always truncate below, and never extends to the inner floor of the basal plates, The walls are coiled without touching at any point, and ‘Meek and Worthen, Geol. Rep. IIL, v., p. 328, call it a convoluted sup- port of the digestive sac. Wachsmuth, Am, Jour. Sci. Aug., 1878, p. 125, terms it the ‘‘ alimentary canal.” 1881. ] NATURAL SCIENCES OF PHILADELPHIA. 209 the convolutions are directed outward from left to right, varying in number from two to four in different species. In the usual state of preservation the walls are perfectly solid, almost like a “ convoluted plate” as which it was described by Hall. In transverse sections, they are seen to be strong, and appear to be constructed of two partitions closely fitted together and united at the edges. The unusual thickness and apparent double nature of the walls in these specimens misled Wachsmuth in 1877, who considered the walls to be the body of the alimentary canal. This is evidently a mistake. We now know, from a num- ber of other specimens, that the wall was simple in all cases, very delicate, and constructed of an extremely fine filigree work, which generally in the fossil became thickly incrusted with silicious matter on both sides, thus producing the apparent duplication of the wall. In good specimens, a magnifier shows the wall to he composed of an extremely fine network of minute pieces or bars, with intervening meshes. These bars, according to Meek and Worthen, “ do not intersect each other at any uniform angle, but anastomose so as to impart a kind of irregular regularity to the form and size of the meshes.” That this network was in some way connected with the diges- tive organs, is no doubt true, but whether it formed a mere sup- port for the digestive sac,as Meek and Worthen suggested, or was an extensive plexus of blood vessels surrounding the ambu- lacral canal, is a question we areas yet unable to solve. It should, however, receive a more appropriate name than any yet given, and we propose to call it the “ csophageal network,’ which may be changed when its special functions and affinities are discovered. One of the writers found a specimen of Actinocrinus, in which the conyolutions were nearly intact, and by removing the outer fold, the inner or upper end, as distinguished from the outer or terminal part, could be examined (Pl. XIX, fig. 12). The organ has the usualdense structure, and where it comes into view, is an elongate tube, which passing downward widens at first gradually to near the middle of the visceral cavity, then rapidly until it attains a width equal to two-thirds the entire length of the cavity. The upper part descending spirally turns from right to left, but on becoming wider the whorls are abruptly reversed, and there- after the convolutions are from left to right. The outer end also tapers rapidly, assuming the form of a flattened tube, and ascends 15 210 PROCEEDINGS OF TILE ACADEMY OF [1881. spirally on the outside, while the other end is directed toward the oral centre, but a connection with the food grooves has not yet been observed. In all cases where the cesophageal network has been examined, among the Actinocrinide and Platycrinide, it varies only in out- line and in the number of whorls; while in Ollacrinus the entire skeleton resembles a large spiral. In the only specimen observed, it apparently consisted of a large round canal, which turned spirally on its axis, and which near the basal plates turned upward, but the organ is in a fragmentary condition, and it is quite possible that this part was surrounded by other conyolu- tions. Until recently this organ had been observed only in crinoids from the Burlington group. Angelin, however, in the Icononographia Crinoideorum Suecie, figures several examples from the Upper Silurian of Sweden (PI. 26, figs. 12,12 a,b, ). It is well preserved in these specimens, and resembles that of later Actinocrinide. It differs, however, in being closed at the outer side, while the inner parts, as in Burlington specimens, are distinctly coiled. The outer wall is pentangular in outline, open toward the basal disk, and consists of a very delicate porous texture, appearing like an envelope for the inner or coiled parts, and as such possibly represents the perivisceral plexus, which in some cases almost equals the cesophageal network in delicacy of structure. 8. CoLUMN. The column in the Spheroidocrinide is generally circular transversely, though sometimes elliptical or pentagonal and even quadrangular. It is elliptical only in Platyerinus, and the pentagonal form occurs only in Refeocrinus and some few species of the Glyptocrinites. The central canal varies from large to extremely small, and is round or pentagonal. In Platycrinus it is so minute that in columns of an inch or more in thickness on their long diameters, the opening will scarcely admit the point of a needle. In the Rhodocrinide it is irregularly pentagonal, and as arule small. Among the Actinocrinide also, the passage is generally not above medium size, but in Megistocrinus it*is remarkably large throughout the column and all its branches. By the earlier writers, new species were often based upon frag- mentary columns, a practice which has fortunately been aban- 1881. ] NATURAL SCIENCES OF PHILADELPHIA. 211 doned, since it became known that the different parts of the column in the same specimens are often widely dissimilar. In the growing animal,new columnar joints were continually introduced by interpolation between the older segments, and these younger joints, which are found throughout the column in all stages of development, produce striking changes in the general aspect. The column matured from the root upward, and the upper part remained throughout the life of the crinoid in a kind of immature state, wherein the intercalated joints did not attain the width of the others. The uppermost joints, however, although they were probably among some of the earliest developed parts of the col- umn, are not separated by smaller joints. Gradually, sometimes between the third and fourth joints, new plates make their appear- ance ; the first one so thin as to be scarcely visible, the next which lies between the succeeding joints much larger, the third probably reaching full size. Secondary intercalations follow between the new pieces, the intervening spaces between the larger joints increasing gradually to a maximum, from which point down the column seems to be mature, for all succeeding spaces have a like number of intercalated joints. As a general rule, the column decreases somewhat in thickness from the calyx for a certain dis- . tance down, after which it increases again towards the root. In some species the primary joints are only longer, but not of greater - diameter than the others. This is the case in Platycrinus where the new joints seem to have been formed directly beneath the calyx, their number increasing in length gradually along the stem, and not in sections as in the Actinocrinidz and the Rhodocrinide. Lateral cirrhi along the column have been rarely observed, and in this family probably existed only toward the root. The form of the root is exceedingly variable, and depended evidently upon the conditions of its place of attachment. When living in a soft or sandy soil, it seems to have been provided with a great number of small rootlets which are given off both vertically and horizon- tally ; but when it was attached to a rock or other hard substance, the lower surface grew entirely flat and was often deeply grooved. The grooves pass out from the root, and apparently took the place of the vertical rootlets. The central passage extends to the smallest rootlets and is often of considerable size. We have already noted this fact in Part I,and suggested that probably the 212, PROCEEDINGS OF THE ACADEMY OF [1881. rootlets may have had respiratory functions by introducing water into the body. In adult specimens of Actinocrinide and Rhodocrinide the column was long. We have never seen its full length, but have in several instances traced it three to four feet without seeing either root or body, and we suppose that it was in some genera very much longer. Of Platyerinus, however, we have examined five complete speci- mens measuring from the tips of the arms to the extreme ends of the fine rootlets from 7 to 27 inches—the latter in a large species. In all these specimens, the column gives off for some distance large lateral branches, which decrease in size toward the end of the root, each one with irregular branches which divide again and terminate in hair-like tubes. We never saw a Platycrinus in which the root was flattened, as in some of the Actinocrinide, and it seems possible that the crinoids of this genus only grew on a soft bottom, or possibly floated about with their column like an anchor. The same was evidently the case in the genus Glyp- tocrinus, in which the column was short, tapering to almost a needle’s point, without lateral branches, 9. MopE oF GROWTH AND PALZONTOLOGICAL DEVELOPMENT. In the Pentacrinoid larva of Antedon, the calyx is composed chiefly of very distinct, rather large basals, alternating with which are five dots, which represent minute radials. The crinoid at this stage consists only of five columnar joints, the large basals, the rudimentary radials, and of five large oral plates which cover the entire peristome. The succeeding radials, at first unrepresented, develop afterwards, and the arms make their appearance at a much later period. Of the Palzocrinoids, the first stages are, of course, unknown, all the specimens we have discovered—even the very youngest— being already provided with arms, and hence were considerably advanced in the scale of growth. It can be ascertained, however, by a comparison of larger and smaller specimens, that their mode of growth must have been similar to that of An/edon. In the smaller, and as we consider them, younger specimens, the basals, compared with the other plates, are much larger, being almost the same size as in mature individuals. Next in size are the first radials, which are larger than-the second and third. In the inter- 1881.] NATURAL SCIENCES OF PHILADELPHIA. 213 radial series the first plate is much the largest, and the number of interradial and anal plates is considerably less in young specimens than in the adult, thus indicating that the calyx in these crinoids was developed from the basals up, as in their living representa- tives. With the development of the first interradial, apparently simul- taneous with the second and third radials, the Platycrinoid form, the simplest of the Spheroidocrinidz was complete. The earliest types of the Platycrinide known to us, but evidently not the earliest representatives of that group, are from the Upper Silurian. In Coccocrinus the body is composed of three basals, two by five radials—the first very much the larger—a single interradial, and five large oral plates, exactly as we must expect from analogy to find the Actinocrinoid in its earlier phases. The rays in the Platycrinoid are free from the primary radials up, but the first joints of the two main divisions are simple and constructed similar to the radials in the body of the Actinocrinide. To transform the Platycrinoid into an Actinocrinoid, it only requires the inter- polation of one or more interradial pieces between the proximal plates of the first division of the ray. By this simple process, the plates which were before free in the Platycrinoid, were incorpo- rated into the body, and raised to the dignity of secondary radiais. Many of the earlier Rhodocrinide and Actinocrinidz are char- acterized by highly elevated ridges, which extend all along the radial series of the body. They run vertically along the middle of the primary radials, divide upon the third plate, and branch to the secondary and tertiary radials, whence they pass very gradu- ally into the arms. The ridges are very prominent, rounded exteriorly, and as they approach the arm bases, assume nearly the shape and size of the arms. The plates upon which they are extended, in their upper series, scarcely differ in length from the first free arm plates, and all gradually diminish upward. The longitudinal ridges are evidently not accidental, nor a mere orna- mentation, but represent the arm joints as they were when first developed in the young animal. In this early stage they were round joints, the lateral wing-like extensions being developed afterward, when by reason of the upward growth of interradial and interaxillary pieces, the plates became parts of the body. We find on the surface of many internal casts of forms belonging to this group similar but narrow ridges, which follow the same direction 214 PROCEEDINGS OF THE ACADEMY OF [1881. as those upon the test. The ridges in these cases are evidently the impressions of rudimentary grooves, indicating that the plates at that stage were provided with ambulacral furrows like regular arm plates. They also had pinnules attached, which like the arm plates were by the growth of the animal absorbed into the calyx. The fixed pinnules, which Wetherby describes in Glyptocrinus Richardsoni, Cincinnati Soc. Nat. Hist., 1880, and which we found also in Gl. decadactylus and Reteocrinus O’Nealli, confirms our views as to the mode of growth of the Paleocrinoid, and throws light upon the paleontological development of Crinoids generally. The number of secondary radials varies considerably with age. In Reteocrinus O’ Nealli we have observed as many as seven, but the number may be even more in some cases; and on the other hand we found in some younger specimens only three or four. The first and second of these plates, and the third and fourth seem to have been united by syzygies, at least the first and third bear no pinnules, while beyond the fourth pinnules are given off regularly from alternate sides as in the free arms. The proximal pinnule is given off toward the outer side of the ray, or, which is the same thing, toward the interradial area, the next one toward the interaxillary space. The plate which gives rise to the first pinnule has almost the form of a bifurcating plate, but instead of supporting radials of a higher order, it bears on its inner sloping side a third secondary radial, and on the outer a stout pinnule. The first fixed pinnule is highly elevated above all other plates of the interradial space, is rounded like the radials, and almost as conspicuous. In one of our specimens it consists of five plates, three of which are soldered into the body, and the fourth appar- ently free. The fixed plates are nearly as strong as the radials and may be easily taken for them, having like them winged exten- sions by which they are laterally connected with the interradial plates. The upper joint is much smaller and constructed like the joints of the free pinnules. The second pinnule has only two joints in the body, the third but one, which in either case are larger than any of the free joints, but which already attain some- what more the aspect of regular pinnules, and are given off ina similar manner. In a specimen of Glyptocrinus Richardsont, kindly loaned to us by Prof. Wetherby, the first fixed pinnule is given off from the second plate above the first bifurcation, and consists of seven plates within the body, the third and fourth 1881.] NATURAL SCIENCES OF PHILADELPHIA. 215 plates were joined by syzygies. The second pinnule, with but four joints within the body, springs from the opposite side of the fourth plate. The fifth plate, instead of a pinnule, supports a regular arm, and the sixth again a pinnule, but from the same side as the preceding pinnule. This is important as it suggests the idea that the pinnule may have here developed into an arm. That a transformation of this kind took place in some groups, is more than probable, as will be shown presently. In Glyptocrinus decadactylus only the second and third secondary radials are joined by syzygies, all succeeding pinnules being given off regularly. The number of arms has been considered of specific importance among the Palzocrinidze, and even genera have been based upon this character. The greatest variation in the arm formula is found among species of the typical Actinocrinide, in which we include, besides the genus Actinocrinus, also Strotocrinus, Teleiocrinus, Physetocrinus and Steganocrinus. These genera agree in the style of their ornamentation, and in the construction of the anal area, which differs somewhat from that of all other Actinocrinide. It is to be noted that in this group the specimens have, as a rule, the same number of arms in the different rays, the few exceptions being due to deficient or abnormal development of these parts. The genus Actinocrinus has been very appropriately separated by Meek and Worthen into two sections, The one, with Actino- erinus proboscidialis Hall, as a type, has the arm-bases arranged in a continuous series all round; the other, with Actinocr. multi- radiatus as type, has the rays formed into more or less protuberant lobes. A. proboscidialis of the first section, which represents the simplest form of this interesting group, has only four arms to the ray. The first departure is A. reticulatus with four arms in all but the two posterior rays, which have five; the fifth arm where it exists, being placed below the line of the others and appearing somewhat crowded. Next in order are A. limabrachiatus and A. clarus with five arms, in which one of the divisions in each ray divides again. A. sexarmatus and A. opusculus have six arms or three to each division of the ray. there are twelve which have sixteen simple arms: 23, two others have at one side of the right posterior ray @ pair of arms instead of a single arm—the formula may be graphically represented thus: 141,112 1111.1111—, and in one specimen we find one of the anterior arms represented by a pair, while all the others are simple. The last three specimens no doubt are abnormal cases, but they are inter- esting as showing a tendency of the species to an increase of arms in the postero-lateral and anterior rays. All the above specimens can be safely referred to a new species which we call Hretmocrinus originarius. There are thirteen other specimens, for which we propose the name Hretmocrinus intermedius, which agree with the former in all essential points, having the same peculiar ornamentation, the same form and size, and being derived from the same layers ; but they differ in having, as a rule, in the anterior ray two, and in both postero-lateral rays three additional arms, while the antero- lateral rays are unchanged. The additional arms are given off alternately from the two main divisions of the rays as in Actino- crinus. They are simple and in most cases included within the body walls, except in the two posterior rays, in which the arms arising from the last bifurcation are arranged in pairs. 224 PROCEEDINGS OF THE ACADEMY OF [1881. In this lot of crinoids there are two more specimens, one having wh 2 DAD twenty-one arms distributed thus: 1111 1122, and the other twenty- 221,52 : i 2 22 Both agree with the two preceding _ bo - iS seven, thus: 2 at species except in the arm formula, but even this is fundamentally identical with 2. originarius, the simple arms of the latter being in part replaced by pairs. None of our Bono specimens have the double arm structure throughout all the rays, but we obtained from Edwardsville and Canton several specimens in which that feature prevails, and for which we propose the name Hretmocrinus adultus. The two irregular Bono specimens may not be the intermediate forms between that species and LZ. originarius, but this is probable, and they show how the double arm structure became introduced. Searcely less interesting are some specimens of Batocrinus. Among sixteen examples of a form which we call B. Whitei—eleven from Bono, two from Canton, and three from Edwardsville—there are fourteen with the arm formula: 24, while two of them have four arms to the posterior ray. No double arm structure has been observed in this species at Bono, but its apparent representative at Crawfordsville has always two arms from each opening. JB. Indianensis has the same form and ornamentation as B. Whitei, arms—and it evidently bears the same relation to that species as B. Chrystito B. Lovei,and E£. adultus to E. originarius. It would be interesting to pursue this line of examination further, and trace the relations subsisting among other groups similarly connected. But we have perhaps gone far enough to serve our present purpose. It is to be observed, however, that the import- ance of this kind of investigation, in its bearing upon systematic classification, can scarcely be overestimated. It has to do with the principles which lie at the very bottom, and it is only by the study of these relations, of the exact anatomical changes which produced individual variation, and in time permanent modification of forms, that we can hope to arrive at a correct understanding of the groups in nature, or be able to make scientific discrimination of families, genera, species and varieties. Many species have been made, upon mere differences of srowth; some upon unimportant variations in the arm formula; some upon abnormal development in certain parts of the body; others upon 1881. | NATURAL SCIENCES OF PHILADELPHIA. 225 slight modifications in ornamentation; while still others were founded upon material so imperfect that neither figure nor de- scription sufficiently defines the form. Our literature is so over- burdened with synonyms that we fear a very large percentage of so-called species ought to be eliminated. We have undertaken to point out such cases among the Burlington and Keokuk Crinoids, and although we have been obliged to throw out a considerable number of species, we have only done so where necessity seemed to require it, and we suspect we should not have gone amiss by reducing the list still more. Schultze undertook the same task for the Crinoids of the Eifel, and, although we cannot agree with him in retiring certain genera, we concur in his determination of Ssynonymic species. There are no doubt, also many synonyms among the Subcarboniferous Crinoids of Belgium and England. One of us had an opportunity, several years ago, of studying the original collection of De Koninck in the Museum of Cambridge, . and became convinced that the eleven Belgian species of Actino-. erinus, described in the Recherches Crin. Carb. Belg., might be safely reduced to four or five. 10. THE SO-CALLED ‘‘ RESPIRATORY PORES.” In the first part of this work, on page 11, we called attention to certain pores, located in the body at the arm regions, on either side of the ambulacral openings, and we endeavored to show that they correspond in position with the so-called ovarian openings of the Blastoids. At that time we asserted that the pores were in some genera fixed at a definite number, independent of the number of arms in the species; that Batocrinus, for instance, had always twenty pores, whether the species had twenty arms or more, and that one-half the pores were located radially and the rest interradially. In this we were evidently in error; the pores probably always agree in number with the arms, and are really neither radial nor interradial, but are placed at the base of the arms. as \ sae Ys GDH PH, WA S405 eS = Y PROC, A. N.S. PHILA., 1881. PL. XV. PL. XV, PROC. A. N. S. PHILA., 1881. “PROG. ALN. S. PHILA., 1881. PL. XVI. SS . ES SESS een iy vy TN, ii i MAM Ait 4 4, rah 1A Nate KN Ui PROC. A. N. S. PHILA., 1881. PL.XVII. 7 PROC, A. N. S, PHILA, 1881. PL. XVII, 7 io O 9 CR Or to ae Wy) a ~ 6 C SKA, Os rl! AQ -\ ei SOM \v M Ad | ra PL. XIX PROC. A. N. S. PHILA., 1881. AeA ye ee cha qi nhs Peokeen oem )_geeep fe 1 na | _ i “ae Te= i ae QH Academy of natural sciences 1 of Philadelphia A2 Proceedings Biological & Medics] Serials PLEASE DO NOT REMOVE CARDS OR SLIPS FROM THIS POCKET eee ne UNIVERSITY OF TORONTO LIBRARY oe ee ee Pile IBosRN pews pi by a Hy it att or iit: haart ret OF reese 7 Jot i, nee Rhes “re He av sh or?) ieee Hyer Bayt! , ih ? its 43} Valeonah SS Asstitattitts te i} AS ha iit i (} rik ities ie e 3 rear ae 233